Quark x1000 Datasheet

Intel® Quark™ SoC X1000
Datasheet
August 2015
Document Number: 329676-005US

You may not use or facilitate the use of this document in connection with any infringement or other legal analysis concerning Intel products described
Legal Lines and Disclaimers
herein. You agree to grant Intel a non-exclusive, royalty-free license to any patent claim thereafter drafted which includes subject matter disclosed
herein.
No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.
All information provided here is subject to change without notice. Contact your Intel representative to obtain the latest Intel product specifications and
roadmaps.
The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications.
Current characterized errata are available on request.
Copies of documents which have an order number and are referenced in this document may be obtained by calling 1-800-548-4725 or by visiting:
http://www.intel.com/design/literature.htm
Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Learn more
at http://www.intel.com/ or from the OEM or retailer.
No computer system can be absolutely secure.
Intel, Quark, and the Intel logo are trademarks of Intel Corporation in the U.S. and/or other countries.
*Other names and brands may be claimed as the property of others.
Copyright © 2015, Intel Corporation. All rights reserved.

Datasheet August 2015
2 Document Number: 329676-005US
Contents—Intel® Quark™ SoC X1000
Contents
1.0 Introduction ............................................................................................................ 37

1.1 About This Manual ............................................................................................. 37
1.2 Component Overview......................................................................................... 37
1.2.1 SoC CPU Core Features ........................................................................... 38
1.2.2 System Memory Controller Features.......................................................... 39
1.2.3 Embedded SRAM Features ....................................................................... 39
1.2.4 Power Management Features ................................................................... 39
1.2.5 Security Features ................................................................................... 39
1.2.6 PCI Express* Features ............................................................................ 39
1.2.7 Ethernet Features................................................................................... 40
1.2.8 USB2 Host Controller Features ................................................................. 40
1.2.9 USB2 Device Controller Features .............................................................. 40
1.2.10 SD/SDIO/eMMC Controller Features .......................................................... 40
1.2.11 I2C* Master Controller ............................................................................ 40
1.2.12 GPIO Features ....................................................................................... 41
1.2.13 SPI Master Controller .............................................................................. 41
1.2.14 High Speed UART Controller with DMA ...................................................... 41
1.2.15 Legacy Bridge ........................................................................................ 41
1.2.16 Package ................................................................................................ 41
1.3 Component Identification ................................................................................... 41
2.0 Physical Interfaces .................................................................................................. 45
2.1 Pin States Through Reset ................................................................................... 47
2.2 System Memory Signals ..................................................................................... 47
2.3 PCI Express* 2.0 Signals .................................................................................... 48
2.4 Ethernet Interface Signals .................................................................................. 49
2.5 USB 2.0 Interface Signals................................................................................... 49
2.6 Integrated Clock Interface Signals ....................................................................... 50
2.7 SDIO/SD/MMC Signals ....................................................................................... 50
2.8 High Speed UART Interface Signals...................................................................... 51
2.9 I2C* Interface Signals........................................................................................ 51
2.10 Legacy Serial Peripheral Interface (SPI) Signals..................................................... 52
2.11 Serial Peripheral Interface (SPI) .......................................................................... 52
2.12 Real Time Clock (RTC) Interface Signals ............................................................... 53
2.13 Power Management Signals ................................................................................ 53
2.14 JTAG and Debug Interface Signals ....................................................................... 53
2.15 Legacy Interface Signals .................................................................................... 54
2.16 General Purpose I/O Interface Signals.................................................................. 54
2.17 Power And Ground Pins ...................................................................................... 55
2.18 Hardware Straps ............................................................................................... 56
3.0 Ballout and Package Information............................................................................. 59
3.1 Package Diagram .............................................................................................. 59
3.2 Ball Listings ...................................................................................................... 60
4.0 Electrical Characteristics ......................................................................................... 69
4.1 Absolute Maximum Ratings ................................................................................. 69
4.2 Recommended Power Supply Ranges ................................................................... 70
4.3 Maximum Supply Current ................................................................................... 71
4.4 Configurable IO Characteristics ........................................................................... 72
4.5 RTC DC Characteristics....................................................................................... 74
4.6 PCI Express* 2.0 DC/AC Characteristics ............................................................... 74
4.7 USB 2.0 DC/AC Characteristics............................................................................ 77

August 2015 Datasheet
Document Number: 329676-005US 3
Intel® Quark™ SoC X1000—Contents
4.8 General Interface Timing ....................................................................................79

4.8.1 Legacy SPI Interface Timing .....................................................................79
4.8.2 SPI0/1 Interface Timing...........................................................................80
4.8.3 SDIO Interface Timing.............................................................................81
4.9 Clock AC Timing ................................................................................................82
4.9.1 Reference Clock AC Characteristics............................................................82
5.0 Register Access Methods.........................................................................................85
5.1 Fixed I/O Register Access ...................................................................................85
5.2 Fixed Memory Mapped Register Access .................................................................85
5.3 I/O Referenced Register Access ...........................................................................85
5.4 Memory Referenced Register Access.....................................................................86
5.5 PCI Configuration Register Access ........................................................................86
5.5.1 PCI Configuration Access - CAM: I/O Indexed Scheme .................................86
5.5.2 PCI Configuration Access - ECAM: Memory Mapped Scheme .........................87
5.6 Message Bus Register Access ..............................................................................88
5.7 Register Field Access Types.................................................................................89
6.0 Mapping Address Spaces..........................................................................................91
6.1 Physical Address Space Mappings.........................................................................91
6.1.1 Bridge Memory Map ................................................................................91
6.1.1.1 MMIO ......................................................................................93
6.1.1.2 DOS DRAM ...............................................................................94
6.1.1.3 Additional Mappings...................................................................94
6.1.2 MMIO Map .............................................................................................95
6.2 I/O Address Space .............................................................................................95
6.2.1 Host Bridge I/O Map................................................................................96
6.2.2 I/O Fabric I/O Map..................................................................................96
6.2.2.1 Legacy Bridge Fixed I/O Address Ranges ......................................96
6.2.2.2 Variable I/O Address Ranges.......................................................96
6.3 PCI Configuration Space .....................................................................................97
6.4 Message Bus Space............................................................................................99
7.0 Clocking ................................................................................................................. 101
7.1 Clocking Features ............................................................................................ 101
7.2 Platform/System Clock Domains ........................................................................ 102
8.0 Power Management ............................................................................................... 105
8.1 Power Management Features............................................................................. 105
8.2 Signal Descriptions .......................................................................................... 105
8.3 ACPI Supported States ..................................................................................... 106
8.3.1 S-State Definition ................................................................................. 106
8.3.1.1 S0 - Full On ............................................................................ 106
8.3.1.2 S3 - Suspend to RAM (Standby) ................................................ 106
8.3.1.3 S4 - Suspend to Disk (Hibernate) .............................................. 106
8.3.1.4 S5 - Soft Off ........................................................................... 107
8.3.2 System States...................................................................................... 107
8.3.3 Processor Idle States............................................................................. 108
8.3.4 Integrated Memory Controller States ....................................................... 108
8.3.5 PCIe* States ........................................................................................ 108
8.3.6 Interface State Combinations ................................................................. 109
8.4 Processor Core Power Management .................................................................... 109
8.4.1 Low-Power Idle States........................................................................... 109
8.4.1.1 Clock Control and Low-Power States .......................................... 109
8.4.2 Processor Core C-States Description ........................................................ 109
8.4.2.1 Core C0 State ......................................................................... 109
8.4.2.2 Core C1 State ......................................................................... 110

8.4.2.3 Core C2 State......................................................................... 110

8.5 Memory Controller Power Management............................................................... 110
8.5.1 Disabling Unused System Memory Outputs .............................................. 110
8.5.2 DRAM Power Management and Initialization ............................................. 110
8.5.2.1 Initialization Role of CKE .......................................................... 110
8.5.2.2 Dynamic Self-Refresh .............................................................. 110
8.5.2.3 Dynamic Power Down Operation ............................................... 111
8.5.2.4 Functional Clock Gating ........................................................... 111
9.0 Power Up and Reset Sequence............................................................................... 113
9.1 Intel® Quark™ SoC X1000 System States ........................................................... 113
9.1.1 System Sleep States Control (S-States) .................................................. 113
9.2 Power Up and Down Sequences......................................................................... 113
9.2.1 Power Up, Wake and Reset Overview ...................................................... 113
9.2.2 RTC Power Well Transition: G5 to G3 State Transition ............................... 114
9.2.3 AC Power Applied: G3 to S4/S5 State Transition ....................................... 115
9.2.4 Using PWR_BTN_B: Transition from S4/S5 to S0 ...................................... 116
9.2.5 Power-Up Sequence without G2/G3: No Coin-Cell Battery .......................... 118
9.2.6 Going to Sleep: Transitions from S0 to S3 or S4/S5 .................................. 120
9.2.7 Wake Events: Transition from S3 to S0 ................................................... 120
9.2.8 System Reset Sequences....................................................................... 121
9.2.8.1 Cold Boot Sequence ............................................................... 121
9.2.8.2 Cold Reset Sequence............................................................... 121
9.2.8.3 Warm Reset Sequence (Internal) .............................................. 122
9.2.8.4 Externally Initiated Warm Reset Sequence ................................. 122
9.2.9 Handling Power Failures ........................................................................ 122
10.0 Thermal Management ............................................................................................ 123
10.1 Overview ....................................................................................................... 123
10.2 Thermal Sensor............................................................................................... 123
11.0 Processor Core ...................................................................................................... 125
12.0 Host Bridge ........................................................................................................... 127
12.1 Embedded SRAM (eSRAM)................................................................................ 127
12.1.1 Initialization ........................................................................................ 127
12.1.2 Configuration ....................................................................................... 127
12.1.2.1 4KB Page Mode ...................................................................... 127
12.1.2.2 512KB Block Page Mode........................................................... 128
12.1.3 Configuration Locking ........................................................................... 129
12.1.4 ECC Protection ..................................................................................... 130
12.1.5 Flush to DRAM ..................................................................................... 130
12.2 Isolated Memory Regions (IMR)......................................................................... 130
12.2.1 IMR Violation ....................................................................................... 131
12.2.2 IMR Locking......................................................................................... 131
12.3 Remote Management Unit DMA ......................................................................... 131
12.4 Register Map .................................................................................................. 132
12.5 PCI Configuration Registers .............................................................................. 132
12.5.1 PCI Device ID and Vendor ID Fields (PCI_DEVICE_VENDOR)—Offset 0h....... 133
12.5.2 PCI Status and Command Fields (PCI_STATUS_COMMAND)—Offset 4h ........ 133
12.5.3 PCI Class Code and Revision ID Fields (PCI_CLASS_REVISION)—Offset 8h ... 134
12.5.4 PCI Miscellaneous Fields (PCI_MISC)—Offset Ch ....................................... 134
12.5.5 PCI Subsystem ID and Subsystem Vendor ID Fields
(PCI_SUBSYSTEM)—Offset 2Ch .............................................................. 135
12.5.6 Message Bus Control Register (MCR) (SB_PACKET_REG)—Offset D0h.......... 135
12.5.7 Message Data Register (MDR) (SB_DATA_REG)—Offset D4h ...................... 136

12.5.8 Message Control Register eXtension (MCRX) (SB_ADDR_EXTN_REG)—Offset

D8h .................................................................................................... 136
12.5.9 Manufacturer ID (PCI_MANUFACTURER)—Offset F8h ................................. 137
12.6 IO Mapped Register ......................................................................................... 138
12.6.1 ACPI Processor Block............................................................................. 138
12.6.1.1 Processor Control (P_CNT)—Offset 0h ........................................ 138
12.6.1.2 Level 2 Register (P_LVL2)—Offset 4h ......................................... 138
12.6.1.3 C6 Control Register (P_C6C)—Offset Ch ..................................... 139
12.6.2 SPI DMA Block ..................................................................................... 140
12.6.2.1 Option Register 1(P_CFG_72) —Offset 72h ................................. 140
12.7 Message Bus Register....................................................................................... 140
12.7.1 Host Bridge Arbiter (Port 0x00) .............................................................. 140
12.7.1.1 Enhanced Configuration Space (AEC_CTRL)—Offset 0h ................ 141
12.7.1.2 STATUS—Offset 21h ............................................................... 141
12.7.1.3 Requester ID Match Control (ASUBCHAN_CTRL)—Offset 50h ........ 142
12.7.1.4 Requester ID Match Sub-Channel 1 (ASUBCHAN1_MATCH)—Offset
51h ....................................................................................... 143
52h ....................................................................................... 143
53h ....................................................................................... 144
12.7.2 Host Bridge (Port 0x03) ......................................................................... 145
12.7.2.1 Host Miscellaneous Controls 2 (HMISC2)—Offset 3h .................... 146
12.7.2.2 Host System Management Mode Controls (HSMMCTL)—Offset 4h.. 147
12.7.2.3 Host Memory I/O Boundary (HMBOUND)—Offset 8h .................... 148
12.7.2.4 Extended Configuration Space (HECREG)—Offset 9h ................... 148
12.7.2.5 Miscellaneous Legacy Signal Enables (HLEGACY)—Offset Ah ......... 149
12.7.2.6 Host Bridge Write Flush Control (HWFLUSH)—Offset Ch ............... 149
12.7.2.7 MTRR Capabilities (MTRR_CAP)—Offset 40h ............................... 150
12.7.2.8 MTRR Default Type (MTRR_DEF_TYPE)—Offset 41h ..................... 151
12.7.2.9 MTRR Fixed 64KB Range 0x00000 (MTRR_FIX64K_00000)—Offset
42h ....................................................................................... 151
43h ....................................................................................... 152
44h ....................................................................................... 152
45h ....................................................................................... 153
12.7.2.13 MTRR Fixed 16KB Range 0xA0000 (MTRR_FIX16K_A0000)—Offset
46h ....................................................................................... 154
12.7.2.14 MTRR Fixed 16KB Range 0xB0000 (MTRR_FIX16K_B0000)—Offset
47h ....................................................................................... 154
12.7.2.15 MTRR Fixed 4KB Range 0xC0000 (MTRR_FIX4K_C0000)—Offset
48h ....................................................................................... 155
49h ....................................................................................... 155
4Ah ....................................................................................... 156
12.7.2.18 MTRR Fixed 4KB Range 0xCC000 (MTRR_FIX4K_CC000)—Offset
4Bh ....................................................................................... 156
12.7.2.19 MTRR Fixed 4KB Range 0xD0000 (MTRR_FIX4K_D0000)—Offset
4Ch ....................................................................................... 157
4Dh....................................................................................... 158
4Eh ....................................................................................... 158
12.7.2.22 MTRR Fixed 4KB Range 0xDC000 (MTRR_FIX4K_DC000)—Offset
4Fh ....................................................................................... 159

12.7.2.23 MTRR Fixed 4KB Range 0xE0000 (MTRR_FIX4K_E0000)—Offset

50h....................................................................................... 159
51h....................................................................................... 160
52h....................................................................................... 160
12.7.2.26 MTRR Fixed 4KB Range 0xEC000 (MTRR_FIX4K_EC000)—Offset
53h....................................................................................... 161
12.7.2.27 MTRR Fixed 4KB Range 0xF0000 (MTRR_FIX4K_F0000)—Offset
54h....................................................................................... 161
55h....................................................................................... 162
56h....................................................................................... 163
12.7.2.30 MTRR Fixed 4KB Range 0xFC000 (MTRR_FIX4K_FC000)—Offset
57h....................................................................................... 163
12.7.2.31 System Management Range Physical Base
(MTRR_SMRR_PHYSBASE)—Offset 58h ...................................... 164
12.7.2.32 System Management Range Physical Mask
(MTRR_SMRR_PHYSMASK)—Offset 59h...................................... 164
12.7.2.33 MTRR Variable Range Physical Base 0
(MTRR_VAR_PHYSBASE0)—Offset 5Ah....................................... 165
12.7.2.34 MTRR Variable Range Physical Mask 0 (MTRR_VAR_PHYSMASK0)—
Offset 5Bh ............................................................................. 165
(MTRR_VAR_PHYSBASE1)—Offset 5Ch....................................... 166
Offset 5Dh ............................................................................. 167
(MTRR_VAR_PHYSBASE2)—Offset 5Eh ....................................... 167
Offset 5Fh.............................................................................. 168
(MTRR_VAR_PHYSBASE3)—Offset 60h ....................................... 168
Offset 61h ............................................................................. 169
Offset 63h ............................................................................. 170
Offset 65h ............................................................................. 171
Offset 67h ............................................................................. 172
Offset 69h ............................................................................. 173
12.7.3 Remote Management Unit (Port 0x04) .................................................... 174
12.7.3.1 SPI DMA Count Register (P_CFG_60)—Offset 60h........................ 174
12.7.3.2 SPI DMA Destination Register (P_CFG_61)—Offset 61h ................ 175
12.7.3.3 SPI DMA Source Register (P_CFG_62)—Offset 62h ...................... 175
12.7.3.4 Processor Register Block (P_BLK) Base Address
(P_CFG_70)—Offset 70h .......................................................... 176
12.7.3.5 Control Register (P_CFG_71)—Offset 71h................................... 176
12.7.3.6 Watchdog Control Register (P_CFG_74)—Offset 74h .................... 177

12.7.3.7 Thermal Sensor Mode Register (P_CFG_B0)—Offset B0h............... 178

12.7.3.8 Thermal Sensor Temperature Register (P_CFG_B1)—Offset B1h .... 178
12.7.3.9 Thermal Sensor Programmable Trip Point Register
(P_CFG_B2)—Offset B2h .......................................................... 179
12.7.4 Memory Manager (Port 0x05) ................................................................. 180
12.7.4.1 Control (BCTRL)—Offset 1h ...................................................... 181
12.7.4.2 Write Flush Policy (BWFLUSH)—Offset 2h ................................... 182
12.7.4.3 Isolated Memory Region Violation Control (BIMRVCTL)—Offset 19h 183
12.7.4.4 Debug 1 (DEBUG1)—Offset 31h ................................................ 184
12.7.4.5 Isolated Memory Region 0 Low Address (IMR0L)—Offset 40h ........ 186
12.7.4.6 Isolated Memory Region 0 High Address (IMR0H)—Offset 41h ....... 186
12.7.4.7 Isolated Memory Region 0 Read Mask (IMR0RM)—Offset 42h ........ 187
12.7.4.8 Isolated Memory Region 0 Write Mask (IMR0WM)—Offset 43h ....... 189
12.7.4.9 Isolated Memory Region 1 Low Address (IMR1L)—Offset 44h ........ 190
12.7.4.10 Isolated Memory Region 1 High Address (IMR1H)—Offset 45h.... 191
12.7.4.11 Isolated Memory Region 1 Read Mask (IMR1RM)—Offset 46h..... 191
12.7.4.12 Isolated Memory Region 1 Write Mask (IMR1WM)—Offset 47h ... 193
12.7.4.13 Isolated Memory Region 2 Low Address (IMR2L)—Offset 48h ..... 195
12.7.4.15 Isolated Memory Region 2 Read Mask (IMR2RM)—Offset 4Ah..... 196
12.7.4.16 Isolated Memory Region 2 Write Mask (IMR2WM)—Offset 4Bh ... 198
12.7.4.17 Isolated Memory Region 3 Low Address (IMR3L)—Offset 4Ch ..... 200
12.7.4.18 Isolated Memory Region 3 High Address (IMR3H)—Offset 4Dh ... 200
12.7.4.19 Isolated Memory Region 3 Read Mask (IMR3RM)—Offset 4Eh..... 201
12.7.4.20 Isolated Memory Region 3 Write Mask (IMR3WM)—Offset 4Fh.... 203
12.7.4.31 Isolated Memory Region 6 Read Mask (IMR6RM)—Offset 5Ah..... 215
12.7.4.32 Isolated Memory Region 6 Write Mask (IMR6WM)—Offset 5Bh ... 217
12.7.4.33 Isolated Memory Region 7 Low Address (IMR7L)—Offset 5Ch ..... 218
12.7.4.34 Isolated Memory Region 7 High Address (IMR7H)—Offset 5Dh ... 219
12.7.4.35 Isolated Memory Region 7 Read Mask (IMR7RM)—Offset 5Eh..... 219
12.7.4.36 Isolated Memory Region 7 Write Mask (IMR7WM)—Offset 5Fh.... 221
12.7.4.37 eSRAM Control (ESRAMCTRL)—Offset 81h ............................... 223
12.7.4.38 eSRAM Block Page Control (ESRAMPGCTRL_BLOCK)—Offset 82h 224
12.7.4.39 eSRAM Correctable Error (ESRAMCERR)—Offset 83h................. 226
12.7.4.40 eSRAM Uncorrectable Error (ESRAMUERR)—Offset 84h ............. 226
12.7.4.41 eSRAM ECC Error Syndrome (ESRAMSDROME)—Offset 88h ....... 227
12.7.5 Memory Manager eSRAM (Port 0x05) ...................................................... 228
12.7.5.1 eSRAM Page Control Register[0-127]
(ESRAMPGCTRL[0-127])—Offset 0h, Count 128, Stride 4h ............ 228
12.7.6 SoC Unit (Port 0x31) ............................................................................. 229
12.7.6.1 Thermal Sensor Configuration 4 (SCU_TSCFG4_Config)—Offset
34h ....................................................................................... 229
12.7.6.2 Sticky Write Once (CFGSTICKY_W1)—Offset 50h ......................... 230
12.7.6.3 Sticky Read/Write (CFGSTICKY_RW)—Offset 51h......................... 231
12.7.6.4 Non-Sticky Read/Write Once (CFGNONSTICKY_W1)—Offset 52h .... 231
13.0 System Memory Controller ..................................................................................... 233
13.2 Features ......................................................................................................... 234

13.2.1 System Memory Technology Supported ................................................... 234

13.2.2 Rules for Populating Memory Down Ranks................................................ 235
13.2.3 DRAM Error Detection & Correction (EDC)................................................ 235
13.2.4 DRAM Data Scrambling ......................................................................... 236
13.2.5 Power Management .............................................................................. 236
13.3 Register Map .................................................................................................. 236
13.4 Message Bus Registers..................................................................................... 236
13.4.1 DRAM Rank Population (DRP)—Offset 0h ................................................. 237
13.4.2 DRAM Timing Register 0 (DTR0)—Offset 1h ............................................. 238
13.4.7 DRAM Power Management Control 0 (DPMC0)—Offset 6h........................... 245
13.4.8 DRAM Refresh Control (DRFC)—Offset 8h ................................................ 247
13.4.9 DRAM Scheduler Control (DSCH)—Offset 9h............................................. 248
13.4.10 DRAM Calibration Control (DCAL)—Offset Ah .......................................... 249
13.4.11 DRAM Reset Management Control (DRMC)—Offset Bh.............................. 250
13.4.12 Power Management Status (PMSTS)—Offset Ch ...................................... 251
13.4.13 DRAM Control Operation (DCO)—Offset Fh ............................................. 252
13.4.14 Sticky Scratchpad 0 (SSKPD0)—Offset 4Ah ............................................ 252
13.4.15 Sticky Scratchpad 1 (SSKPD1)—Offset 4Bh ............................................ 253
13.4.16 DRAM ECC Control Register (DECCCTRL)—Offset 60h .............................. 253
13.4.17 DRAM ECC Status (DECCSTAT)—Offset 61h............................................ 254
13.4.18 DRAM ECC Single Bit Error Count (DECCSBECNT)—Offset 62h .................. 254
13.4.19 DRAM Single Bit ECC Error Captured Address (DECCSBECA)—Offset 68h.... 255
13.4.20 DRAM Single Bit ECC Error Captured Syndrome (DECCSBECS)—Offset 69h. 256
13.4.21 DRAM Double Bit ECC Error Captured Address (DECCDBECA)—Offset 6Ah .. 256
13.4.22 DRAM Double Bit ECC Error Captured Syndrome (DECCDBECS)—Offset 6Bh 257
13.4.23 Memory Controller Fuse Status (DFUSESTAT)—Offset 70h ........................ 257
13.4.24 Scrambler Seed (DSCRMSEED)—Offset 80h............................................ 258
13.5 Message Bus Commands .................................................................................. 259
14.0 PCI Express* 2.0 ................................................................................................... 261
14.2 Features ........................................................................................................ 261
14.2.1 Interrupts and Events ........................................................................... 262
14.2.1.1 Express Card Hot Plug Events ................................................... 262
14.2.1.2 System Error (SERR)............................................................... 263
14.2.2 Power Management .............................................................................. 263
14.3 References ..................................................................................................... 263
14.4 Register Map .................................................................................................. 263
14.5.1 Identifiers (ID)—Offset 0h ..................................................................... 266
14.5.2 Primary Status (CMD_PSTS)—Offset 4h................................................... 266
14.5.3 Class Code (RID_CC)—Offset 8h............................................................. 268
14.5.4 Header Type (CLS_PLT_HTYPE)—Offset Ch .............................................. 268
14.5.5 Secondary Latency Timer (BNUM_SLT)—Offset 18h .................................. 269
14.5.6 Secondary Status (IOBL_SSTS)—Offset 1Ch ............................................ 269
14.5.7 Memory Base and Limit (MBL)—Offset 20h .............................................. 270
14.5.8 Prefetchable Memory Base and Limit (PMBL)—Offset 24h........................... 271
14.5.9 Prefetchable Memory Base Upper 32 Bits (PMBU32)—Offset 28h ................. 271
14.5.10 Prefetchable Memory Limit Upper 32 Bits (PMLU32)—Offset 2Ch ............... 272
14.5.11 Capabilities List Pointer (CAPP)—Offset 34h............................................ 272
14.5.12 Bridge Control (INTR_BCTRL)—Offset 3Ch.............................................. 273

14.5.13 PCI Express Capabilities (CLIST_XCAP)—Offset 40h ................................. 274

14.5.14 Device Capabilities (DCAP)—Offset 44h .................................................. 275
14.5.15 Device Status (DCTL_DSTS)—Offset 48h ................................................ 276
14.5.16 Link Capabilities (LCAP)—Offset 4Ch ...................................................... 277
14.5.17 Link Status (LCTL_LSTS)—Offset 50h..................................................... 279
14.5.18 Slot Capabilities (SLCAP)—Offset 54h..................................................... 280
14.5.19 Slot Status (SLCTL_SLSTS)—Offset 58h ................................................. 281
14.5.20 Root Control (RCTL)—Offset 5Ch ........................................................... 283
14.5.21 Root Status (RSTS)—Offset 60h ............................................................ 283
14.5.22 Device Capabilities 2 (DCAP2)—Offset 64h.............................................. 284
14.5.23 Device Status 2 (DCTL2_DSTS2)—Offset 68h.......................................... 285
14.5.24 Link Capability 2 (LCAP2)—Offset 6Ch.................................................... 286
14.5.25 Link Status 2 (LCTL2_LSTS2)—Offset 70h .............................................. 286
14.5.26 Slot Capabilities 2 (SLCAP2)—Offset 74h ................................................ 288
14.5.27 Slot Status 2 (SLCTL2_SLSTS2)—Offset 78h ........................................... 288
14.5.28 Message Signaled Interrupt Message Control (MID_MC)—Offset 80h .......... 289
14.5.29 Message Signaled Interrupt Message Address (MA)—Offset 84h................. 289
14.5.30 Message Signaled Interrupt Message Data (MD)—Offset 88h..................... 290
14.5.31 Subsystem Vendor Capability (SVCAP)—Offset 90h.................................. 290
14.5.32 Subsystem Vendor IDs (SVID)—Offset 94h ............................................. 291
14.5.33 PCI Power Management Capabilities (PMCAP_PMC)—Offset A0h................. 291
14.5.34 PCI Power Management Control And Status (PMCS)—Offset A4h................ 292
14.5.35 Channel Configuration (CCFG)—Offset D0h ............................................. 293
14.5.36 Miscellaneous Port Configuration 2 (MPC2)—Offset D4h ............................ 294
14.5.37 Miscellaneous Port Configuration (MPC)—Offset D8h ................................ 295
14.5.38 SMI / SCI Status (SMSCS)—Offset DCh.................................................. 296
14.5.39 Message Bus Control (PHYCTL_PHYCTL2_IOSFSBCTL)—Offset F4h ............. 297
14.5.40 Advanced Error Reporting Capability Header (AECH)—Offset 100h ............. 298
14.5.41 Uncorrectable Error Status (UES)—Offset 104h ....................................... 299
14.5.42 Uncorrectable Error Mask (UEM)—Offset 108h......................................... 300
14.5.43 Uncorrectable Error Severity (UEV)—Offset 10Ch..................................... 301
14.5.44 Correctable Error Status (CES)—Offset 110h........................................... 302
14.5.45 Correctable Error Mask (CEM)—Offset 114h ............................................ 303
14.5.46 Advanced Error Capabilities and Control (AECC)—Offset 118h ................... 304
14.5.47 Header Log (HL_DW1)—Offset 11Ch ...................................................... 304
14.5.48 Header Log (HL_DW2)—Offset 120h ...................................................... 305
14.5.51 Root Error Command (REC)—Offset 12Ch ............................................... 306
14.5.52 Root Error Status (RES)—Offset 130h .................................................... 306
14.5.53 Error Source Identification (ESID)—Offset 134h ...................................... 307
15.0 10/100 Mbps Ethernet ........................................................................................... 309
15.2 Features:........................................................................................................ 309
15.3 References...................................................................................................... 310
15.4 Register Map ................................................................................................... 311
15.5 PCI Configuration Registers ............................................................................... 311
15.5.1 Vendor ID (VENDOR_ID)—Offset 0h ........................................................ 312
15.5.2 Device ID (DEVICE_ID)—Offset 2h .......................................................... 313
15.5.3 Command Register (COMMAND_REGISTER)—Offset 4h .............................. 313
15.5.4 Status Register (STATUS)—Offset 6h....................................................... 314
15.5.5 Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h .................. 315
15.5.6 Cache Line Size (CACHE_LINE_SIZE)—Offset Ch ....................................... 315
15.5.7 Latency Timer (LATENCY_TIMER)—Offset Dh ............................................ 315

15.5.8 Header Type (HEADER_TYPE)—Offset Eh ................................................. 316

15.5.9 BIST (BIST)—Offset Fh ......................................................................... 316
15.5.10 Base Address Register (BAR0)—Offset 10h ............................................. 317
15.5.11 Cardbus CIS Pointer (CARDBUS_CIS_POINTER)—Offset 28h..................... 317
15.5.12 Subsystem Vendor ID (SUB_SYS_VENDOR_ID)—Offset 2Ch ..................... 318
15.5.13 Subsystem ID (SUB_SYS_ID)—Offset 2Eh.............................................. 318
15.5.14 Expansion ROM Base Address (EXP_ROM_BASE_ADR)—Offset 30h ............ 318
15.5.15 Capabilities Pointer (CAP_POINTER)—Offset 34h ..................................... 319
15.5.16 Interrupt Line Register (INTR_LINE)—Offset 3Ch .................................... 319
15.5.17 Interrupt Pin Register (INTR_PIN)—Offset 3Dh ....................................... 320
15.5.18 MIN_GNT (MIN_GNT)—Offset 3Eh......................................................... 320
15.5.19 MAX_LAT (MAX_LAT)—Offset 3Fh ......................................................... 320
15.5.20 Capability ID (PM_CAP_ID)—Offset 80h ................................................. 321
15.5.21 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h .......................... 321
15.5.22 Power Management Capabilities (PMC)—Offset 82h ................................. 321
15.5.23 Power Management Control/Status Register (PMCSR)—Offset 84h ............. 322
15.5.24 PM CSR PCI-to-PCI Bridge Support Extension (PMCSR_BSE)—Offset 86h.... 323
15.5.25 Power Management Data Register (DATA_REGISTER)—Offset 87h............. 323
15.5.26 Capability ID (MSI_CAP_ID)—Offset A0h................................................ 324
15.5.27 Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h ........................ 324
15.5.28 Message Control (MESSAGE_CTRL)—Offset A2h ...................................... 324
15.5.29 Message Address (MESSAGE_ADDR)—Offset A4h .................................... 325
15.5.30 Message Data (MESSAGE_DATA)—Offset A8h......................................... 325
15.5.31 Mask Bits for MSI (PER_VEC_MASK)—Offset ACh .................................... 326
15.5.32 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h ................................. 326
15.6 Memory Mapped Registers ................................................................................ 327
15.6.1 MAC Configuration Register (Register 0) (GMAC_REG_0)—Offset 0h............ 331
15.6.2 MAC Frame Filter (Register 1) (GMAC_REG_1)—Offset 4h .......................... 334
15.6.3 Hash Table High Register (Register 2) (GMAC_REG_2)—Offset 8h ............... 336
15.6.4 Hash Table Low Register (Register 3) (GMAC_REG_3)—Offset Ch ............... 336
15.6.5 GMII Address Register (Register 4) (GMAC_REG_4)—Offset 10h ................. 337
15.6.6 GMII Data Register (Register 5) (GMAC_REG_5)—Offset 14h ..................... 338
15.6.7 Flow Control Register (Register 6) (GMAC_REG_6)—Offset 18h .................. 339
15.6.8 VLAN Tag Register (Register 7) (GMAC_REG_7)—Offset 1Ch ...................... 340
15.6.9 Version Register (Register 8) (GMAC_REG_8)—Offset 20h ......................... 341
15.6.10 Debug Register (Register 9) (GMAC_REG_9)—Offset 24h......................... 342
15.6.11 Interrupt Register (Register 14) (GMAC_REG_14)—Offset 38h.................. 343
15.6.12 Interrupt Mask Register (Register 15) (GMAC_REG_15)—Offset 3Ch ......... 344
15.6.13 MAC Address0 High Register (Register 16) (GMAC_REG_16)—Offset 40h ... 345
15.6.14 MAC Address0 Low Register (Register 17) (GMAC_REG_17)—Offset 44h .... 345
15.6.15 MMC Control Register (Register 64) (GMAC_REG_64)—Offset 100h ........... 346
15.6.16 MMC Receive Interrupt Register (MMC_INTR_RX)—Offset 104h ................ 347
15.6.17 MMC Transmit Interrupt Register (MMC_INTR_TX)—Offset 108h ............... 349
15.6.18 MMC Receive Interrupt Mask Register
(MMC_INTR_MASK_RX)—Offset 10Ch ................................................... 351
15.6.19 MMC Transmit Interrupt Mask Register
(MMC_INTR_MASK_TX)—Offset 110h.................................................... 353
15.6.20 MMC Transmit Good Bad Octet Counter Register
(TXOCTETCOUNT_GB)—Offset 114h........................................................ 355
15.6.21 MMC Transmit Good Bad Frame Counter Register
(TXFRAMECOUNT_GB)—Offset 118h ....................................................... 355
15.6.22 MMC Transmit Broadcast Good Frame Counter Register
(TXBROADCASTFRAMES_G)—Offset 11Ch................................................ 356
15.6.23 MMC Transmit Multicast Good Frame Counter Register
(TXMULTICASTFRAMES_G)—Offset 120h ................................................. 356

15.6.24 MMC Transmit 64 Octet Good Bad Frame Counter Register

(TX64OCTETS_GB)—Offset 124h ............................................................ 357
15.6.25 MMC Transmit 65 to 127 Octet Good Bad Frame Counter Register
(TX65TO127OCTETS_GB)—Offset 128h ................................................... 357
(TX128TO255OCTETS_GB)—Offset 12Ch ................................................. 358
(TX256TO511OCTETS_GB)—Offset 130h.................................................. 358
(TX512TO1023OCTETS_GB)—Offset 134h ................................................ 358
15.6.29 MMC Transmit 1024 to Maximum Octet Good Bad Frame Counter Register
(TX1024TOMAXOCTETS_GB)—Offset 138h ............................................... 359
15.6.30 MMC Transmit Unicast Good Bad Frame Counter Register
(TXUNICASTFRAMES_GB)—Offset 13Ch ................................................... 359
15.6.31 MMC Transmit Multicast Good Bad Frame Counter Register
(TXMULTICASTFRAMES_GB)—Offset 140h................................................ 360
15.6.32 MMC Transmit Broadcast Good Bad Frame Counter Register
(TXBROADCASTFRAMES_GB)—Offset 144h .............................................. 360
15.6.33 MMC Transmit Underflow Error Frame Counter Register
(TXUNDERFLOWERROR)—Offset 148h ..................................................... 361
15.6.34 MMC Transmit Single Collision Good Frame Counter Register
(TXSINGLECOL_G)—Offset 14Ch............................................................. 361
15.6.35 MMC Transmit Multiple Collision Good Frame Counter Register
(TXMULTICOL_G)—Offset 150h............................................................... 362
15.6.36 MMC Transmit Deferred Frame Counter Register (TXDEFERRED)—Offset
154h ................................................................................................... 362
15.6.37 MMC Transmit Late Collision Frame Counter Register (TXLATECOL)—Offset
158h ................................................................................................... 362
15.6.38 MMC Transmit Excessive Collision Frame Counter Register
(TXEXESSCOL)—Offset 15Ch .................................................................. 363
15.6.39 MMC Transmit Carrier Error Frame Counter Register
(TXCARRIERERROR)—Offset 160h........................................................... 363
15.6.40 MMC Transmit Good Octet Counter Register (TXOCTETCOUNT_G)—Offset
164h ................................................................................................... 364
15.6.41 MMC Transmit Good Frame Counter Register (TXFRAMECOUNT_G)—Offset
168h ................................................................................................... 364
15.6.42 MMC Transmit Excessive Deferral Frame Counter Register (TXEXCESSDEF)—
Offset 16Ch ......................................................................................... 365
15.6.43 MMC Transmit Pause Frame Counter Register (TXPAUSEFRAMES)—Offset
170h ................................................................................................... 365
15.6.44 MMC Transmit VLAN Good Frame Counter Register
(TXVLANFRAMES_G)—Offset 174h .......................................................... 366
15.6.45 MMC Transmit Oversize Good Frame Counter Register
(TXOVERSIZE_G)—Offset 178h............................................................... 366
15.6.46 MMC Receive Good Bad Frame Counter Register
(RXFRAMECOUNT_GB)—Offset 180h........................................................ 366
15.6.47 MMC Receive Good Bad Octet Counter Register
(RXOCTETCOUNT_GB)—Offset 184h........................................................ 367
15.6.48 MMC Receive Good Octet Counter Register (RXOCTETCOUNT_G)—Offset
188h ................................................................................................... 367
15.6.49 MMC Receive Broadcast Good Frame Counter Register
(RXBROADCASTFRAMES_G)—Offset 18Ch ................................................ 368
15.6.50 MMC Receive Multicast Good Frame Counter Register
(RXMULTICASTFRAMES_G)—Offset 190h ................................................. 368
15.6.51 MMC Receive CRC Error Frame Counter Register (RXCRCERROR)—Offset
194h ................................................................................................... 369

15.6.52 MMC Receive Alignment Error Frame Counter Register

(RXALIGNMENTERROR)—Offset 198h ...................................................... 369
15.6.53 MMC Receive Runt Frame Counter Register (RXRUNTERROR)—Offset 19Ch 370
15.6.54 MMC Receive Jabber Error Frame Counter Register
(RXJABBERERROR)—Offset 1A0h ............................................................ 370
15.6.55 MMC Receive Undersize Good Frame Counter Register
(RXUNDERSIZE_G)—Offset 1A4h............................................................ 370
15.6.56 MMC Receive Oversize Good Frame Counter Register
(RXOVERSIZE_G)—Offset 1A8h.............................................................. 371
15.6.57 MMC Receive 64 Octet Good Bad Frame Counter Register
(RX64OCTETS_GB)—Offset 1ACh ........................................................... 371
15.6.58 MMC Receive 65 to 127 Octet Good Bad Frame Counter Register
(RX65TO127OCTETS_GB)—Offset 1B0h................................................... 372
(RX128TO255OCTETS_GB)—Offset 1B4h ................................................. 372
(RX256TO511OCTETS_GB)—Offset 1B8h ................................................. 373
(RX512TO1023OCTETS_GB)—Offset 1BCh ............................................... 373
15.6.62 MMC Receive 1024 to Maximum Octet Good Bad Frame Counter Register
(RX1024TOMAXOCTETS_GB)—Offset 1C0h .............................................. 374
15.6.63 MMC Receive Unicast Good Frame Counter Register (RXUNICASTFRAMES_G)—
Offset 1C4h ......................................................................................... 374
15.6.64 MMC Receive Length Error Frame Counter Register
(RXLENGTHERROR)—Offset 1C8h ........................................................... 374
15.6.65 MMC Receive Out Of Range Error Frame Counter Register
(RXOUTOFRANGETYPE)—Offset 1CCh...................................................... 375
15.6.66 MMC Receive Pause Frame Counter Register (RXPAUSEFRAMES)—Offset
1D0h .................................................................................................. 375
15.6.67 MMC Receive FIFO Overflow Frame Counter Register
(RXFIFOOVERFLOW)—Offset 1D4h.......................................................... 376
15.6.68 MMC Receive VLAN Good Bad Frame Counter Register (RXVLANFRAMES_GB)—
Offset 1D8h......................................................................................... 376
15.6.69 MMC Receive Watchdog Error Frame Counter Register
(RXWATCHDOGERROR)—Offset 1DCh ..................................................... 377
15.6.70 MMC Receive Error Frame Counter Register (RXRCVERROR)—Offset 1E0h .. 377
15.6.71 MMC Receive Control Frame Counter Register (RXCTRLFRAMES_G)—Offset
1E4h .................................................................................................. 378
15.6.72 MMC IPC Receive Checksum Offload Interrupt Mask Register
(MMC_IPC_INTR_MASK_RX)—Offset 200h ............................................... 378
15.6.73 MMC Receive Checksum Offload Interrupt Register
(MMC_IPC_INTR_RX)—Offset 208h......................................................... 380
15.6.74 MMC Receive IPV4 Good Frame Counter Register
(RXIPV4_GD_FRMS)—Offset 210h .......................................................... 382
15.6.75 MMC Receive IPV4 Header Error Frame Counter Register
(RXIPV4_HDRERR_FRMS)—Offset 214h ................................................... 383
15.6.76 MMC Receive IPV4 No Payload Frame Counter Register
(RXIPV4_NOPAY_FRMS)—Offset 218h ..................................................... 383
15.6.77 MMC Receive IPV4 Fragmented Frame Counter Register
(RXIPV4_FRAG_FRMS)—Offset 21Ch ....................................................... 384
15.6.78 MMC Receive IPV4 UDP Checksum Disabled Frame Counter Register
(RXIPV4_UDSBL_FRMS)—Offset 220h ..................................................... 384
(RXIPV6_GD_FRMS)—Offset 224h .......................................................... 384
(RXIPV6_HDRERR_FRMS)—Offset 228h ................................................... 385


(RXIPV6_NOPAY_FRMS)—Offset 22Ch ..................................................... 385
15.6.82 MMC Receive UDP Good Frame Counter Register (RXUDP_GD_FRMS)—Offset
230h ................................................................................................... 386
15.6.83 MMC Receive UDP Error Frame Counter Register
(RXUDP_ERR_FRMS)—Offset 234h .......................................................... 386
15.6.84 MMC Receive TCP Good Frame Counter Register (RXTCP_GD_FRMS)—Offset
238h ................................................................................................... 387
15.6.85 MMC Receive TCP Error Frame Counter Register (RXTCP_ERR_FRMS)—Offset
23Ch................................................................................................... 387
15.6.86 MMC Receive ICMP Good Frame Counter Register
(RXICMP_GD_FRMS)—Offset 240h .......................................................... 388
15.6.87 MMC Receive ICMP Error Frame Counter Register
(RXICMP_ERR_FRMS)—Offset 244h......................................................... 388
15.6.88 MMC Receive IPV4 Good Octet Counter Register
(RXIPV4_GD_OCTETS)—Offset 250h ....................................................... 388
15.6.89 MMC Receive IPV4 Header Error Octet Counter Register
(RXIPV4_HDRERR_OCTETS)—Offset 254h ................................................ 389
15.6.90 MMC Receive IPV4 No Payload Octet Counter Register
(RXIPV4_NOPAY_OCTETS)—Offset 258h .................................................. 389
15.6.91 MMC Receive IPV4 Fragmented Octet Counter Register
(RXIPV4_FRAG_OCTETS)—Offset 25Ch .................................................... 390
15.6.92 MMC Receive IPV4 UDP Checksum Disabled Octet Counter Register
(RXIPV4_UDSBL_OCTETS)—Offset 260h .................................................. 390
(RXIPV6_GD_OCTETS)—Offset 264h ....................................................... 391
15.6.94 MMC Receive IPV6 Good Octet Counter Register (RXIPV6_HDRERR_OCTETS)—
Offset 268h.......................................................................................... 391
(RXIPV6_NOPAY_OCTETS)—Offset 26Ch .................................................. 392
15.6.96 MMC Receive IPV6 No Payload Octet Counter Register (RXUDP_GD_OCTETS)—
Offset 270h.......................................................................................... 392
15.6.97 MMC Receive UDP Good Octet Counter Register
(RXUDP_ERR_OCTETS)—Offset 274h....................................................... 392
15.6.98 MMC Receive TCP Good Octet Counter Register
(RXTCP_GD_OCTETS)—Offset 278h ........................................................ 393
15.6.99 MMC Receive TCP Error Octet Counter Register
(RXTCP_ERR_OCTETS)—Offset 27Ch ....................................................... 393
15.6.100 MMC Receive ICMP Good Octet Counter Register
(RXICMP_GD_OCTETS)—Offset 280h....................................................... 394
15.6.101 MMC Receive ICMP Error Octet Counter Register
(RXICMP_ERR_OCTETS)—Offset 284h ..................................................... 394
15.6.102 VLAN Tag Inclusion or Replacement Register (Register 353)
(GMAC_REG_353)—Offset 584h.............................................................. 395
15.6.103 VLAN Hash Table Register (Register 354) (GMAC_REG_354)—Offset 588h 396
15.6.104 Timestamp Control Register (Register 448)
(GMAC_REG_448)—Offset 700h ......................................................... 396
15.6.105 Sub-Second Increment Register (Register 449) (GMAC_REG_449)—Offset
704h ................................................................................................... 398
15.6.106 System Time - Seconds Register (Register 450) (GMAC_REG_450)—Offset
708h ................................................................................................... 398
15.6.107 System Time - Nanoseconds Register (Register 451)
(GMAC_REG_451)—Offset 70Ch ............................................................. 399
15.6.108 System Time - Seconds Update Register (Register 452) (GMAC_REG_452)—
Offset 710h.......................................................................................... 399
15.6.109 System Time - Nanoseconds Update Register (Register 453)
(GMAC_REG_453)—Offset 714h.............................................................. 400

15.6.110 Timestamp Addend Register (Register 454)

(GMAC_REG_454)—Offset 718h ......................................................... 400
15.6.111 Target Time Seconds Register (Register 455) (GMAC_REG_455)—Offset
71Ch .................................................................................................. 401
15.6.112 Target Time Nanoseconds Register (Register 456)
(GMAC_REG_456)—Offset 720h ............................................................. 401
15.6.113 System Time - Higher Word Seconds Register (Register 457)
(GMAC_REG_457)—Offset 724h ............................................................. 402
15.6.114 Timestamp Status Register (Register 458)
(GMAC_REG_458)—Offset 728h.......................................................... 403
15.6.115 Bus Mode Register (Register 0) (DMA_REG_0)—Offset 1000h ................. 404
15.6.116 Transmit Poll Demand Register (Register 1) (DMA_REG_1)—Offset 1004h 406
15.6.117 Receive Poll Demand Register (Register 2) (DMA_REG_2)—Offset 1008h.. 406
15.6.118 Receive Descriptor List Address Register (Register 3) (DMA_REG_3)—Offset
100Ch ................................................................................................ 407
15.6.119 Transmit Descriptor List Address Register (Register 4)
(DMA_REG_4)—Offset 1010h ................................................................. 407
15.6.120 Status Register (Register 5) (DMA_REG_5)—Offset 1014h...................... 408
15.6.121 Operation Mode Register (Register 6) (DMA_REG_6)—Offset 1018h......... 411
15.6.122 Interrupt Enable Register (Register 7) (DMA_REG_7)—Offset 101Ch........ 414
15.6.123 Missed Frame and Buffer Overflow Counter Register (Register 8)
(DMA_REG_8)—Offset 1020h ................................................................. 415
15.6.124 Receive Interrupt Watchdog Timer Register (Register 9)
(DMA_REG_9)—Offset 1024h ................................................................. 416
15.6.125 AHB Status Register (Register 11) (DMA_REG_11)—Offset 102Ch ........... 416
15.6.126 Current Host Transmit Descriptor Register (Register 18)
(DMA_REG_18)—Offset 1048h ............................................................... 417
15.6.127 Current Host Receive Descriptor Register (Register 19)
(DMA_REG_19)—Offset 104Ch ............................................................... 417
15.6.128 Current Host Transmit Buffer Address Register (Register 20)
(DMA_REG_20)—Offset 1050h ............................................................... 418
15.6.129 Current Host Receive Buffer Address Register (Register 21) (DMA_REG_21)—
Offset 1054h ....................................................................................... 418
15.6.130 HW Feature Register (Register 22) (DMA_REG_22)—Offset 1058h ........... 419
15.7 MAC Descriptor Details..................................................................................... 421
15.7.1 Descriptor Overview ............................................................................. 421
15.7.2 Descriptor Endianness .......................................................................... 421
15.7.3 Transmit Descriptor ............................................................................. 421
15.7.4 Receive Descriptor ............................................................................... 427
16.0 USB 2.0 ................................................................................................................. 435
16.2 Features ........................................................................................................ 435
16.2.1 USB2.0 Host Controller Features ............................................................ 435
16.2.2 USB2.0 Device Features ........................................................................ 436
16.3 References ..................................................................................................... 436
16.4 Register Map .................................................................................................. 437
16.5.1 USB Device ......................................................................................... 437
16.5.1.1 Vendor ID (VENDOR_ID)—Offset 0h .......................................... 438
16.5.1.2 Device ID (DEVICE_ID)—Offset 2h ............................................ 439
16.5.1.3 Command Register (COMMAND_REGISTER)—Offset 4h ................ 439
16.5.1.4 Status Register (STATUS)—Offset 6h......................................... 440
16.5.1.5 Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h .... 440
16.5.1.6 Cache Line Size (CACHE_LINE_SIZE)—Offset Ch ......................... 441
16.5.1.7 Latency Timer (LATENCY_TIMER)—Offset Dh .............................. 441
16.5.1.8 Header Type (HEADER_TYPE)—Offset Eh ................................... 442

16.5.1.9 BIST (BIST)—Offset Fh ............................................................ 442

16.5.1.10 Base Address Register (BAR0)—Offset 10h ............................... 443
16.5.1.11 Cardbus CIS Pointer (CARDBUS_CIS_POINTER)—Offset 28h ....... 443
16.5.1.12 Subsystem Vendor ID (SUB_SYS_VENDOR_ID)—Offset 2Ch........ 444
16.5.1.13 Subsystem ID (SUB_SYS_ID)—Offset 2Eh ................................ 444
16.5.1.14 Expansion ROM Base Address
(EXP_ROM_BASE_ADR)—Offset 30h ........................................ 444
16.5.1.15 Capabilities Pointer (CAP_POINTER)—Offset 34h ........................ 445
16.5.1.16 Interrupt Line Register (INTR_LINE)—Offset 3Ch ....................... 445
16.5.1.17 Interrupt Pin Register (INTR_PIN)—Offset 3Dh .......................... 446
16.5.1.18 MIN_GNT (MIN_GNT)—Offset 3Eh............................................ 446
16.5.1.19 MAX_LAT (MAX_LAT)—Offset 3Fh ............................................ 446
16.5.1.20 Capability ID (PM_CAP_ID)—Offset 80h .................................... 447
16.5.1.21 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h ............. 447
16.5.1.22 Power Management Capabilities (PMC)—Offset 82h .................... 447
16.5.1.23 Power Management Control/Status Register (PMCSR)—Offset 84h 448
16.5.1.24 PM CSR PCI-to-PCI Bridge Support Extension (PMCSR_BSE)—Offset
86h ....................................................................................... 449
16.5.1.25 Power Management Data Register (DATA_REGISTER)—Offset 87h 449
16.5.1.26 Capability ID (MSI_CAP_ID)—Offset A0h................................... 450
16.5.1.27 Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h ........... 450
16.5.1.28 Message Control (MESSAGE_CTRL)—Offset A2h ......................... 450
16.5.1.29 Message Address (MESSAGE_ADDR)—Offset A4h ....................... 451
16.5.1.30 Message Data (MESSAGE_DATA)—Offset A8h............................ 451
16.5.1.31 Mask Bits for MSI (PER_VEC_MASK)—Offset ACh ....................... 452
16.5.1.32 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h .................... 452
16.5.2 USB EHCI ............................................................................................ 453
16.5.2.4 Status Register (STATUS)—Offset 6h ......................................... 455
16.5.2.8 Header Type (HEADER_TYPE)—Offset Eh .................................... 457
16.5.2.12 Subsystem Vendor ID (SUB_SYS_VENDOR_ID)—Offset 2Ch ....... 459
16.5.2.15 Capabilities Pointer (CAP_POINTER)—Offset 34h ....................... 460
16.5.2.16 Interrupt Line Register (INTR_LINE)—Offset 3Ch....................... 460
16.5.2.17 Interrupt Pin Register (INTR_PIN)—Offset 3Dh.......................... 461
16.5.2.18 MIN_GNT (MIN_GNT)—Offset 3Eh ........................................... 461
16.5.2.19 MAX_LAT (MAX_LAT)—Offset 3Fh............................................ 462
16.5.2.20 Serial Bus Release Number Register (SBRN)—Offset 60h ............ 462
16.5.2.21 Frame Length Adjustment Register (FLADJ)—Offset 61h............. 462
16.5.2.22 Capability ID (PM_CAP_ID)—Offset 80h ................................... 463
16.5.2.23 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h ............ 463
16.5.2.24 Power Management Capabilities (PMC)—Offset 82h.................... 463
16.5.2.25 Power Management Control/Status Register
(PMCSR)—Offset 84h............................................................. 464
86h ....................................................................................... 465
16.5.2.27 Power Management Data Register
(DATA_REGISTER)—Offset 87h ............................................... 465
16.5.2.28 Capability ID (MSI_CAP_ID)—Offset A0h .................................. 465
16.5.2.29 Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h........... 466

16.5.2.30 Message Control (MESSAGE_CTRL)—Offset A2h........................ 466

16.5.2.31 Message Address (MESSAGE_ADDR)—Offset A4h...................... 467
16.5.2.32 Message Data (MESSAGE_DATA)—Offset A8h........................... 467
16.5.2.33 Mask Bits for MSI (PER_VEC_MASK)—Offset ACh ...................... 468
16.5.2.34 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h................... 468
16.5.2.35 USB Legacy Support Extended Capability (USBLEGSUP)—Offset
C0h ...................................................................................... 468
16.5.2.36 USB Legacy Support Control/Status
(USBLEGCTLSTS)—Offset C4h ................................................ 469
16.5.3 USB OHCI ........................................................................................... 471
16.5.3.4 Status Register (STATUS)—Offset 6h......................................... 473
16.5.3.8 Header Type (HEADER_TYPE)—Offset Eh ................................... 475
16.5.3.12 Subsystem Vendor ID (SUB_SYS_VENDOR_ID)—Offset 2Ch........ 477
16.5.3.15 Capabilities Pointer (CAP_POINTER)—Offset 34h........................ 479
16.5.3.16 Interrupt Line Register (INTR_LINE)—Offset 3Ch ....................... 479
16.5.3.17 Interrupt Pin Register (INTR_PIN)—Offset 3Dh .......................... 479
16.5.3.18 MIN_GNT (MIN_GNT)—Offset 3Eh ........................................... 480
16.5.3.19 MAX_LAT (MAX_LAT)—Offset 3Fh ............................................ 480
16.5.3.20 Capability ID (PM_CAP_ID)—Offset 80h.................................... 480
16.5.3.21 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h ............ 481
16.5.3.22 Power Management Capabilities (PMC)—Offset 82h .................... 481
16.5.3.23 Power Management Control/Status Register (PMCSR)—Offset 84h 482
86h....................................................................................... 483
16.5.3.25 Power Management Data Register (DATA_REGISTER)—Offset 87h 483
16.5.3.26 Capability ID (MSI_CAP_ID)—Offset A0h .................................. 483
16.5.3.27 Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h ........... 484
16.5.3.28 Message Control (MESSAGE_CTRL)—Offset A2h ........................ 484
16.5.3.29 Message Address (MESSAGE_ADDR)—Offset A4h ...................... 484
16.5.3.30 Message Data (MESSAGE_DATA)—Offset A8h ........................... 485
16.5.3.31 Mask Bits for MSI (PER_VEC_MASK)—Offset ACh....................... 485
16.5.3.32 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h ................... 486
16.6.1 USB Device ......................................................................................... 486
16.6.1.1 IN Endpoint 0 Control Register (ep0_in_ctrl_udc_reg)—Offset 0h .. 489
16.6.1.2 IN Endpoint 0 Status Register (ep0_in_sts_udc_reg)—Offset 4h .... 490
16.6.1.3 IN Endpoint 0 Buffer Size Register
(ep0_in_bufsize_udc_reg)—Offset 8h ........................................ 492
16.6.1.4 IN Endpoint 0 Maximum Packet Size Register
(ep0_in_mpkt_sz_reg)—Offset Ch............................................. 493
16.6.1.5 IN Endpoint 0 Data Descriptor Pointer Register
(ep0_in_desptr_udc_reg)—Offset 14h ....................................... 493
16.6.1.6 IN Endpoint 0 Write Confirmation register (for Slave-Only mode)
(ep0_wr_cfrm_udc_reg)—Offset 1Ch ......................................... 494
16.6.1.7 IN Endpoint 1 Control Register (ep1_in_ctrl_udc_reg)—Offset 20h. 494
16.6.1.8 IN Endpoint 1 Status Register (ep1_in_sts_udc_reg)—Offset 24h .. 495
(ep1_in_bufsize_udc_reg)—Offset 28h....................................... 497


(ep1_in_mpkt_sz_reg)—Offset 2Ch ........................................... 498
(ep1_in_desptr_udc_reg)—Offset 34h ........................................ 498
16.6.1.13 IN Endpoint 2 Control Register (ep2_in_ctrl_udc_reg)—Offset 40h499
16.6.1.14 IN Endpoint 2 Status Register (ep2_in_sts_udc_reg)—Offset 44h . 500
(ep2_in_bufsize_udc_reg)—Offset 48h ....................................... 502
16.6.1.19 IN Endpoint 3 Control Register (ep3_in_ctrl_udc_reg)—Offset 60h504
16.6.1.20 IN Endpoint 3 Status Register (ep3_in_sts_udc_reg)—Offset 64h . 505
(ep3_in_bufsize_udc_reg)—Offset 68h ....................................... 507
16.6.1.25 OUT Endpoint 0 Control Register (ep0_out_ctrl_udc_reg)—Offset
200h ..................................................................................... 509
16.6.1.26 OUT Endpoint 0 Status Register (ep0_out_sts_udc_reg)—Offset
204h ..................................................................................... 510
16.6.1.27 OUT Endpoint 0 Receive Packet Frame Number Register
(ep0_out_rpf_udc_reg)—Offset 208h ......................................... 512
16.6.1.28 OUT Endpoint 0 Buffer Size Register (ep0_out_bufsize_udc_reg)—
Offset 20Ch ............................................................................ 513
16.6.1.29 OUT Endpoint 0 SETUP Buffer Pointer Register
(ep0_subptr_udc_reg)—Offset 210h .......................................... 513
16.6.1.30 OUT Endpoint 0 Data Descriptor Pointer Register
(ep0_out_desptr_udc_reg)—Offset 214h .................................... 514
16.6.1.31 OUT Endpoint 0 Read Confirmation Register for zero-length OUT
data (for Slave-Only mode) (ep0_rd_cfrm_udc_reg)—Offset 21Ch . 514
220h ..................................................................................... 515
224h ..................................................................................... 516
Offset 22Ch ............................................................................ 519
(ep1_out_desptr_udc_reg)—Offset 234h .................................... 520
data (for Slave-Only mode) (ep1_rd_cfrm_udc_reg)—Offset 23Ch . 520
240h ..................................................................................... 521
244h ..................................................................................... 522


Offset 24Ch ........................................................................... 525
(ep2_out_desptr_udc_reg)—Offset 254h.................................... 526
data (for Slave-Only mode) (ep2_rd_cfrm_udc_reg)—Offset 25Ch. 526
260h ..................................................................................... 527
264h ..................................................................................... 528
(ep3_out_rpf_udc_reg)—Offset 268h......................................... 530
Offset 26Ch ........................................................................... 531
(ep3_out_desptr_udc_reg)—Offset 274h.................................... 532
data (for Slave-Only mode) (ep3_rd_cfrm_udc_reg)—Offset 27Ch. 532
16.6.1.53 Device Configuration Register (d_cfg_udc_reg)—Offset 400h ...... 533
16.6.1.54 Device Control Register (d_ctrl_udc_reg)—Offset 404h .............. 534
16.6.1.55 Device Status Register (d_sts_udc_reg)—Offset 408h................ 536
16.6.1.56 Device Interrupt Register (d_intr_udc_reg)—Offset 40Ch ........... 537
16.6.1.57 Device Interrupt Mask Register (d_intr_msk_udc_reg)—Offset
410h ..................................................................................... 538
16.6.1.58 Endpoints Interrupt Register (ep_intr_udc_reg)—Offset 414h ..... 539
16.6.1.59 Endpoints Interrupt Mask Register (ep_intr_msk_udc_reg)—Offset
418h ..................................................................................... 539
16.6.1.60 Test Mode Register (test_mode_udc_reg)—Offset 41Ch ............. 540
16.6.1.61 Product Release Number Register
(revision_udc_reg)—Offset 420h............................................. 541
16.6.1.62 SETUP command address pointer register
(udc_desc_addr_udc_reg)—Offset 500h..................................... 541
16.6.1.63 Physical Endpoint 0 Register
(udc_ep_ne_udc_reg_0)—Offset 504h ..................................... 542
16.6.1.65 Physical Endpoint 2 Register (udc_ep_ne_udc_reg_2)—Offset
50Ch..................................................................................... 543
16.6.1.69 Physical Endpoint 6 Register (udc_ep_ne_udc_reg_6)—Offset
51Ch..................................................................................... 546
16.6.1.70 RxFIFO Array[0-511] (udc_rx_fifo_reg_array[0-511])—Offset
800h, Count 512, Stride 4h ...................................................... 547
16.6.1.71 TxFIFO 0 Array[0-255] (udc_tx_fifo_reg_0_array[0-255])—Offset
1000h, Count 256, Stride 4h .................................................... 548
1400h, Count 256, Stride 4h .................................................... 548
1800h, Count 256, Stride 4h .................................................... 548
1C00h, Count 256, Stride 4h .................................................... 549

16.6.2 USB EHCI ............................................................................................ 549

16.6.2.1 Host Controller Interface Version Number and Capability Registers
Length (HCCAPBASE)—Offset 0h ............................................... 550
16.6.2.2 Host Controller Structural Parameters (HCSPARAMS)—Offset 4h .... 550
16.6.2.3 Host Controller Capability Parameters (HCCPARAMS)—Offset 8h .... 552
16.6.2.4 USB Command (USBCMD)—Offset 10h....................................... 553
16.6.2.5 USB Status (USBSTS)—Offset 14h ............................................. 555
16.6.2.6 USB Interrupt Enable (USBINTR)—Offset 18h.............................. 557
16.6.2.7 USB Frame Index (FRINDEX)—Offset 1Ch................................... 558
16.6.2.8 4 Gigabyte Memory Segment Selector (CTRLDSSEGMENT)—Offset
20h ....................................................................................... 559
16.6.2.9 Periodic Frame List Base Address (PERIODICLISTBASE)—Offset
24h ....................................................................................... 559
16.6.2.10 Asynchronous List Address (ASYNCLISTADDR)—Offset 28h ......... 560
16.6.2.11 Configure Flag (CONFIGFLAG)—Offset 50h................................ 560
16.6.2.12 Port Status/Control[0-1] (PORTSC[0-1])—Offset 54h, Count 2,
Stride 4h................................................................................ 561
16.6.2.13 Programmable Microframe Base Value (INSNREG00)—Offset 90h. 564
16.6.2.14 Programmable Packet Buffer OUT/IN Thresholds
(INSNREG01)—Offset 94h ........................................................ 565
16.6.2.15 Programmable Packet Buffer Depth (INSNREG02)—Offset 98h..... 565
16.6.2.16 Programmable Controller Settings (INSNREG03)—Offset 9Ch ...... 566
16.6.2.17 Programmable Controller Settings (INSNREG04)—Offset A0h ...... 567
16.6.2.18 UTMI Configuration (INSNREG05)—Offset A4h........................... 568
16.6.3 USB OHCI............................................................................................ 569
16.6.3.1 OHCI Revision (HCREVISION)—Offset 0h.................................... 570
16.6.3.2 Host Controller Control (HCCONTROL)—Offset 4h ........................ 570
16.6.3.3 Host Controller Command Status (HCCMDSTATUS)—Offset 8h ...... 571
16.6.3.4 Host Controller Interrupt Status (HCINTRSTATUS)—Offset Ch ....... 573
16.6.3.5 Host Controller Interrupt Enable (HCINTRENABLE)—Offset 10h...... 574
16.6.3.6 Host Controller Interrupt Disable (HCINTRDISABLE)—Offset 14h ... 575
16.6.3.7 Host Controller Communication Area (HCHCCA)—Offset 18h ......... 576
16.6.3.8 Host Controller Current Isochronous or Interrupt Endpoint
(HCPRDCURED)—Offset 1Ch ..................................................... 577
16.6.3.9 Host Controller Current First Control Endpoint (HCCTRLHEADED)—
Offset 20h .............................................................................. 577
16.6.3.10 Host Controller Current Control Endpoint (HCCTRLCURED)—Offset
24h ....................................................................................... 578
16.6.3.11 Host Controller First Bulk Endpoint (HCBULKHEADED)—Offset 28h578
16.6.3.12 Host Controller Current Bulk Endpoint (HCBULKCURED)—Offset
2Ch ....................................................................................... 579
16.6.3.13 Host Controller Last Completed Descriptor (HCDONEHEAD)—Offset
30h ....................................................................................... 580
16.6.3.14 Host Controller Frame Interval (HCFMINTERVAL)—Offset 34h ...... 580
16.6.3.15 Host Controller Remaining Frame (HCFMREMAINING)—Offset 38h 581
16.6.3.16 Host Controller Frame Number (HCFMNUMBER)—Offset 3Ch........ 582
16.6.3.17 Host Controller Periodic List Start (HCPERIODICSTART)—Offset
40h ....................................................................................... 583
16.6.3.18 Host Controller LS Threshold (HCLSTHRESHOLD)—Offset 44h ...... 583
16.6.3.19 Host Controller Root Hub Descriptor A (HCRHDESPA)—Offset 48h 584
16.6.3.20 Host Controller Root Hub Descriptor B (HCRHDESPB)—Offset 4Ch 585
16.6.3.21 Host Controller Root Hub Status (HCRHSTATUS)—Offset 50h....... 586
16.6.3.22 Host Controller Root Hub Port Status (HCRHPORTSTS)—Offset 54h ...
587
17.0 SDIO/SD/eMMC ..................................................................................................... 591
17.2 Features ......................................................................................................... 592
17.2.1 SDIO/SD/eMMC Features ....................................................................... 592
17.2.2 SD 3.0/ SDIO 3.0 / eMMC 4.41 Interfaces ................................................ 592

17.2.2.1 SD 3.0 Bus Topology ............................................................... 592

17.2.2.2 SDIO 3.0 Interface.................................................................. 593
17.2.2.3 eMMC Interface ...................................................................... 594
17.2.3 SDIO/SD/eMMC Host Controller.............................................................. 594
17.2.3.1 SD DMA................................................................................. 595
17.3 References ..................................................................................................... 595
17.4 Register Map .................................................................................................. 595
17.5.1 Vendor ID (VENDOR_ID)—Offset 0h ....................................................... 597
17.5.2 Device ID (DEVICE_ID)—Offset 2h ......................................................... 598
17.5.3 Command Register (COMMAND_REGISTER)—Offset 4h ............................. 598
17.5.4 Status Register (STATUS)—Offset 6h ...................................................... 599
17.5.5 Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h ................. 599
17.5.6 Cache Line Size (CACHE_LINE_SIZE)—Offset Ch ...................................... 600
17.5.7 Latency Timer (LATENCY_TIMER)—Offset Dh ........................................... 600
17.6.1 SDMA System Address Register (SYS_ADR)—Offset 0h.............................. 613
17.6.2 Block Size Register (BLK_SIZE)—Offset 4h .............................................. 614
17.6.3 Block Count Register (BLK_COUNT)—Offset 6h......................................... 615
17.6.4 Argument Register (ARGUMENT)—Offset 8h............................................. 616
17.6.5 Transfer Mode Register (TX_MODE)—Offset Ch......................................... 616
17.6.6 Command Register (CMD)—Offset Eh...................................................... 618
17.6.7 Response Register 0 (RESPONSE0)—Offset 10h........................................ 619
17.6.10 Response Register 6 (RESPONSE6)—Offset 1Ch ...................................... 621
17.6.11 Buffer Data Port Register (BUF_DATA_PORT)—Offset 20h......................... 621
17.6.12 Present State Register (PRE_STATE)—Offset 24h .................................... 622
17.6.13 Host Control Register (HOST_CTL)—Offset 28h ....................................... 627
17.6.14 Power Control Register (PWR_CTL)—Offset 29h ...................................... 628

17.6.15 Block Gap Control Register (BLK_GAP_CTL)—Offset 2Ah ........................... 628

17.6.16 Clock Control Register (CLK_CTL)—Offset 2Ch......................................... 630
17.6.17 Timeout Control Register (TIMEOUT_CTL)—Offset 2Eh.............................. 632
17.6.18 Software Reset Register (SW_RST)—Offset 2Fh ....................................... 633
17.6.19 Normal Interrupt Status Register (NML_INT_STATUS)—Offset 30h ............. 634
17.6.20 Error Interrupt Status Register (ERR_INT_STATUS)—Offset 32h ................ 636
17.6.21 Normal Interrupt Status Enable (NRM_INT_STATUS_EN)—Offset 34h ......... 638
17.6.22 Error Interrupt Status Enable Register (ERR_INT_STAT_EN)—Offset 36h .... 639
17.6.23 Normal Interrupt Signal Enable Register (NRM_INT_SIG_EN)—Offset 38h ... 640
17.6.24 Error Interrupt Signal Enable Register (ERR_INT_SIG_EN)—Offset 3Ah....... 642
17.6.25 Auto CMD12 Error Status Register (CMD12_ERR_STAT)—Offset 3Ch .......... 643
17.6.26 Host Control 2 Register (HOST_CTRL_2)—Offset 3Eh ............................... 644
17.6.27 Capabilities Register (CAPABILITIES)—Offset 40h .................................... 645
17.6.28 Capabilities Register 2 (CAPABILITIES_2)—Offset 44h .............................. 647
17.6.29 Maximum Current Capabilities Register (MAX_CUR_CAP)—Offset 48h ......... 648
17.6.30 Force Event Register for Auto CMD12 Error Status
(FORCE_EVENT_CMD12_ERR_STAT)—Offset 50h ...................................... 649
17.6.31 Force Event Register for Error Interrupt Status
(FORCE_EVENT_ERR_INT_STAT)—Offset 52h ........................................... 650
17.6.32 ADMA Error Status Register (ADMA_ERR_STAT)—Offset 54h ..................... 651
17.6.33 ADMA System Address Register (ADMA_SYS_ADDR)—Offset 58h ............... 652
17.6.34 initialization Preset Values Register (3.3v or 1.8v)
(PRESET_VALUE_0)—Offset 60h ............................................................. 653
17.6.35 Default Speed Preset Values Register (PRESET_VALUE_1)—Offset 62h ....... 653
17.6.36 High Speed Preset Values Register (PRESET_VALUE_2)—Offset 64h ........... 654
17.6.37 SDR12 Preset Values Register (PRESET_VALUE_3)—Offset 66h.................. 654
17.6.38 SDR25 Preset Values Register (PRESET_VALUE_4)—Offset 68h.................. 655
17.6.39 SDR50 Preset Values Register (PRESET_VALUE_5)—Offset 6Ah ................. 656
17.6.40 SDR104 Preset Values Register (PRESET_VALUE_6)—Offset 6Ch................ 656
17.6.41 DDR50 Preset Values Register (PRESET_VALUE_7)—Offset 6Eh ................. 657
17.6.42 Boot Time-out control register (BOOT_TIMEOUT_CTRL)—Offset 70h........... 658
17.6.43 Debug Selection Register (DEBUG_SEL)—Offset 74h ................................ 658
17.6.44 Shared Bus Control Register (SHARED_BUS)—Offset E0h.......................... 659
17.6.45 SPI Interrupt Support Register (SPI_INT_SUP)—Offset F0h ....................... 660
17.6.46 Slot Interrupt Status Register (SLOT_INT_STAT)—Offset FCh .................... 661
17.6.47 Host Controller Version Register (HOST_CTRL_VER)—Offset FEh................ 661
18.0 High Speed UART ................................................................................................... 663
18.2 Features ......................................................................................................... 664
18.2.1 UART Function...................................................................................... 664
18.2.2 Baud Rate Generator............................................................................. 664
18.3 Usage ............................................................................................................ 665
18.3.1 DMA Mode Operation............................................................................. 665
18.3.1.1 Receiver DMA ......................................................................... 665
18.3.1.2 Transmitter DMA ..................................................................... 666
18.3.2 FIFO Interrupt-Mode Operation............................................................... 666
18.3.2.1 Receiver Interrupt ................................................................... 666
18.3.2.2 Transmitter Interrupt............................................................... 666
18.3.3 FIFO Polled-Mode Operation ................................................................... 667
18.3.3.1 Receive Data Service ............................................................... 667
18.3.3.2 Transmit Data Service.............................................................. 667
18.3.4 Autoflow Control ................................................................................... 667
18.3.4.1 RTS (UART Output) ................................................................. 667
18.3.4.2 CTS (UART Input) ................................................................... 668
18.4 Register Map ................................................................................................... 668


18.5.10 Base Address Register (BAR0)—Offset 10h............................................. 674
18.5.12 Cardbus CIS Pointer (CARDBUS_CIS_POINTER)—Offset 28h .................... 675
18.5.14 Subsystem ID (SUB_SYS_ID)—Offset 2Eh ............................................. 676
18.5.15 Expansion ROM Base Address (EXP_ROM_BASE_ADR)—Offset 30h............ 676
18.5.22 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h.......................... 679
18.5.24 Power Management Control/Status Register (PMCSR)—Offset 84h............. 680
18.5.25 PM CSR PCI-to-PCI Bridge Support Extension (PMCSR_BSE)—Offset 86h ... 681
18.5.26 Power Management Data Register (DATA_REGISTER)—Offset 87h ............ 681
18.5.27 Capability ID (MSI_CAP_ID)—Offset A0h ............................................... 681
18.5.29 Message Control (MESSAGE_CTRL)—Offset A2h...................................... 682
18.5.30 Message Address (MESSAGE_ADDR)—Offset A4h.................................... 682
18.5.33 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h................................. 684
18.6.1 UART Registers .................................................................................... 684
18.6.1.1 Receive Buffer / Transmit Holding / Divisor Latch Low
(RBR_THR_DLL)—Offset 0h ...................................................... 685
18.6.1.2 Interrupt Enable / Divisor Latch High (IER_DLH)—Offset 4h.......... 685
18.6.1.3 Interrupt Identification / FIFO Control (IIR_FCR)—Offset 8h ......... 686
18.6.1.4 Line Control (LCR)—Offset Ch ................................................... 688
18.6.1.5 MODEM Control (MCR)—Offset 10h ........................................... 688
18.6.1.6 Line Status (LSR)—Offset 14h .................................................. 689
18.6.1.7 MODEM Status (MSR)—Offset 18h............................................. 691
18.6.1.8 Scratchpad (SCR)—Offset 1Ch .................................................. 692
18.6.1.9 UART Status (USR)—Offset 7Ch ................................................ 693
18.6.1.10 Halt Transmission (HTX)—Offset A4h....................................... 693
18.6.1.11 DMA Software Acknowledge (DMASA)—Offset A8h .................... 694
18.6.2 DMA Controller Registers....................................................................... 694
18.6.2.1 Channel 0 Source Address (SAR0)—Offset 0h ............................. 696
18.6.2.2 Channel 0 Destination Address (DAR0)—Offset 8h ....................... 696
18.6.2.3 Channel 0 Linked List Pointer (LLP0)—Offset 10h ........................ 697
18.6.2.4 Channel 0 Control LOWER (CTL0_L)—Offset 18h ......................... 697
18.6.2.5 Channel 0 Control UPPER (CTL0_U)—Offset 1Ch.......................... 699
18.6.2.6 Channel 0 Source Status (SSTAT0)—Offset 20h .......................... 700
18.6.2.7 Channel 0 Destination Status (DSTAT0)—Offset 28h .................... 700

18.6.2.8 Channel 0 Source Status Address (SSTATAR0)—Offset 30h ........... 701

18.6.2.9 Channel 0 Destination Status Address (DSTATAR0)—Offset 38h..... 701
18.6.2.10 Channel 0 Configuration LOWER (CFG0_L)—Offset 40h .............. 702
18.6.2.11 Channel 0 configuration UPPER (CFG0_U)—Offset 44h ............... 703
18.6.2.12 Channel 0 Source Gather (SGR0)—Offset 48h ........................... 704
18.6.2.13 Channel 0 Destination Scatter (DSR0)—Offset 50h ..................... 705
18.6.2.14 Channel 1 Source Address (SAR1)—Offset 58h .......................... 705
18.6.2.15 Channel 1 Destination Address (DAR1)—Offset 60h .................... 706
18.6.2.16 Channel 1 Linked List Pointer (LLP1)—Offset 68h ....................... 706
18.6.2.17 Channel 1 Control LOWER (CTL1_L)—Offset 70h ........................ 707
18.6.2.18 Channel 1 Control UPPER (CTL1_U)—Offset 74h......................... 709
18.6.2.19 Channel 1 Source Status (SSTAT1)—Offset 78h ......................... 710
18.6.2.20 Channel 1 Destination Status (DSTAT1)—Offset 80h................... 710
18.6.2.21 Channel 1 Source Status Address (SSTATAR1)—Offset 88h ......... 711
18.6.2.22 Channel 1 Destination Status Address (DSTATAR1)—Offset 90h ... 711
18.6.2.23 Channel 1 Configuration LOWER (CFG1_L)—Offset 98h ............... 712
18.6.2.24 Channel 1 configuration UPPER (CFG1_U)—Offset 9Ch ................ 713
18.6.2.25 Channel 1 Source Gather (SGR1)—Offset A0h ........................... 714
18.6.2.26 Channel 1 Destination Scatter (DSR1)—Offset A8h ..................... 714
18.6.2.27 Raw Status for IntTfr Interrupt (RAW_TFR)—Offset 2C0h ............ 715
18.6.2.28 Raw Status for IntBlock Interrupt (RAW_BLOCK)—Offset 2C8h .... 715
18.6.2.29 Raw Status for IntSrcTran Interrupt (RAW_SRC_TRAN)—Offset
2D0h ..................................................................................... 716
18.6.2.30 Raw Status for IntDstTran Interrupt (RAW_DST_TRAN)—Offset
2D8h ..................................................................................... 716
18.6.2.31 Raw Status for IntErr Interrupt (RAW_ERR)—Offset 2E0h............ 717
18.6.2.32 Status for IntTfr Interrupt (STATUS_TFR)—Offset 2E8h .............. 717
18.6.2.33 Status for IntBlock Interrupt (STATUS_BLOCK)—Offset 2F0h ....... 718
18.6.2.34 Status for IntSrcTran Interrupt
(STATUS_SRC_TRAN)—Offset 2F8h.......................................... 718
18.6.2.35 Status for IntDstTran Interrupt
(STATUS_DST_TRAN)—Offset 300h.......................................... 719
18.6.2.36 Status for IntErr Interrupt (STATUS_ERR)—Offset 308h .............. 719
18.6.2.37 Mask for IntTfr Interrupt (MASK_TFR)—Offset 310h ................... 720
18.6.2.38 Mask for IntBlock Interrupt (MASK_BLOCK)—Offset 318h............ 720
18.6.2.39 Mask for IntSrcTran Interrupt (MASK_SRC_TRAN)—Offset 320h... 721
18.6.2.40 Mask for IntDstTran Interrupt (MASK_DST_TRAN)—Offset 328h... 722
18.6.2.41 Mask for IntErr Interrupt (MASK_ERR)—Offset 330h................... 722
18.6.2.42 Clear for IntTfr Interrupt (CLEAR_TFR)—Offset 338h .................. 723
18.6.2.43 Clear for IntBlock Interrupt (CLEAR_BLOCK)—Offset 340h........... 723
18.6.2.44 Clear for IntSrcTran Interrupt (CLEAR_SRC_TRAN)—Offset 348h.. 724
18.6.2.45 Clear for IntDstTran Interrupt (CLEAR_DST_TRAN)—Offset 350h . 724
18.6.2.46 Clear for IntErr Interrupt (CLEAR_ERR)—Offset 358h.................. 725
18.6.2.47 Combined Interrupt Status (STATUS_INT)—Offset 360h.............. 725
18.6.2.48 Source Software Transaction Request (REQ_SRC_REG)—Offset
368h ..................................................................................... 726
18.6.2.49 Destination Software Transaction Request register
(REQ_DST_REG)—Offset 370h .................................................. 726
18.6.2.50 Source Single Transaction Request (SGL_REQ_SRC_REG)—Offset
378h ..................................................................................... 727
18.6.2.51 Destination Single Software Transaction Request
(SGL_REQ_DST_REG)—Offset 380h ........................................... 728
18.6.2.52 Source Last Transaction Request (LST_SRC_REG)—Offset 388h ... 728
18.6.2.53 Destination Single Transaction Request (LST_DST_REG)—Offset
390h ..................................................................................... 729
18.6.2.54 DMA Configuration (DMA_CFG_REG)—Offset 398h ..................... 729
18.6.2.55 Channel Enable (CH_EN_REG)—Offset 3A0h .............................. 730
19.0 I2C* Controller/GPIO Controller ............................................................................ 731
19.1 I2C Controller.................................................................................................. 731

19.1.1 Signal Descriptions ............................................................................... 731

19.1.2 Features ............................................................................................. 731
19.1.2.1 I2C* Protocol.......................................................................... 731
19.1.2.2 I2C* Modes of Operation .......................................................... 732
19.1.2.3 Functional Description ............................................................. 732
19.1.3 Use .................................................................................................... 737
19.1.3.1 Master Mode Operation............................................................ 737
19.1.3.2 Disabling I2C* Controller.......................................................... 737
19.1.4 References .......................................................................................... 738
19.2 GPIO Controller............................................................................................... 738
19.2.2 Features ............................................................................................. 738
19.3 Register Map .................................................................................................. 738
19.4.12 Cardbus CIS Pointer (CARDBUS_CIS_POINTER)—Offset 28h .................... 746
19.4.14 Subsystem ID (SUB_SYS_ID)—Offset 2Eh ............................................. 747
19.4.15 Expansion ROM Base Address (EXP_ROM_BASE_ADR)—Offset 30h............ 747
19.4.22 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h.......................... 750
19.4.24 Power Management Control/Status Register (PMCSR)—Offset 84h............. 751
19.4.25 PM CSR PCI-to-PCI Bridge Support Extension (PMCSR_BSE)—Offset 86h ... 752
19.4.26 Power Management Data Register (DATA_REGISTER)—Offset 87h ............ 752
19.4.27 Capability ID (MSI_CAP_ID)—Offset A0h ............................................... 752
19.4.29 Message Control (MESSAGE_CTRL)—Offset A2h...................................... 753
19.4.30 Message Address (MESSAGE_ADDR)—Offset A4h.................................... 754
19.4.33 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h................................. 755
19.5.1 I2C* Controller Memory Mapped Registers............................................... 755
19.5.1.1 Control Register (IC_CON)—Offset 0h........................................ 756
19.5.1.2 Master Target Address (IC_TAR)—Offset 4h................................ 757
19.5.1.3 Data Buffer and Command (IC_DATA_CMD)—Offset 10h .............. 758
19.5.1.4 Standard Speed Clock SCL High Count (IC_SS_SCL_HCNT)—Offset
14h....................................................................................... 759

19.5.1.5 Standard Speed Clock SCL Low Count (IC_SS_SCL_LCNT)—Offset

18h ....................................................................................... 760
19.5.1.6 Fast Speed Clock SCL High Count (IC_FS_SCL_HCNT)—Offset 1Ch 760
19.5.1.7 Fast Speed Clock SCL Low Count (IC_FS_SCL_LCNT)—Offset 20h .. 761
19.5.1.8 Interrupt Status (IC_INTR_STAT)—Offset 2Ch............................. 761
19.5.1.9 Interrupt Mask (IC_INTR_MASK)—Offset 30h .............................. 763
19.5.1.10 Raw Interrupt Status (IC_RAW_INTR_STAT)—Offset 34h ........... 764
19.5.1.11 Receive FIFO Threshold Level (IC_RX_TL)—Offset 38h ............... 766
19.5.1.12 Transmit FIFO Threshold Level (IC_TX_TL)—Offset 3Ch.............. 766
19.5.1.13 Clear Combined and Individual Interrupt (IC_CLR_INTR)—Offset
40h ....................................................................................... 767
19.5.1.14 Clear RX_UNDER Interrupt (IC_CLR_RX_UNDER)—Offset 44h ..... 767
19.5.1.15 Clear RX_OVER Interrupt (IC_CLR_RX_OVER)—Offset 48h ......... 768
19.5.1.16 Clear TX_OVER Interrupt (IC_CLR_TX_OVER)—Offset 4Ch.......... 768
19.5.1.17 Clear RD_REQ Interrupt (IC_CLR_RD_REQ)—Offset 50h............. 769
19.5.1.18 Clear TX_ABRT Interrupt (IC_CLR_TX_ABRT)—Offset 54h........... 769
19.5.1.19 Clear ACTIVITY Interrupt (IC_CLR_ACTIVITY)—Offset 5Ch.......... 770
19.5.1.20 Clear STOP_DET Interrupt (IC_CLR_STOP_DET)—Offset 60h....... 770
19.5.1.21 Clear START_DET Interrupt (IC_CLR_START_DET)—Offset 64h ... 771
19.5.1.22 Enable (IC_ENABLE)—Offset 6Ch ............................................ 771
19.5.1.23 Status (IC_STATUS)—Offset 70h............................................. 772
19.5.1.24 Transmit FIFO Level (IC_TXFLR)—Offset 74h ............................ 773
19.5.1.25 Receive FIFO Level (IC_RXFLR)—Offset 78h.............................. 774
19.5.1.26 SDA Hold (IC_SDA_HOLD)—Offset 7Ch .................................... 774
19.5.1.27 Transmit Abort Source (IC_TX_ABRT_SOURCE)—Offset 80h ....... 775
19.5.1.28 Enable Status (IC_ENABLE_STATUS)—Offset 9Ch...................... 776
19.5.1.29 SS and FS Spike Suppression Limit (IC_FS_SPKLEN)—Offset A0h 777
19.5.2 GPIO Controller Memory Mapped Registers............................................... 777
19.5.2.1 Port A Data (GPIO_SWPORTA_DR)—Offset 0h ............................. 778
19.5.2.2 Port A Data Direction (GPIO_SWPORTA_DDR)—Offset 4h.............. 778
19.5.2.3 Interrupt Enable (GPIO_INTEN)—Offset 30h ............................... 779
19.5.2.4 Interrupt Mask (GPIO_INTMASK)—Offset 34h ............................. 779
19.5.2.5 Interrupt Type (GPIO_INTTYPE_LEVEL)—Offset 38h ..................... 780
19.5.2.6 Interrupt Polarity (GPIO_INT_POLARITY)—Offset 3Ch .................. 781
19.5.2.7 Interrupt Status (GPIO_INTSTATUS)—Offset 40h......................... 781
19.5.2.8 Raw Interrupt Status (GPIO_RAW_INTSTATUS)—Offset 44h.......... 782
19.5.2.9 Debounce Enable (GPIO_DEBOUNCE)—Offset 48h ....................... 782
19.5.2.10 Clear Interrupt (GPIO_PORTA_EOI)—Offset 4Ch ........................ 783
19.5.2.11 Port A External Port (GPIO_EXT_PORTA)—Offset 50h.................. 784
19.5.2.12 Synchronization Level (GPIO_LS_SYNC)—Offset 60h .................. 784
20.0 SPI Interface ........................................................................................................ 787
20.2 Features ......................................................................................................... 787
20.2.1 SPI Controller....................................................................................... 787
20.2.1.1 Processor-Initiated Data Transfer .............................................. 788
20.2.1.2 Data Format ........................................................................... 788
20.2.1.3 FIFO Operation ....................................................................... 789
20.2.1.4 Baud Rate Generation .............................................................. 789
20.3 Register Map ................................................................................................... 791
20.4 PCI Configuration Registers ............................................................................... 792
20.4.1 Vendor ID (VENDOR_ID)—Offset 0h ........................................................ 792
20.4.2 Device ID (DEVICE_ID)—Offset 2h .......................................................... 793
20.4.3 Command Register (COMMAND_REGISTER)—Offset 4h .............................. 793
20.4.4 Status Register (STATUS)—Offset 6h....................................................... 794
20.4.5 Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h .................. 795
20.4.6 Cache Line Size (CACHE_LINE_SIZE)—Offset Ch ....................................... 795
20.4.7 Latency Timer (LATENCY_TIMER)—Offset Dh ............................................ 796


20.5.1 SPI Control Register 0 (SSCR0)—Offset 0h .............................................. 807
20.5.2 SPI Control Register 1 (SSCR1)—Offset 4h .............................................. 808
20.5.3 SPI Status Register (SSSR)—Offset 8h .................................................... 810
20.5.4 SPI Data Register (SSDR)—Offset 10h .................................................... 811
20.5.5 DDS Clock Rate Register (DDS_RATE)—Offset 28h.................................... 812
21.0 Legacy Bridge........................................................................................................ 815
21.1 Features ........................................................................................................ 815
21.2 Register Map .................................................................................................. 816
21.3.1 PCI Device ID and Vendor ID Fields (PCI_DEVICE_VENDOR)—Offset 0h....... 818
21.3.2 PCI Status and Command Fields (PCI_STATUS_COMMAND)—Offset 4h ........ 818
21.3.3 PCI Class Code and Revision ID Fields (PCI_CLASS_REVISION)—Offset 8h ... 819
21.3.4 PCI Miscellaneous Fields (PCI_MISC)—Offset Ch ....................................... 819
(PCI_SUBSYSTEM)—Offset 2Ch .............................................................. 820
21.3.6 GPIO Base Address (GBA)—Offset 44h .................................................... 821
21.3.7 PM1_BLK Base Address (PM1BLK)—Offset 48h ......................................... 821
21.3.8 GPE0_BLK Base Address (GPE0BLK)—Offset 4Ch ...................................... 821
21.3.9 ACPI Control (ACTL)—Offset 58h ............................................................ 822
21.3.10 PIRQA, PIRQB, PIRQC and PIRQD Routing Control (PABCDRC)—Offset 60h. 822
21.3.11 PIRQE, PIRQF, PIRQG and PIRQH Routing Control (PEFGHRC)—Offset 64h . 824
21.3.12 Watch Dog Timer Base Address (WDTBA)—Offset 84h ............................. 824
21.3.13 BIOS Decode Enable (BCE)—Offset D4h ................................................ 825
21.3.14 BIOS Control (BC)—Offset D8h............................................................. 826
21.3.15 Root Complex Base Address (RCBA)—Offset F0h..................................... 827
21.4.1 Root Complex Register Block ................................................................. 827
21.4.1.1 Root Complex Topology Capabilities List (RCTCL)—Offset 0h......... 828
21.4.1.2 Element Self Description (ESD)—Offset 4h ................................. 828

21.4.1.3 Interrupt Queue Agent 0 (IRQAGENT0)—Offset 3140h.................. 829

21.4.1.7 RTC Configuration (RC)—Offset 3400h ....................................... 831
21.5 IO Registers.................................................................................................... 832
21.5.1 Fixed IO Registers ................................................................................ 832
21.5.1.1 NMI Status and Control Register (NSC)—Offset 61h ..................... 832
21.5.1.2 NMI Enable and RTC Index Register (NMIE)—Offset 70h ............... 833
21.5.1.3 Software SMI Control Port (SWSMICTL)—Offset B2h .................... 833
21.5.1.4 Software SMI Status Port (SWSMISTS)—Offset B3h ..................... 834
21.5.1.5 Reset Control Register (RSTC)—Offset CF9h................................ 834
21.5.2 ACPI GPE0 Block................................................................................... 835
21.5.2.1 GPE0 Status Register (GPE0STS)—Offset 0h ............................... 835
21.5.2.2 GPE0 Enable Register (GPE0EN)—Offset 4h................................. 836
21.5.2.3 SMI Enable Register (SMIEN)—Offset 10h................................... 837
21.5.2.4 SMI Status Register (SMISTS)—Offset 14h ................................. 838
21.5.2.5 General Purpose Event Control Register (GPEC)—Offset 18h ......... 839
21.5.2.6 Power Management Configuration Core Well Register
(PMCW)—Offset 28h ................................................................ 839
21.5.2.7 Power Management Configuration Suspend Well Register (PMSW)—
Offset 2Ch.............................................................................. 840
21.5.2.8 Power Management Configuration RTC Well Register
(PMRW)—Offset 30h ................................................................ 841
21.5.3 ACPI PM1 Block .................................................................................... 841
21.5.3.1 PM1 Status Register (PM1S)—Offset 0h ...................................... 841
21.5.3.2 PM1 Enable Register (PM1E)—Offset 2h...................................... 842
21.5.3.3 PM1 Control Register (PM1C)—Offset 4h ..................................... 843
21.5.3.4 Power Management 1 Timer Register (PM1T)—Offset 8h ............... 844
21.6 Legacy GPIO ................................................................................................... 845
21.6.2 Features .............................................................................................. 845
21.6.3 Use..................................................................................................... 845
21.6.4 Register Map ........................................................................................ 846
21.6.5 IO Mapped Registers ............................................................................. 846
21.6.5.1 Core Well GPIO Enable (CGEN)—Offset 0h .................................. 847
21.6.5.2 Core Well GPIO Input/Output Select (CGIO)—Offset 4h ................ 847
21.6.5.3 Core Well GPIO Level for Input or Output (CGLVL)—Offset 8h........ 848
21.6.5.4 Core Well GPIO Trigger Positive Edge Enable (CGTPE)—Offset Ch .. 848
21.6.5.5 Core Well GPIO Trigger Negative Edge Enable (CGTNE)—Offset
10h ....................................................................................... 849
21.6.5.6 Core Well GPIO GPE Enable (CGGPE)—Offset 14h ........................ 849
21.6.5.7 Core Well GPIO SMI Enable (CGSMI)—Offset 18h ........................ 850
21.6.5.8 Core Well GPIO Trigger Status (CGTS)—Offset 1Ch ...................... 850
21.6.5.9 Resume Well GPIO Enable (RGEN)—Offset 20h............................ 851
21.6.5.10 Resume Well GPIO Input/Output Select (RGIO)—Offset 24h ........ 851
21.6.5.11 Resume Well GPIO Level for Input or Output (RGLVL)—Offset 28h 852
21.6.5.12 Resume Well GPIO Trigger Positive Edge Enable (RGTPE)—Offset
2Ch ....................................................................................... 852
21.6.5.13 Resume Well GPIO Trigger Negative Edge Enable (RGTNE)—Offset
30h ....................................................................................... 852
21.6.5.14 Resume Well GPIO GPE Enable (RGGPE)—Offset 34h .................. 853
21.6.5.15 Resume Well GPIO SMI Enable (RGSMI)—Offset 38h .................. 854
21.6.5.16 Resume Well GPIO Trigger Status (RGTS)—Offset 3Ch................ 854
21.6.5.17 Core Well GPIO NMI Enable (CGNMIEN)—Offset 40h................... 855
21.6.5.18 Resume Well GPIO NMI Enable (RGNMIEN)—Offset 44h .............. 855
21.7 Legacy SPI Controller ....................................................................................... 856

21.7.2 Features ............................................................................................. 856

21.7.3 Register Map ....................................................................................... 856
21.7.4 Legacy SPI Host Interface Registers........................................................ 857
21.7.4.1 SPI Status (SPISTS)—Offset 3020h ........................................... 859
21.7.4.2 SPI Control (SPICTL)—Offset 3022h .......................................... 859
21.7.4.3 SPI Address (SPIADDR)—Offset 3024h ...................................... 860
21.7.4.4 SPI Data 0 - Lower 32 Bits (SPID0_1)—Offset 3028h................... 861
21.7.4.5 SPI Data 0 - Upper 32 Bits (SPID0_2)—Offset 302Ch................... 861
21.7.4.7 SPI Data 1 - Upper 32 Bits (SPID1_2)—Offset 3034h ................... 862
21.7.4.9 SPI Data 2 - Upper 32 Bits (SPID2_2)—Offset 303Ch................... 863
21.7.4.10 SPI Data 3 - Lower 32 Bits (SPID3_1)—Offset 3040h ................. 863
21.7.4.11 SPI Data 3 - Upper 32 Bits (SPID3_2)—Offset 3044h ................. 863
21.7.4.13 SPI Data 4 - Upper 32 Bits (SPID4_2)—Offset 304Ch ................. 864
21.7.4.17 SPI Data 6 - Upper 32 Bits (SPID6_2)—Offset 305Ch ................. 866
21.7.4.20 BIOS Base Address (BBAR)—Offset 3070h ................................ 867
21.7.4.21 Prefix Opcode Configuration (PREOP)—Offset 3074h .................. 867
21.7.4.22 Opcode Type Configuration (OPTYPE)—Offset 3076h .................. 868
21.7.4.23 Opcode Menu Configuration - Lower 32 Bits (OPMENU_1)—Offset
3078h ................................................................................... 869
21.7.4.24 Opcode Menu Configuration - Upper 32 Bits (OPMENU_2)—Offset
307Ch ................................................................................... 869
21.7.4.25 Protected BIOS Range 0 (PBR0)—Offset 3080h ......................... 870
21.8 8254 Programmable Interval Timer.................................................................... 873
21.8.1 Features ............................................................................................. 873
21.8.1.1 Counter 0, System Timer ......................................................... 873
21.8.1.2 Counter 1, Refresh Request Signal ............................................ 873
21.8.1.3 Counter 2, Speaker Tone ......................................................... 873
21.8.2 Use .................................................................................................... 873
21.8.2.1 Timer Programming ................................................................ 873
21.8.2.2 Reading from the Interval Timer ............................................... 874
21.8.3 Register Map ....................................................................................... 875
21.8.4 Timer I/O Registers .............................................................................. 876
21.8.4.1 Counter 0 Interval Time Status Byte Format (C0TS)—Offset 40h ... 877
21.8.4.4 Timer Control Word Register (TCW)—Offset 43h ......................... 879
21.8.4.5 Counter 0 Counter Access Port Register (C0AP)—Offset 50h ......... 879
21.9 High Precision Event Timer (HPET)..................................................................... 880
21.9.1 Features ............................................................................................. 881
21.9.1.1 Non-Periodic Mode - All Timers ................................................. 881
21.9.1.2 Periodic Mode - Timer 0 Only.................................................... 881
21.9.1.3 Interrupts .............................................................................. 882
21.9.2 Register Map ....................................................................................... 882
21.9.3 Memory Mapped Registers..................................................................... 883
21.9.3.1 General Capabilities and ID Register - Lower 32 Bits
(GCID_1)—Offset 0h ............................................................... 884

21.9.3.2 General Capabilities and ID Register - Upper 32 Bits

(GCID_2)—Offset 4h................................................................ 885
21.9.3.3 General Configuration (GC)—Offset 10h ..................................... 885
21.9.3.4 General Interrupt Status Register (GIS)—Offset 20h .................... 885
21.9.3.5 Main Counter Value Register - Lower 32 Bits (MCV_1)—Offset F0h . 886
21.9.3.6 Main Counter Value Register - Upper 32 Bits (MCV_2)—Offset F4h . 886
21.9.3.7 Timer 0 Config and Capabilities Register - Lower 32 Bits
(T0C_1)—Offset 100h .............................................................. 887
21.9.3.8 Timer 0 Config and Capabilities Register - Upper 32 Bits
(T0C_2)—Offset 104h .............................................................. 888
21.9.3.9 Timer 0 Comparator Value Register - Lower 32 Bits
(T0CV_1)—Offset 108h ............................................................ 888
21.9.3.10 Timer 0 Comparator Value Register - Upper 32 Bits
(T0CV_2)—Offset 10Ch ............................................................ 888
21.9.3.11 Timer 1 Config and Capabilities Register - Lower 32 Bits (T1C_1)—
Offset 120h ............................................................................ 889
21.9.3.12 Timer 1 Config and Capabilities Register - Upper 32 Bits (T1C_2)—
Offset 124h ............................................................................ 890
21.9.3.13 Timer 1 Comparator Value Register (T1CV_1)—Offset 128h......... 890
Offset 140h ............................................................................ 890
Offset 144h ............................................................................ 891
21.9.3.16 Timer 2 Comparator Value Register (T2CV_1)—Offset 148h......... 892
21.9.4 References........................................................................................... 892
21.10 Real Time Clock (RTC)...................................................................................... 892
21.10.1 Signal Descriptions .............................................................................. 892
21.10.2 Features ............................................................................................ 893
21.10.2.1 Update Cycles ....................................................................... 893
21.10.2.2 Interrupts............................................................................. 893
21.10.2.3 Lockable RAM Ranges ............................................................ 893
21.10.3Register Map ........................................................................................ 894
21.10.4I/O Registers ....................................................................................... 894
21.10.5Indexed Registers ................................................................................. 895
21.10.5.1 Offset 0Ah: Register A............................................................ 896
21.10.5.2 Offset 0Bh: Register B - General Configuration .......................... 896
21.10.5.3 Offset 0Ch: Register C - Flag Register ...................................... 897
21.10.5.4 Offset 0Dh: Register D - Flag Register ...................................... 898
21.10.6References........................................................................................... 898
21.11 Interrupt Decoding & Routing ............................................................................ 898
21.11.1Features .............................................................................................. 898
21.11.1.1 Interrupt Decoder .................................................................. 898
21.11.1.2 Interrupt Router .................................................................... 899
21.12 8259 Programmable Interrupt Controllers (PIC) ................................................... 899
21.12.1Features .............................................................................................. 900
21.12.1.1 Interrupt Handling ................................................................. 900
21.12.1.2 Initialization Command Words (ICWx) ...................................... 902
21.12.1.3 Operation Command Words (OCW) .......................................... 903
21.12.1.4 Modes of Operation................................................................ 903
21.12.1.5 Masking Interrupts ................................................................ 905
21.12.1.6 Steering of PCI Interrupts....................................................... 905
21.12.2Register Map ........................................................................................ 905
21.12.3I/O Registers ....................................................................................... 906
21.12.3.1 Master Initialization Command Word 1 (MICW1)—Offset 20h ....... 908
21.12.3.3 Master Operational Control Word 2 (MOCW2)—Offset 24h ........... 909
21.12.3.5 Master Operational Control Word 3 (MOCW3)—Offset 28h ........... 910

21.12.3.6 Master Initialization Command Word 4 (MICW4)—Offset 29h ...... 911

21.12.3.7 Master Operational Control Word 1 (MOCW1)—Offset 2Dh .......... 912
21.12.3.8 Slave Initialization Command Word 1 (SICW1)—Offset A0h ........ 912
21.12.3.9 Slave Initialization Command Word 2 (SICW2)—Offset A1h ........ 913
21.12.3.10 Slave Operational Control Word 2 (SoCW2)—Offset A4h............ 913
21.12.3.11 Slave Initialization Command Word 3 (SICW3)—Offset A5h ....... 914
21.12.3.12 Slave Operational Control Word 3 (SoCW3)—Offset A8h............ 914
21.12.3.13 Slave Initialization Command Word 4 (SICW4)—Offset A9h ....... 915
21.12.3.14 Slave Operational Control Word 1 (SoCW1)—Offset ADh ........... 916
21.12.3.15 Master Edge/Level Control (ELCR1)—Offset 4D0h..................... 916
21.12.3.16 Slave Edge/Level Control (ELCR2)—Offset 4D1h ...................... 916
21.13 I/O APIC ........................................................................................................ 918
21.13.1Features ............................................................................................. 918
21.13.2Use .................................................................................................... 919
21.13.3Unsupported Modes .............................................................................. 919
21.13.4Register Map ....................................................................................... 920
21.13.5Memory Mapped Registers..................................................................... 920
21.13.5.1 Index Register (IDX)—Offset FEC00000h .................................. 921
21.13.5.2 Window Register (WDW)—Offset FEC00010h............................. 921
21.13.5.3 End of Interrupt Register (EOI)—Offset FEC00040h.................... 921
21.13.6Index Registers.................................................................................... 922
21.13.6.1 Identification Register (ID)—Offset 0h ..................................... 922
21.13.6.2 Version Register (VS)—Offset 1h ............................................. 923
21.13.6.3 Redirection Table Entry Lower (RTE[0-23]L)—Offset 10h - 3Eh.... 923
21.13.6.4 Redirection Table Entry Upper (RTE[0-23]U)—Offset 11h - 3Fh ... 924
21.14 Watchdog Timer.............................................................................................. 926
21.14.1Features ............................................................................................. 926
21.14.2Use .................................................................................................... 926
21.14.3Register Map ....................................................................................... 927
21.14.4I/O Mapped Registers ........................................................................... 927
21.14.4.1 Preload Value 1 Register 0 (PV1R0)—Offset 0h ......................... 928
21.14.4.7 Reload Register 0 (RR0)—Offset Ch ......................................... 930
21.14.4.8 Reload Register 1 (RR1)—Offset Dh......................................... 931
21.14.4.9 WDT Configuration Register (WDTCR)—Offset 10h .................... 931
21.14.4.10 WDT Lock Register (WDTLR)—Offset 18h ................................ 932
22.0 Debug Port and JTAG/TAP ..................................................................................... 935
22.2 Features ........................................................................................................ 936
22.2.1 OpenOCD ............................................................................................ 936
Figures
1 Block Diagram ......................................................................................................... 38
2 Intel® Quark™ SoC X1000 PCI View............................................................................ 43
3 Signals In Default System Pin List .............................................................................. 46
4 Intel® Quark™ SoC X1000 Package Dimensions............................................................ 59
5 PCI Express Transmitter Eye...................................................................................... 76
6 PCI Express Receiver Eye .......................................................................................... 76
7 USB Rise and Fall Time ............................................................................................. 78

8 USB Jitter................................................................................................................79
9 USB EOP Width ........................................................................................................79
10 SPI Interface Timing .................................................................................................81
11 SDIO Interface Timing ..............................................................................................82
12 Measurement Points for Differential Clocks...................................................................84
13 Physical Address Space - Low DRAM & MMIO................................................................92
14 Physical Address Space - MMIO ..................................................................................93
15 Physical Address Space - DOS DRAM...........................................................................94
16 Physical Address Space - SMM Range ..........................................................................95
17 Bus 0 PCI Devices and Functions ................................................................................98
18 Message Bus with PCI Space ......................................................................................99
19 SoC Platform Clocking ............................................................................................. 102
20 RTC Power Well Timing Diagrams ............................................................................. 115
21 Power Up Sequence ................................................................................................ 117
22 Power-Up Sequence without G2/G3........................................................................... 118
23 eSRAM 4KB Page Mapping ....................................................................................... 128
24 eSRAM 512KB Page Mapping.................................................................................... 129
25 Intel® Quark™ SoC X1000 Host Bridge Register Map ................................................... 132
26 Register Map.......................................................................................................... 236
27 PCI Express Register Map ........................................................................................ 264
28 Ethernet Register Map............................................................................................. 311
29 Transmit Descriptor Fields ....................................................................................... 422
30 Transmit Descriptor Fetch (Read) ............................................................................. 423
31 Receive Descriptor Fields ......................................................................................... 427
32 USB Register Map................................................................................................... 437
33 SD Memory Card Bus Topology................................................................................. 593
34 SDIO Card Bus Topology ......................................................................................... 593
35 eMMC Interface ...................................................................................................... 594
36 SDIO/SD/eMMC Register Map................................................................................... 596
37 UART Data Transfer Flow ......................................................................................... 664
38 HSUART Register Map ............................................................................................. 668
39 Data Transfer on the I2C* Bus.................................................................................. 733
40 START and STOP Conditions..................................................................................... 733
41 7-Bit Address Format .............................................................................................. 734
42 10-bit Address Format ............................................................................................ 734
43 Master Transmitter Protocol ..................................................................................... 735
44 Master Receiver Protocol ......................................................................................... 736
45 START Byte Transfer ............................................................................................... 736
46 I2C*/GPIO Register Map.......................................................................................... 739
47 Generic SPI Waveform ............................................................................................ 788
48 SPI Register Map .................................................................................................... 791
49 Legacy Bridge Register Map ..................................................................................... 817
50 Legacy GPIO Register Map ....................................................................................... 846
51 Legacy SPI Register Map ......................................................................................... 857
52 8254 Timers Register Map ....................................................................................... 876
53 HPET Register Map ................................................................................................. 883
54 RTC Register Map ................................................................................................... 894
55 8259 Register Map.................................................................................................. 906
56 Detailed I/O APIC Block Diagram .............................................................................. 918
57 MSI Address and Data............................................................................................. 919
58 I/O APIC Register Map ............................................................................................ 920
59 Watchdog Timer Register Map .................................................................................. 927
Tables

1 Industry Specifications.............................................................................................. 37
2 Component Identification .......................................................................................... 42
3 Intel® Quark™ SoC X1000 Device ID .......................................................................... 44
4 I/O Power Well Definitions......................................................................................... 47
5 Buffer Type Definitions ............................................................................................. 47
6 Default Buffer State Definitions .................................................................................. 47
7 System Memory Signals............................................................................................ 48
8 PCI Express* 2.0 Signals........................................................................................... 48
9 Ethernet Interface Signals ......................................................................................... 49
10 USB 2.0 Interface Signals ......................................................................................... 49
11 Integrated Clock Interface Signals.............................................................................. 50
12 SD/SDIO/MMC Signals .............................................................................................. 50
13 High Speed UART Signals .......................................................................................... 51
14 I2C* Signals ............................................................................................................ 51
15 Legacy SPI Signals ................................................................................................... 52
16 SPI Signals ............................................................................................................. 52
17 Real Time Clock (RTC) Interface Signals...................................................................... 53
18 Power Management Interface Signals.......................................................................... 53
19 JTAG and Debug Interface Signals .............................................................................. 54
20 Legacy Interface Signals ........................................................................................... 54
21 General Purpose I/O Signals ..................................................................................... 54
22 Power and Ground Pins ............................................................................................. 55
23 Hardware Straps ...................................................................................................... 57
24 Alphabetical Ball Listing ............................................................................................ 60
25 Alphabetical Signal Listing ......................................................................................... 64
26 Intel® Quark™ SoC X1000 Absolute Maximum Voltage Ratings ....................................... 69
27 Power Supply Rail Ranges ......................................................................................... 70
28 Maximum Supply Current: ICC Max ............................................................................ 71
30 Configurable IO (CFIO) Bi-directional Signal Groupings ................................................. 72
31 CFIO DC Characteristics ............................................................................................ 73
32 CFIO AC Characteristics ............................................................................................ 73
33 RTC DC Characteristics ............................................................................................. 74
34 PCI Express* 2.0 Differential Signal DC Characteristics ................................................. 74
35 PCI Express* 2.0 Interface Timings ............................................................................ 75
36 USB 2.0 Differential Signal DC Characteristics .............................................................. 77
37 USB 2.0 Interface Timings......................................................................................... 77
38 Legacy SPI Interface Timings (20 MHz) ....................................................................... 79
39 SPI0/1 Interface Timings (25 MHz)............................................................................. 80
40 SDIO Timing ........................................................................................................... 81
41 Reference Clocks AC Characteristics ........................................................................... 82
42 Fixed I/O Register Access Method Example (NSC Register) ............................................ 85
43 Fixed Memory Mapped Register Access Method Example (IDX Register)........................... 85
44 Referenced I/O Register Access Method Example (PM1S Register) .................................. 86
45 Memory Mapped Register Access Method Example (ESD Register) .................................. 86
46 PCI Register Access Method Example (PCI_DEVICE_VENDOR Register)............................ 86
47 PCI CONFIG_ADDRESS Register (I/O PORT CF8h) Mapping ............................................ 87
48 PCI Configuration Memory Bar Mapping....................................................................... 87
49 MCR Description ...................................................................................................... 88
50 MCRX Description..................................................................................................... 88
51 Register Access Types and Definitions......................................................................... 89
52 Fixed Memory Ranges in the Legacy Bridge ................................................................. 95
53 Fixed I/O Ranges in the Legacy Bridge ........................................................................ 96
54 Movable I/O Ranges Decoded by PCI Devices on the I/O Fabric ...................................... 96
55 PCI Devices and Functions......................................................................................... 97

56 Message Types.........................................................................................................99
57 Intel® Quark™ SoC X1000 Clock Inputs ..................................................................... 103
58 Intel® Quark™ SoC X1000 Clock Outputs ................................................................... 103
59 Power Management ................................................................................................ 105
60 General Power States for System.............................................................................. 107
61 ACPI PM State Transition Rules................................................................................. 107
62 Processor Core/ States Support ................................................................................ 108
63 Main Memory States ............................................................................................... 108
64 PCIe* States.......................................................................................................... 108
65 G, S and C State Combinations................................................................................. 109
66 RTC Power Well Timing Parameters ........................................................................... 115
67 S4/S5 to S0 Timing Parameters................................................................................ 119
68 Intel® Quark™ SoC X1000 S3 Wake Events ................................................................ 121
69 SoC Reset Events ................................................................................................... 121
70 Thermal Sensor Signals ........................................................................................... 123
71 Summary of PCI Configuration Registers—0/0/0 ......................................................... 132
72 Summary of I/O Registers—PMBA............................................................................. 138
73 Summary of Message Bus Registers—0x00 ................................................................ 140
79 Memory Signals...................................................................................................... 233
80 Supported DDR3 DRAM Devices................................................................................ 235
81 Supported DDR3 Memory Configurations ................................................................... 235
83 PCI Express* 2.0 Signals ......................................................................................... 261
84 Possible Interrupts Generated From Events/Packets .................................................... 262
85 Summary of PCI Configuration Registers—0/23/0 ....................................................... 264
86 10/100 Ethernet Interface Signals ............................................................................ 309
88 Summary of Memory Mapped I/O Registers—BAR0 ..................................................... 327
89 Transmit Descriptor Word 0 (TDES0)......................................................................... 423
90 Transmit Descriptor Word 1 (TDES1)......................................................................... 426
91 Transmit Descriptor 2 (TDES2) ................................................................................. 426
95 Receive Descriptor Fields (RDES0) ............................................................................ 428
96 Receive Descriptor Fields 1 (RDES1) ......................................................................... 429
102 Signals.................................................................................................................. 435
109 SDIO/SD/eMMC Interface Signals ............................................................................. 591
110 SDIO/SD/eMMC Features......................................................................................... 592


113 UART 0 Interface Signals......................................................................................... 663
114 UART 1 Interface Signals......................................................................................... 663
115 Baud Rates Achievable with Different DLAB Settings ................................................... 664
119 I2C* Signals .......................................................................................................... 731
120 I2C* Definition of Bits in First Byte ........................................................................... 734
121 GPIO Signals ......................................................................................................... 738
125 SPI Interface Signals .............................................................................................. 787
126 SPI Clock Frequency Settings .................................................................................. 790
129 Miscellaneous Legacy Signals................................................................................... 816
131 Summary of Memory Mapped I/O Registers—RCBA..................................................... 827
132 Summary of I/O Registers....................................................................................... 832
133 Summary of I/O Registers—GPE0BLK ....................................................................... 835
134 Summary of I/O Registers—PM1BLK ......................................................................... 841
135 Legacy GPIO Signals............................................................................................... 845
136 Summary of I/O Registers—GBA .............................................................................. 846
137 Legacy SPI Signals ................................................................................................. 856
138 Summary of Memory Mapped I/O Registers—RCBA..................................................... 858
139 Counter Operating Modes........................................................................................ 874
140 Register Aliases ..................................................................................................... 876
141 8254 Interrupt Mapping .......................................................................................... 882
142 Summary of Memory Mapped I/O Registers—0xFED00000........................................... 883
143 RTC Signals........................................................................................................... 893
144 I/O Registers Alias Locations ................................................................................... 895
145 Indexed Registers .................................................................................................. 895
146 IRQAGENT Description ............................................................................................ 899
147 Interrupt Controller Connections .............................................................................. 900
148 Interrupt Status Registers ....................................................................................... 901
149 Content of Interrupt Vector Byte .............................................................................. 901
150 8259 I/O Registers Alias Locations ........................................................................... 907
151 Summary of I/O Registers....................................................................................... 907
152 I/O APIC Memory Mapped Registers ......................................................................... 921
153 Index Registers ..................................................................................................... 922
154 Summary of I/O Registers—WDTBA.......................................................................... 927
155 Debug Port and JTAG/TAP Signals ............................................................................ 935

Intel® Quark™ SoC X1000—Revision History
Revision History
Date Revision Description
• Updated Section 18.3.1, “DMA Mode Operation” on page 665

August 2015 005
• Updated Section 18.3.4, “Autoflow Control” on page 667
• Updated Section 9.2.3, “AC Power Applied: G3 to S4/S5 State Transition” on page 115
November 2014 004
• Updated Section 9.2.4, “Using PWR_BTN_B: Transition from S4/S5 to S0” on page 116
• Updated Table 23, “Hardware Straps” on page 57
• Updated subsection numbering under Section 12.6.2, “SPI DMA Block” on page 140
• Updated Table 27, “Power Supply Rail Ranges” on page 70
• Updated Table 28, “Maximum Supply Current: ICC Max” on page 71
August 2014 003
• Updated Table 33, “RTC DC Characteristics” on page 74
• Updated Table 36, “USB 2.0 Differential Signal DC Characteristics” on page 77
• Updated Table 66, “RTC Power Well Timing Parameters” on page 115
• Updated Table 110, “SDIO/SD/eMMC Features” on page 592
Updated Chapter 3 Ballout and Package Information
Updated Table 18 Power Management Interface Signals
Updated Table 23 Hardware Straps
Added Table 30, “Configurable IO (CFIO) Bi-directional Signal Groupings” on page 72 and Table 31,
“CFIO DC Characteristics” on page 73
Replaced Figure 19, “SoC Platform Clocking” on page 102
Removed ECC Scrubbing (Sections 12.7.3.1 - 12.7.3.10, Section 12.3.1)
Removed SPI DMA (Updated Section 12.3, Removed Sections 12.6.2.1 - 12.6.2.3, Added Register
May 2014 002 Option Register 1(P_CFG_72) —Offset 72h)
Updated Table 49
Added Table 54 Message Types to Section 6.4 Message Bus Space
Updated Table 67
Added 12.7.9.2 Miscellaneous Legacy Signal Enables (HLEGACY)—Offset 0Ah
Updated Section 13.4.1 DRAM Rank Population (DRP)—Offset 0h
Added Table 78 Message Opcode Definition (Section 13.5)
Added Section 15.7, “MAC Descriptor Details”
Other changes are marked with change bars
October 2013 001 Initial Public Release

Introduction—Intel® Quark™ SoC X1000
1.0 Introduction
1.1 About This Manual

This document is intended for Original Equipment Manufacturers and BIOS vendors
creating products based on the Intel® Quark™ SoC X1000 application processor.
Note: Throughout this document, SoC is used as a general term and refers to all Intel®
Quark™ SoC X1000 SKUs, unless specifically noted otherwise.
This manual assumes a working knowledge of the vocabulary and principles of

interfaces and architectures such as PCI Express*, USB, SDIO/MMC, and ACPI.
Although some details of these features are described within this manual, refer to the
individual industry specifications listed in Table 1 for the complete details.
All PCI buses, devices and functions in this manual are abbreviated using the following
nomenclature; Bus:Device:Function. This manual abbreviates buses as Bn, devices as
Dn and functions as Fn. For example, Device 31 Function 0 is abbreviated as D31:F0,
Bus 1 Device 8 Function 0 is abbreviated as B1:D8:F0. Generally, the bus number is not
used, and can be considered to be Bus 0.
Table 1. Industry Specifications

Specification Location
PCI Express* Base Specification, Revision 2.0 http://www.pcisig.com/specifications
PCI Local Bus Specification, Revision 2.3 (PCI) http://www.pcisig.com/specifications
PCI Power Management Specification, Revision 1.2 http://www.pcisig.com/specifications
Universal Serial Bus Specification (USB), Revision 2.0 http://www.usb.org/developers/docs
Advanced Configuration and Power Interface, Version 3.0 (ACPI) http://www.acpi.info/spec.htm
Enhanced Host Controller Interface Specification for Universal http://developer.intel.com/technology/usb/

Serial Bus, Revision 1.0 (EHCI) ehcispec.htm
IEEE 802.3 Fast Ethernet http://standards.ieee.org/getieee802/
AT Attachment - 6 with Packet Interface (ATA/ATAPI - 6) http://T13.org (T13 1410D)
http://www.intel.com/content/www/us/en/
IA-PC HPET (High Precision Event Timers) Specification,
software-developers/software-developers-
Revision 1.0a
hpet-spec-1-0a.html
1.2 Component Overview

The Intel® Quark™ SoC X1000 processor is the next generation secure, low-power
Intel® Architecture (IA) SoC for deeply embedded applications. The SoC integrates the
Inte® QuarkTM SoC X1000 Core plus all the required hardware components to run off-
the-shelf operating systems and to leverage the vast x86 software ecosystem.

Intel® Quark™ SoC X1000—Introduction
To enable secure applications, the SoC secure SKUs feature an on-die Boot ROM that is
used to establish a hardware Root of Trust (RoT). The immutable code located within
the Boot ROM is used to initiate an iterative firmware authentication process ensuring
only trusted code is executed when taking the platform out of reset.
To facilitate low-cost platforms with sensitive Bill of Material (BOM) requirements, all
SoC clocks can be generated from a single crystal oscillator while all the required SoC
voltage levels can be derived from a single commercial off-the-shelf (COTS) voltage
regulator. In addition, the SoC provides an ECC-protected DRAM solution using only
standard x8 DDR3 devices.
The SoC also features a 512 Kbyte on-die embedded SRAM (eSRAM) that can be
configured to overlay regions of DRAM to provide low latency access to critical portions
of system memory. For robustness, the contents of this on-die eSRAM are also ECC
protected.
Figure 1. Block Diagram
CPU Core
Clock
eSRAM
JTAG Host Bridge
DDR3
Memory
Controller
AMBA Fabric Legacy Bridge

PCIe*
USB 2.0
10/100 ETH
I2C*/GPIO
UART
HPET
GPIO
SDIO
APIC
ROM
PMC
8254
8259
RTC
SPI
SPI
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
I/O
1.2.1 SoC CPU Core Features

• 400 MHz maximum operating frequency
• Low power options to run at half or at quarter of maximum CPU frequency

• 32-bit address bus, 32-bit data bus

• 16 Kbyte shared instruction and data L1 cache.
1.2.2 System Memory Controller Features

• Single channel DDR3 memory controller with ECC support
• 16-bit data bus
• Supports up to two ranks total
• Supports DDR3 with 800 MT/s data rates
• x8 DRAM device data width
• 1 Gbit, 2 Gbit, and 4 Gbit DRAM device densities
• Total memory size from 128 Mbyte to 2 Gbyte
• Supports different physical mappings of bank addresses to optimize performance
• Out-of-order request processing to increase performance
• Aggressive power management to reduce power consumption
• Proactive page closing policies to close unused pages
• Supports soldered down DRAM devices
1.2.3 Embedded SRAM Features

• Low Latency 512 Kbyte on-die embedded SRAM
• Configurable to either overlay a 512 Kbyte block or overlay individual 4 Kbyte
pages of system memory
• ECC protected
1.2.4 Power Management Features

• Supports ACPI 3.0 specification
• Supports C0, C1, and C2 processor power states
• Supports S0, S3, and S4/S5 system power states
1.2.5 Security Features

• On-die Boot ROM provides Hardware Root of Trust (RoT) for firmware
authentication
1.2.6 PCI Express* Features

The SoC has two PCI Express* root ports, each supporting the PCI Express Base
specification Rev 2.0 at a maximum of 2.5 GT/s data transfer rates. Each root port is
configured as a x1 link.
• 128 Byte max payload size with the capability of splitting the request at 64 Byte
granularity
• Software-Initiated Link Power Management (D1, D2, D3Hot, and L1 States)
• PME event generation

1.2.7 Ethernet Features

• 10 and 100 Mbps data transfer rates with RMII interface to communicate with an
external Fast Ethernet PHY
• Full-duplex operation:
— IEEE 802.3x flow control support
— Optional forwarding of received pause control frames to the user application
• Half-duplex operation:
— CSMA/CD Protocol support
• Flexible address filtering modes:
— 64-bit Hash filter for multicast and unicast (DA) addresses
— Option to pass all multicast addressed frames
— Promiscuous mode to pass all frames without any filtering for network
monitoring
— Pass all incoming packets (as per filter) with a status report
• Programmable frame length to support Standard Ethernet frames with size up to
1522 bytes
• Enhanced Receive module for checking IPv4 header checksum and TCP, UDP, or
ICMP checksum encapsulated in IPv4 or IPv6 datagrams (Type 2)
• Support Ethernet frame time stamping as described in IEEE 1588-2002 and IEEE
1588-2008. The 64-bit timestamps are given in the transmit or receive status of
each frame.
1.2.8 USB2 Host Controller Features

• 2 host ports that support high-speed (480 Mbps), full-speed (12 Mbps), and low-
speed (1.5 Mbps) operation
• EHCI and OHCI host controllers
1.2.9 USB2 Device Controller Features

• Single device port that supports high-speed (480 Mbps) and full-speed (12 Mbps)
operation
1.2.10 SD/SDIO/eMMC Controller Features

• Host Controller provides a single port configurable as an SD, SDIO, or eMMC
interface
• SD Clock Frequency up to 50 MHz
• Supports SD Host Controller Standard Specification 3.0
• Supports SDIO card specification 3.0
• Supports SD Memory Card Specification 3.0
• Supports SD Memory Card Security Specification 1.01
• Supports eMMC Specification 4.41
1.2.11 I2C* Master Controller

• Two-wire I2C serial bus interface

• Two I2C speeds supported: Standard (100 Kbit/s) and Fast (400 Kbit/s) data rates
• Fully asynchronous I2C clock signal
• Master I2C operation
1.2.12 GPIO Features

• 16 GPIO pins provided
• 6 GPIO pins remain active during S3 and can be used to wake the system from the
Suspend state.
• Remaining 10 GPIO pins are powered during S0 state only and are not available in
S3
1.2.13 SPI Master Controller

• Two SPI Master controllers
• One Chip Select per master controller
• Configurable SCLK frequency from 1 kHz up to 25 MHz
1.2.14 High Speed UART Controller with DMA

• Two 16550 compliant UART controllers
• Supported Baud rates from 300 to 2764800
• Integrated DMA capability with hardware flow control
1.2.15 Legacy Bridge

The Legacy Bridge is a collection of hardware blocks critical to implement an Intel
Architecture compatible platform. Some of its key features are:
• A 20 MHz Serial Peripheral Interface (SPI) for Flash only - stores boot FW and
system configuration data
• A Power Management Controller (PMC) that controls many of the power
management features present in the SoC
• Legacy Bridge Components - Provides hardware blocks required to support legacy
PC platform features. The legacy bridge components include the RTC, Interrupt
Controllers, Timers and General Purpose I/Os (GPIO).
1.2.16 Package
The SoC is packaged in a Flip-Chip Ball Grid Array (FCBGA) package with 393 solder
balls with 0.593 mm ball pitch. The package dimensions are 15mm x 15mm.
1.3 Component Identification

The Intel® Quark™ SoC X1000 stepping is identified by both:
• Processor Family/Model/Stepping returned by the CPUID instruction. This always
returns 0x590 for SoC.
• Revision ID register of the Host Bridge, located at D0:F0. Reads of the register
reflect the stepping.

Table 2. Component Identification

Vendor ID1 Device ID2 Revision ID3 Stepping
8086h 0958h 00h A0h
Notes:
1. The Vendor ID corresponds to bits 15-0 of the Vendor ID Register located at offset 00-01h in the PCI
configuration space of the device.
2. The Device ID corresponds to bits 15-0 of the Device ID Register located at offset 02-03h in the PCI
3. The Revision ID corresponds to bits 7-0 of the Revision ID Register located at offset 08h in the PCI
The SoC incorporates a variety of PCI functions as listed in Table 3. All devices reside
on PCI Bus 0 as shown in Figure 2.

Figure 2. Intel® Quark™ SoC X1000 PCI View
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem)
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1
IO Fabric
D:21
SPI1 F:1
I2C*/GPIO F:2 PMC
GPIO
Legacy Bridge
RTC
D:31,F:0
8254
SDIO/eMMC F:0 8259
HPET
HSUART0 F:1
IO APIC
IO Fabric D:20
USB Device F:2

SPI
EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7

Table 3. Intel® Quark™ SoC X1000 Device ID

Device Function Description Device ID A0 SRID
D0:F0 Host Bridge 0958h 00h
D31:F0 Legacy Bridge 095Eh 00h
D23:F0 PCIe* Root Port 0 11C3h 00h
D23:F1 PCIe* Root Port 1 11C4h 00h
D20:F0 SDIO / eMMC Controller 08A7h 10h
D20:F1 HS-UART 0 0936h 10h
D20:F2 USB 2.0 Device 0939h 10h
D20:F3 USB EHCI Host Controller 0939h 10h
D20:F4 USB OHCI Host Controller 093Ah 10h
D20:F5 HS-UART 1 0936h 10h
D20:F6 10/100 Ethernet MAC 0 0937h 10h
D20:F7 10/100 Ethernet MAC 1 0937h 10h
D21:F0 SPI Controller 0 0935h 10h
D21:F1 SPI Controller 1 0935h 10h

2
D21:F2 I C* Controller and GPIO Controller 0934h 10h
§§

Physical Interfaces—Intel® Quark™ SoC X1000
2.0 Physical Interfaces
Many interfaces contain physical pins. These groups of pins make up the physical
interfaces. This chapter summarizes the physical interfaces.

November 2014 Datasheet
Intel® Quark™ SoC X1000—Physical Interfaces
Figure 3. Signals In Default System Pin List
DDR3_DQ[15:0]
DDR3_DQS[1:0] CPU
DDR3_DQSB[1:0] Core
DDR3_DM[1:0]
DDR3_MA[15:0]
DDR3_BS[2:0]
DDR3_RASB
DDR3_CASB
DDR3_WEB IVCCRTCEXT
DDR3_CSB[1:0] DDR3
Interface RTCX1
DDR3_ODT[1:0] Legacy
Components RTCX2
DDR3_CKE[1:0]
DDR3_CK[1:0] RTCRST_B
DDR3_CKB[1:0] THERM_B
DDR3_VREF SMI_B
DDR3_ODTPU/DQPU/CMDPU
DDR3_IDRAM_PWROK
DDR3_ISYSPWRGOOD
DDR3_DRAMRSTB
LSPI_SS_B
MAC[0/1]_TXEN LSPI_SCK
MAC[0/1]_TXDATA[1:0] Legacy Bridge LSPI_MISO
MAC[0/1]_RXDV Ethernet LSPI_MOSI
MAC[0/1]_RXDATA[1:0] RMII
MAC[0/1]_MDC Interface
MAC[0/1]_MDIO
RMII_REF_CLK RESET_BTN_B
PWR_BTN_B
PCIE_PETP[1:0]
WAKE_B
PCIE_PETN[1:0]
PCIE_PERP[1:0] GPE_B
PCIE_PERN[1:0] PCI Express* 2.0 S5_PG
PCIE_REFCLKN Interface S3_3V3_EN
PCIE_REFCLKP Power S3_1V5_EN
PCIE_IRCOMP Management
S3_PG
PCIE_RBIAS Controller
Interface S0_3V3_EN
S0_1V5_EN
SD_DATA[7:0]
S0_1V0_EN
SD_CMD
SD_CLK S0_1P0_PG
SD/MMC
SD_WP Interface S0_PG
SD_CD_B ODRAM_PWROK
SD_LED
SD_PWR OSYSPWRGOOD
USBD_DP SIU0_DCD_B
USBD_DN SIU0_DSR_B
USBH[1:0]_DP
Universal Serial SIU0_DTR_B
USBH[1:0]_DN
Bus 2.0 HSUART SIU0_RI_B
USBH[1:0]_OC_B
Interface Interface SIU[0/1]_CTS_B
IUSBCOMP
OUSBCOMP SIU[0/1]_RTS_B
USBH[1:0]_PWR_EN SIU[0/1]_RXD
SIU[0/1]_TXD
FLEX_CLK[2:0]
CKSYS25OUT
RMII_REF_CLK_OUT
REF[0/1]_OUTCLK_P
REF[0/1]_OUTCLK_N SPI[0/1]_SS_B
OSC_COMP Internal Clocking
SPI SPI[0/1]_SCK
XTAL_IN Interface SPI[0/1]_MISO
XTAL_OUT
SPI[0/1]_MOSI
RTC_EXT_CLK_EN
I2C* I2C_CLK
PRDY_B Interface I2C_DATA
PREQ_B
TCK
TDI JTAG Port
TDO GPIO GPIO_SUS[5:0]
Interface GPIO[9:0]
TMS
TRST_B

Datasheet November 2014
2.1 Pin States Through Reset

This chapter describes the each signal state before, during, and directly after reset.
Additionally, some signals have internal pull-up/pull-down termination resistors, and
their values are also provided.
Table 4. I/O Power Well Definitions

Power Type Power Well Description
CORE Core I/O, and everything else uses the CORE power well.
SUS Devices outside of memory that must remain on in the S3 state use the SUS power well.
RTC Devices that must be on in the S4/S5 state use the RTC power well.
Table 5. Buffer Type Definitions

Buffer Type Buffer Description
PCIe* PCIe*, differential buffer type
SSTL-15 DDR3, 1.5V tolerant SSTL buffer type
DDI DDR (TMDS, DP) 1.0V tolerant differential buffer type
CMOS[Voltage] CMOS buffer type. [Voltage] can be of the following types: 1.05, 1.5, 1.8, and 3.3.
Open drain CMOS buffer type [Voltage] can be of the following types: 1.05, 1.5, 1.8 and
CMOS[Voltage]_OD
3.3.
Analog pins that do not have specific digital requirements. Often used for circuit
Analog
calibration or monitoring.
Table 6. Default Buffer State Definitions

Buffer State Description
The SoC places this output in a high-impedance state. For inputs, external drivers are not
High-Z
expected.
The state of the input (driven or tristated) does not affect the SoC. For outputs, it is
Do Not Care
assumed that the output buffer is in a high-impedance state.
VOH The SoC drives this signal high.
VOL The SoC drives this signal low.
Unknown The SoC drives or expects an indeterminate value.
VIH The SoC expects/requires the signal to be driven high.
VIL The SoC expects/requires the signal to be driven low.
This signal is pulled high by a pull-up resistor (internal or external — internal value
Pull-up
specified in “Term” column).
This signal is pulled low by a pull-down resistor (internal or external — internal value
Pull-down
specified in “Term” column).
Running The clock is toggling, or the signal is transitioning.
The power plane for this signal is powered down. The SoC does not drive outputs, and
Off
inputs should not be driven to the SoC. (VSS on output)
2.2 System Memory Signals

See Section 6.0 for more details of the DDR3 interface signals. Termination not listed.

Table 7. System Memory Signals

Default Buffer State
Signal Name Dir Term Power Type S4/S5 S3 Reset Enter S0
DDR3_BS[2:0] O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_CASB O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_RASB O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_WEB O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_MA[15:0] I/O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_CK[1:0] I/O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_CKB[1:0] I/O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_CKE[1:0] I/O - 1.5V SSTL-15 Off VOL VOL VOL
DDR3_CSB[1:0] I/O - 1.5V SSTL-15 Off High-Z High-Z VOH
DDR3_ODT[1:0] I/O - 1.5V SSTL-15 Off High-Z High-Z VOL
DDR3_DQ[15:0] I/O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_DM[1:0] I/O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_DQS[1:0] I/O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_DQSB[1:0] I/O - 1.5V SSTL-15 Off High-Z High-Z High-Z
DDR3_IDRAM_PWROK I Ext 1.5V CMOS-15 VIL VIH Pull-up VIH
DDR3_ISYSPWRGOOD I Ext 1.5V CMOS-15 VIL VIL Pull-up VIH
VOL/VOH(S3
DDR3_DRAMRSTB O - 1.5V CMOS-15 Off VOH VOL
Exit)
DDR3_VREF I/O - 1.5V Reference Off Reference Reference Reference
DDR3_ODTPU I/O - 1.5V Analog Off Analog Analog Analog
DDR3_DQPU I/O - 1.5V Analog Off Analog Analog Analog
DDR3_CMDPU I/O - 1.5V Analog Off Analog Analog Analog
2.3 PCI Express* 2.0 Signals

See “PCI Express* 2.0” on page 261 for more details of the interface signals.
Table 8. PCI Express* 2.0 Signals

Running/ Running/
PCIE_REFCLKP I - 1.05V PCIe Off Off
Unknown Unknown
Running/ Running/
PCIE_REFCLKN I - 1.05V PCIe Off Off
Unknown Unknown
PCIE_PETP[1:0] O - 1.05V PCIe Off Off VOL VOL
PCIE_PETN[1:0] O - 1.05V PCIe Off Off VOL VOL
PCIE_PERP[1:0] I - 1.05V PCIe Off Off High-Z High-Z
PCIE_PERN[1:0] I - 1.05V PCIe Off Off High-Z High-Z
PCIE_IRCOMP I/O - 1.5V Analog Off Off Analog Analog
PCIE_RBIAS I - 1.5V Analog Off Off Analog Analog

2.4 Ethernet Interface Signals

See Chapter 15.0, “10/100 Mbps Ethernet” for more details of the Ethernet interface
signals.
Table 9. Ethernet Interface Signals

Running/
RMII_REF_CLK I - 3.3V CMOS3.3 Off Off Running
Unknown
MAC0_TXEN O - 3.3V CMOS3.3 Off Off VOL VOL
MAC0_TXDATA[1:0] O - 3.3V CMOS3.3 Off Off VOL VOL
MAC0_RXDV I 20k(L) 3.3V CMOS3.3 Off Off Pull-down Pull-down
MAC0_RXDATA[1:0] I - 3.3V CMOS3.3 Off Off Unknown Unknown
MAC0_MDC O Ext 3.3V CMOS3.3 Off Off Pull-up Pull-up
MAC0_MDIO I/O Ext 3.3V CMOS3.3_OD Off Off Pull-up Pull-up
MAC1_TXEN O - 3.3V CMOS3.3 Off Off VOL VOL
MAC1_TXDATA[1:0] O - 3.3V CMOS3.3 Off Off VOL VOL
MAC1_RXDV I 20k(L) 3.3V CMOS3.3 Off Off Pull-down Pull-down
MAC1_RXDATA[1:0] I - 3.3V CMOS3.3 Off Off Unknown Unknown
MAC1_MDC O Ext 3.3V CMOS3.3 Off Off Pull-up Pull-up
MAC1_MDIO I/O Ext 3.3V CMOS3.3_OD Off Off Pull-up Pull-up
2.5 USB 2.0 Interface Signals

See Chapter 16.0, “USB 2.0” for more details of the USB 2.0 interface signals.
Table 10. USB 2.0 Interface Signals (Sheet 1 of 2)

USBH0_OC_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
USBH1_OC_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
USBH0_PWR_EN O Ext 3.3V CMOS3.3 Off Off Pull-down Pull-down
USBH1_PWR_EN O Ext 3.3V CMOS3.3 Off Off Pull-down Pull-down
USBH0_DP I/O - 3.3V USB Off Off High-Z High-Z
USBH0_DN I/O - 3.3V USB Off Off High-Z High-Z
USBH1_DP I/O - 3.3V USB Off Off High-Z High-Z
USBH1_DN I/O - 3.3V USB Off Off High-Z High-Z
USBD_DP I/O - 3.3V USB Off Off High-Z High-Z
USBD_DN I/O - 3.3V USB Off Off High-Z High-Z
Running/ Running/
USB_CLK96P I - 1.05V USB Off Off
Unknown Unknown

Table 10. USB 2.0 Interface Signals (Sheet 2 of 2)

Running/ Running/
USB_CLK96N I - 1.05V USB Off Off
Unknown Unknown
OUSBCOMP_P18 O - 1.8V Analog Off Off Analog Analog
IUSBCOMP_N18 I - 1.8V Analog Off Off Analog Analog
2.6 Integrated Clock Interface Signals

See Chapter 7.0, “Clocking” for more details of the Integrated Clock interface signals.
Table 11. Integrated Clock Interface Signals

XTAL_IN I - 1.05V Analog Running Running Running Running
XTAL_OUT O - 1.05V Analog Running Running Running Running
CKSYS25OUT O - 3.3V CMOS3.3 Running Running Running Running
REF0_OUTCLK_P O - 1.05V CMOS1.05 VOL VOL VOL Running
REF0_OUTCLK_N O - 1.05V CMOS1.05 VOL VOL VOL Running
REF1_OUTCLK_P O - 1.05V CMOS1.05 VOL VOL VOL Running
REF1_OUTCLK_N O - 1.05V CMOS1.05 VOL VOL VOL Running
FLEX0_CLK O - 3.3V CMOS3.3 VOL VOL VOL Running
RMII_REF_CLK_OUT O - 3.3V CMOS3.3 VOL VOL VOL Running
OSC_COMP I - 1.5V Analog Analog Analog Analog Analog
Running/ Running/
HPLL_REFCLK_P I - 1.05V CMOS1.05 Off Off
Unknown Unknown
Running/ Running/
HPLL_REFCLK_N I - 1.05V CMOS1.05 Off Off
Unknown Unknown
Running/ Running/
PAD_BYPASS_CLK I - 1.05V CMOS1.05 Off Off
Unknown Unknown
2.7 SDIO/SD/MMC Signals

See Chapter 17.0, “SDIO/SD/eMMC” for more details of the interface signals, including
different options based on port configuration.
Table 12. SD/SDIO/MMC Signals (Sheet 1 of 2)

SD_DATA[7:0] I/O 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SD_CMD I/O 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SD_CLK O - 3.3V CMOS3.3 Off Off VOL VOL

Table 12. SD/SDIO/MMC Signals (Sheet 2 of 2)

SD_WP I 20k(L) 3.3V CMOS3.3 Off Off Pull-down Pull-down
SD_CD_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SD_LED O - 3.3V CMOS3.3 Off Off VOL VOL
SD_PWR O - 3.3V CMOS3.3 Off Off VOL VOL
2.8 High Speed UART Interface Signals

The SoC features two separate High Speed UARTs. However, only UART0 provides the
Modem Control pins DCD, DSR, DTR and RI.
See Chapter 18.0, “High Speed UART” for more details of the HSUART interface signals.
Table 13. High Speed UART Signals

SIU0_CTS_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SIU0_DCD_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SIU0_DSR_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SIU0_DTR_B O - 3.3V CMOS3.3 Off Off VOH VOH
SIU0_RI_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SIU0_RTS_B O - 3.3V CMOS3.3 Off Off VOH VOH
SIU0_RXD I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SIU0_TXD O - 3.3V CMOS3.3 Off Off VOH VOH
SIU1_CTS_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SIU1_RTS_B O - 3.3V CMOS3.3 Off Off VOH VOH
SIU1_RXD I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SIU1_TXD O - 3.3V CMOS3.3 Off Off VOH VOH
2.9 I2C* Interface Signals

See Chapter 19.0, “I2C* Controller/GPIO Controller” for more details of the I2C
Interface signals.
Table 14. I2C* Signals

I2C_DATA I/O Ext 3.3V CMOS3.3_OD Off Off Pull-up Pull-up
I2C_CLK I/O Ext 3.3V CMOS3.3_OD Off Off Pull-up Pull-up

2.10 Legacy Serial Peripheral Interface (SPI) Signals

See Section 21.7 for more details of the SPI signals.
Table 15. Legacy SPI Signals

LSPI_MOSI O - 3.3V CMOS3.3 Off Off VOH VOH
LSPI_MISO I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
LSPI_SS_B O - 3.3V CMOS3.3 Off Off VOH VOH
LSPI_SCK O - 3.3V CMOS3.3 Off Off VOL VOL
2.11 Serial Peripheral Interface (SPI)

See Chapter 20.0, “SPI Interface” for more details of the SPI signals.
Table 16. SPI Signals

SPI0_MOSI O - 3.3V CMOS3.3 Off Off VOL VOL
SPI0_MISO I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SPI0_SS_B O - 3.3V CMOS3.3 Off Off VOH VOH
SPI0_SCK O - 3.3V CMOS3.3 Off Off VOL VOL
SPI1_MOSI O - 3.3V CMOS3.3 Off Off VOL VOL
SPI1_MISO I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SPI1_SS_B O - 3.3V CMOS3.3 Off Off VOH VOH
SPI1_SCK O - 3.3V CMOS3.3 Off Off VOL VOL

2.12 Real Time Clock (RTC) Interface Signals

See Section 21.10 for more details of the RTC interface signals.
Table 17. Real Time Clock (RTC) Interface Signals

RTCX1 I/O - <1V Analog Running Running Running Running
RTCX2 I/O - <1V Analog Running Running Running Running
IVCCRTCEXT I/O - 1.5V Analog Analog Analog Analog Analog
RTCRST_B I - 3.3V CMOS3.3 ViH ViH ViH ViH
RTC_EXT_CLK_EN_B I - 3.3V CMOS3.3 ViH/ViL ViH/ViL ViH/ViL ViH/ViL
2.13 Power Management Signals

See Chapter 8.0, “Power Management” for more details of the Power Management
interface signals.
Table 18. Power Management Interface Signals

PWR_BTN_B I - 3.3V CMOS3.3 ViH ViH ViH ViH
RESET_BTN_B I 20k(H) 3.3V CMOS3.3 Off Pull-up Pull-up Pull-up
S5_PG I - 3.3V CMOS3.3 ViH ViH ViH ViH
S3_PG I - 3.3V CMOS3.3 ViL ViH ViH ViH
S0_PG I - 3.3V CMOS3.3 ViL ViL ViH ViH
S0_1P0_PG I - 3.3V CMOS3.3 ViL ViL ViH ViH
S3_3V3_EN O Ext 3.3V CMOS3.3 VOL VOH VOH VOH
S3_1V5_EN O Ext 3.3V CMOS3.3 VOL VOH VOH VOH
S0_3V3_EN O Ext 3.3V CMOS3.3 Pull-down VOL VOL VOiH
S0_1V5_EN O Ext 3.3V CMOS3.3 Pull-down VoL VoH VoH
S0_1P0_EN O Ext 3.3V CMOS3.3 Pull-down VoL VoH VoH
ODRAM_PWROK O Ext 3.3V CMOS3.3_OD Pull-up Pull-up Pull-up Pull-up
OSYSPWRGOOD O Ext 3.3V CMOS3.3_OD Pull-up Pull-up Pull-up Pull-up
VNNSENSE I/O - 1.05V Analog Off Off Analog Analog
VSSSENSE I/O - GND Analog Off Off Analog Analog
2.14 JTAG and Debug Interface Signals

See Chapter 1.0, “Debug Port and JTAG/TAP” for more details of the JTAG interface
signals.

Table 19. JTAG and Debug Interface Signals

TCK I 20k(L) 3.3V CMOS3.3 Pull-down Pull-down Pull-down Pull-down
TDI I 20k(H) 3.3V CMOS3.3 Pull-up Pull-up Pull-up Pull-up
TDO O Ext 3.3V CMOS3.3 Pull-up Pull-up Pull-up Pull-up
TMS I 20k(H) 3.3V CMOS3.3 Pull-up Pull-up Pull-up Pull-up
TRST_B I 20k(H) 3.3V CMOS3.3 Pull-up Pull-up Pull-up Pull-up
PRDY_B O - 3.3V CMOS3.3 Off Off VoH VoH
PREQ_B I 20k(H) 3.3V CMOS3.3 Pull-up Pull-up Pull-up Pull-up
2.15 Legacy Interface Signals

See Chapter 19.0, “I2C* Controller/GPIO Controller” and Chapter 21.0, “Legacy Bridge”
for more details of the legacy interface signals.
Table 20. Legacy Interface Signals

Unknown/ Unknown/
CLK14 I - 3.3V CMOS3.3 Off Off
Running Running
WAKE_B I 20k(H) 3.3V CMOS3.3_OD Off Pull-up Pull-up Pull-up
GPE_B I 20k(H) 3.3V CMOS3.3_OD Off Pull-up Pull-up Pull-up
THRM_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
SMI_B I 20k(H) 3.3V CMOS3.3 Off Off Pull-up Pull-up
2.16 General Purpose I/O Interface Signals

All GPIOs default to inputs. GPIO_SUS[5:0] are suspend well GPIOs and remain
available in S3. The default buffer state of these GPIOs while in S3 and when entering
S0 from S3 is configuration dependent.
Table 21. General Purpose I/O Signals

GPIO_SUS[5:0] I/O - 3.3V CMOS3.3 Off Unknown ViX ViX/Unknown
GPIO[9:0] I/O - 3.3V CMOS3.3 Off Off ViX ViX

2.17 Power And Ground Pins
Table 22. Power and Ground Pins (Sheet 1 of 2)

Nominal Lowest
Signal Name Description/Notes
Voltage Active State
OVOUT_1P0_S5 Unused output from internal LDO. Leave this pin No Connect
VCC1P0_S5 Standard 1.0V Rail for S5 Logic
VCCAICLKCB_1P0 ICLK Control Supply

1.0V
VCCAICLKDBUFF_1P0 ICLK Differential Output Buffer Supply
VCCAICLKSSC1_1P0 ICLK SSC Supply
VCCDICLKDIG_1P0 S5 ICLK Digital Supply
VCCAICLKSFR_1P5 1.5V ICLK SFR (for Oscillator, IPLL)
OVOUT_1P8_S5 S5 1.8V Rail Standby LDO Output

1.8V
VCC1P8_S5 S5 1.8V CFIO Supply
VCC3P3_S5 S5 3.3V Rail Standby LDO Input

3.3V
VCCAICLKSE_3P3 ICLK Single Ended Output Buffer Supply

1.0V
VCC1P0_S3 Standard 1.0V Rail for S3 logic
VCCCLKDDR_1P5 DDR IO Clock Analog Thick Gate Isolated Quiet Supply

1.5V
VCCDDR_1P5 DDR IO Analog Thick Gate Supply

S3
VCC1P8_S3 1.8V S3 1.8V CFIO Supply
VCCAUSB_1P8_S3 USB 1.8V Supply
3.3V USB 3.3V Supply

VCC3P3_USB_S3 The USB PHY resides in the S3 power domain. However, from a
functional point of view USB is only active in S0.

Table 22. Power and Ground Pins (Sheet 2 of 2)

Nominal Lowest
Signal Name Description/Notes
Voltage Active State
VCC1P0_S0 Standard 1.0V Rail for HPLL (Host PLL) and USB Logic
VCCACLKDDR_1P0 DDR IO Digital Clock Isolated Quiet Supply
VCCADDR_1P0 DDR IO Digital Supply
VCCADLLDDR_1P0 DDR IO Digital Isolated Quiet Supply

1.0V
VCCAPCIE_1P0 PCIe Analog Supply
VCCAVISA_1P0 VISA IO Analog Supply
VCCPLLDDR_1P0 DDR IO Digital PLL High Voltage
VNN Default 1.0V Standard Cell Rail including Core and Uncore Logic
VCCFHVSOC_1P05 1.05V SoC Fuses Supply

S0
VCCFSOC_1P05 Fuse Digital Sensing
VCC1P5_S0 PCIe Band-Gap Supply

1.5V
VCCSFRPLLDDR_1P5 DDR IO PLL High Voltage
OVOUT_1P8_SLDO S0 1.8V Rail Standby LDO Output (Currently Unused)
VCC1P8_S0 1.8V S0 1.8V CFIO Supply
VCCAA_1P8 1.8V Analog Supply
VCCAUSB_1P8 USB 1.8V Analog Supply
VCC3P3_A S0 3.3V Rail Standby LDO Input (Currently Unused)

3.3V
RTC Well Supply

To be active in G3, the source supply for VCCRTC3P3 must be a 3.0V
VCCRTC3P3 3.3V G3 coin cell battery or equivalent. If the target application does not
require the RTC to be operational in G3, VCCRTC3P3 should be
supplied from a 3.3V supply that is active in S5.
VSS Ground
0V -
VSSA_USB USB Low-Noise Ground
2.18 Hardware Straps

The pins used for hardware straps are output pins in functional mode.Initially during a
cold boot, these strap pins are configured as inputs. These pins remain inputs until the
external pull up or pull down values are sampled during S0 Power OK. Once sampled,
the pins are enabled as outputs only.

Table 23. Hardware Straps

Signal Name Default Strap Description
Reports the Strap status of Recovery Mode:

SPI0_MOSI 1b 0 = Recovery Mode
1 = Normal Mode
Defines Memory Device Density

00b = Reserved
{SPI0_SCK,
11b 01b = 1Gb
SPI1_MOSI}
10b = 2Gb
11b = 4Gb
Defines the Number of Ranks Enabled

LSPI_MOSI 1b 0b = 1 Rank
1b = 2 Ranks
Frequency SKU Power Optimize Mode

[2:1] CPU Clock/DDR Clock
00b = Reserved
{MAC0_TXDATA[1], 01b = 400MHz/800MHz
MAC0_TXDATA[0], 010b 10b = 200MHz/800MHz
MAC1_TXDATA[1] 11b = 100MHz/800MHz
[0]
0b = Low Latency
1b = Low Power
0b = FFF0_0000h
MAC1_TXDATA[0] 0b
1b = FFD0_0000h
SDIO Slot Type

00b = Removable Card Slot
{LSPI_SCK, SD_CLK} 00b 01b = Embedded Slot for One Device
10b = Shared Bus Slot
11b = Reserved
Power Button Disable

PWR_BTN_B 0b 0b = Power Button Disabled
1b = Power Button Enabled

§§

Ballout and Package Information—Intel® Quark™ SoC X1000
3.0 Ballout and Package Information
The Intel® Quark™ SoC X1000 package comes in a 373 ball, 15mm x 15mm FCBGA based on a
0.593 mm pitch.
3.1 Package Diagram
Figure 4. Intel® Quark™ SoC X1000 Package Dimensions

Intel® Quark™ SoC X1000—Ballout and Package Information
3.2 Ball Listings

Table 24. Alphabetical Ball Listing
C7 I2C_CLK E32 VSS

A2 RESERVED C10 SMI_B E35 VSS
A4 S0_PG C12 MAC1_TXEN F2 S0_1V5_EN
A5 WAKE_B C14 MAC1_RXDV F4 S0_1V0_EN
A7 VSS C16 RMII_REF_CLK F6 VSS
A10 VSS C18 MAC0_RXDV F30 VCC3P3_A
A12 VSS C21 MAC0_TXEN F32 VSS
A14 VSS C23 SD_DATA[0] F34 USBH0_PWR_EN
A16 VSS C26 SD_DATA[3] G1 VSS
A18 VSS C28 SD_DATA[7] G3 RTCX2
A21 MAC0_TXDATA[0] C35 VSS G7 S5_PG
A23 VSS D2 S0_1P0_PG G11 GPIO_SUS[0]
A26 SD_DATA[4] D5 VSS G12 VSS
A28 VSS D6 RESET_BTN_B G15 GPIO_SUS[4]
A31 VSS D9 CLK14 G17 SD_PWR
A32 VSS D11 MAC1_MDIO G19 SD_WP
A34 VSS D13 MAC1_RXDATA[0] G33 USBH1_PWR_EN
B1 RESERVED D15 MAC1_TXDATA[0] G35 USBH0_OC_B
B2 PREQ_B D17 MAC0_MDC H6 REF1_OUTCLK_N
B4 GPE_B D20 MAC0_RXDATA[0] H17 VSS
B6 I2C_DATA D22 MAC0_TXDATA[1] H30 VSSSENSE
B9 THRM_B D24 SD_DATA[1] J2 RTCX1
B11 MAC1_MDC D27 SD_DATA[5] J4 S3_PG
B13 MAC1_RXDATA[1] D29 VSS J6 REF1_OUTCLK_P
B15 MAC1_TXDATA[1] D31 VSS J7 RTCRST_B
B17 MAC0_MDIO D33 VSS J11 IVCCRTCEXT
B20 MAC0_RXDATA[1] E1 VCCRTC_3P3 J12 GPIO_SUS[2]
B22 SD_CMD E4 S0_3V3_EN J15 GPIO_SUS[5]
B24 SD_DATA[2] E7 PRDY_B J19 SD_CLK
B27 SD_DATA[6] RTC_EXT_- J25 VSS
E11
B29 VSS CLK_EN_B
J30 VNNSENSE
B31 VSS E12 GPIO_SUS[1]
J32 USBH1_OC_B
B33 VSS E15 GPIO_SUS[3]
J34 OUSBCOMP_P18
B35 VSS E17 SD_LED
K1 S3_1V5_EN
C1 VSS E19 SD_CD_B
K3 S3_3V3_EN

K11 OVOUT_1P8_S5 P11 VCCDICLKDIG_1P0 V9 VSS
K13 VCC1P8_S5 P13 VSS VCCAICLKD-

V11
BUFF_1P0
K14 VCC3P3_S3 P14 VCC1P0_S3
VCCAI-
K16 VCC3P3_USB_S3 P16 VSSA_USB V13
CLKSSC1_1P0
K18 VCC1P8_S3 P18 OVOUT_1P8_S0
V14 VCC1P0_S0
K20 VCCAUSB_1P8_S3 P20 VCC1P8_S0
V16 VCC1P0_S0
K22 VCCAA_1P8 P22 VCCAUSB_1P8
V18 VNN
K24 OVOUT_1P8_SLDO P24 OVOUT_1P05_S0
V20 VNN
K33 IUSBCOMP_N18 P27 VSS
V22 VSS
K35 VSS P29 USBH1_DP
V27 VSS
L2 PWR_BTN_B P30 USBH1_DN
V29 USBH0_DN
L4 TRST_B R1 TDO
V30 USBH0_DP
L32 RESERVED R3 ODRAM_PWROK
V32 GPIO[7]
L34 RESERVED R10 VCCAICLKSFR_1P5
V34 GPIO[8]
M1 VSS R26 VCCFSOC_1P05
W1 VSS
M3 TCK R33 GPIO[2]
W3 OSC_COMP
M5 VSS R35 GPIO[3]
W5 RESERVED
M6 REF0_OUTCLK_N T2 OSYSPWRGOOD
W7 CKSYS25OUT
M8 REF0_OUTCLK_P RMII_REF_-
T4 W9 TS_IREF_N
CLK_OUT
M11 VSS
W27 VSS
T11 VCCAICLKCB_1P0
M13 VSS
W29 SIU0_DTR_B
T13 VCC1P0_S5
M14 OVOUT_1P0_S3
W30 SIU0_RTS_B
T14 OVOUT_1P0_S5
M16 VSS
W33 GPIO[9]
T16 VSS
M18 OVOUT_1P8_S3
W35 SIU0_TXD
T18 VCCAVISA_1P0
M22 VSS
Y11 VSS
T20 VCC1P8_S0
M28 VSS
Y13 VSS
T22 VSS
M29 USBD_DP
Y14 VCCAPCIE_1P0
T24 VCCFHVSOC_1P05
M31 USBD_DN
Y16 VNN
T32 GPIO[4]
M33 USB_CLK96N
Y18 VNN
T34 GPIO[5]
M35 USB_CLK96P
Y20 VNN
U1 VSS
N2 TMS
Y22 VSS
U3 FLEX2_CLK
N4 TDI
AA2 XTAL_IN
U33 GPIO[6]
N32 GPIO[0]
AA4 RESERVED
U35 VSS
N34 GPIO[1]
AA10 RESERVED
V2 FLEX1_CLK
P5 VCC3P3_S5
AA26 VCC3P3_S0
V4 FLEX0_CLK
P7 VCCAICLKSE_3P3
AA32 SIU0_DSR_B
V5 HPLL_REFCLK_N
P9 VSS
AA34 SIU0_RXD
V7 HPLL_REFCLK_P

AB1 XTAL_OUT AD35 VSS AK2 PCIE_PERP_1
AB3 VSS AE2 PCIE_REFCLKP AK4 PCIE_PERN_1

AB11 VSS AE4 PCIE_REFCLKN AK7 DDR3_DQS[0]
AB13 VCCACLKDDR_1P0 AE10 RESERVED AK11 DDR3_DQ[1]
AB14 VCCADDR_1P0 AE26 VSS AK12 DDR3_DQ[7]

AB16 VSS AE32 SPI0_SCK AK15 VSS
AB18 VNN AE34 SPI0_SS_B AK17 RESERVED

AB20 VNN AF11 RESERVED AK19 RESERVED
AB24 VCC3P3_S0 AF13 VSS AK23 DDR3_CMDPU

AB33 SIU1_CTS_B AF20 VSS DDR3_IDRAM_P-
AK25
WROK
AB35 VSS AF22 VSS
AK29 VSS
AC2 VSS AF24 VSS
AK32 LSPI_MOSI
AC4 PAD_BYPASS_CLK AG1 VSS
AK34 SPI1_MISO
AC5 TS_TDC AG3 VSS
AL6 VSS
AC7 TS_TDA AG6 DDR3_DQ[4]
AL11 DDR3_DQ[0]
AC9 VSS AG30 DDR3_CK[1]
AL12 DDR3_DM[0]
AC27 SIU0_DCD_B AG33 SPI0_MISO
AL15 DDR3_DQ[2]
AC29 VSS AG35 SPI0_MOSI
AL17 RESERVED
AC30 VSS AH2 PCIE_PETN_1
AL19 DDR3_ODTPU
AC32 SIU1_TXD AH4 PCIE_PETP_1
AL23 VSS
AC34 SIU1_RXD AH7 DDR3_DQSB[0]
AL25 DDR3_DRAMRSTB
AD1 PCIE_PETN_0 AH11 VSS
AL30 DDR3_CKB[0]
AD3 PCIE_PETP_0 AH12 DDR3_DQ[6]
AM1 VSS
AD5 VCC1P5_S0 AH15 DDR3_DQ[3]
AM2 PCIE_PERP_0
AD6 VSS AH17 VSS
AM4 VSS
AD8 VSS AH19 RESERVED
AM7 VSS
AD11 RESERVED AH23 DDR3_DQPU
AM29 DDR3_VREF
AD13 VCCPLLDDR_1P0 DDR3_ISYSPWR-
AH25
GOOD AM32 VSS
AD14 VCCADLLDDR_1P0
AH29 DDR3_CKB[1] AM33 LSPI_SS_B
AD16 VSS
AH32 SPI1_MOSI AM35 LSPI_SCK
AD18 VSS
AH34 SPI1_SCK AN1 PCIE_PERN_0
AD20 VSS
AJ1 PCIE_IRCOMP AN5 DDR3_DQ[13]
AD24 VCC3P3_S0
AJ3 PCIE_RBIAS AN6 DDR3_DQS[1]
AD28 SIU0_RI_B
AJ6 DDR3_DQ[5] AN9 VSS
AD29 SIU0_CTS_B
AJ30 DDR3_CK[0] AN11 DDR3_CKE[0]
VCCSFRPLLD-
AD31
DR_1P5 AJ33 SPI1_SS_B AN13 VSS
AD33 SIU1_RTS_B AJ35 VSS AN15 DDR3_MA[11]

AN17 DDR3_MA[7] AT5 DDR3_DQ[12]
AN20 DDR3_MA[5] AT7 VSS

AN22 DDR3_MA[1] AT10 DDR3_DQ[14]
AN24 DDR3_MA[10] AT12 VCCDDR_1P5
AN27 VSS AT14 DDR3_MA[15]

AN29 DDR3_CASB AT16 VCCDDR_1P5
AN31 DDR3_CSB[0] AT18 VCCDDR_1P5

AN35 LSPI_MISO AT21 DDR3_MA[4]
AP2 VSS AT23 VCCDDR_1P5

AP4 DDR3_DQ[9] AT26 DDR3_BS[1]
AP7 DDR3_DM[1] AT28 VCCCLKDDR_1P5

AP10 DDR3_DQ[11] AT31 DDR3_ODT[0]
AP12 DDR3_CKE[1] AT32 DDR3_ODT[1]
AP14 DDR3_MA[14] AT34 RESERVED
AP16 DDR3_MA[9] NOTE: The following balls are
No Connect (NC):
AP18 VSS M26
AP21 DDR3_MA[3] E25
J23
AP23 VSS
M24
AP26 DDR3_BS[0] Y24
AP28 DDR3_WEB G25
AF16
AP31 DDR3_CSB[1]
G23
AP33 VSS M20
AR1 RESERVED AD22
AF14
AR6 DDR3_DQSB[1]
AB22
AR9 DDR3_DQ[10] V24
AR11 DDR3_DQ[15] E23
G29
AR13 DDR3_BS[2]
AF18
AR15 DDR3_MA[12] J29
AR17 DDR3_MA[8] E29
AR20 DDR3_MA[6]
AR22 DDR3_MA[2]
AR24 DDR3_MA[0]
AR27 DDR3_RASB
AR29 DDR3_MA[13]
AR35 RESERVED
AT2 RESERVED
AT3 DDR3_DQ[8]

Table 25. Alphabetical Signal Listing
DDR3_DQSB[0] AH7 GPIO_SUS[5] J15

CKSYS25OUT W7
DDR3_DQSB[1] AR6 GPIO[0] N32
CLK14 D9
DDR3_DRAMRSTB AL25 GPIO[1] N34
DDR3_BS[0] AP26
DDR3_IDRAM_PWROK AK25 GPIO[2] R33
DDR3_BS[1] AT26
DDR3_ISYSPWR- GPIO[3] R35
AH25
DDR3_BS[2] AR13 GOOD
GPIO[4] T32
DDR3_CASB AN29 DDR3_MA[0] AR24
GPIO[5] T34
DDR3_CK[0] AJ30 DDR3_MA[1] AN22
GPIO[6] U33
DDR3_CK[1] AG30 DDR3_MA[10] AN24
GPIO[7] V32
DDR3_CKB[0] AL30 DDR3_MA[11] AN15
GPIO[8] V34
DDR3_CKB[1] AH29 DDR3_MA[12] AR15
GPIO[9] W33
DDR3_CKE[0] AN11 DDR3_MA[13] AR29
HPLL_REFCLK_N V5
DDR3_CKE[1] AP12 DDR3_MA[14] AP14
HPLL_REFCLK_P V7
DDR3_CMDPU AK23 DDR3_MA[15] AT14
I2C_CLK C7
DDR3_CSB[0] AN31 DDR3_MA[2] AR22
I2C_DATA B6
DDR3_CSB[1] AP31 DDR3_MA[3] AP21
IUSBCOMP_N18 K33
DDR3_DM[0] AL12 DDR3_MA[4] AT21
IVCCRTCEXT J11
DDR3_DM[1] AP7 DDR3_MA[5] AN20
LSPI_MISO AN35
DDR3_DQ[0] AL11 DDR3_MA[6] AR20
LSPI_MOSI AK32
DDR3_DQ[1] AK11 DDR3_MA[7] AN17
LSPI_SCK AM35
DDR3_DQ[10] AR9 DDR3_MA[8] AR17
LSPI_SS_B AM33
DDR3_DQ[11] AP10 DDR3_MA[9] AP16
MAC0_MDC D17
DDR3_DQ[12] AT5 DDR3_ODT[0] AT31
MAC0_MDIO B17
DDR3_DQ[13] AN5 DDR3_ODT[1] AT32
MAC0_RXDATA[0] D20
DDR3_DQ[14] AT10 DDR3_ODTPU AL19
MAC0_RXDATA[1] B20
DDR3_DQ[15] AR11 DDR3_RASB AR27
MAC0_RXDV C18
DDR3_DQ[2] AL15 DDR3_VREF AM29
MAC0_TXDATA[0] A21
DDR3_DQ[3] AH15 DDR3_WEB AP28
MAC0_TXDATA[1] D22
DDR3_DQ[4] AG6 FLEX0_CLK V4
MAC0_TXEN C21
DDR3_DQ[5] AJ6 FLEX1_CLK V2
MAC1_MDC B11
DDR3_DQ[6] AH12 FLEX2_CLK U3
MAC1_MDIO D11
DDR3_DQ[7] AK12 GPE_B B4
MAC1_RXDATA[0] D13
DDR3_DQ[8] AT3 GPIO_SUS[0] G11
MAC1_RXDATA[1] B13
DDR3_DQ[9] AP4 GPIO_SUS[1] E12
MAC1_RXDV C14
DDR3_DQPU AH23 GPIO_SUS[2] J12
MAC1_TXDATA[0] D15
DDR3_DQS[0] AK7 GPIO_SUS[3] E15
MAC1_TXDATA[1] B15
DDR3_DQS[1] AN6 GPIO_SUS[4] G15

MAC1_TXEN C12 RESERVED AK17 SD_DATA[5] D27
ODRAM_PWROK R3 RESERVED AT2 SD_DATA[6] B27

OSC_COMP W3 RESERVED AR1 SD_DATA[7] C28
OSYSPWRGOOD T2 RESERVED AD11 SD_LED E17
OUSBCOMP_P18 J34 RESERVED AF11 SD_PWR G17

OVOUT_1P0_S3 M14 RESERVED A2 SD_WP G19
OVOUT_1P0_S5 T14 RESERVED B1 SIU0_CTS_B AD29

OVOUT_1P05_S0 P24 RESERVED AA10 SIU0_DCD_B AC27
OVOUT_1P8_S0 P18 RESERVED AE10 SIU0_DSR_B AA32

OVOUT_1P8_S3 M18 RESERVED AA4 SIU0_DTR_B W29
OVOUT_1P8_S5 K11 RESERVED W5 SIU0_RI_B AD28

OVOUT_1P8_SLDO K24 RESERVED L32 SIU0_RTS_B W30
PAD_BYPASS_CLK AC4 RESERVED L34 SIU0_RXD AA34
PCIE_IRCOMP AJ1 RESET_BTN_B D6 SIU0_TXD W35
PCIE_PERN_0 AN1 RMII_REF_CLK C16 SIU1_CTS_B AB33
PCIE_PERN_1 AK4 RMII_REF_CLK_OUT T4 SIU1_RTS_B AD33
PCIE_PERP_0 AM2 RTC_EXT_CLK_EN_B E11 SIU1_RXD AC34

PCIE_PERP_1 AK2 RTCRST_B J7 SIU1_TXD AC32
PCIE_PETN_0 AD1 RTCX1 J2 SMI_B C10
PCIE_PETN_1 AH2 RTCX2 G3 SPI0_MISO AG33
PCIE_PETP_0 AD3 S0_1P0_PG D2 SPI0_MOSI AG35
PCIE_PETP_1 AH4 S0_1V0_EN F4 SPI0_SCK AE32

PCIE_RBIAS AJ3 S0_1V5_EN F2 SPI0_SS_B AE34
PCIE_REFCLKN AE4 S0_3V3_EN E4 SPI1_MISO AK34
PCIE_REFCLKP AE2 S0_PG A4 SPI1_MOSI AH32
PRDY_B E7 S3_1V5_EN K1 SPI1_SCK AH34

PREQ_B B2 S3_3V3_EN K3 SPI1_SS_B AJ33
PWR_BTN_B L2 S3_PG J4 TCK M3

REF0_OUTCLK_N M6 S5_PG G7 TDI N4
REF0_OUTCLK_P M8 SD_CD_B E19 TDO R1

REF1_OUTCLK_N H6 SD_CLK J19 THRM_B B9
REF1_OUTCLK_P J6 SD_CMD B22 TMS N2

RESERVED AR35 SD_DATA[0] C23 TRST_B L4
RESERVED AT34 SD_DATA[1] D24 TS_IREF_N W9

RESERVED AH19 SD_DATA[2] B24 TS_TDA AC7
RESERVED AK19 SD_DATA[3] C26 TS_TDC AC5
RESERVED AL17 SD_DATA[4] A26 USB_CLK96N M33

USB_CLK96P M35 VCCAUSB_1P8 P22 VSS C1
USBD_DN M31 VCCAUSB_1P8_S3 K20 VSS C35

USBD_DP M29 VCCAVISA_1P0 T18 VSS D5
USBH0_DN V29 VCCCLKDDR_1P5 AT28 VSS D29
USBH0_DP V30 VCCDDR_1P5 AT12 VSS D31

USBH0_OC_B G35 VCCDDR_1P5 AT16 VSS D33
USBH0_PWR_EN F34 VCCDDR_1P5 AT18 VSS E32

USBH1_DN P30 VCCDDR_1P5 AT23 VSS E35
USBH1_DP P29 VCCDICLKDIG_1P0 P11 VSS F6

USBH1_OC_B J32 VCCFHVSOC_1P05 T24 VSS F32
USBH1_PWR_EN G33 VCCFSOC_1P05 R26 VSS G1

VCC1P0_S0 V14 VCCPLLDDR_1P0 AD13 VSS G12
VCC1P0_S0 V16 VCCRTC_3P3 E1 VSS H17
VCC1P0_S3 P14 VCCSFRPLLDDR_1P5 AD31 VSS J25
VCC1P0_S5 T13 VNN V18 VSS K35
VCC1P5_S0 AD5 VNN V20 VSS M1
VCC1P8_S0 P20 VNN Y16 VSS M5

VCC1P8_S0 T20 VNN Y18 VSS M11
VCC1P8_S3 K18 VNN Y20 VSS M13
VCC1P8_S5 K13 VNN AB18 VSS M16
VCC3P3_A F30 VNN AB20 VSS M22
VCC3P3_S0 AA26 VNNSENSE J30 VSS M28

VCC3P3_S0 AB24 VSS A7 VSS P9
VCC3P3_S0 AD24 VSS A10 VSS P13
VCC3P3_S3 K14 VSS A12 VSS P27
VCC3P3_S5 P5 VSS A14 VSS T16

VCC3P3_USB_S3 K16 VSS A16 VSS T22
VCCAA_1P8 K22 VSS A18 VSS U1

VCCACLKDDR_1P0 AB13 VSS A23 VSS U35
VCCADDR_1P0 AB14 VSS A28 VSS V9

VCCADLLDDR_1P0 AD14 VSS A31 VSS V22
VCCAICLKCB_1P0 T11 VSS A32 VSS V27

VCCAICLKDBUFF_1P0 V11 VSS A34 VSS W1
VCCAICLKSE_3P3 P7 VSS B29 VSS W27

VCCAICLKSFR_1P5 R10 VSS B31 VSS Y11
VCCAICLKSSC1_1P0 V13 VSS B33 VSS Y13
VCCAPCIE_1P0 Y14 VSS B35 VSS Y22

VSS AB3 VSS AP23
VSS AB11 VSS AP33

VSS AB16 VSS AT7
VSS AB35 VSSA_USB P16
VSS AC2 VSSSENSE H30

VSS AC9 WAKE_B A5
VSS AC29 XTAL_IN AA2

VSS AC30 XTAL_OUT AB1
VSS AD6
§§
VSS AD8
VSS AD16
VSS AD18
VSS AD20
VSS AD35
VSS AE26
VSS AF13
VSS AF20
VSS AF22
VSS AF24
VSS AG1
VSS AG3
VSS AH11
VSS AH17
VSS AJ35
VSS AK15
VSS AK29
VSS AL6
VSS AL23
VSS AM1
VSS AM4
VSS AM7
VSS AM32
VSS AN9
VSS AN13
VSS AN27
VSS AP2
VSS AP18


Electrical Characteristics—Intel® Quark™ SoC X1000
4.0 Electrical Characteristics
This chapter contains the DC and AC characteristics for Intel® Quark™ SoC X1000. AC
timing diagrams are included.
4.1 Absolute Maximum Ratings

Table 26 specifies the absolute maximum and minimum ratings of the Intel® Quark™
SoC X1000 processor. At conditions outside of the functional operating condition limits,
but within the absolute maximum and minimum ratings, neither functionality nor long-
term reliability can be expected. If a device is returned to conditions within the
functional operating limits after having been subjected to conditions outside these
limits (but within the absolute maximum and minimum ratings) the device may be
functional, but with its lifetime degraded depending on exposure to conditions
exceeding the functional operating condition limits.
At conditions exceeding the absolute maximum and minimum ratings, neither

functionality nor long-term reliability can be expected. Moreover, if a device is
subjected to these conditions for any length of time, it will either not function or its
reliability will be severely degraded when returned to conditions within the functional
operating condition limits.
Although the SoC contains protective circuitry to resist damage from Electrostatic
Discharge (ESD), precautions should always be taken to avoid high static voltages or
electric fields.
All voltage values are given with respect to VSS.
Table 26. Intel® Quark™ SoC X1000 Absolute Maximum Voltage Ratings (Sheet 1 of 2)
Minimum Maximum
Parameter
Limits Limits
Temperature
Junction Temperature (C) 0 110
Storage Temperature Range (C) -55 125
Supplies
3.3V Supply Voltage (V) — 3.7
Signals
Voltage on any 3.3 V Pin (V) — 3.7 V

Intel® Quark™ SoC X1000—Electrical Characteristics
Table 26. Intel® Quark™ SoC X1000 Absolute Maximum Voltage Ratings (Sheet 2 of 2)
Minimum Maximum
Parameter
Limits Limits
Voltage on any 1.8 V Pin (V) — 2.0
Voltage on any 1.5 V Pin (V) — 1.65
Voltage on any 1.0V Tolerant Pin (V) — 1.3
4.2 Recommended Power Supply Ranges

Table 27 shows the recommended operating voltage ranges for each SoC supply pin.
Typically total tolerance (DC+AC +Ripple) is +/- 5% of the nominal value unless
otherwise stated.
Table 27. Power Supply Rail Ranges (Sheet 1 of 2)

Package Ball Description Min (V) Nom (V) Max (V) Notes
VCC3P3_S5 S5 3.3V rail Standby LDO input 3.20 3.30 3.40 +/-3%
VCC3P3_A S0 3.3V rail Standby LDO input 3.20 3.30 3.40 +/-3%
Default 1.0V standard cell rail including Core Active in state

VNN 0.95 1.00 1.10
and Uncore logic S0 only
VCC1P0_S3 Standard 1.0V rail for S3 logic 0.95 1.00 1.05 +/-5%
VCC1P0_S5 Standard 1.0V rail for S5 logic 0.95 1.00 1.05 +/-5%
VCCPLLDDR_1P0 DDR IO digital PLL high voltage 0.95 1.00 1.05 +/-5%
VCCADLLDDR_1P0 DDR IO digital isolated quiet supply 0.95 1.00 1.05 +/-5%
VCCACLKDDR_1P0 DDR IO digital clock isolated quiet supply 0.95 1.00 1.05 +/-5%
VCCADDR_1P0 DDR IO Digital supply 0.95 1.00 1.05 +/-5%
VCCDDR_1P5 DDR IO analog thick gate supply 1.42 1.50 1.57 +/-5%
VCCSFRPLLDDR_1P5 DDR IO PLL high voltage 1.42 1.50 1.57 +/-5%
DDR IO clock analog thick gate isolated quiet

VCCCLKDDR_1P5 1.42 1.50 1.57 +/-5%
supply
VCCAPCIE_1P0 PCIe analog supply 0.95 1.00 1.05 +/-5%
VCC1P5_S0 PCIe band-gap supply 1.42 1.50 1.57 +/-5%
VCC1P8_S0 S0 1.8V CFIO supply 1.71 1.80 1.89 +/-5%
VCCRTC_3P3 RTC well supply 2.00 - 3.40 +/-5%
VCCAUSB_1P8_S3 USB 1.8V analog supply - suspend rail 1.71 1.80 1.89 +/-5%
VCC3P3_USB_S3 USB 3.3V supply - suspend rail 3.13 3.30 3.46 +/-5%
VCCAUSB_1P8 USB 1.8V supply 1.71 1.80 1.89 +/-5%
VSSA_USB USB low-noise ground 0.00 0.00 0.00 +/-5%
VCCAA_1P8 1.8V analog supply 1.71 1.80 1.89 +/-5%
Active in S0-
only; can in
Standard 1.0V rail for HPLL (host PLL) and
VCC1P0_S0 0.95 1.00 1.05 general be
USB logic
connected to
VNN

Table 27. Power Supply Rail Ranges (Sheet 2 of 2)

Package Ball Description Min (V) Nom (V) Max (V) Notes
VCCAVISA_1P0 VISA IO analog supply 0.95 1.00 1.05 +/-5%
VCCFHVSOC_1P05 SoC Fuses supply (sensing) 1.0 1.05 1.10 +/-5%
VCCFSOC_1P05 Fuse digital sensing 1.0 1.05 1.10 +/-5%
VCCAICLKCB_1P0 ICLK control supply 0.95 1.00 1.05 +/-5%
VCCAICLKSSC1_1P0 ICLK SSC supply 0.95 1.00 1.05 +/-5%
VCCAICLKDBUFF_1P0 ICLK differential output buffer supply 0.95 1.00 1.05 +/-5%
VCCDICLKDIG_1P0 ICLK digital supply 0.95 1.00 1.05 +/-5%
VCCAICLKSFR_1P5 ICLK SFR (for oscillator, IPLL) 1.42 1.50 1.57 +/-5%
VCCAICLKSE_3P3 ICLK single ended output buffer supply 3.13 3.30 3.46 +/-5%
4.3 Maximum Supply Current
Table 28. Maximum Supply Current: ICC Max (Sheet 1 of 2)

Voltage
Voltage Rail ICCMax (mA)
(V)
VCC3P3_S5 3.3 180
VCC3P3_S3 3.3 280
VCC3P3_S0 3.3 680
VCC3P3_A 3.3 5
VNN 1.0 480
VCC1P0_S3 1.0 70
VCC1P0_S5 1.0 40
VCCPLLDDR_1P0 1.0 25
VCCADLLDDR_1P0 1.0 100
VCCACLKDDR_1P0 1.0 15
VCCADDR_1P0 1.0 300
VCCDDR_1P5 1.5 580
VCCSFRPLLDDR_1P5 1.5 35
VCCCLKDDR_1P5 1.5 90
VCCAPCIE_1P0 1.0 330
VCC1P5_S0 1.5 120
VCC1P8_S0 1.8 50
VCC1P8_S3 1.8 110
VCC1P8_S5 1.8 15
VCCRTC_3P3 3.3 3
VCCRTC_3P3 (Battery) 3.0 6uA
VCCAUSB_1P8_S3 1.8 20
VCC3P3_USB_S3 3.3 40
VSSA_USB 0.0 20

Table 28. Maximum Supply Current: ICC Max (Sheet 2 of 2)

Voltage
Voltage Rail ICCMax (mA)
(V)
VCCAA_1P8 1.8 10
VCC1P0_S0 1.0 120
VCCAVISA_1P0 1.0 280
VCCFHVSOC_1P05 1.05 100
VCCFSOC_1P05 1.05 40
VCCAICLKCB_1P0 1.0 40
VCCAICLKSSC1_1P0 1.0 15
VCCAICLKDBUFF_1P0 1.0 30
VCCDICLKDIG_1P0 1.0 15
VCCAICLKSFR_1P5 1.5 40
VCCAICLKSE_3P3 3.3 15
VCCAUSB_1P8 1.8 20
4.4 Configurable IO Characteristics

The signals in Table 30 are brought on- and off-chip via standard configurable IO
groups. This section describes the DC and AC characteristics associated with these
signals.
Table 30. Configurable IO (CFIO) Bi-directional Signal Groupings

Related Supply
Group Name Interfaces Signals
(VCC)
SPI0_MOSI, SPI0_MISO, SPI0_SS_B, SPI0_SCK,

SPI
S0 CFIO Group 0 VCCCFIO_0_3P3 SPI1_MOSI, SPI1_MISO, SPI1_SS_B, SPI1_SCK
Legacy SPI
LSPI_MOSI, LSPI_MISO, LSPI_SS_B, LSPI_SCK
SIU0_CTS_B, SIU0_DCD_B, SIU0_DSR_B, SIU0_DTR_B,

S0 CFIO Group 1 UART VCCCFIO_1_3P3 SIU0_RI_B, SIU0_RTS_B, SIU0_RXD, SIU0_TXD,
SIU1_CTS_B, SIU1_RTS_B, SIU1_RXD, SIU1_TXD
USB_OC_0, USB_OC_1, USB_PWR_EN[0], USB_PWR_EN[1]

USB
S0 CFIO Group 2 VCCCFIO_2_3P3 GPIO[0], GPIO[1], GPIO[2], GPIO[3], GPIO[4], GPIO[5],
GPIO
GPIO[6], GPIO[7], GPIO[8], GPIO[9]
SD_DATA[0], SD_DATA[1], SD_DATA[2], SD_DATA[3],

S0 CFIO Group 3 SDIO VCCCFIO_3_3P3 SD_DATA[4], SD_DATA[5], SD_DATA[6], SD_DATA[7],
SD_CMD , SD_CLK, SD_WP, SD_CD_B,
SD_LED, SD_PWR_B
SDIO MAC0_TXDATA[1], MAC0_TXDATA[0], MAC0_RXDV,
S0 CFIO Group 4 VCCCFIO_4_3P3 MAC0_RXDATA[1], MAC0_RXDATA[0], MAC1_TXDATA[1],
Ethernet MAC
MAC1_TXDATA[0], MAC1_RXDV, MAC1_RXDATA[1],
MAC1_RXDATA[0]

Table 30. Configurable IO (CFIO) Bi-directional Signal Groupings

Related Supply
Group Name Interfaces Signals
(VCC)
MAC0_TXEN, MAC0_MDC, MAC0_MDIO, MAC1_TXEN,

Ethernet MAC
MAC1_MDC, MAC1_MDIO
S0 CFIO Group 5 I2C VCCCFIO_5_3P3
I2C_DATA, I2C_CLK
Legacy
THRM_B, SMI_B, RMII_REF_CLK, CLK14
RESET_BTN_B, WAKE_B, GPE_B

Legacy
S0_3V3_EN, S0_1V5_EN, S0_1V0_EN
S3 CFIO Group Power Management VCCCFIO_S3_3P3
GPIO_SUS[0], GPIO_SUS[1], GPIO_SUS[2], GPIO_SUS[3],
Suspend GPIOs
GPIO_SUS[4], GPIO_SUS[5]
TCK, TDI, TDO, TMS, TRST_B

JTAG/Debug
S5 CFIO Group VCCCFIO_S5_3P3 PWR_BTN_B, S3_3V3_EN, S3_1V5_EN, ODRAM_PWROK,
Power Management
OSYSPWRGOOD, PRDY_B, PREQ_B
Table 31. CFIO DC Characteristics

Type Symbol Parameter Min Max Unit Condition Notes
This data applies to all signals in Table 30.

To determine the correct VCC supply for a signal use the related supply shown in Table 30
VCC Supply Voltage Reference 3.13 3.49 V
VIH Input High Voltage 0.625 X VCC VCC + 0.3 V 2
VIL Input Low Voltage — 0.25 X VCC V 2
Input
IIL Input Leakage Current — 35 uA
CIN Input Pin Capacitance — 10 pF
VOH Output High Voltage 0.75 x VCC — V Iout=2mA 1, 2
Output
VOL Output Low Voltage — 0.125 x VCC V Iout=-2mA 2
Notes:
1. The VOH specification does not apply to open-collector or open-drain drivers. Signals of this type must have an external
pull-up resistor, and that’s what determines the high-output voltage level. Refer to Chapter 2 for details on signal types.
2. Input characteristics apply when a signal is configured as Input or to signals that are only Inputs. Output characteristics
apply when a signal is configured as an Output or to signals that are only Outputs. Refer to Chapter 2 for details on
signal types.
Table 32. CFIO AC Characteristics

Type Symbol Parameter Min Max Units Conditions
This data applies to all signals in Table 30.

To determine the correct VCC supply for a signal use the related supply shown in Table 30
VCC Supply Voltage Reference 3.13 3.49 V
SRRISE Slew Rate Rise 0.5 3.0 V/ns CLOAD = 10pF
SRFALL Slew Rate Fall 0.5 3.0 V/ns CLOAD = 10pF
Output 0.10*VCC -
TRISE Output Rise Time 0.88 5.28 ns
0.90*VCC
0.90*VCC -
TFALL Output Fall Time 0.88 5.28 ns
0.10*VCC

4.5 RTC DC Characteristics
Table 33. RTC DC Characteristics

Type Symbol Parameter Min Max Unit Condition Notes
Associated Signals: RTCRST_B

Related Supply (VCC): VCCRTC_3P3
VCC Supply Voltage Reference 2.00 3.40 V N/A
VIH Input High Voltage .7 x VCC VCC + 0.5 V N/A 2

Input
VIL Input Low Voltage -0.5 0.78 V N/A 2
CIN Input Pin Capacitance — 3 pF N/A
Associated Signals: S5_PG, RTC_EXT_CLK_EN_B, S0_PG

Related Supply (VCC): VCCRTC_3P3
VCC Supply Voltage Reference 2.0 3.40 V N/A
VIH Input High Voltage 2.0 VCC + 0.5 V N/A 2

Input
VIL Input Low Voltage -0.5 0.78 V N/A 2
CIN Input Pin Capacitance — 3 pF N/A
4.6 PCI Express* 2.0 DC/AC Characteristics
Table 34. PCI Express* 2.0 Differential Signal DC Characteristics

Symbol Parameter Min Max Unit Figures Notes
Associated Signals:
PCIE_PERN[1], PCIE_PERP[1],
PCIE_PETN[1], PCIE_PETP[1],
PCIE_PERN[0], PCIE_PERP[0],
PCIE_PETN[0], PCIE_PETP[0]
Related Supply (VCC):
VCCAPCIE_1P0
VTX-DIFF P-P Differential Peak to Peak Output Voltage 0.8 1.2 V 1
Low power differential Peak to Peak Output

VTX-DIFF P-P - Low 0.4 1.2 V
Voltage
TX AC Common Mode Output Voltage (2.5GT/

VTX_CM-ACp — 20 mV
s)
ZTX-DIFF-DC DC Differential TX Impedance 80 120 
VRX-DIFF p-p Differential Input Peak to Peak Voltage 0.175 1.2 V 1
VRX_CM-ACp AC peak Common Mode Input Voltage — 150 mV
Notes:
1. PCI Express mVdiff p-p = 2*|PETP[x] – PETN[x]|; PCI Express mVdiff p-p = 2*|PERP[x] – PERN[x]|

Table 35. PCI Express* 2.0 Interface Timings

Symbol Parameter Min Max Unit Figures Notes
Transmitter and Receiver Timings
UI Unit Interval – PCI Express* 399.88 400.12 ps 5,6
TTX-EYE Minimum Transmission Eye Width 0.7 — UI 5 1,2,6
TTX-RISE/Fall D+/D- TX Out put Rise/Fall time 0.125 UI 1,2,6
TRX-EYE Minimum Receiver Eye Width 0.40 — UI 6 3,4,6
Notes:
1. Specified at the measurement point into a timing and voltage compliance test load and measured over any 250
consecutive TX UIs. (Also refer to the Transmitter compliance eye diagram)
2. A TTX-EYE = 0.70 UI provides for a total sum of deterministic and random jitter budget of TTXJITTER-MAX = 0.30 UI for the
Transmitter collected over any 250 consecutive TX UIs. The TTXEYE-MEDIAN-to-MAX-JITTER specification ensures a jitter
distribution in which the median and the maximum deviation from the median is less than half of the total TX jitter
budget collected over any 250 consecutive TX UIs. It should be noted that the median is not the same as the mean. The
jitter median describes the point in time where the number of jitter points on either side is approximately equal as
opposed to the averaged time value.
3. Specified at the measurement point and measured over any 250 consecutive UIs. The test load documented in the PCI
Express* specification 2.0 should be used as the RX device when taking measurements (also refer to the Receiver
compliance eye diagram). If the clocks to the RX and TX are not derived from the same reference clock, the TX UI
recovered from 3500 consecutive UI must be used as a reference for the eye diagram.
4. A TRX-EYE = 0.40 UI provides for a total sum of 0.60 UI deterministic and random jitter budget for the Transmitter and
interconnect collected any 250 consecutive UIs. The TRX-EYE-MEDIAN-to--MAX-JITTER specification ensures a jitter
distribution in which the median and the maximum deviation from the median is less than half of the total 0.6 UI jitter
budget collected over any 250 consecutive TX UIs. It should be noted that the median is not the same as the mean. The
jitter median describes the point in time where the number of jitter points on either side is approximately equal as
opposed to the averaged time value. If the clocks to the RX and TX are not derived from the same reference clock, the
TX UI recovered from 3500 consecutive UI must be used as the reference for the eye diagram.
5. Nominal Unit Interval is 400 ps for 2.5 GT/s.
6. Intel® Quark™ SoC X1000 supports PCI Gen 1 timing only: 2.5 GT/s

Figure 5. PCI Express Transmitter Eye
Figure 6. PCI Express Receiver Eye
VTS-Diff = 0mV
D+/D- Crossing point
VRS-Diffp-p-Min>175mV
.4 UI =TRX-EYE min

4.7 USB 2.0 DC/AC Characteristics
Table 36. USB 2.0 Differential Signal DC Characteristics

Symbol Parameter Min Max Unit Conditions Notes
Associated Signals:
USBD_DP, USBD_DN,
USBH1_DP, USBH1_DN,
USBH0_DP, USBH0_DN,
USBH(1:0)_OC_B, USBH(1:0)_PWR_EN
Related Supply (VCC):
VCCUSBSUS_3P3
VDI Differential Input Sensitivity 0.2 — V N/A 1,3
VCM Differential Common Mode Range 0.8 2.5 V N/A 2,3
VSE Single-Ended Receiver Threshold 0.8 2 V N/A 3
VCRS Output Signal Crossover Voltage 1.3 2 V N/A 3
VOL Output Low Voltage — 0.4 V Iol=5 mA 3
VOH Output High Voltage VCC – 0.5 — V Ioh=-2mA 3
VHSSQ HS Squelch Detection Threshold 100 150 mV N/A 4
VHSDSC HS Disconnect Detection Threshold 525 625 mV N/A 4
VHSCM HS Data Signaling Common Mode Voltage Range -50 500 mV N/A 4
VHSOI HS Idle Level -10 10 mV N/A 4
VHSOH HS Data Signaling High 360 440 mV N/A 4
VHSOL HS Data Signaling Low -10 10 mV N/A 4
VCHIRPJ Chirp J Level 700 1100 mV N/A 4
VCHIRPK Chirp K Level -900 -500 mV N/A 4
Notes:
1. VDI = | USBHx_DP – USBHx_DN |
2. Includes VDI range
3. Applies to Low-Speed/Full-Speed USB
4. Applies to High-Speed USB 2.0
Table 37. USB 2.0 Interface Timings (Sheet 1 of 2)

Symbol Parameter Min Max Units Notes Fig
Full-speed Source (Note 7)
t100 USBHx_DP, USBHx_DN - Driver Rise Time 4 20 ns 1,6 CL = 50 pF 8
t101 USBHx_DP, USBHx_DN Driver Fall Time 4 20 ns 1,6 CL = 50 pF 8
Source Differential Driver Jitter

–3.5 3.5 ns
t102 - To Next Transition 2, 3 9
–4 4 ns
- For Paired Transitions
t103 Source SE0 interval of EOP 160 175 ns 4 10
Source Jitter for Differential Transition to SE0

t104 –2 5 ns 5
Transition
Receiver Data Jitter Tolerance

–18.5 18.5 ns
t105 - T o Next Transition 3 9
–9 9 ns

Table 37. USB 2.0 Interface Timings (Sheet 2 of 2)

Symbol Parameter Min Max Units Notes Fig
Full-speed Source (Note 7)
t106 EOP Width: Must accept as EOP 82 — ns 4 10
t107 Width of SE0 interval during differential transition — 14 ns
Low-speed Source (Note 8)
1, 6
t108 USBHx_DP, USBHx_DN - Driver Rise Time 75 300 ns CL = 200 pF 8
CL = 600 pF
1,6
t109 USBHx_DP, USBHx_DN Driver Fall Time 75 300 ns CL = 200 pF 8
CL = 600 pF
Source Differential Driver Jitter

–25 25 ns
t110 To Next Transition 2, 3 9
–14 14 ns
For Paired Transitions
t111 Source SE0 interval of EOP 1.25 1.50 µs 4 10
Source Jitter for Differential Transition to SE0

t112 –40 100 ns 5
Transition
Receiver Data Jitter Tolerance

–152 152 ns
t113 - To Next Transition 3 9
–200 200 ns
t114 EOP Width: Must accept as EOP 670 — ns 4 10
t115 Width of SE0 interval during differential transition — 210 ns
Notes:
1. Driver output resistance under steady state drive is specified at 28  at minimum and 43  at maximum.
2. Timing difference between the differential data signals.
3. Measured at crossover point of differential data signals.
4. Measured at 50% swing point of data signals.
5. Measured from last crossover point to 50% swing point of data line at leading edge of EOP.
6. Measured from 10% to 90% of the data signal.
7. Full-speed Data Rate has minimum of 11.97 Mb/s and maximum of 12.03 Mb/s.
8. Low-speed Data Rate has a minimum of 1.48 Mb/s and a maximum of 1.52 Mb/s.
Figure 7. USB Rise and Fall Time
Rise Time Fall Time

90% 90%
CL
Differential
Data Lines
10% 10%
CL
tR tF
Low-speed: 75 ns at CL = 50 pF, 300 ns at CL = 350 pF

Full-speed: 4 to 20 ns at C L = 50 pF
High-speed: 0.8 to 1.2 ns at C L = 10 pF

Figure 8. USB Jitter
T period
Crossover
Points
Differential
Data Lines
Jitter
Consecutive
Transitions
Paired
Transitions
Figure 9. USB EOP Width
Tperiod
Data
Crossover
Differential Level
Data Lines
EOP
Width
4.8 General Interface Timing
4.8.1 Legacy SPI Interface Timing
Table 38. Legacy SPI Interface Timings (20 MHz) (Sheet 1 of 2)

Sym Parameter Min Max Units Fig
F Serial Clock Frequency - 20M Hz Operation — 20 MHz
TCH LSPI_SCK high time 20 — ns 10
TCL LSPI_SCK low time 30 — ns 10
Setup of LSPI_MISO with respect to

TDSCR 11.7 — ns 10
LPSI_SCK rising edge
Hold time of LSPI_MISO with respect to

TCRDH -3.0 — ns 10
LPSI_SCK rising edge
TCFDV LSPI_SCK falling edge to LSPI_MOSI valid -1.9 2.5 ns 10

Table 38. Legacy SPI Interface Timings (20 MHz) (Sheet 2 of 2)

TCFSF LSPI_SCK falling edge to LSPI_SS_B low -1.6 2.4 ns 10
TCFSR LSPI_SCK falling edge to LSPI_SS_B high -1.7 2.3 ns 10
Notes:
1. All input signals have a slope of 1.0ns measured between 20% and 80% VCC values.
2. All output signals are loaded with 20pF.
3. Measurements are made at 50% VCC levels.
4.8.2 SPI0/1 Interface Timing
Table 39. SPI0/1 Interface Timings (25 MHz)

F Serial Clock Frequency - 25MHz Operation — 25 MHz
TCH SPI0/1_SCK high time 20 — ns 10
TCL SPI0/1_SCK low time 20 — ns 10
Setup of SPI0/1_MISO with respect to SPI0/

TDSCR 10.7 — ns 10
1_SCK capturing edge
Hold time of SPI0/1_MISO with respect to

TCRDH -2.5 — ns 10
SPI0/1_SCK capturing edge
SPI0/1_SCK driving edge to SPI0/1_MOSI

TCFDV -0.8 3.5 ns 10
valid
Setup of SPI0/1_SS_B with respect to first

TSSCF 20 — ns 10
edge out of inactive state of SPI0/1_SCK
Hold of SPI0/1_SS_B with respect to last

TCLSH 20 — ns 10
edge into inactive state of SPI0/1_SCK
Note:
1. All input signals have a slope of 1.0ns measured between 20%and 80% VCC values
2. All output signals are loaded with 20pF
3. Measurements are made at 50% VCC levels
4. Driving edge and capturing edge of SPI0/1_SCK are determined by SPI Control Register 1
settings SSCR1.SPH and SSCR1.SPO; Figure 10 shows SPI_SCK rising edge as the driving
edge and SPI_SCK falling edge as the capturing edge by way of example

Figure 10. SPI Interface Timing
TCH TCL
SPI_SCK
TCDDV
SPI_MOSI
TSSCF TCLSH
SPI_SS_B
TDSCC TCCDH
SPI_MISO
4.8.3 SDIO Interface Timing
Table 40. SDIO Timing

Sym Parameter Min Max Units Notes Fig
F Operating Frequency SD_CLK - 50 MHz
TCH Clock High Time SD_CLK 10 - ns 11
TCL Clock Low Time SD_CLK 10 - ns 11
SD_DATA[7:0]/SD_CMD setup time

TDSCR 0.6 - ns 11
with respect to SD_CLK rising
SD_DATA[7:0]/SD_CMD hold time

TCRDH 1.6 - 11
with respect to SD_CLK rising
SD_CLK falling to data valid on

TCFDV -2.0 2.7 ns 11
SD_DATA[7:0]/SD_CMD
Note:
1. All input signals have a slope of 1.0ns measured between 20%and 80% VCC values
2. All output signals are loaded with 20pF
3. Measurements are made at 50% VCC levels

Figure 11. SDIO Interface Timing
TCL TCH
SD_CLK
TCFDV
SD_DATA[7:0]
SD_CMD
(outputs)
TDSCR TCRDH
SD_DATA[7:0]
SD_CMD
(inputs)
4.9 Clock AC Timing
4.9.1 Reference Clock AC Characteristics
Table 41. Reference Clocks AC Characteristics (Sheet 1 of 2)

Parameter Description Min Max Units Notes Fig
Associated Signals:
REF0_OUTCLK_P, REF0_OUTCLK_N,
REF1_OUTCLK_P, REF1_OUTCLK_N
Related Supply:
VCCAICLKDBUFF_1P0
TSLEW_RISE Rising slew rate 1.5 8.0 V/ns 2,3,10 12
TSLEW_FALL Falling slew rate 1.5 8.0 V/ns 2,3,10 12
PSLEW_VAR Slew rate matching - 20 % 1,9,10 12
VSWING Differential output swing 300 - mV 2,11 12
VCROSS Crossing point voltage 300 550 mV 1,4,5,11 12
VCROSS_DELTA VCROSS variation - 140 mV 1,4,8,11 12
VMAX Maximum output voltage - 1.15 V 1,6,11 12

Table 41. Reference Clocks AC Characteristics (Sheet 2 of 2)

Parameter Description Min Max Units Notes Fig
VMIN Minimum output voltage -0.3 - V 1,7,11 12
PDTY_CYC Duty Cycle 40 60 % 2,10 12
Note:
1. Measurement taken from a single-ended waveform on a component test board
2. Measurement taken from a differential waveform on a component test board
3. Slew rate measured through VSWING voltage measured at differential zero
4. VCROSS is defined as the voltage where CLK_P = CLK_N
5. Only applies to differential rising edge (CLK_P rising, CLK_N falling)
6. The maximum voltage including over-shoot
7. The minimum voltage including under-shoot
8. The total variation of all VCROSS measurements in any particular system
9. Matching applies to rising edge rate for CLK_P and falling edge rate for CLK_N; It is measured using a
±75mV window centered on the average cross point where CLK_P rising meets CLK_N falling. The
median cross point is used to calculate the voltage thresholds the oscilloscope is to use for the edge
rate calculations
10. Average measurement
11. Instantaneous measurement

Figure 12. Measurement Points for Differential Clocks1
Differential Clock – Single Ended Measurements

V max = 1.15V V max = 1.15V
Clock#
Vcross max = Vcross max =
550mV 550mV
Vcross min = 300 mV Vcross min = 300 mV
Clock
V min = -0.30V V min = -0.30V
Clock#
Vcross delta = 140mV Vcross delta = 140mV
Clock
Clock# Clock#
Vcross median
+75mV
i se
T
fa
Tr
Vcross median Vcross median
ll
Vcross median -75mV
Clock Clock
Differential Clock – Differential Measurements

Clock Period (Differential )
Positive Duty Cycle (Differential ) Negative Duty Cycle (Differential )
0.0V
Clock-Clock#
Rise Fall
Edge Edge
Rate Rate
Vih_min = +150 mV
0.0V
Vil_max = -150 mV
Clock-Clock#
1. Clock == CLK_P; Clock# == CLK_N
§§

Register Access Methods—Intel® Quark™ SoC X1000
5.0 Register Access Methods
There are six different common register access methods:

• Fixed I/O Register Access
• Fixed Memory Mapped Register Access
• I/O Referenced Register Access
• Memory Referenced Register Access
• PCI Configuration Register Access (Indirect - via Memory or I/O registers)
• Message Bus Register Access (Indirect - via PCI Configuration Registers)
5.1 Fixed I/O Register Access

Fixed I/O registers are accessed by specifying their 16-bit address in a PORT IN and/or
PORT OUT transaction from the CPU core. This allows direct manipulation of the
registers. Fixed I/O registers are unmovable registers in I/O space.
Table 42. Fixed I/O Register Access Method Example (NSC Register)
Type: I/O Register
NSC: 61h
(Size: 8 bits)
5.2 Fixed Memory Mapped Register Access

Fixed Memory Mapped I/O (MMIO) registers are accessed by specifying their 32-bit
address in a memory transaction from the CPU core. This allows direct manipulation of
the registers. Fixed MMIO registers are unmovable registers in memory space.
Table 43. Fixed Memory Mapped Register Access Method Example (IDX Register)
Type: Memory Mapped I/O Register
IDX: FEC00000h
(Size: 32 bits)
5.3 I/O Referenced Register Access

I/O referenced registers use programmable base address registers (BARs) to select a
range of I/O addresses that it uses to decode PORT IN and/or PORT OUT transactions
from the CPU to directly access a register. Thus, the I/O BARs act as pointers to blocks
of actual I/O registers. To access an I/O referenced register for a specific I/O base
address, start with that base address and add the register’s offset. Example
pseudocode for an I/O referenced register read is shown below:
Register_Snapshot = IOREAD([IO_BAR]+Register_Offset)
Base address registers are often located in the PCI configuration space and are
programmable by the BIOS/OS. Other base address register types may include fixed
memory registers, fixed I/O registers or message bus registers.

Intel® Quark™ SoC X1000—Register Access Methods
Table 44. Referenced I/O Register Access Method Example (PM1S Register)
Type: I/O Register
PM1S: [PM1BLK] + 0h
(Size: 16 bits)
PM1BLK Type: PCI Configuration Register (Size: 32 bits)

PM1BLK Reference: [B:0, D:31, F:0] + 48h
5.4 Memory Referenced Register Access

The SoC uses programmable base address registers (BARs) to set a range of physical
address (memory) locations that it uses to decode memory reads and writes from the
CPU to directly access a register. These BARs act as pointers to blocks of actual
memory mapped I/O (MMIO) registers. To access a memory referenced register for a
specific base address, start with that base address and add the register’s offset.
Example pseudocode for a read is shown below:
Register_Snapshot = MEMREAD([Mem_BAR]+Register_Offset)
Base address registers are often located in the PCI configuration space and are
programmable by the BIOS/OS. Other common base address register types include
fixed memory registers and I/O registers that point to MMIO register blocks.
Table 45. Memory Mapped Register Access Method Example (ESD Register)
ESD: [RCBA] + 4h
(Size: 32 bits)
RCBA Type: PCI Configuration Register (Size: 32 bits)

RCBA Reference: [B:0, D:31, F:0] + F0h
5.5 PCI Configuration Register Access

Access to PCI configuration space registers is performed through one of two different
configuration access methods (CAMs):
• I/O indexed - PCI CAM
• Memory mapped - PCI Enhanced CAM (ECAM)
Each PCI function (see Section 6.3, “PCI Configuration Space” on page 97) has a
standard PCI header consisting of 256 bytes for the I/O access scheme (CAM), or 4096
bytes for the enhanced memory access method (ECAM). Invalid read accesses return
binary strings of 1s.
Table 46. PCI Register Access Method Example (PCI_DEVICE_VENDOR Register)

Type: PCI Configuration Register PCI_DEVICE_VENDOR: [B:0, D:31, F:0] + 0h
(Size: 32 bits)
5.5.1 PCI Configuration Access - CAM: I/O Indexed Scheme

Accesses to configuration space using the I/O method rely on two 32-bit I/O registers:
• CONFIG_ADDRESS - I/O Port CF8h
• CONFIG_DATA - I/O Port CFCh
These two registers are both 32-bit registers in I/O space. Using this indirect access
mode, software uses CONFIG_ADDRESS (CF8h) as an index register, indicating which
configuration space register to access, and CONFIG_DATA (CFCh) acts as a window to

the register pointed to in CONFIG_ADDRESS. Accesses to CONFIG_ADDRESS (CF8h)

are internally captured. Upon a read or write access to CONFIG_DATA (CFCh),
configuration cycles are generated to the PCI function specified by the address
captured in CONFIG_ADDRESS. The format of the address is shown in Table 47.
Table 47. PCI CONFIG_ADDRESS Register (I/O PORT CF8h) Mapping

Field CONFIG_ADDRESS Bits
Enable PCI Config. Space Mapping 31
Reserved 30:24
Bus Number 23:16
Device Number 15:11
Function Number 10:08
Register/Offset Number 07:02
Note: Bit 31 of CONFIG_ADDRESS must be set for a configuration cycle to be generated.
Pseudocode for a PCI register read is shown below:
MyCfgAddr[23:16] = bus; MyCfgAddr[15:11] = device; MyCfgAddr[10:8] = funct;
MyCfgAddr[7:2] = dWordMask(offset); MyCfgAddr[31] = 1;
IOWRITE(0xCF8, MyCfgAddr)
Register_Snapshot = IOREAD(0xCFC)
5.5.2 PCI Configuration Access - ECAM: Memory Mapped Scheme

A flat, 256 Mbyte memory space may also be allocated to perform configuration
transactions. This is enabled through the HECREG message bus register (Port: 3h,
Register: 09h) found in the Host Bridge. HECREG allows remapping this 256 Mbyte
region anywhere in physical memory space. Memory accesses within the programmed
MMIO range result in configuration cycles to the appropriate PCI devices specified by
the memory address as shown below.
Table 48. PCI Configuration Memory Bar Mapping

ECAM Memory Address Field ECAM Memory Address Bits
Use from BAR: HECREG[31:28] 31:28
Bus Number 27:20
Device Number 19:15
Function Number 14:12
Register Number 11:02
Note: ECAM accesses are only possible when HECREG.EC_ENABLE (bit 0) is set.
Pseudocode for an enhanced PCI configuration register read is shown below:
MyCfgAddr[27:20] = bus; MyCfgAddr[19:15] = device; MyCfgAddr[14:12] = funct;
MyCfgAddr[11:2] = dw_offset; MyCfgAddr[31:28] = HECREG[31:28];
Register_Snapshot = MEMREAD(MyCfgAddr)

5.6 Message Bus Register Access

Accesses to the message bus space are through the Host Bridge’s PCI configuration
registers. This unit relies on three 32-bit PCI configuration registers to generate
messages:
• Message Bus Control Register (MCR) - PCI[B:0,D:0,F:0] + D0h
• Message Data Register (MDR) - PCI[B:0,D:0,F:0] + D4h
• Message Control Register eXtension (MCRX) - PCI[B:0,D:0,F:0] + D8h
This indirect access mode is similar to PCI CAM. Software uses the MCR/MCRX as an
index register, indicating which message bus space register to access (MCRX only when
required), and MDR as the data register. Writes to the MCR trigger message bus
transactions.
Writes to MCRX and MDR are captured. Writes to MCR generates an internal ‘message
bus’ transaction with the opcode and target (port, offset, byte enable) specified in the
MCR and the captured MCRX. When a write opcode is specified in MCR, the data that
was captured by MDR is used for the write. When a data read opcode is specified in
MCR, the data is available in the MDR register after the MCR write completes (non-
posted). The format of MCR and MCRX are shown in Table 49 and Table 50.
Table 49. MCR Description

Field MBPR Bits
OpCode (typically 10h for read, 11h for write) 31:24
Port 23:16
Offset/Register 15:08
Byte Enable 07:04
Table 50. MCRX Description

Field MBPER Bits
Offset/Register Extension. This is used for messages sent to end points that require more
31:08
than 8 bits for the offset/register. These bits are a direct extension of MCR[15:8].
Most message bus registers are located in the Host Bridge. The default opcode
messages for those registers are as follows:
• Message ‘Read Register’ Opcode: 10h
• Message ‘Write Register’ Opcode: 11h
Registers with different opcodes are specified as applicable. Pseudocode of a message

bus register read is shown below (where ReadOp==0x10):
MyMCR[31:24] = ReadOp; MyMCR[23:16] = port; MyMCR[15:8] = offset;
MyMCR[7:4] = 0xf
PCIWRITE(0, 0, 0, 0xD0, MyMCR)
Register_Snapshot = PCIREAD(0, 0, 0, 0xD4)

5.7 Register Field Access Types
Table 51. Register Access Types and Definitions

Access Type Meaning Description
In some cases, if a register is read only, writes to this register

location have no effect. However, in other cases, two separate
RO Read Only registers are located at the same location where a read accesses
one of the registers and a write accesses the other register. See the
I/O and memory map tables for details.
In some cases, if a register is write only, reads to this register

location have no effect. However, in other cases, two separate
WO Write Only registers are located at the same location where a read accesses
one of the registers and a write accesses the other register. See the
I/O and memory map tables for details.
RW Read/Write A register with this attribute can be read and written.
A register bit with this attribute can be read and written. However, a
RW/C Read/Write Clear write of 1 clears (sets to 0) the corresponding bit and a write of 0
has no effect.
A register bit with this attribute can be written only once after power
RW/O Read/Write-Once
up. After the first write, the bit becomes read only.
A register bit with the attribute can be read at any time but writes
may only occur if the associated lock bit is set to unlock. If the
RW/L Read/Write Lockable
associated lock bit is set to lock, this register bit becomes RO unless
otherwise indicated.
A register bit with this attribute can be written to the non-locked

Read/Write, Lock-
RW/L/O value multiple times, but to the locked value only once. After the
Once
locked value has been written, the bit becomes read only.
RW/SN Read/Write Read/Write register initial value loaded from NVM.
Reserved Reserved The value of reserved bits must never be changed.
When the processor is reset, it sets its registers to predetermined

default states. The default state represents the minimum
functionality feature set required to successfully bring up the
system. Hence, it does not represent the optimal system
Default Default
configuration. It is the responsibility of the system initialization
software to determine configuration, operating parameters, and
optional system features that are applicable, and to program the
processor registers accordingly.

§§

Mapping Address Spaces—Intel® Quark™ SoC X1000
6.0 Mapping Address Spaces
The Intel® Quark™ SoC X1000 supports four different address spaces:
• Physical Address Space Mappings (Memory Space)
• I/O Address Space
• PCI Configuration Space
• Message Bus Space
The CPU core can only directly access memory space through memory reads and writes
and I/O space through the IN and OUT I/O port instructions. PCI configuration space is
indirectly accessed through I/O or memory space, and the Message Bus space is
accessed through PCI configuration space. See Chapter 5.0, “Register Access Methods”
for details.
This chapter describes how the memory, I/O, PCI, and Message Bus spaces are mapped
to interfaces in the SoC.
Note: See Chapter 12.0, “Host Bridge” for registers specified in the chapter.
6.1 Physical Address Space Mappings

There are 4 Gbyte (32-bits) of physical address space that can be used as:
• Memory Mapped I/O (MMIO - I/O fabric)
• Physical Memory (DRAM)
The CPU core can access the full physical address space, while downstream devices can
only access SoC DRAM, and the CPU core’s local APIC. Peer to peer transactions are not
supported.
Most devices map their registers and memory to the physical address space. This
chapter summarizes the possible mappings.
6.1.1 Bridge Memory Map

The Host Bridge maps the physical address space as follows:
• CPU core to DRAM
• CPU core to I/O fabric (MMIO)
• CPU core to extended PCI registers (ECAM accesses)
• I/O fabric to CPU cores (local APIC interrupts)
This SoC has the following distinct memory regions:

• DOS DRAM + Low DRAM
• MMIO
The HMBOUND register is used to create these memory regions, as shown in Figure 13.

Intel® Quark™ SoC X1000—Mapping Address Spaces
Figure 13. Physical Address Space - Low DRAM & MMIO
4 Gbyte
MMIO
HMBOUND
Low DRAM Low DRAM
1 Mbyte
DOS DRAM DOS DRAM
Physical Address DRAM Address

Space Space

6.1.1.1 MMIO
The MMIO mappings are shown in Figure 14.
Figure 14. Physical Address Space - MMIO
4 Gbyte
- 1 (FFFFFFFFh)
Boot Vector
- 128 Kbyte (FFFE0000h)
- 17 Mbyte (FEEFFFFFh)
MMIO
Local APIC
- 18 Mbyte (FEE0 0000h)
HMBOUND
HECREG + 256 Mbyte
Low DRAM
PCI ECAM
HECREG
1 Mbyte
DOS DRAM
Physical Address
Space
By default, CPU core reads targeting the Boot Vector range (FFFFFFFFh-FFFE0000h)
are sent to the Legacy Bridge, and write accesses target DRAM. For secure SKU’s, reads
targeting the Boot Vector are decoded and routed to a Secure Root of Trust Boot ROM.
For non-secure SKU’s, reads targeting this region are routed to a boot SPI flash device
connected to the Legacy Bridge.
Upstream writes from the I/O fabric to the Local APIC range (FEE00000h-FEEFFFFFh)
are sent to the CPU core’s APIC.
Accesses in the 256 Mbyte PCI ECAM range starting at HECREG generate enhanced
PCI configuration register accesses when enabled (HECREG.EC_ENABLE). Unlike
traditional memory writes, writes to this range are non-posted when enabled. See
Chapter 5.0, “Register Access Methods” for more details.
All other downstream accesses in the MMIO range are decoded based on PCI resource
allocations. The subtractive agent (for unclaimed accesses) is the I/O Fabric. The I/O
Fabric returns an UNSUPPORTED REQUEST for unclaimed accesses.

6.1.1.2 DOS DRAM

The DOS DRAM is the memory space below 1 MByte. In general, accesses from a
processor targeting DOS DRAM target system DRAM. Exceptions are shown in
Figure 15.
Figure 15. Physical Address Space - DOS DRAM
4 Gbyte
MMIO
HMBOUND
F Segment 64 Kbyte (F0000h to F FFFFh)
E Segment 64 Kbyte (E 0000h to E FFFFh)
Low DRAM
VGA/CSEG 128 Kbyte (A0000h to BFFFFh)
1 Mbyte
DOS DRAM
Physical Address
Space
Processor writes to the 64 Kbyte (each) E and F segments (E0000h-EFFFFh and

F0000h-FFFFFh) always target DRAM. The HMISC2 register is used to direct CPU core
reads in these two segments to DRAM or the I/O fabric (MMIO).
CPU core accesses to the 128 Kbyte VGA/CSEG range (A0000h-BFFFFh) can target
DRAM or the MMIO space depending on the setting of HMISC2.ABSEG_IN_DRAM. When
targeting MMIO space, requests are sent to the PCIe* port if legacy VGA is enabled in
the PCIe controller.
6.1.1.3 Additional Mappings

There is one additional mapping available in the Host Bridge:
• SMM range
Figure 16 shows these mappings.

Figure 16. Physical Address Space - SMM Range
L o w o r H ig h
D R A M in P h y s ic a l
Space
H S M M C T L .S M M _ E N D
SM M Range
H S M M C T L .S M M _ S T A R T
SMI handlers running on a CPU core execute out of SMRAM. To protect this memory
from non-CPU core access, the SMM Range (HSMMCTL.SMM_START -
HSMMCTL.SMM_END) may be programmed anywhere in low DRAM space (1 Mbyte
aligned). This range only allows accesses from the CPU core while in SMM.
6.1.2 MMIO Map

Memory accesses targeting MMIO are routed by the programmed PCI ranges.
Fixed MMIO is claimed by the Legacy Bridge. The default regions are listed below.
Movable ranges are not shown. See the register maps of all Legacy Bridge components
for details.
Table 52. Fixed Memory Ranges in the Legacy Bridge

Device Start Address End Address Comments
Starts 128 Kbyte below 1 Mbyte; Firmware/

Low BIOS (Flash Boot) 000E0000h 000FFFFFh
BIOS
I/O APIC FEC00000h FEC00040h Starts 20 Mbyte below 4 Gbyte
HPET FED00000h FED003FFh Starts 19 Mbyte below 4 Gbyte
Starts 128 Kbyte below 4 Gbyte; Firmware/

High BIOS/Boot Vector FFFE0000h FFFFFFFFh
BIOS
PCI devices may also claim memory resources in MMIO space. For details see each
device’s interface chapter.
Warning: Variable memory ranges should not be set to conflict with other memory ranges. There
may be unpredictable results if the configuration software allows conflicts to occur.
Hardware does not check for conflicts.
6.2 I/O Address Space

There are 64 Kbyte + 3 bytes of I/O space (0h-10002h) for accessing I/O registers.
Most I/O registers exists for legacy functions in the Legacy Bridge or for PCI devices,
while some are claimed by the Host Bridge for the PCI configuration space access
registers.

6.2.1 Host Bridge I/O Map

The Host Bridge claims I/O transactions for VGA/Extended VGA found in the display/
graphics interface. It also claims the two 32-bit registers at port CF8h and CFCh used to
access PCI configuration space.
6.2.2 I/O Fabric I/O Map
6.2.2.1 Legacy Bridge Fixed I/O Address Ranges

Table 53 shows the fixed I/O space ranges seen by a processor.
Table 53. Fixed I/O Ranges in the Legacy Bridge

Device I/O Address Comments
8259 Master 20h-3Dh
8254s 40h-43h, 50h-53h
NMI Controller 61h, 63h, 65h, 67h
RTC 70h-73h
Scratch Pad 80h-83h
8259 Slave A0h-BDh
Reset Control CF9h Overlaps PCI I/O registers
6.2.2.2 Variable I/O Address Ranges

Table 54 shows the variable I/O decode ranges. They are set using base address
registers (BARs) or other similar means. Plug-and-play (PnP) software (PCI/ACPI) can
use their configuration mechanisms to set and adjust these values.
Warning: The variable I/O ranges should not be set to conflict with other I/O ranges. There may
be unpredictable results if the configuration software allows conflicts to occur. Hardware
does not check for conflicts.
Table 54. Movable I/O Ranges Decoded by PCI Devices on the I/O Fabric
Device Size (bytes) Target
ACPI Power Management 16 PM1BLK: PCI[B:0,D:31,F:0] + 48h
ACPI General Purpose Event 0 64 GPE0BLK: PCI[B:0,D:31,F:0] + 4Ch
GPIO 128 GBA: PCI[B:0,D:31,F:0] + 44h
Watchdog Timer 64 WDTBA: PCI[B:0,D:31,F:0] + 84h
ACPI Processor Block 16 PMBA: Port[0x04] + 70h
SPI DMA Block 16 SPI_DMA_BAR: Port[0x04] + 7Ah

6.3 PCI Configuration Space

All PCI devices/functions are shown below.
Table 55. PCI Devices and Functions

Bus Device Function Device Description Function Description
0 0 Host Bridge
0 SDIO / eMMC
1 HS-UART 0
2 USB 2.0 Device
3 USB EHCI
20 I/O Fabric
4 USB OHCI
5 HS-UART 1
0 6 10/100 Ethernet MAC 0
7 10/100 Ethernet MAC 1
0 SPI 0
21 1 I/O Fabric SPI 1
2 I2C* / GPIO
0 Root Port 0
23 PCI Express*
1 Root Port 1
31 0 Legacy Bridge Legacy Components

Figure 17. Bus 0 PCI Devices and Functions
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem)
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1
IO Fabric
D:21
SPI1 F:1
I2C*/GPIO F:2 PMC
GPIO
Legacy Bridge
RTC
D:31,F:0
8254
SDIO/eMMC F:0 8259
HSUART0 F:1 HPET
IO APIC
IO Fabric D:20
USB Device F:2

SPI
EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7

6.4 Message Bus Space

Message bus space provides access to different units within the host bridge. These
units are useful in configuring the memory map, power management, and more.
Figure 18. Message Bus with PCI Space
CPU
Core
Host
Bridge Message Bus Space
B:0,D:0,F0
Port: 0x00 Port: 0x01 Port: 0x03 Port: 0x04 Port: 0x05
Remote
Memory Memory Host Memory
Management
Arbiter Controller Bridge Unit Manager
Table 56. Message Types

Msg Type Message Description
Register Write message - used to write to the 32-bit

RegWr
Dunit registers.
Register Read message - used to read from the 32-bit

RegRd
Dunit registers.
Simple message without data - used to send atomic

Msg
commands, such as Wake and Suspend.
Simple Message with 4 bytes of data - used to

MsgD communicate with the DRAM devices during
initialization.

§§

Clocking—Intel® Quark™ SoC X1000
7.0 Clocking
The SoC has a variable frequency, multiple clock domain, and multiple power plane
clocking system. This clock architecture achieves a low power clocking solution that
supports the various clocking requirements of the different IPs on the SoC. This is
achieved by using an Integrated Clock module (iClock) that supplies the clocks to the
entire platform.
7.1 Clocking Features

The SoC provides a complete system clocking solution through integrated clocking. All
the required platform clocks are provided by the SoC using only one input: a 25 MHz
primary reference for the integrated clock block. An optional 32 KHz reference for the
Real Time Clock (RTC) block may be provided if required.

Intel® Quark™ SoC X1000—Clocking
Figure 19. SoC Platform Clocking
Platform
Clocks
DDR3_CK/CKB[1:0] DRAM Ranks [0-1]

25 MHz
Primary
Reference
XTAL_IN/OUT
REF[0/1]_OUTCLK_P/N PCIe* Connector(s)
Ethernet
RMII_REF_CLK_OUT PHY
(RMII)
RMII_REF_CLK
I2C_CLK I2C* Interface

32 kHz
RTC_X1/X2
RTC Reference
(optional)
LSPI_SCK SPI Flash
SPI[0/1]_SCK SPI Device
FLEX0_CLK
FLEX1_CLK External Device(s)
FLEX2_CLK
The reference clocks required for the various interface PLLs (e.g., USB/PCIe*) and the
processor are internally generated by the Integrated Clocking unit.
7.2 Platform/System Clock Domains

The SoC contains multiple clock domains to support its various interfaces. Table 57 and
Table 58 summarize the different clock inputs and outputs in the system.

Clocking—Intel® Quark™ SoC X1000
Table 57. Intel® Quark™ SoC X1000 Clock Inputs

Clock Domain Signal Name Frequency Usage/Description
XTAL_IN
Main 25 MHz 25 MHz reference for the iCLK PLL
XTAL_OUT
RTC Crystal I/O for RTC block.

RTCX1
RTC 32 kHz This clock is optional and may be
RTCX2
generated internally by the iCLK PLL.
RMII 50MHz Clock

This clock is a loopback of
Ethernet PHY RMII_REF_CLK 50 MHz
RMII_REF_CLK_OUT
JTAG TCK 25 MHz JTAG Test Clock
CPU/PLL OSC_COMP Static current Ext. precision R 10.5 kOhm to 1.5V
Table 58. Intel® Quark™ SoC X1000 Clock Outputs

Clock Domain Signal Name Frequency Usage/Description
DDR3_CK[1:0] Drives the Memory ranks 0-1. Data rate is

DDR 400 MHz
DDR3_CKB[3:0] 2x the clock rate.
REF[0/1]_OUTCLK_N Differential Clocks supplied to external PCI

PCI Express* 100 MHz
REF[0/1]_OUTCLK_P Express* devices
FLEX0_CLK 33 MHz
Flex Clocks FLEX1_CLK 33 MHz Output clock for External devices
FLEX2_CLK 48 MHz
Legacy SPI LSPI_SCK 20 MHz Clock for external SPI Flash
SPI SPI[0/1]_SCK 25 MHz SPI serial clocks
Ethernet PHY RMII_REF_CLK_OUT 50 MHz Reference clock for RMII interface
I2C* I2C_CLK 400 kHz I2C clocks
SD SD_CLK 50 MHz SD Clock
Main CKSYS25OUT 25 MHz 25 MHz Oscillator Output

Intel® Quark™ SoC X1000—Clocking
§§

Power Management—Intel® Quark™ SoC X1000
8.0 Power Management
This chapter provides information on the following power management topics:

• ACPI States
• Processor Core
• PCI Express*
8.1 Power Management Features

• ACPI 3.0 specification support
• ACPI Processor C States (C0, C1, and C2)
• ACPI Sleep State Support (S0, S3, S4, and S5)
• PCI Express L0, L1, L2, and L3
8.2 Signal Descriptions

See Chapter 2.0, “Physical Interfaces” for additional details.
The signal description table has the following headings:

• Signal Name: The name of the signal/pin
• Direction: The buffer direction can be either input, output, or I/O (bidirectional)
• Type: The buffer type found in Chapter 4.0, “Electrical Characteristics”
• Description: A brief explanation of the signal’s function
Table 59. Power Management (Sheet 1 of 2)

Direction/
Signal Name Description
Type
I Reset Button: With the SOC in S0 an activation of this input will

RESET_BTN_B
PwrMgmt result in a 'Warm Reset'. Active LOW
PCI Express Wake Event:

I This signal indicates a PCI Express port wants to wake the
WAKE_B system. Can optionally be used by an external device to wake
PwrMgmt the system if the WAKE_B functionality is not required by PCI
Express.
I General Purpose Event:

GPE_B GPE_B is asserted by an external device to log an event in the
PwrMgmt system's ACPI space and cause an SCI (if enabled).
Power Button: Two modes of operation.

I 1. A power button press is required to complete cold boot.
PWR_BTN_B Active LOW.
PwrMgmt 2. The button is tied low, results in an automated start at power
on.
O
S3_3V3_EN S3 Domain 3.3v platform rail enable. Active HIGH.
PwrMgmt

Intel® Quark™ SoC X1000—Power Management
Table 59. Power Management (Sheet 2 of 2)

Direction/
Type
O
PwrMgmt
I
S3_PG S3 Power Good
PwrMgmt
I
S0_1P0_PG S0 Domain 1.0V Power Good
PwrMgmt
O
PwrMgmt
O
PwrMgmt
O
PwrMgmt
O
ODRAM_PWROK DRAM Power Okay: Active HIGH.
PwrMgmt
O
OSYSPWRGOOD System Power Good: S0 power is good. Active HIGH
PwrMgmt
IO
VNNSENSE VNN sense voltage for IMVP
PwrMgmt
IO
VSSSENSE VSS sense voltage for IMVP
PwrMgmt
8.3 ACPI Supported States

The ACPI states supported by the processor are described in this section.
8.3.1 S-State Definition
8.3.1.1 S0 - Full On
This is the normal operating state of the processor. In S0, the core processor transitions
in and out of the various processor C-States.
Note: The processor core does not support P-states.
8.3.1.2 S3 - Suspend to RAM (Standby)

S3 is a suspend state in which the core power planes of the processor are turned off
and the suspend wells remain powered.
• All power wells are disabled, except for the suspend and RTC wells.
• The core processor’s macro-state is saved in memory.
• Memory is held in self-refresh and the memory interface is disabled, except the
CKE pin as it is powered from the memory voltage rail. CKE is driven low.
8.3.1.3 S4 - Suspend to Disk (Hibernate)

S4 is a suspend state in which most power planes of the processor are turned off,
except for the suspend and RTC well. In this ACPI state, system context is saved to the
mass storage device attached to SDIO/eMMC interface.

Note: This is a software based state that is the same as S5 to hardware. On S4 entry, the
system saves the entire contents of data off to NVRAM. On S4 resume, the system
restores the entire contents of memory after performing the a typical S5-S0 boot.
Key features:
• No activity is allowed.
• All power wells are disabled, except for the suspend and RTC well.
8.3.1.4 S5 - Soft Off

From a hardware perspective the S5 state is identical to the S4 state. The difference is
purely software; software does not write system context to OS storage when entering
S5.
8.3.2 System States
Table 60. General Power States for System

States/Sub-states Legacy Name / Description
FULL ON: CPU operating. Individual devices may be shut down to save power. The
G0/S0/C0
different CPU operating levels are defined by Cx states.
G0/S0/Cx Cx State: CPU manages C-state itself.
Suspend-To-RAM (STR): The system context is maintained in system DRAM, but

power is shut to non-critical circuits. Memory is retained, and refreshes continue. All
G1/S3
external clocks are shut off; RTC clock and internal ring oscillator clocks are still
toggling.
Suspend-To-Disk (STD): The context of the system is maintained on the disk. All
G1/S4
power is shut down except power for the logic to resume.
Soft-Off: System context is not maintained. All power is shut down except power for
G2/S5 the logic to restart. A full boot is required to restart. A full boot is required when
waking.
Mechanical OFF. System content is not maintained. All power shutdown except for
the RTC. No “Wake” events are possible, because the system does not have any
power. This state occurs if the user removes the batteries, turns off a mechanical
G3
switch, or if the system power supply is at a level that is insufficient to power the
“waking” logic. When system power returns, transition depends on the state just prior
to the entry to G3.
Table 61 shows the transition rules among the various states. Note that transitions
among the various states may appear to temporarily transition through intermediate
states. These intermediate transitions and states are not listed in the table.
Table 61. ACPI PM State Transition Rules (Sheet 1 of 2)

Present
Transition Trigger Next State
State
P_LVL2 Read G0/S0/C2
G1/Sx or G2/S5 state (specified by

PM1C.SLP_EN bit set
G0/S0/C0 PM1C.SLP_TYPE)
Power Button Override G2/S5
Mechanical Off/Power Failure G3
C2 break events which include: MSI, Legacy

G0/S0/C0
Interrupt
G0/S0/C2
Power Button Override G2/S5
Resume Well Power Failure G3

Table 61. ACPI PM State Transition Rules (Sheet 2 of 2)

Present
Transition Trigger Next State
State
Any Enabled Wake Event G0/S0/C0
G1/S3, G1/S4 Power button Override G2/S5
Resume Well Power Failure G3
Any Enabled Wake Event G0/S0/C0

G2/S5
Power Failure or Removal G3
Option to go to S0/C0 (reboot) or G2/S5 (stay

off until power button pressed or other enabled
G3 Power Returns wake event) or G1/S4 (if system state was S4
prior to the power failure). Some wake events
are preserved through a power failure.
8.3.3 Processor Idle States
Table 62. Processor Core/ States Support

State Description
C0 Active mode, processor executing code
C1 AutoHALT state
C2 Stop Grant state
8.3.4 Integrated Memory Controller States
Table 63. Main Memory States

States Description
Powerup CKE asserted. Active mode.
Precharge Powerdown CKE de-asserted (not self-refresh) with all banks closed.
Active Powerdown CKE de-asserted (not self-refresh) with at least one bank active.
Self-Refresh CKE de-asserted using device self-refresh
8.3.5 PCIe* States
Table 64. PCIe* States

States Description
L0 Full on – Active transfer state
L0s First Active Power Management low power state – Low exit latency
L1 Lowest Active Power Management - Longer exit latency
L3 Lowest power state (power-off) – Longest exit latency

8.3.6 Interface State Combinations
Table 65. G, S and C State Combinations

Processor
Global (G) Sleep System
Core Processor State Description
State (S) State Clocks
(C) State
G0 S0 C0 Full On On Full On
G0 S0 C1 Auto-Halt On Auto-Halt
G0 S0 C2 Stop Grant On Stop Grant
Off except RTC

G1 S3 Power Off & internal ring Suspend to RAM
OSC
Off except RTC

G1 S4 Power Off & internal ring Suspend to Disk
OSC
Off except RTC

G2 S5 Power Off & internal ring Soft Off
OSC
G3 NA Power Off Power Off Hard Off
8.4 Processor Core Power Management

When the processor is not executing code, it is idle. A low-power idle state is defined by
ACPI as a C-state. In general, lower power C-states have longer entry and exit
latencies.
8.4.1 Low-Power Idle States

When the processor core is idle, low-power idle states (C-states) are used to save
power. More power savings actions are taken for numerically higher C-state. However,
higher C-states have longer exit and entry latencies.
8.4.1.1 Clock Control and Low-Power States

The processor core supports low power states at core level. States for processor core
include Normal (C0), Auto-Halt (C1) and Stop Grant (C2).
Transition to processor core power states higher than C1 are triggered by initiating a
P_LVLx (P_LVL2) I/O read.
The Cx state ends due to a break event. Based on the break event, the processor
returns the system to C0. The following are examples of such break events:
• Any unmasked interrupt goes active
• Any internal event that will cause an NMI or SMI_B
• CPU Pending Break Event (PBE_B)
• MSI
8.4.2 Processor Core C-States Description
8.4.2.1 Core C0 State

The normal operating state of a core where code is being executed.


C1 is a low power state entered when the core executes a HLT instruction.
A System Management Interrupt (SMI) handler returns execution to either Normal

state or the C1 state. See the Intel® 64 and IA-32 Architecture Software Developer’s
Manual, Volume 3A/3B: System Programmer’s Guide for more information.
While the core is in C1 state, it still processes snoops.

C2 is entered when the processor reads the P_LVL2 register to trigger a transition from
C0 to C2. While the core is in the C2 state, it processes snoops.
An interrupt or a reset is required to exit the C2 state and return to the C0 state.
8.5 Memory Controller Power Management

The main memory is power managed during normal operation and in low-power ACPI
Cx states.
8.5.1 Disabling Unused System Memory Outputs

When a given rank is not populated, the corresponding chip select and CKE signals are
not driven.
At reset, all rows must be assumed to be populated, until it can be proven that they are
not populated. This is due to the fact that when CKE is tri-stated the memory module is
not guaranteed to maintain data integrity.
8.5.2 DRAM Power Management and Initialization

The SoC implements extensive support for power management on the SDRAM
interface. There are four SDRAM operations associated with the Clock Enable (CKE)
signals, SRE, SRX, PDE and PDX, which the SDRAM controller supports. The SoC drives
two CKE pins to perform these operations.
8.5.2.1 Initialization Role of CKE

During power-up, CKE is the only input to the SDRAM that is recognized (other than the
DDR3 reset pin) once power is applied. It must be driven LOW by the DDR controller to
make sure the SDRAM components float DQ and DQS during power-up.
CKE signals remain LOW (while any reset is active) until the BIOS writes to a
configuration register. Using this method, CKE is guaranteed to remain inactive for
much longer than the specified 200 micro-seconds after power and clocks to SDRAM
devices are stable.
8.5.2.2 Dynamic Self-Refresh

When Dynamic Self-Refresh (SR) is enabled, via DPMC0.DYNSREN, the Memory
Controller places the SDRAM in SR mode when the following conditions are true:
1. No requests are pending
2. Internal Request Status is low priority
3. No SR exit requests from the DDRIO (for RCOMP updates)

If one of the above conditions change prior to the SR Entry command being sent to the
DRAM the process is terminated.
When Dynamic SR is enabled the Memory Controller exits SR mode when one of the
following is true:
1. Requests are pending and the Internal Request Status is normal or urgent
2. A SR exit request from the DDRIO
8.5.2.3 Dynamic Power Down Operation

Dynamic power-down of memory is employed during normal operation. Based on idle
conditions, a given memory rank may be powered down. The Memory Controller
implements aggressive CKE control to dynamically put the DRAM devices in a power
down state. The Memory Controller can be configured to put the devices in active
power down (CKE de-assertion with open pages) or precharge power down (CKE de-
assertion with all pages closed). Precharge power down provides greater power savings
but has a bigger performance impact, since all pages will first be closed before putting
the devices in power down mode.
If dynamic power-down is enabled, all ranks are powered up before doing a refresh
cycle and all ranks are powered down at the end of refresh.
8.5.2.4 Functional Clock Gating

The Memory Controller has internal clock gating for the majority of its clocked logic.
When enabled the clock gating is activated when all inputs are inactive and all
commands are complete and DDR3 timing trackers are flushed. When Dynamic SR is
enabled, clock gating is only applied when the SDRAM is in Self-Refresh.
§§


Power Up and Reset Sequence—Intel® Quark™ SoC X1000
9.0 Power Up and Reset Sequence
This chapter provides information on the following topics:

• Power up and down sequences, including wake events
• Reset behavior
9.1 Intel® Quark™ SoC X1000 System States
9.1.1 System Sleep States Control (S-States)

The SoC supports the S0, S3, S4, and S5 sleep states. S4 and S5 states are identical
from a hardware perspective.
The SoC integrates a Power Management Controller (PMC). No external power

controller IC is required.
The SoC sleep states are described in Chapter 8.0, “Power Management”.
9.2 Power Up and Down Sequences

Note: Delays in power sequences are dependent on components outside the SoC. As long as
the sequencing is preserved, the SoC will operate.
9.2.1 Power Up, Wake and Reset Overview

SoC power up is dependent on two supplies:
• VCC3P3_S5, which is generated from AC power
• VCCRTC_3P3, which powers the RTC well only
VCCRTC_3P3 is derived directly from VCC3P3_S5, if present. Otherwise it can be driven

by a coin-cell battery.
• In the case where the coin-cell battery is present but not AC power, only the RTC
well is powered up. The SoC can move to state G3 only. The SoC can subsequently
be transitioned to state S4/S5 by applying AC power.
• In the case where AC power is present but there is no coin-cell battery, power up is
initiated directly by the ramping of the VCC3P3_S5 supply. The SoC transitions
directly to state S4/S5.
Subsequent transition from S4/S5 to S0 is governed by activity on the power button

pin, PWR_BTN_B:
• If PWR_BTN_B is strapped low (auto power button mode) when AC power is
applied, the SoC transitions directly to S0 from S4/S5 via a transitional S3 state.
• If PWR_BTN_B is high when AC power is applied, the SoC transitions to S4/S5 only.
A subsequent falling edge on PWR_BTN_B, with the low value being maintained for
2.5ms or more, is required to initiate a transition to S0 via the transitional S3 state.

Intel® Quark™ SoC X1000—Power Up and Reset Sequence
Once in state S0 the SoC can be put to sleep, i.e., transitioned to sleep states S3 or S4/
S5, through appropriate settings of the Legacy Bridge ACPI registers PM1C.SLPTYPE
and PM1C.SLPEN.
A wake event is defined as a transition from state S3 to state S0. The chip can be
woken up via a number of mechanisms including specific register settings, or by
asserting specific SoC pins. A watchdog function in the Legacy Bridge can also trigger a
wake event.
In auto power button mode, if the SoC is placed in sleep state S4/S5, the system can
only be woken by the removal and reapplication of AC power. It does not resume from
S4, rather it is a new start with context loss. Since PWR_BTN_B is low it will power up
and transition directly back to S0 as described in the power up sequence.
There are two classes of reset associated with Intel® Quark™ SoC X1000:
• A cold reset means transitioning from S0 to S4/S5 and back to S0 again,
independent of the PWR_BTN_B value. This can only be initiated from state S0
through the register RSTC.COLD_RST. All registers except those driven by the RTC
supply are effectively reset.
• A warm reset resets CPU and peripheral blocks without the removal of the power
supplies. This can be initiated via a write to the register RSTC.WARM_RST or by
asserting the SoC pin, RESET_BTN_B (active low). It can occur only in state S0 and
after reset the SoC remains in state S0. RTC well and suspend well registers are left
unaffected.
A cold boot is the sequence where AC power is applied followed by an immediate

transition to S0 using the PWR_BTN_B signal or directly in auto power button mode.
Catastrophic shutdown can be carried out by holding PWR_BTN_B low for at least
3s. This results in a direct return to the S4/S5 state. It can also be initiated by software
under specific error conditions. See the Intel® Quark™ SoC X1000 UEFI Firmware
Writer’s Guide (Document # 330236) for more information.
The following sections provide more detail on these power-related functions.
9.2.2 RTC Power Well Transition: G5 to G3 State Transition

The transition to the G3 state is initiated when VCCRTC_3P3 is ramped. The sequence
is as follows:
1. VCCRTC_3P3 ramps. RTCRST_B should be low.
2. The SoC starts the real time clock oscillator.
3. A minimum of t1 units after VCCRTC_3P3 ramps external circuitry deasserts
RTCRST_B. The system is now in the G3 state.

Figure 20. RTC Power Well Timing Diagrams
Initial
State G3
Power Up
Insert coin-cell battery
VCCRTC_3P3
RTCRST_B t1 [ext]
t2
32kHz Oscillator [int] Oscillator Start-Up Clock Valid
Table 66. RTC Power Well Timing Parameters

Parameter Description Min Max Units Notes
VCCRTC_3P3 to RTCRST_B
t1 9 N/A ms 1
deassertion
t2 Oscillator Startup Time -N/A -N/A s 2
Notes:
1. This delay is typically created from an RC circuit.
2. The oscillator startup times are component and design specific. A crystal oscillator can take as long as
2 s to reach a large enough voltage swing. Whereas, a silicon oscillator can have startups times
<10 ms.
9.2.3 AC Power Applied: G3 to S4/S5 State Transition

The timings shown in Figure 21 and Table 67 occur when AC power is applied. The
following occurs:
1. The supplies VCC3P3_S5, VCC1P0_S5 and VCC1P5_S5 are generated by the
platform regulator. These voltages start ramping at [t1,t2,t3].
Note: It is required that the platform power sequence ensures that the 3 voltages
ramp in this order (VCC3P3_S5 --> VCC1P0_S5 --> VCC1P5_S5) All S5
voltages (3P3,1P8,1P0) must be stable prior to VCC1P5_S5.
2. VCC3P3_S5 drives an internal S5 LDO regulator that generates an internal 1.0V
and 1.8V supplies. The 1.8V is driven off chip at time t4 on pin OVOUT_1P8_S5.
Note: It is intended the 1P8V generated by the LDO is connected back into the
VCC1P8_S5 input in order to eliminate the need to generate this voltage on
the platform.
Note: The 1.0V supply output at pin OVOUT_1P0_S5 is unused and should remain
not connected at the platform.
3. When the internal supplies are stable an internal S5 power-good signal is
generated.

4. A platform S5 power good signal is generated when all Platform Generated S5

supplies are stable at 95% of nominal voltage. This signal is applied at the S5_PG
pin [t6]
Note: If the VCC1P8_S5 is generated by the platform then this voltage must also
be stable prior to assertion of the S5_PG."
5. When both the internal S5 power-good and S5_PG signals are asserted the system
is in state S4/S5.
9.2.4 Using PWR_BTN_B: Transition from S4/S5 to S0

1. The internal power management controller detects an event: either PWR_BTN_B is
actively brought low and remains low for at least 2.5ms [t7], or it is tied low for an
automatic start from the S5 state (auto power button mode). This initiates the
transition from S4/S5 to S0.
2. After a PLL settling time the internal PMC generates a switch enable S3_3V3_EN
which should be used to switch on of the platform S3 supply VCC3P3_S3 [t8, t8’].
3. This is followed by the assertion of S3_1V5_EN to switch on VCC1P5_S3 [t9].
Note: The switch enables are active high and should be used to drive PFET
switches or regulator enables on the platform.
Note: The switch enables are staggered because of SoC rail sequencing
constraints.
4. After a switch-on time [t10] the S3 supply VCC3P3_S3 starts ramping. This is
followed in turn by VCC1P5_S3 [t12].
5. VCC3P3_S3 drives an internal S3 LDO regulator which generates internal 1.0V and
1.8V supplies. These are driven off chip via the OVOUT_1P0_S3 and
OVOUT_1P8_S3 pins [t13, t14].
6. When the internal supplies are stable an internal S3 power-good signal is generated
[t15].
7. The Platform S3_PG pin [t16] is asserted when the Platform generated VCC1P5_S3
Power rails is stable. Based on the power rail sequencing this will ensure that all
platform generated S3 rails will be stable.
8. When both the internal S3 power-good and S3_PG signals are asserted, the SoC is
in a transitional S3 state.
9. The PMC now generates the switch enable S0_3V3_EN which should be used to
switch on the S0 supply VCC3P3_S0 [t17].
10. This is followed by the assertion of S0_1V0_EN to switch on VCC1P0_S0 [t18].
11. After a switch-on time [t19], the S0 supply VCC3P3_S0 starts ramping. This is
followed in turn by VCC1P0_S0 [t21].
12. VCC3P3_S0 drives an internal S0 LDO regulator which generates internal 1.05V
and 1.8V supplies. These are driven off chip via the OVOUT_1P05_S0 and
OVOUT_1P8_S0 pins [t22, t23].
13. Once VCC1P0_S0 is stable the platform should generate an active high power good
signal which is applied to the S0_1P0_PG pin [t25].
14. The PMC now generates the switch enable S0_1V5_EN after S0_1P0_PG asserted.
15. This followed by the assertion of S0_1V5_EN to switch on VCC1P5_S0 [t27].
16. The Platform S0_PG pin [t28] is asserted when the Platform generated VCC1P5_S0
Power rails is stable. Based on the power rail sequencing this will ensure that all
platform generated S0 rails will be stable.
17. All supplies all now on and the system us in state S0.

Figure 21. Power Up Sequence
If power button is used S5 – S3/S0

Power In (t0) transition occurs upon de‐assertion
of PW R_BTN_B.
`
t7 – debounce time
X1000 Input – PW R_BTN
RC Delay – EN_3P3
RC Delay – EN_1P0
S5 Rails
RC Delay – EN_1P5
t8
VCC3P3_S5 (t1)
VCC1P0_S5 (t2)
VCC1P5_S5 (t3)
t4
X1000 Vout – OVOUT_1P8_S5
t6
X1000 Input – S5_PGOOD
X1000 Output – S3_3V3_EN t8'
t9
X1000 Output – S3_1V5_EN
t10
`
VCC3P3_S3
S3 Rails
t12
VCC1P5_S3
t13
t14
t16
X1000 Input – S3_PGOOD
X1000 Out – S0_3V3_EN t17
t18
X1000 Out – S0_1V0_EN
X1000 Out – S0_1V5_EN

(t26)
t19
G
1P0_P
VCC3P3_S0
S0_
t21
VCC1P0_S0
after
S0 Rails
t27
erted
VCC1P5_S0
EN ass
t22
5_
S0_1V
t23
t25
X1000 Input – S0_1P0_PG
t28
X1000 Input – S0_PG

9.2.5 Power-Up Sequence without G2/G3: No Coin-Cell Battery

This sequence must be adhered to in cases where one of the following conditions apply:
1. The system does not implement an RTC battery (coin cell) or a main battery.
2. The coin cell is drained with no main or a dead main battery.
3. No coin cell implemented or dead coin cell and main battery is being swapped.
AND one of the following conditions also applies:

1. The platform does not implement a power button to initiate a sequence to S0, and
AC power becomes available.
2. The platform does use a power button, but the default first sequence when power is
available is entry into S0.
In these cases, the relative timing between RTC and suspend wells becomes important.
The key point is that, as well as a minimum time, there is a maximum time by which
RTCRST_B must be deasserted. It must happen before an internal reset associated with
the suspend well is deasserted. This is shown in Figure 22.
Figure 22. Power-Up Sequence without G2/G3
Apply AC Power
t0
VCCRTC_3P3
[ext]
t1min
RTCRST_B
t1max
32kHz Oscillator [int] t2 Clock Valid

Oscillator Start-Up
t0'
VCC3P3_S5
[ext]
Resume well reset [int]
Initial Power State G3

State S4/S5
Up (transitional)
Notes:
1. This delay is typically created from an RC circuit.
2. The oscillator startup times are component and design specific. A crystal oscillator can take as long as
2 s to reach a large enough voltage swing. Whereas, a silicon oscillator can have startups times
<10 ms.
3. System transitions automatically through S4/S5 to S0. See Section 9.2.4 for S0 power on sequence.

Table 67. S4/S5 to S0 Timing Parameters (Sheet 1 of 2)

Internal
(SoC)/
Time Description Min Max Units Notes
External
(Platform)
Apply AC Assume SoC

t0 N/A N/A N/A N/A
power is in state G3
t1 ext VCC3P3_S5
Platform related and are specific to the
t2 ext VCC1P0_S5
voltage regulator selected.
t3 ext VCC1P5_S5
OVOUT_1P8_ tracks the VCC3P3_S5 with negligible delay.

t4 int
S5 (Delay) There is no specific value.
Platform detection of all platform generated

Assertion of S5 rails to only activate once they are at
t6 ext 1
S5_PG ~95%, the delay is an RC so is platform
controlled.
Must hold
PWR_BTN_B
PWR_BTN_B low for at
t7 int debounce 2.5 - ms least this
time time for
falling edge
to take effect
With respect
S3_3V3_EN to
t8 int N/A N/A N/A
(Delay) PWR_BTN_B
event
auto power
S3_3V3_EN button mode:
t8’ int N/A N/A N/A
(Delay) With respect
to S5_PG
Offset from
S3_3V3_EN
S3_1V5_EN due to SoC
t9 int 90 1800 s
(Delay) rail
sequencing
requirements
VCC3P3_S3
t10 ext (Switch
Delay)
Platform delays based on component delays.
VCC1P5_S3
t12 ext (Switch
Delay)
OVOUT_1P8_
t13 int
S3 (Delay)
Track the VCC3P3_S3 with negligible delay.
OVOUT_1P0_
t14 int
S3 (Delay)
Platform detection of all S3 rails to only

Assertion of
t16 ext activate once they are at ~95%, the delay is 1
S3_PG
a RC so is platform controlled.
S0_3V3_EN
t17 int N/A N/A N/A
(Delay)

Table 67. S4/S5 to S0 Timing Parameters (Sheet 2 of 2)

Internal
(SoC)/
Time Description Min Max Units Notes
External
(Platform)
Offset from
S3_3V3_EN
S0_1V0_EN due to SoC
t18 int N/A N/A N/A
(Delay) rail
sequencing
requirements
VCC3P3_S0
t19 ext (Switch
Delay)
Platform delays based on component delays.
VCC1P0_S0
t21 ext (Switch
Delay)
OVOUT_1P8_
t22 int
S0 (Delay)
Track the VCC3P3_S0 with negligible delay.
OVOUT_1P05
t23 int
_S0 (Delay)
Assertion of
t25 ext N/A N/A N/A
S0_1P0_PG
S0_1V5_EN
t26 int N/A N/A N/A
(Delay)
VCC1P5_S0
t27 ext Platform delays based on component delays.
(Delay)
Platform detection of all S0 rails to only

Assertion of
t28 ext activate once they are at ~95%, the delay is 1
S0_PG
a RC so is platform controlled
1 - Must be asserted after internal power good assertions from respective [S5/S3/S0] LDO regulators
9.2.6 Going to Sleep: Transitions from S0 to S3 or S4/S5

Entry to sleep states (S3, S4/S5) is initiated by any of the following methods:
• Setting the desired type in PM1C.SLPTYPE and setting PM1C.SLPEN in the Legacy
Bridge ACPI registers.
• Detection of a catastrophic event causes a direct transition to S4/S5. This can occur
when the main power well is on, i.e. in the S0 state. Such an event can be initiated
by pressing PWR_BTN_B low for more than 3s. It can also be initiated by software
after detection of a temperature (see Chapter 10.0) or other error event.
9.2.7 Wake Events: Transition from S3 to S0

Wake events are used to return the system to S0 and can only be initiated in state S3.
They are controlled completely by the PMC. Upon exit from sleep states, the
PM1S.WAKE bit in the Legacy Bridge ACPI registers will be set.
Table 68 describes these events:

Table 68. Intel® Quark™ SoC X1000 S3 Wake Events

Event How Enabled Description
Set both PM1S.RTC and PM1E.RTC in Triggered by RTC asserting IRQ#8 in

RTC Alarm
Legacy Bridge Legacy Block
Set both GPE0STS.GPIO and Generates SCI/SMI via ACPI GPE0

Resume GPIOSUS
GPE0EN.GPIO in Legacy Bridge registers in Legacy Bridge
Set both GPE0S.EGPE and GPE0E.EGPE Generates SCI/SMI via General Purpose
External GPE_B (pin)
in Legacy Bridge Event Register in Legacy Bridge
Input to SoC that indicates a PCI

WAKE_B (pin) N/A
Express port wants to wake the system.
PWR_BTN_B press (active low) triggers

Power button Press PWR_BTN_B
transition to S0
RESET_BTN_B press (active low)

Reset button Press RESET_BTN_B
triggers transition to S0
9.2.8 System Reset Sequences

There are two types of reset:
• Cold Reset: Results in power cycling of all rails except the RTC well. The entire chip
is reset except for the RTC well.
• Warm Reset: Results in a reset without the removal of power. All core logic gets
reset. The suspend and RTC wells are not reset. The system remains in state S0.
Table 69. SoC Reset Events

Sequence Type How Initiated
Switch AC power off and then on again

Cold Boot When not in auto power button mode PWR_BTN_B is asserted (low) for at
least 2.5ms
Cold Reset (Internal) RSTC.COLD_RST bit set
Warm Reset (External) RESET_BTN_B pin asserted (low) for at least 2.5ms
Warm Reset (Internal) RSTC.WARM_RST bit set
9.2.8.1 Cold Boot Sequence

Cold boot happens when AC power is turned off and on again. The power button is used
to wake up the system: assert PWR_BTN_B low for at least 2.5ms. In auto power
button mode (PWR_BTN_B strapped low) the system proceeds straight to state S0 once
the AC power is applied, as documented in the power up sequence.
9.2.8.2 Cold Reset Sequence

Cold reset is initiated by the CPU writing to Reset Control Register bit RSTC.COLD_RST
in the Legacy Bridge. Cold reset causes a full cycling of power. The chip is transitioned
to state S4/S5 and then back to S0, independent of the setting of PWR_BTN_B. All
functions are reset except for those powered from the RTC well.
The Watchdog Timer in the Legacy Bridge can be enabled to generate a cold reset in
the event of a timeout event. This is indistinguishable from a cold reset due to
RSTC.COLD_RST being set.

9.2.8.3 Warm Reset Sequence (Internal)

Warm reset is initiated by the CPU writing to Reset Control Register bit
RSTC.WARM_RST in the Legacy Bridge. Warm reset causes reset of the CPU and
peripherals without switching off their power supplies.
The Watchdog Timer in the Legacy Bridge can be enabled to generate a warm reset in
the event of a timeout event. This is indistinguishable from a warm reset due to
RSTC.WARM_RST being set.
9.2.8.4 Externally Initiated Warm Reset Sequence

Warm Reset can also be externally initiated by asserting the reset button
RESET_BTN_B. It results in reset without removal of power on any of the supplies.
9.2.9 Handling Power Failures

When the power supply to SoC is removed in a disorderly fashion, either through
removal of the coin-cell battery or a failure on the AC supply, a normal cold boot
sequence should be initiated.
§§

Thermal Management—Intel® Quark™ SoC X1000
10.0 Thermal Management
10.1 Overview
The Intel® Quark™ SoC X1000 thermal management feature helps in managing the
overall thermal profile of the system to prevent overheating and system breakdown.
The architecture implements various proven methods of maintaining maximum
performance while remaining within the thermal specification.
The thermal management features are:

• On-die thermal sensor
• Supports a hardware trip point and programmable trip points based on the
temperature indicated by thermal sensor.
— Hot trip point is usually used to indicate that the system has reached a
threshold temperature at which damage may occur. A possible action might be
to turn on a fan to cool the system.
— Catastrophic trip point is usually used to indicate that the system has reached
the maximum possible temperature. A possible action might be to shut down
the system.
— See Section 12.7.3.9 for the register description.
10.2 Thermal Sensor

The SoC provides an on-die Thermal Sensor that can be read via Message Bus
registers. The Thermal Sensor provides an 8-bit temperature reading with a resolution
of 1 degree Celsius.
To use the Thermal Sensor:

1. It first must be taken out of reset by setting bit 0 of Msg Port 31:R34h (see
Section 12.7.6.1) to 0 and subsequently enabled via bit 15 of Msg Port 04:RB0h
(see Section 12.7.3.7).
2. Once enabled, registers within the Remote Management Unit can be used to
configure trip points and read the current temperature value.
Table 70. Thermal Sensor Signals

Direction/
Type
I/O
TS_TDA (Reference current – thermal diode anode) max voltage 0.7
Analog
I/O
TS_TDC (Reference current – thermal diode cathode) max voltage 0.7
Analog
I/O
TS_IREF_N (Reference current) max voltage 0.7
Analog

Intel® Quark™ SoC X1000—Thermal Management
§§

Processor Core—Intel® Quark™ SoC X1000
11.0 Processor Core
Processor core features on the Intel® Quark™ SoC X1000 include:
• Single processor core

• Single Instruction 5-stage pipeline
• 32-bit processor with 32-bit data bus
• Integrated Floating Point Unit
• 16 KByte, 4-way shared instruction and data L1 write-back cache
• Integrated local APIC
• Support for IA 32-bit Pentium x86 ISA compatibility
• Supports C0, C1, and C2 states
• Supports Supervisor Mode Execution Protection (SMEP)
• Supports Execute-Disable Page Protection (PAE.XD)
Note: The processor core provides an integrated Local APIC but does not support the
IA32_APIC_BASE MSR. As a result, the Local APIC is always globally enabled and the
Local APIC base address is fixed at FEE00000h. Attempting to access the
IA32_APIC_BASE MSR causes a general protection fault.
See the Intel® Quark™ SoC X1000 Core Hardware Reference Manual (Order #329678)
for more information.
§§

Intel® Quark™ SoC X1000—Processor Core

Host Bridge—Intel® Quark™ SoC X1000
12.0 Host Bridge
The Host Bridge is a central hub that routes transactions to and from the Intel® Quark™
SoC X1000’s CPU core, DRAM controller, and other functional blocks. In general, it
handles:
• CPU Core Interface: Requests for CPU Core-initiated memory and I/O read and
write operations and processor-initiated message-signaled interrupt transactions
• Device MMIO and PCI configuration access routing
• Buffering and memory arbitration
• PCI Config and MMIO accesses to host device (0/0/0)
12.1 Embedded SRAM (eSRAM)

The Host Bridge contains an interface to 512KB of on-chip, low latency, embedded
SRAM (eSRAM). The eSRAM memory may be used as either 128 x 4KB pages, or in
block mode as a single contiguous 512KB block page. The eSRAM pages may be
mapped anywhere in the physical address space as a DRAM overlay.
The eSRAM is a volatile memory and functionality is provided to flush eSRAM pages to
DRAM as part of entry to an S3 system state. Sections of DRAM overlaid by eSRAM are
inaccessible to all system agents.
12.1.1 Initialization
Immediately on coming out of a warm or cold reset, the Host Bridge initializes eSRAM
data to 0. While this is taking place the register fields ESRAMPGCTRLx.INIT_IN_PROG
(where x=0-127) and ESRAMPGCTRL_BLOCK.BLOCK_INIT_IN_PROG are 1. Software
may map and enable eSRAM pages during this time, but accesses to an eSRAM 4KB
page will not complete until the page has completed initialization.
12.1.2 Configuration
Once an eSRAM page (4KB page or 512KB block page) is enabled (see Section 12.1.2.1
and Section 12.1.2.2), it may only be flushed or disabled as part of an entry to an S3
system state. In order to re-configure an eSRAM page, the Host Bridge must be warm
or cold reset.
12.1.2.1 4KB Page Mode

The Host Bridge provides 128 registers (ESRAMPGCTRLx, where x=0-127) that allow
individual configuration of the 128*4KB eSRAM pages. If the block page is already
enabled, it is not possible to individually map 4KB pages.
To map and enable 4KB page x, the following steps should be followed:
• Set ESRAMPGCTRLx.PG_SYSTEM_ADDRESS_4K to the required address value
• Set ESRAMPGCTRLx.ENABLE_PG to 1

Intel® Quark™ SoC X1000—Host Bridge
Software must be careful not to map different eSRAM pages to the same system
address. There is no hardware protection against this.
Figure 23. eSRAM 4KB Page Mapping
4 Gbyte
MMIO
HMBOUND
2*4KB unused
Low DRAM
2*4KB unused
10*4KB unused
1 Mbyte
DOS DRAM 1*4KB unused
Physical Address Space DRAM Address

eSRAM DRAM Overlay
Space
12.1.2.2 512KB Block Page Mode

To map the eSRAM as a single 512KB block page, the register ESRAMPGCTRL_BLOCK is
used. If any of the 4KB pages are already enabled, it is not possible to enable the block
page.
To map and enable the 512KB block page, the following steps should be followed

• Set ESRAMPGCTRL_BLOCK.BLOCK_PG_SYSTEM_ADDRESS_16MB to the required

address value
• Set ESRAMPGCTRL_BLOCK.BLOCK_ENABLE_PG to 1
Figure 24. eSRAM 512KB Page Mapping
4 Gbyte
MMIO
HMBOUND
Low DRAM
512KB unused
1 Mbyte
DOS DRAM
Physical Address Space DRAM Address

eSRAM DRAM Overlay
Space
12.1.3 Configuration Locking

Once an eSRAM page is enabled, the page configuration is implicitly locked and any
further configuration change attempts will fail.

eSRAM page configuration may be explicitly locked on a per page basis with the
ESRAMPGCTRLx. PAGE_CSR_LOCK and ESRAMPGCTRL_BLOCK.
BLOCK_PAGE_CSR_LOCK fields. Locked eSRAM pages may still be flushed to DRAM.
All eSRAM configuration registers may be locked with the

ESRAMCTRL.eSRAM_GLOBAL_CSR_LOCK field.
12.1.4 ECC Protection

The Host Bridge implements ECC protection for the eSRAM. The eSRAM ECC provides
single bit error correction and double bit error detection (SECDED). It is enabled by
default, but may be disabled/enabled by setting the register field
ESRAMCTRL.SECDED_ENABLE to 0/1.
The ESRAMCERR register provides debug information on the most recent single bit ECC
error. Software may configure a threshold number of correctable ECC errors with the
ESRAMCTRL.ECC_THRESH field. If the ECC_THRESH_SB_MSG_EN field is set to 1, and
the threshold number of correctable ECC errors is reached, the Memory Manager will
send an interrupt to the Remote Management Unit with the opcode 0xD8.
The ESRAMUERR register provides debug information on un-correctable ECC errors. If

an un-correctable eSRAM ECC error occurs the Memory Manager will send a message to
inform the Remote Management Unit. A warm reset of the Host Bridge is required in
this case.
The ESRAMSDROME register can be used to decode where in the eSRAM data word the
most recent ECC error occurred.
12.1.5 Flush to DRAM

In order to flush a page to DRAM, software must set the ESRAMPGCTRLx.FLUSH_PG_ENABLE
field to 1 for 4KB pages or the ESRAMPGCTRL_BLOCK.BLOCK_FLUSH_PG_ENABLE to 1 for the
512KB block page. On an S3 entry or warm reset, firmware will flush pages configured in this way to
DRAM.
12.2 Isolated Memory Regions (IMR)

The Host Bridge provides support for Isolated Memory Regions (IMRs). An IMR is an
area of system memory that is accessible only to certain system agents. The range and
access rights of an IMR are software configurable. There are 3 types of IMR
• General IMR
• Host Memory I/O Boundary (HMBOUND) IMR
• System Management Mode (SMM) IMR
There are 8 general IMRs available. The general IMRs allow any location of system
memory- with a 1KB granularity, to have software controlled access rights. The upper
and lower boundaries of a general IMR are set via the IMRxL.IMRL and IMRxH.IMRH
register fields (where x=one of the 8 IMRs). Read access rights are controlled via the
IMRxRM registers, write access rights are controlled via the IMRx.WM register fields.
General IMRs may be overlapping. In this case, in order be allowed access a particular
region in memory, an agent will need to have access rights to all the IMRs which
contain that region.
The HMBOUND IMR prevents access by non host agents to any region of memory above
HMBOUND. HMBOUND is software configured in the HMBOUND register.

The SMM IMR prevents access by non host agents to any region of memory contained
within the SMM region. The SMM region access rights are configured in the HSMMCTL
register.
12.2.1 IMR Violation

If an agent is blocked attempting to write to a region covered by an IMR, its write data
is dropped, and system memory is not updated.
If an agent is blocked attempting to read to a region covered by an IMR, all 0s are

returned for read data. The region of memory is still read (with ECC checking/
correction performed if enabled), but false (all 0) data is returned to the agent instead
of the real data from system memory.
If the register field BIMRVCTL.EnableIMRInt is set to 1, and an IMR violation occurs, an

interrupt will be sent to the Remote Management Unit with opcode 0xC0. In this case,
software may use the BIMRVCTL register to debug the cause of the violation.
12.2.2 IMR Locking

The following settings lock the relevant IMR
• General IMR: set IMRxL.IMR_LOCK to 1
• HMBOUND IMR: set HMBOUND.HMBOUND_LOCK to 1
• SMM IMR: set HSMMCTL.SMM_LOCK to 1
Until HMBOUND is configured and locked, any General IMR region that is programmed
will only be applied if the General IMR’s register set is not locked. This allows software
to configure a General IMR region and test it without locking it’s register set. Once a
General IMR register set is locked, however, HMBOUND is required to be configured and
locked or the security mechanism will deny all accesses to that General IMR region.
Until HMBOUND is configured and locked, the SMM IMR region that is programmed will
only be applied if the SMM IMR’s register set is not locked. This allows software to
configure the SMM IMR region and test it without locking it’s register set. Once the
SMM IMR register set is locked, however, HMBOUND is required to be configured and
locked or the security mechanism will deny all accesses to the SMM IMR region.
12.3 Remote Management Unit DMA

The Remote Management Unit supports DMA transfers between System Memory and
Legacy SPI Flash. The DMA engine is used on boot-up to perform the initial firmware
fetch from SPI Flash. In addition, this can be used for shadowing firmware to DRAM or
eSRAM.
Remote Management Unit message bus registers - SPI DMA Count Register
(P_CFG_60), SPI DMA Destination Register (P_CFG_61) and SPI DMA Source Register
(P_CFG_62) are used to control DMA transfers. These registers are managed by the
Remote Management Unit firmware.
The SPI DMA Count Register (P_CFG_60) should be programmed after the SPI DMA
Source Register (P_CFG_62) and the SPI DMA Destination Register (P_CFG_61) as
writing to the SPI DMA Count Register (P_CFG_60) will trigger the start of the DMA
transfer.
See Option Register 1(P_CFG_72) bit [0] for details on how to disable DMA
functionality.

12.4 Register Map

See Chapter 5.0, “Register Access Methods” for additional information.
Figure 25. Intel® Quark™ SoC X1000 Host Bridge Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem)
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1
IO Fabric
D:21
SPI1 F:1
I2C*/GPIO F:2
Legacy Bridge
D:31, F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7
12.5 PCI Configuration Registers
Table 71. Summary of PCI Configuration Registers—0/0/0

Default
Offset Start Offset End Register ID—Description
Value
“PCI Device ID and Vendor ID Fields (PCI_DEVICE_VENDOR)—Offset 0h” on

0h 3h 09588086h
page 133
“PCI Status and Command Fields (PCI_STATUS_COMMAND)—Offset 4h” on

4h 7h 00000007h
page 133
“PCI Class Code and Revision ID Fields (PCI_CLASS_REVISION)—Offset 8h” on

8h Bh 06000000h
page 134
Ch Fh “PCI Miscellaneous Fields (PCI_MISC)—Offset Ch” on page 134 00000000h
“PCI Subsystem ID and Subsystem Vendor ID Fields (PCI_SUBSYSTEM)—Offset

2Ch 2Fh 00000000h
2Ch” on page 135

Table 71. Summary of PCI Configuration Registers—0/0/0 (Continued)

Default
Value
“Message Bus Control Register (MCR) (SB_PACKET_REG)—Offset D0h” on

D0h D3h 00000000h
page 135
D4h D7h “Message Data Register (MDR) (SB_DATA_REG)—Offset D4h” on page 136 00000000h
“Message Control Register eXtension (MCRX) (SB_ADDR_EXTN_REG)—Offset

D8h DBh 00000000h
D8h” on page 136
F8h FBh “Manufacturer ID (PCI_MANUFACTURER)—Offset F8h” on page 137 00000FB1h
12.5.1 PCI Device ID and Vendor ID Fields (PCI_DEVICE_VENDOR)—

Offset 0h
PCI Device ID
Access Method
(Size: 32 bits)
Default: 09588086h
31 28 24 20 16 12 8 4 0
0 0 0 0 1 0 0 1 0 1 0 1 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
VENDOR_ID
DEVICE_ID
Bit Default &

Description
Range Access
0958h
31: 16 Device ID (DEVICE_ID): PCI Device ID
RO
8086h
15: 0 Vendor ID (VENDOR_ID): PCI Vendor ID for Intel
RO
12.5.2 PCI Status and Command Fields (PCI_STATUS_COMMAND)—

Offset 4h
Access Method
Type: PCI Configuration Register PCI_STATUS_COMMAND: [B:0, D:0, F:0] + 4h
(Size: 32 bits)
Default: 00000007h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1
STATUS
COMMAND

Bit Default &

Description
Range Access
0000h
31: 16 Status (STATUS): Hardwired to 0.
RO
0007h
15: 0 Command (COMMAND): Hardwired to 0.
RO
12.5.3 PCI Class Code and Revision ID Fields

(PCI_CLASS_REVISION)—Offset 8h
PCI Revision ID
Access Method
Type: PCI Configuration Register PCI_CLASS_REVISION: [B:0, D:0, F:0] + 8h
(Size: 32 bits)
Default: 06000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLASS_CODE
REVISION_ID
Bit Default &
Description
Range Access
060000h
31: 8 Class Code (CLASS_CODE): PCI Class Code for Chipset.
RO
00h
7: 0 Revision ID (REVISION_ID): PCI Revision ID
RO
12.5.4 PCI Miscellaneous Fields (PCI_MISC)—Offset Ch

Access Method
Type: PCI Configuration Register PCI_MISC: [B:0, D:0, F:0] + Ch
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LATENCY
CACHE_LINE_SIZE
BIST
HEADER

Bit Default &

Description
Range Access
00h
31: 24 BIST (BIST): PCI BIST Field
RO
00h
23: 16 Header Type (HEADER): PCI Header Type Field
RO
00h
15: 8 Latency Timer (LATENCY): PCI Latency Timer Field
RO
00h
7: 0 Cache Line Size (CACHE_LINE_SIZE): PCI Cache Line Size Field
RO

(PCI_SUBSYSTEM)—Offset 2Ch
PCI Subsystem ID
Access Method
Type: PCI Configuration Register PCI_SUBSYSTEM: [B:0, D:0, F:0] + 2Ch
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SUBSYSTEM_VENDOR_ID
SUBSYSTEM_ID
Bit Default &

Description
Range Access
0000h
31: 16 Subsystem Vendor ID (SUBSYSTEM_VENDOR_ID): PCI Subsystem Vendor ID
RO
0000h
15: 0 Subsystem ID (SUBSYSTEM_ID): PCI Subsystem ID
RO
12.5.6 Message Bus Control Register (MCR) (SB_PACKET_REG)—Offset

D0h
Access Method
Type: PCI Configuration Register SB_PACKET_REG: [B:0, D:0, F:0] + D0h
(Size: 32 bits)
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SB_ADDR
SB_OPCODE
SB_PORT
SB_BE
RSV
Bit Default &
Description
Range Access
00h
31: 24 OpCode (SB_OPCODE): The operation to be performed on the target port.
WO
00h
23: 16 Port (SB_PORT): The device or unit to be targeted by the message bus transaction.
WO
00h Offset/Register (SB_ADDR): Bits 7:0 of the private register offset to be targeted by
15: 8 the message bus transaction. This field applies only to register read and write
WO operations.
0h Byte Enable (SB_BE): The byte enables to be used by the triggered transaction. This
7: 4
WO field applies only to register read and write operations.
0h
3: 0 Reserved (RSV): Reserved.
WO
12.5.7 Message Data Register (MDR) (SB_DATA_REG)—Offset D4h

Access Method
Type: PCI Configuration Register
(Size: 32 bits) SB_DATA_REG: [B:0, D:0, F:0] + D4h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SB_DATA
Bit Default &

Description
Range Access
0h Data (SB_DATA): Used as the place to store the data when the operation triggered is
31: 0 a read semantic, or the place to get the data if the triggered operation is a data write
RW semantic.
12.5.8 Message Control Register eXtension (MCRX)

(SB_ADDR_EXTN_REG)—Offset D8h
Access Method
(Size: 32 bits) SB_ADDR_EXTN_REG: [B:0, D:0, F:0] + D8h
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
SB_ADDR_EXTN
Bit Default &
Description
Range Access
000000h Offset/Register Extension (SB_ADDR_EXTN): This is used for messages sent to

31: 8 end points that require more than 8 bits for the offset/register. These bits are a direct
RW/S extension of MCR[15:8].
00h
RO
12.5.9 Manufacturer ID (PCI_MANUFACTURER)—Offset F8h

Manufacturer ID
Access Method
Type: PCI Configuration Register PCI_MANUFACTURER: [B:0, D:0, F:0] + F8h
(Size: 32 bits)
Default: 00000FB1h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 1 1 0 0 0 1
RSV
MANUFACTURER_ID
Bit Default &

Description
Range Access
00h
RO
000FB1h
23: 0 Manufacturer ID (MANUFACTURER_ID): Manufacturer ID
RO

12.6 IO Mapped Register
12.6.1 ACPI Processor Block
Table 72. Summary of I/O Registers—PMBA

Offset Default
Offset End Register ID—Description
Start Value
0h 3h “Processor Control (P_CNT)—Offset 0h” on page 138 00000000h
4h 7h “Level 2 Register (P_LVL2)—Offset 4h” on page 138 00000000h
Ch Fh “C6 Control Register (P_C6C)—Offset Ch” on page 139 00000000h
12.6.1.1 Processor Control (P_CNT)—Offset 0h

Access Method
Type: I/O Register P_CNT: [PMBA] + 0h
(Size: 32 bits)
PMBA Type: Message Bus Register (Size: 32 bits)

PMBA Reference: [Port: 0x04] + 70h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
P_CNT_RSV2
P_CNT_RSV1
THROTL_EN
THROTL_DUTY_CYCLE
Bit Default &
Description
Range Access
0b
31:5 Reserved (P_CNT_RSV2): Reserved.
RO
0b Throttle Enable (THROTL_EN): When set and the processor is in C0, it enables
4 software-controlled STPCLK# throttling. The duty cycle is selected via
RW THROTL_DUTY_CYCLE. It remains in effect on each re-entry to C0 as long as enabled.
Throttle Duty Cycle (THROTL_DUTY_CYCLE): This field determines the duty cycle of
throttling (percentage of time STPCLK# is asserted) when throttling is enabled.
000b : 50% (Default)
001b : 87.5%
000b 010b : 75%
3:1
RW 011b : 62.5%
100b : 50%
101b : 37.5%
110b : 35%
111b : 12.5%
0b
0 Reserved (P_CNT_RSV1): Reserved.
RO
12.6.1.2 Level 2 Register (P_LVL2)—Offset 4h

Level 2 Register is an 8-bit register that is used to generate requests to enter C2.

Access Method
Type: I/O Register P_LVL2: [PMBA] + 4h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
GO_TO_C2
P_LVL2_RSV
Bit Default &
Description
Range Access
0b
31:8 Reserved (P_LVL2_RSV): Reserved.
RW
00h Go to C2 (GO_TO_C2): Reads to this register return all zeroes, writes have no effect.
7:0
RO Reads to this register generate a C2 request.
12.6.1.3 C6 Control Register (P_C6C)—Offset Ch

This is a read only register. It provides information on the last C-state entered and
residency in the last entered C-state
Access Method
Type: I/O Register P_C6C: [PMBA] + Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
P_C6C_RSV
LAST_CSTATE
RESIDENCY_COUNT
Bit Default &

Description
Range Access
0b
31 Reserved (P_C6C_RSV): Reserved.
RO
Last Entered C-State (LAST_CSTATE): Once CPU transitions from C0 to C2, it

0b updates this register.
30:27 0000b : C0
RO 0010b : C2
All other values are reserved.

Bit Default &

Description
Range Access
0b Residency Count (RESIDENCY_COUNT): This register reports the residency in the

26:0
RO last entered C-state. The granularity used is microseconds
12.6.2 SPI DMA Block

DMA functionality must be disabled on boot completion to prevent an attacker from
using it to take control of the system.
12.6.2.1 Option Register 1(P_CFG_72) —Offset 72h

Access Method
Type: I/O Register
Offset: [Port: 0x04] + 72h
(Size: 32 bits)
SPI_DMA_BAR Type: Message Bus Register (Size: 32 bits)

SPI_DMA_BAR Reference: [Port: 0x04] + 72h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Reserved
DMA_DISABLE
Bit Default &
Field Name (ID): Description
Range Access
0
31:1 Reserved (RSVD): Reserved.
RO
0 DMA_DISABLE: Remote Management Unit DMA disable. Once set RMU DMA
0
RW functionality is disabled until system reset
12.7 Message Bus Register
12.7.1 Host Bridge Arbiter (Port 0x00)
Table 73. Summary of Message Bus Registers—0x00

Default
Offset Register ID—Description
Value
0h “Enhanced Configuration Space (AEC_CTRL)—Offset 0h” on page 141 00000000h
21h “STATUS—Offset 21h” on page 141 00000000h
50h “Requester ID Match Control (ASUBCHAN_CTRL)—Offset 50h” on page 142 00000000h
51h “Requester ID Match Sub-Channel 1 (ASUBCHAN1_MATCH)—Offset 51h” on page 143 00000000h

12.7.1.1 Enhanced Configuration Space (AEC_CTRL)—Offset 0h

Access Method
Type: Message Bus Register AEC_CTRL: [Port: 0x00] + 0h
(Size: 32 bits)
Op Codes:
10h - Read, 11h - Write
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EC_BASE
RSV
EC_ENABLE
Bit Default &
Description
Range Access
Enhanced Configuration Space Base Address (EC_BASE): When EC_BASE

0b matches bits [31:28] of a system memory address that has been forwarded to the Host
31:28
RW Bridge Arbiter and Enhanced Configuration operation is enabled, the corresponding
operation is treated as an Enhanced Configuration Space access.
0b
27:1 Reserved (RSV): Reserved.
RO
0b
0 Enable (EC_ENABLE): Enables Enhanced Configuration operation
RW
12.7.1.2 STATUS—Offset 21h

Access Method
Type: Message Bus Register ASTATUS: [Port: 0x00] + 21h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
STATUS1_RAISED_VALUE
STATUS0_RAISED_VALUE
STATUS1_DEFAULT_VALUE
STATUS0_DEFAULT_VALUE
RSV2
RSV1
Bit Default &

Description
Range Access
0b
31:12 RSV2: Reserved
RO

Bit Default &

Description
Range Access
STATUS1_RAISED_VALUE: Elevated Status value presented to the Memory Manager

for Virtual Channel 1 (VC1) accesses from the Host Bridge Arbiter.
0b Encodings as follows:
11:10
RW 2'b00 = Casual - Channel is requesting low priority, will be serviced only if convenient.
2'b01 = Reserved.
2'b10 = Normal - Channel is requesting normal priority for servicing.
2'b11 = Urgent - Channel is requesting high priority for servicing.
STATUS0_RAISED_VALUE: Elevated Status value presented to the Memory Manager

9:8
2'b01 = Reserved.
0b
7:4 RSV1: Reserved
RO
STATUS1_DEFAULT_VALUE: Default Status value presented to the Memory Manager

3:2
2'b01 = Reserved.
STATUS0_DEFAULT_VALUE: Default Status value presented to the Memory Manager

1:0
2'b01 = Reserved.
12.7.1.3 Requester ID Match Control (ASUBCHAN_CTRL)—Offset 50h

Access Method
Type: Message Bus Register ASUBCHAN_CTRL: [Port: 0x00] + 50h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SUB_CHAN_MATCH_LOCK
SUB_CHAN3_MRQID_ENABLE
RSV4
RSV3
RSV2
RSV1

Bit Default &

Description
Range Access
0b
31:7 Reserved (RSV4): Reserved.
RO
0b
6 Reserved (RSV3): Reserved.
RO
0b Enable Sub-Channel 3 Matching (SUB_CHAN3_MRQID_ENABLE): When set,

5
RW/L register 53h will be enabled for Sub-Channel 3 generation on a Requester ID match.
0b
RO

3
0b
RO

1
0b Lock Requester ID Matching Registers (SUB_CHAN_MATCH_LOCK): When set,

0 registers 50h-53h will be set to read-only, in order to preserve the integrity of the IMR
RW/O Sub-Channel mechanism.
12.7.1.4 Requester ID Match Sub-Channel 1 (ASUBCHAN1_MATCH)—Offset 51h

Access Method
Type: Message Bus Register ASUBCHAN1_MATCH: [Port: 0x00] + 51h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RSV1
SUB_CHAN1_MRQID
Bit Default &

Description
Range Access
0b
RO
0b
RO
0b Requester ID Match Value (SUB_CHAN1_MRQID): Value compared to an incoming

15:0
RW/L Requester ID value to determine its Sub-Channel.

Access Method


(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RSV1
SUB_CHAN2_MRQID
Bit Default &
Description
Range Access
0b
RO
0b
RO

15:0

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RSV1
SUB_CHAN3_MRQID
Bit Default &

Description
Range Access
0b
RO
0b
RO

15:0

12.7.2 Host Bridge (Port 0x03)

Default
Value
3h “Host Miscellaneous Controls 2 (HMISC2)—Offset 3h” on page 146 00170001h
4h “Host System Management Mode Controls (HSMMCTL)—Offset 4h” on page 147 00060006h
8h “Host Memory I/O Boundary (HMBOUND)—Offset 8h” on page 148 40000000h
9h “Extended Configuration Space (HECREG)—Offset 9h” on page 148 00000000h
Ah “Miscellaneous Legacy Signal Enables (HLEGACY)—Offset Ah” on page 149 00008000h
Ch “Host Bridge Write Flush Control (HWFLUSH)—Offset Ch” on page 149 00010000h
40h “MTRR Capabilities (MTRR_CAP)—Offset 40h” on page 150 00000908h
41h “MTRR Default Type (MTRR_DEF_TYPE)—Offset 41h” on page 151 00000000h
42h “MTRR Fixed 64KB Range 0x00000 (MTRR_FIX64K_00000)—Offset 42h” on page 151 00000000h
46h “MTRR Fixed 16KB Range 0xA0000 (MTRR_FIX16K_A0000)—Offset 46h” on page 154 00000000h
47h “MTRR Fixed 16KB Range 0xB0000 (MTRR_FIX16K_B0000)—Offset 47h” on page 154 00000000h
48h “MTRR Fixed 4KB Range 0xC0000 (MTRR_FIX4K_C0000)—Offset 48h” on page 155 00000000h
49h “MTRR Fixed 4KB Range 0xC4000 (MTRR_FIX4K_C4000)—Offset 49h” on page 155 00000000h
4Ah “MTRR Fixed 4KB Range 0xC8000 (MTRR_FIX4K_C8000)—Offset 4Ah” on page 156 00000000h
4Bh “MTRR Fixed 4KB Range 0xCC000 (MTRR_FIX4K_CC000)—Offset 4Bh” on page 156 00000000h
4Ch “MTRR Fixed 4KB Range 0xD0000 (MTRR_FIX4K_D0000)—Offset 4Ch” on page 157 00000000h
4Dh “MTRR Fixed 4KB Range 0xD40000 (MTRR_FIX4K_D4000)—Offset 4Dh” on page 158 00000000h
4Eh “MTRR Fixed 4KB Range 0xD8000 (MTRR_FIX4K_D8000)—Offset 4Eh” on page 158 00000000h
4Fh “MTRR Fixed 4KB Range 0xDC000 (MTRR_FIX4K_DC000)—Offset 4Fh” on page 159 00000000h
50h “MTRR Fixed 4KB Range 0xE0000 (MTRR_FIX4K_E0000)—Offset 50h” on page 159 00000000h
53h “MTRR Fixed 4KB Range 0xEC000 (MTRR_FIX4K_EC000)—Offset 53h” on page 161 00000000h
54h “MTRR Fixed 4KB Range 0xF0000 (MTRR_FIX4K_F0000)—Offset 54h” on page 161 00000000h
57h “MTRR Fixed 4KB Range 0xFC000 (MTRR_FIX4K_FC000)—Offset 57h” on page 163 00000000h
58h “System Management Range Physical Base (MTRR_SMRR_PHYSBASE)—Offset 58h” on page 164 00000000h
59h “System Management Range Physical Mask (MTRR_SMRR_PHYSMASK)—Offset 59h” on page 164 00000000h
5Ah “MTRR Variable Range Physical Base 0 (MTRR_VAR_PHYSBASE0)—Offset 5Ah” on page 165 00000000h
5Bh “MTRR Variable Range Physical Mask 0 (MTRR_VAR_PHYSMASK0)—Offset 5Bh” on page 165 00000000h
5Ch “MTRR Variable Range Physical Base 1 (MTRR_VAR_PHYSBASE1)—Offset 5Ch” on page 166 00000000h
5Dh “MTRR Variable Range Physical Mask 1 (MTRR_VAR_PHYSMASK1)—Offset 5Dh” on page 167 00000000h
5Eh “MTRR Variable Range Physical Base 2 (MTRR_VAR_PHYSBASE2)—Offset 5Eh” on page 167 00000000h

Table 74. Summary of Message Bus Registers—0x03 (Continued)

Default
Value
5Fh “MTRR Variable Range Physical Mask 2 (MTRR_VAR_PHYSMASK2)—Offset 5Fh” on page 168 00000000h
60h “MTRR Variable Range Physical Base 3 (MTRR_VAR_PHYSBASE3)—Offset 60h” on page 168 00000000h
61h “MTRR Variable Range Physical Mask 3 (MTRR_VAR_PHYSMASK3)—Offset 61h” on page 169 00000000h
12.7.2.1 Host Miscellaneous Controls 2 (HMISC2)—Offset 3h

Access Method
Type: Message Bus Register HMISC2: [Port: 0x03] + 3h
(Size: 32 bits)
Op Codes:
Default: 00170001h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
PBE_STATUS
RSVD
FSEG_RD_DRAM
RSV43
RSV05
ESEG_RD_DRAM
OR_PM
RSV04
ABSEG_IN_DRAM
RSV03
Bit Default &

Description
Range Access
0h
RO
0b PBE Status (PBE_STATUS): Reflects the value of the Pending Break Event pin from
22
RO the processor
0b
RO
OR PM Signals from Legacy Bridge (OR_PM): When set, the Host Bridge will OR the
power management signals driven by the Legacy Bridge with the internal values
generated by the Remote Management Unit. This field specifies, on a signal-by-signal
basis, whether a given bit should be driven via a message from the Remote
10111b Management Unit or via a direct pin from the Legacy Bridge.
20:16
RW [20] Reserved
[19] SMI
[18] NMI
[17] INIT
[16] INTR

Bit Default &

Description
Range Access
0000h
RO
0b A and B Segment in DRAM (ABSEG_IN_DRAM): When this bit is set, memory reads
4
RW and writes targeting A-segment or B-segment are routed to DRAM
0b
RO
0b Read F Segment from DRAM (FSEG_RD_DRAM): When this bit is set, memory
2
RW reads targeting F-segment are routed to DRAM
0b Read E Segment from DRAM (ESEG_RD_DRAM): When this bit is set, memory
1
RW reads targeting E-segment are routed to DRAM
1b
0 Reserved (RSVD): Reserved.
RO
12.7.2.2 Host System Management Mode Controls (HSMMCTL)—Offset 4h

Access Method
Type: Message Bus Register HSMMCTL: [Port: 0x03] + 4h
(Size: 32 bits)
Op Codes:
Default: 00060006h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0
NON_HOST_SMM_WR_OPEN
NON_HOST_SMM_RD_OPEN
RSV42
RSV07
SMM_START
RSV06
SMM_WR_OPEN
SMM_END
SMM_RD_OPEN
Bit Default & SMM_LOCK
Description
Range Access
000h SMM Upper Bound (SMM_END): These bits are compared with bits [31:20] of the
31:20 incoming address to determine the upper 1MB aligned value of the protected SMM
RW/L range.
0b
RO
1b Non-Host SMM Writes Open (NON_HOST_SMM_WR_OPEN): Allow writes to SMM

18 space from non-host devices. The Memory Manager uses this bit to allow non-host
RW/L writes to the SMM space defined by the SMM Start and SMM End fields
1b Non-Host SMM Reads Open (NON_HOST_SMM_RD_OPEN): Allow reads to SMM

17 space from non-host devices. The Memory Manager uses this bit to allow non-host reads
RW/L to the SMM space defined by the SMM Start and SMM End fields
0b
RO

Bit Default &

Description
Range Access
000h SMM Lower Bound (SMM_START): These bits are compared with bits [31:20] of the
15:4 incoming address to determine the lower 1MB aligned value of the protected SMM
RW/L range.
0b
RO
1b SMM Writes Open (SMM_WR_OPEN): Allow non-SMM writes to SMM space. This bit
2 allows processor writes to the SMM space defined by the SMM Start and SMM End fields
RW/L even when the processor is not in SMM mode
1b SMM Reads Open (SMM_RD_OPEN): Allow non-SMM reads to SMM space. This bit
1 allows processor reads to the SMM space defined by the SMM Start and SMM End fields
RW/L even when the processor is not in SMM mode
0b SMM Locked (SMM_LOCK): When set, this bit locks this register and prevents write
0
RW/O access until the system is reset
12.7.2.3 Host Memory I/O Boundary (HMBOUND)—Offset 8h

Access Method
Type: Message Bus Register HMBOUND: [Port: 0x03] + 8h
(Size: 32 bits)
Op Codes:
Default: 40000000h
31 28 24 20 16 12 8 4 0
0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IO_DISABLE
RSV10
HMBOUND_LOCK
HMBOUND
Bit Default &

Description
Range Access
Host IO Boundary (HMBOUND): This register field is compared with the bits [31:12]
of incoming memory accesses to determine if the transaction should be routed to
40000h Memory space or MMIO space. If address bits[31:12] are greater than or equal to the
31:12
RW/L Host IO Boundary then the transaction is routed to MMIO space.
This allows the Host IO Boundary to be set to a 4KB aligned boundary. By default, the
Host IO Boundary is set at 1GB.
000h
RO
0b Host IO Disable (IO_DISABLE): When this bit is set, all accesses will be sent to
1 memory regardless of the address with the exception of accesses to the A, B, E and F
RW/L Segments. Access to the segments is controlled by HMISC2
0b HMBOUND Lock (HMBOUND_LOCK): When this bit is set, the HMBOUND register is
0
RW/O locked and can no longer be modified until Host Bridge is reset
12.7.2.4 Extended Configuration Space (HECREG)—Offset 9h

Access Method

Type: Message Bus Register HECREG: [Port: 0x03] + 9h

(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV11
EC_BASE
EC_ENABLE
Bit Default &
Description
Range Access
0000b Extended Configuration Space Base Address (EC_BASE): This field describes the
31:28 upper 4-bits of the 32-bit address range used to access the memory-mapped
RW configuration space. This field must not be set to 0xF
000000h
RO
Extended Configuration Space Enable (EC_ENABLE): When set, causes the

0b EC_Base range to be compared to incoming memory accesses. If bits [31:28] of the
0
RW memory access match the EC_Base value then a posted memory access is treated as a
non-posted configuration access.
12.7.2.5 Miscellaneous Legacy Signal Enables (HLEGACY)—Offset Ah

Access Method
Type: Message Bus Register HLEGACY: [Port: 0x03] + Ah
(Size: 32 bits)
Op Codes:
Default: 00008000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
RSV
SMI
Bit Default &

Description
Range Access
0b
31: 13 Reserved: Reserved.
RO
0b SMI Pin Value (SMI): Reflects the value of the SMI pin set via message 0x70. Pin
12 value can also be set by writes to this register field.
RW/SE
0b
11: 0 Reserved: Reserved.
RO
12.7.2.6 Host Bridge Write Flush Control (HWFLUSH)—Offset Ch

Access Method

Type: Message Bus Register HWFLUSH: [Port: 0x03] + Ch

(Size: 32 bits)
Op Codes:
Default: 00010000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV18
RSV17
ALL_FLUSHED
HWM
Bit Default &
Description
Range Access
0000h
RO
1b All Entries Flushed (ALL_FLUSHED): Indicates all dirty entries have been flushed
16
RO from the Host Bridge to the Memory Manager
00h
RO
00h High Water Mark (HWM): High Water Mark for Dirty Entries within the Host Bridge.
7:0 When this threshold is exceeded, entries are flushed from the Host Bridge to the
RW Memory Manager. Valid values are 0x00, 0x01 and 0x02
12.7.2.7 MTRR Capabilities (MTRR_CAP)—Offset 40h

Access Method
Type: Message Bus Register MTRR_CAP: [Port: 0x03] + 40h
(Size: 32 bits)
Op Codes:
Default: 00000908h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1 0 0 0
WC
RSV21
RSV20
SMRR
FIX
VCNT
Bit Default &

Description
Range Access
00000h
RO
1b System Management Register Range Supported (SMRR): System Management

11
RO Register Range supported if set
0b Write Combining Memory Type Supported (WC): Write Combining memory

10
RO supported if set
0b
RO

Bit Default &

Description
Range Access
1b Fixed Range Registers Supported (FIX): Indicates fixed range registers are
8
RO supported if set
08h Variable Range Registers Count (VCNT): Indicates the number of variable range
7:0
RO registers implemented in the Host Bridge
12.7.2.8 MTRR Default Type (MTRR_DEF_TYPE)—Offset 41h

Access Method
Type: Message Bus Register MTRR_DEF_TYPE: [Port: 0x03] + 41h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV23
RSV22
E
FE
DEF_TYPE
Bit Default &
Description
Range Access
00000h
RO
MTRR Enable (E): MTRRs are enabled when set and are disabled when clear and the
0b UC memory type is applied to all of physical memory. When this flag is set, the FE flag
11 can disable the fixed range MTRRs. When the flag is clear, the FE flag has no affect.
RW When the E flag is set, the type specified in the default memory type field is used for
areas of memory not already mapped by either a fixed or variable MTRR.
Fixed MTRR Enable (FE): Fixed range MTRRs are enabled when set and are disabled
0b when clear. When the fixed range MTRRs are enabled, they take priority over the
10 variable range MTRRs when overlaps in ranges occur. If the fixed range MTRRs are
RW disabled, the variable range MTRRs can still be used and can map the range ordinarily
covered by the fixed range MTRRs.
00b
RO
00h Default Memory Type (DEF_TYPE): Indicates the default memory type used for
7:0 memory address ranges that do not have a memory type specified for them by an
RW MTRR. Default value is UC (Uncached)
12.7.2.9 MTRR Fixed 64KB Range 0x00000 (MTRR_FIX64K_00000)—Offset 42h

Access Method
Type: Message Bus Register MTRR_FIX64K_00000: [Port: 0x03] + 42h
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX64K_30000
FIX64K_20000
FIX64K_10000
FIX64K_00000
Bit Default &
Description
Range Access
00h Fixed 64KB Range 0x30000 (FIX64K_30000): Maps the 64KB range from 0x30000
31:24
RW to 0x3FFFF
23:16
RW to 0x2FFFF
15:8
RW to 0x1FFFF
7:0
RW to 0x0FFFF

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX64K_70000
FIX64K_60000
FIX64K_50000
FIX64K_40000
Bit Default &

Description
Range Access
31:24
RW to 0x7FFFF
23:16
RW to 0x6FFFF
15:8
RW to 0x5FFFF
7:0
RW to 0x4FFFF

Access Method


(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX16K_8C000
FIX16K_88000
FIX16K_84000
FIX16K_80000
Bit Default &
Description
Range Access
00h Fixed 16KB Range 0x8C000 (FIX16K_8C000): Maps the 16KB range from 0x8C000
31:24
RW to 0x8FFFF
23:16
RW to 0x8BFFF
15:8
RW to 0x87FFF
7:0
RW to 0x83FFF

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX16K_9C000
FIX16K_98000
FIX16K_94000
FIX16K_90000
Bit Default &

Description
Range Access
00h Fixed 16KB Range 0x9C000 (FIX16K_9C000): Maps the 16KB range from 0x9C000
31:24
RW to 0x9FFFF
23:16
RW to 0x9BFFF
15:8
RW to 0x97FFF

Bit Default &

Description
Range Access
7:0
RW to 0x93FFF
12.7.2.13 MTRR Fixed 16KB Range 0xA0000 (MTRR_FIX16K_A0000)—Offset

46h
Access Method
Type: Message Bus Register MTRR_FIX16K_A0000: [Port: 0x03] + 46h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX16K_A8000
FIX16K_A4000
FIX16K_A0000
FIX16K_AC000
Bit Default &

Description
Range Access
00h Fixed 16KB Range 0xAC000 (FIX16K_AC000): Maps the 16KB range from 0xAC000
31:24
RW to 0xAFFFF
00h Fixed 16KB Range 0xA8000 (FIX16K_A8000): Maps the 16KB range from 0xA8000
23:16
RW to 0xABFFF
15:8
RW to 0xA7FFF
7:0
RW to 0xA3FFF
12.7.2.14 MTRR Fixed 16KB Range 0xB0000 (MTRR_FIX16K_B0000)—Offset

47h
Access Method
Type: Message Bus Register MTRR_FIX16K_B0000: [Port: 0x03] + 47h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX16K_BC000
FIX16K_B8000
FIX16K_B4000
FIX16K_B0000

Bit Default &

Description
Range Access
00h Fixed 16KB Range 0xBC000 (FIX16K_BC000): Maps the 16KB range from 0xBC000
31:24
RW to 0xBFFFF
00h Fixed 16KB Range 0xB8000 (FIX16K_B8000): Maps the 16KB range from 0xB8000
23:16
RW to 0xBBFFF
15:8
RW to 0xB7FFF
7:0
RW to 0xB3FFF
12.7.2.15 MTRR Fixed 4KB Range 0xC0000 (MTRR_FIX4K_C0000)—Offset 48h

Access Method
Type: Message Bus Register MTRR_FIX4K_C0000: [Port: 0x03] + 48h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_C3000
FIX4K_C2000
FIX4K_C1000
FIX4K_C0000
Bit Default &
Description
Range Access
00h Fixed 4KB Range 0xC3000 (FIX4K_C3000): Maps the 4KB range from 0xC3000 to
31:24
RW 0xC3FFF
23:16
RW 0xC2FFF
15:8
RW 0xC1FFF
7:0
RW 0xC0FFF
12.7.2.16 MTRR Fixed 4KB Range 0xC4000 (MTRR_FIX4K_C4000)—Offset 49h

Access Method
Type: Message Bus Register MTRR_FIX4K_C4000: [Port: 0x03] + 49h
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_C7000
FIX4K_C6000
FIX4K_C5000
FIX4K_C4000
Bit Default &
Description
Range Access
31:24
RW 0xC7FFF
23:16
RW 0xC6FFF
15:8
RW 0xC5FFF
7:0
RW 0xC4FFF
12.7.2.17 MTRR Fixed 4KB Range 0xC8000 (MTRR_FIX4K_C8000)—Offset 4Ah

Access Method
Type: Message Bus Register MTRR_FIX4K_C8000: [Port: 0x03] + 4Ah
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_CB000
FIX4K_CA000
FIX4K_C9000
FIX4K_C8000
Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xCB000 (FIX4K_CB000): Maps the 4KB range from 0xCB000 to
31:24
RW 0xCBFFF
00h Fixed 4KB Range 0xCA000 (FIX4K_CA000): Maps the 4KB range from 0xCA000 to
23:16
RW 0xCAFFF
15:8
RW 0xC9FFF
7:0
RW 0xC8FFF
12.7.2.18 MTRR Fixed 4KB Range 0xCC000 (MTRR_FIX4K_CC000)—Offset 4Bh

Access Method

Type: Message Bus Register MTRR_FIX4K_CC000: [Port: 0x03] + 4Bh

(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_CF000
FIX4K_CD000
FIX4K_CE000
FIX4K_CC000
Bit Default &
Description
Range Access
00h Fixed 4KB Range 0xCF000 (FIX4K_CF000): Maps the 4KB range from 0xCF000 to
31:24
RW 0xCFFFF
00h Fixed 4KB Range 0xCE000 (FIX4K_CE000): Maps the 4KB range from 0xCE000 to
23:16
RW 0xCEFFF
00h Fixed 4KB Range 0xCD000 (FIX4K_CD000): Maps the 4KB range from 0xCD000 to
15:8
RW 0xCDFFF
00h Fixed 4KB Range 0xCC000 (FIX4K_CC000): Maps the 4KB range from 0xCC000 to
7:0
RW 0xCCFFF
12.7.2.19 MTRR Fixed 4KB Range 0xD0000 (MTRR_FIX4K_D0000)—Offset 4Ch

Access Method
Type: Message Bus Register MTRR_FIX4K_D0000: [Port: 0x03] + 4Ch
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_D3000
FIX4K_D2000
FIX4K_D1000
FIX4K_D0000
Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xD3000 (FIX4K_D3000): Maps the 4KB range from 0xD3000 to
31:24
RW 0xD3FFF
23:16
RW 0xD2FFF
15:8
RW 0xD1FFF
7:0
RW 0xD0FFF

12.7.2.20 MTRR Fixed 4KB Range 0xD40000 (MTRR_FIX4K_D4000)—Offset 4Dh

Access Method
Type: Message Bus Register MTRR_FIX4K_D4000: [Port: 0x03] + 4Dh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_D7000
FIX4K_D6000
FIX4K_D5000
FIX4K_D4000
Bit Default &
Description
Range Access
31:24
RW 0xD7FFF
23:16
RW 0xD6FFF
15:8
RW 0xD5FFF
7:0
RW 0xD4FFF
12.7.2.21 MTRR Fixed 4KB Range 0xD8000 (MTRR_FIX4K_D8000)—Offset 4Eh

Access Method
Type: Message Bus Register MTRR_FIX4K_D8000: [Port: 0x03] + 4Eh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_DB000
FIX4K_DA000
FIX4K_D9000
FIX4K_D8000
Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xDB000 (FIX4K_DB000): Maps the 4KB range from 0xDB000 to
31:24
RW 0xDBFFF
00h Fixed 4KB Range 0xDA000 (FIX4K_DA000): Maps the 4KB range from 0xDA000 to
23:16
RW 0xDAFFF

Bit Default &

Description
Range Access
15:8
RW 0xD9FFF
7:0
RW 0xD8FFF
12.7.2.22 MTRR Fixed 4KB Range 0xDC000 (MTRR_FIX4K_DC000)—Offset 4Fh

Access Method
Type: Message Bus Register MTRR_FIX4K_DC000: [Port: 0x03] + 4Fh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_DF000
FIX4K_DC000
FIX4K_DD000
FIX4K_DE000
Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xDF000 (FIX4K_DF000): Maps the 4KB range from 0xDF000 to
31:24
RW 0xDFFFF
00h Fixed 4KB Range 0xDE000 (FIX4K_DE000): Maps the 4KB range from 0xDE000 to
23:16
RW 0xDEFFF
00h Fixed 4KB Range 0xDD000 (FIX4K_DD000): Maps the 4KB range from 0xDD000 to
15:8
RW 0xDDFFF
00h Fixed 4KB Range 0xDC000 (FIX4K_DC000): Maps the 4KB range from 0xDC000 to
7:0
RW 0xDCFFF
12.7.2.23 MTRR Fixed 4KB Range 0xE0000 (MTRR_FIX4K_E0000)—Offset 50h

Access Method
Type: Message Bus Register MTRR_FIX4K_E0000: [Port: 0x03] + 50h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_E3000
FIX4K_E2000
FIX4K_E1000
FIX4K_E0000

Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xE3000 (FIX4K_E3000): Maps the 4KB range from 0xE3000 to
31:24
RW 0xE3FFF
23:16
RW 0xE2FFF
15:8
RW 0xE1FFF
7:0
RW 0xE0FFF

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_E7000
FIX4K_E6000
FIX4K_E5000
FIX4K_E4000
Bit Default &
Description
Range Access
31:24
RW 0xE7FFF
23:16
RW 0xE6FFF
15:8
RW 0xE5FFF
7:0
RW 0xE4FFF

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_EB000
FIX4K_EA000
FIX4K_E9000
FIX4K_E8000
Bit Default &
Description
Range Access
00h Fixed 4KB Range 0xEB000 (FIX4K_EB000): Maps the 4KB range from 0xEB000 to
31:24
RW 0xEBFFF
00h Fixed 4KB Range 0xEA000 (FIX4K_EA000): Maps the 4KB range from 0xEA000 to
23:16
RW 0xEAFFF
15:8
RW 0xE9FFF
7:0
RW 0xE8FFF
12.7.2.26 MTRR Fixed 4KB Range 0xEC000 (MTRR_FIX4K_EC000)—Offset 53h

Access Method
Type: Message Bus Register MTRR_FIX4K_EC000: [Port: 0x03] + 53h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_EE000
FIX4K_ED000
FIX4K_EF000
FIX4K_EC000
Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xEF000 (FIX4K_EF000): Maps the 4KB range from 0xEF000 to
31:24
RW 0xEFFFF
00h Fixed 4KB Range 0xEE000 (FIX4K_EE000): Maps the 4KB range from 0xEE000 to
23:16
RW 0xEEFFF
00h Fixed 4KB Range 0xED000 (FIX4K_ED000): Maps the 4KB range from 0xED000 to
15:8
RW 0xEDFFF
00h Fixed 4KB Range 0xEC000 (FIX4K_EC000): Maps the 4KB range from 0xEC000 to
7:0
RW 0xECFFF
12.7.2.27 MTRR Fixed 4KB Range 0xF0000 (MTRR_FIX4K_F0000)—Offset 54h

Access Method

Type: Message Bus Register MTRR_FIX4K_F0000: [Port: 0x03] + 54h

(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0FIX4K_F3000 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_F2000
FIX4K_F1000
FIX4K_F0000
Bit Default &
Description
Range Access
00h Fixed 4KB Range 0xF3000 (FIX4K_F3000): Maps the 4KB range from 0xF3000 to
31:24
RW 0xF3FFF
23:16
RW 0xF2FFF
15:8
RW 0xF1FFF
7:0
RW 0xF0FFF

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_F7000
FIX4K_F6000
FIX4K_F5000
FIX4K_F4000
Bit Default &

Description
Range Access
31:24
RW 0xF7FFF
23:16
RW 0xF6FFF
15:8
RW 0xF5FFF
7:0
RW 0xF4FFF


Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_FB000
FIX4K_FA000
FIX4K_F9000
FIX4K_F8000
Bit Default &
Description
Range Access
00h Fixed 4KB Range 0xFB000 (FIX4K_FB000): Maps the 4KB range from 0xFB000 to
31:24
RW 0xFBFFF
00h Fixed 4KB Range 0xFA000 (FIX4K_FA000): Maps the 4KB range from 0xFA000 to
23:16
RW 0xFAFFF
15:8
RW 0xF9FFF
7:0
RW 0xF8FFF
12.7.2.30 MTRR Fixed 4KB Range 0xFC000 (MTRR_FIX4K_FC000)—Offset 57h

Access Method
Type: Message Bus Register MTRR_FIX4K_FC000: [Port: 0x03] + 57h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FIX4K_FF000
FIX4K_FE000
FIX4K_FD000
FIX4K_FC000
Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xFF000 (FIX4K_FF000): Maps the 4KB range from 0xFF000 to
31:24
RW 0xFFFFF
00h Fixed 4KB Range 0xFE000 (FIX4K_FE000): Maps the 4KB range from 0xFE000 to
23:16
RW 0xFEFFF

Bit Default &

Description
Range Access
00h Fixed 4KB Range 0xFD000 (FIX4K_FD000): Maps the 4KB range from 0xFD000 to
15:8
RW 0xFDFFF
00h Fixed 4KB Range 0xFC000 (FIX4K_FC000): Maps the 4KB range from 0xFC000 to
7:0
RW 0xFCFFF
12.7.2.31 System Management Range Physical Base

(MTRR_SMRR_PHYSBASE)—Offset 58h
This register may only be written while in SMM. Attempts to write this register outside
of SMM will be ignored.
Access Method
Type: Message Bus Register MTRR_SMRR_PHYSBASE: [Port: 0x03] + 58h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SMRR_PHYSBASE
RSV24
SMRR_TYPE
Bit Default &
Description
Range Access
00000h SMRR Physical Base (SMRR_PHYSBASE): Specifies the base address for the System
31:12 Management Range. This 20 bit value is extended by 12 bits at the low end to form the
RW base address
0000b
RO
00h SMRR Type (SMRR_TYPE): Specifies the memory type for System Management
7:0
RW Range
12.7.2.32 System Management Range Physical Mask

(MTRR_SMRR_PHYSMASK)—Offset 59h
This register may only be written while in SMM. Attempts to write this register outside
of SMM will be ignored.
Access Method
Type: Message Bus Register MTRR_SMRR_PHYSMASK: [Port: 0x03] + 59h
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SMRR_PHYSMASK
SMRR_VALID
RSV25
Bit Default &
Description
Range Access
00000h SMRR Physical Mask (SMRR_PHYSMASK): Specifies a mask value for the System
31:12 Management Range. The mask determines the range of the region begin mapped. The
RW mask value is extended by 12 bits at the low end to form the mask value.
0b SMRR Valid (SMRR_VALID): Enables the register pair for the System Management
11
RW Range when set and disables the register pair when clear.
0000h
RO
12.7.2.33 MTRR Variable Range Physical Base 0 (MTRR_VAR_PHYSBASE0)—

Offset 5Ah
Access Method
Type: Message Bus Register MTRR_VAR_PHYSBASE0: [Port: 0x03] + 5Ah
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE0
RSV26
VAR_TYPE0
Bit Default &

Description
Range Access
00000h Physical Base (VAR_PHYSBASE0): Specifies the base address for Variable Range 0.
31:12
RW This 20 bit value is extended by 12 bits at the low end to form the base address
0000b
RO
00h
7:0 Type (VAR_TYPE0): Specifies the memory type for Variable Range 0
RW

Offset 5Bh
Access Method

Type: Message Bus Register MTRR_VAR_PHYSMASK0: [Port: 0x03] + 5Bh

(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSMASK0
VAR_VALID0
RSV27
Bit Default &
Description
Range Access
00000h Physical Mask (VAR_PHYSMASK0): Specifies a mask value for Variable Range 0. The
31:12 mask determines the range of the region begin mapped. The mask value is extended by
RW 12 bits at the low end to form the mask value.
0b Valid (VAR_VALID0): Enables the register pair for Variable Range 0 when set and
11
RW disables the register pair when clear.
0000h
RO

Offset 5Ch
Access Method
Type: Message Bus Register MTRR_VAR_PHYSBASE1: [Port: 0x03] + 5Ch
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE1
RSV28
VAR_TYPE1
Bit Default &

Description
Range Access
31:12
0000b
RO
00h
RW


Offset 5Dh
Access Method
Type: Message Bus Register MTRR_VAR_PHYSMASK1: [Port: 0x03] + 5Dh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSMASK1
VAR_VALID1
RSV29
Bit Default &
Description
Range Access
11
0000h
RO

Offset 5Eh
Access Method
Type: Message Bus Register MTRR_VAR_PHYSBASE2: [Port: 0x03] + 5Eh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE2
RSV30
VAR_TYPE2
Bit Default &

Description
Range Access
31:12

Bit Default &

Description
Range Access
0000b
RO
00h
RW

Offset 5Fh
Access Method
Type: Message Bus Register MTRR_VAR_PHYSMASK2: [Port: 0x03] + 5Fh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSMASK2
VAR_VALID2
RSV31
Bit Default &
Description
Range Access
11
0000h
RO

Offset 60h
Access Method
Type: Message Bus Register MTRR_VAR_PHYSBASE3: [Port: 0x03] + 60h
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE3
RSV32
VAR_TYPE3
Bit Default &
Description
Range Access
31:12
0000b
RO
00h
RW

Offset 61h
Access Method
Type: Message Bus Register MTRR_VAR_PHYSMASK3: [Port: 0x03] + 61h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_VALID3
VAR_PHYSMASK3
RSV33
Bit Default &

Description
Range Access
11
0000h
RO

Offset 62h
Access Method


(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE4
RSV34
VAR_TYPE4
Bit Default &
Description
Range Access
31:12
0000b
RO
00h
RW

Offset 63h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSMASK4
VAR_VALID4
RSV35
Bit Default &

Description
Range Access
11
0000h
RO


Offset 64h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE5
RSV36
VAR_TYPE5
Bit Default &
Description
Range Access
31:12
0000b
RO
00h
RW

Offset 65h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSMASK5
VAR_VALID5
RSV37
Bit Default &

Description
Range Access

Bit Default &

Description
Range Access
11
0000h
RO

Offset 66h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE6
RSV38
VAR_TYPE6
Bit Default &
Description
Range Access
31:12
0000b
RO
00h
RW

Offset 67h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSMASK6
VAR_VALID6
RSV39
Bit Default &
Description
Range Access
11
0000h
RO

Offset 68h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSBASE7
RSV40
VAR_TYPE7
Bit Default &

Description
Range Access
31:12
0000b
RO
00h
RW

Offset 69h
Access Method


(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAR_PHYSMASK7
VAR_VALID7
RSV41
Bit Default &
Description
Range Access
11
0000h
RO
12.7.3 Remote Management Unit (Port 0x04)

Default
Offset Register Name (Register Symbol)
Value
60h “SPI DMA Count Register (P_CFG_60)—Offset 60h” on page 174 00000000h
61h “SPI DMA Destination Register (P_CFG_61)—Offset 61h” on page 175 00000000h
62h “SPI DMA Source Register (P_CFG_62)—Offset 62h” on page 175 00000000h
70h “Processor Register Block (P_BLK) Base Address (P_CFG_70)—Offset 70h” on page 176 00000000h
71h “Control Register (P_CFG_71)—Offset 71h” on page 176 00000009h
74h “Watchdog Control Register (P_CFG_74)—Offset 74h” on page 177 00040000h
B0h “Thermal Sensor Mode Register (P_CFG_B0)—Offset B0h” on page 178 00000000h
B1h “Thermal Sensor Temperature Register (P_CFG_B1)—Offset B1h” on page 178 00000000h
B2h “Thermal Sensor Programmable Trip Point Register (P_CFG_B2)—Offset B2h” on page 179 FFFFFFFFh
12.7.3.1 SPI DMA Count Register (P_CFG_60)—Offset 60h

Access Method
Type: Message Bus Register Offset: [Port: 0x04] + 60h
(Size: 32 bits)
Op Codes:

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CFG_SPI_DMA_CNT
Bit Default &
Range Access
SPI DMA Count (CFG_SPI_DMA_CNT): Count of 512 byte block transfers.

0b Writing this register triggers the start of the transfer of the indicated number of blocks.
31:0
RW Reading this register returns the number of blocks that are remaining to be transferred.
A value of 0 indicates the transfer is complete.
12.7.3.2 SPI DMA Destination Register (P_CFG_61)—Offset 61h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CFG_SPI_DMA_DST
Bit Default &

Range Access
00000000h SPI DMA Destination (CFG_SPI_DMA_DST): 32-bit Destination Address of data in

31:0
RW System Memory (eSRAM/DRAM).
12.7.3.3 SPI DMA Source Register (P_CFG_62)—Offset 62h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CFG_SPI_DMA_SRC
Bit Default &
Range Access
00000000h SPI DMA Source (CFG_SPI_DMA_SRC): 32-bit Source Address of data in Legacy
31:0
RW SPI.
12.7.3.4 Processor Register Block (P_BLK) Base Address (P_CFG_70)—Offset

70h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
P_BLK_IO_EN
CFG_70_RSV
P_BLK_IO_BAR
Bit Default &

Range Access
0b Enable (P_BLK_IO_EN): When set to “1”, decode of the IO range pointed to by the
31
RW Base Address is enabled.
0b
30:16 Reserved (CFG_70_RSV): Reserved.
RO
0b Base Address (P_BLK_IO_BAR): IO Base Address for the Processor Register Block
15:0
RW (P_BLK) decode range.
12.7.3.5 Control Register (P_CFG_71)—Offset 71h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000009h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
LOCK_THRM_CTRL_REGS
Bit Default &
Range Access
0b
RO
0b
RO
0b
RO
0b
RO
0b
RO
0b Lock Thermal Control Registers (LOCK_THRM_CTRL_REGS): Setting this bit locks

5
RW/O the thermal control registers (registers 0xB0 and 0xB2).
0b
RO
1b
RO
0b
RO
0b
RO
1b
RO
12.7.3.6 Watchdog Control Register (P_CFG_74)—Offset 74h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00040000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DBL_ECC_BIT_ERR
RSVD
RSVD
Bit Default &
Range Access
0b
RO
Double ECC Bit Error (DBL_ECC_BIT_ERR): Double ECC bit error handling selection:
01b 00b: Do nothing
19:18 01b: Catastrophic Shutdown
RW 10b: Warm Reset
11b: Send SERR
0b
RO
12.7.3.7 Thermal Sensor Mode Register (P_CFG_B0)—Offset B0h

Access Method
Type: Message Bus Register Offset: [Port: 0x04] + B0h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CFG_B0_RSV2
CFG_B0_RSV1
THRM_SNSR_EN
Bit Default &

Range Access
0b
31:16 Reserved (CFG_B0_RSV2): Reserved.
RO
0b
15 Thermal Sensor Enable (THRM_SNSR_EN): Setting to 1 Enables Thermal Sensor
RW/L
0b
RO
12.7.3.8 Thermal Sensor Temperature Register (P_CFG_B1)—Offset B1h

Access Method


(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CFG_B1_RSV2
CFG_B1_RSV1
THRM_SENSR_REL_TEMP
THRM_SENSR_TEMP
Bit Default &
Range Access
0b
RO
Thermal Sensor Relative Temperature (THRM_SENSR_REL_TEMP): The thermal

sensor relative temperature value is an 8-bit signed value relative to the Hot Trip point.
0b If the Hot Trip point minus the current temperature is greater than +127, this value is
23:16
RO clipped at +127. If the Hot Trip point minus the current temperature is less than -127,
this value is clipped at -127. Otherwise this value is a signed sum magnitude where
bit[23] is the sign and bits[22:16] are the magnitude.
0b
RO
Thermal Sensor Temperature (THRM_SENSR_TEMP): 8-bit Thermal Sensor

Temperature. The temperature in degrees Celsius is calculated by subtracting an offset
of 50 from the 8-bit register value.
0b The temperature in degrees Celsius corresponds to: 00h: -50
7:0
RO 01h: -49
...
FEh: 204
FFh: 205
12.7.3.9 Thermal Sensor Programmable Trip Point Register (P_CFG_B2)—

Offset B2h
Access Method
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh

31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
HOT_TRIP_CLEAR_THOLD
CAT_TRIP_CLEAR_THOLD
CAT_TRIP_SET_THOLD
HOT_TRIP_SET_THOLD
Bit Default &
Range Access
FFh Hot Clear Trip Point Threshold (HOT_TRIP_CLEAR_THOLD): Sets the target value
31:24
RW/L for the hot trip clear point
FFh Catastrophic Clear Trip Point Threshold (CAT_TRIP_CLEAR_THOLD): Sets the

23:16
RW/L target value for the catastrophic trip clear point
FFh Hot Set Trip Point Threshold (HOT_TRIP_SET_THOLD): Sets the target value for
15:8
RW/L the hot trip set point
FFh Catastrophic Set Trip Point Threshold (CAT_TRIP_SET_THOLD): Sets the target
7:0
RW/L value for the catastrophic trip set point
12.7.4 Memory Manager (Port 0x05)

Default
Value
1h “Control (BCTRL)—Offset 1h” on page 181 00000800h
2h “Write Flush Policy (BWFLUSH)—Offset 2h” on page 182 0C070408h
19h “Isolated Memory Region Violation Control (BIMRVCTL)—Offset 19h” on page 183 00000000h
31h “Debug 1 (DEBUG1)—Offset 31h” on page 184 4F08C20Ch
40h “Isolated Memory Region 0 Low Address (IMR0L)—Offset 40h” on page 186 00000000h
41h “Isolated Memory Region 0 High Address (IMR0H)—Offset 41h” on page 186 00000000h
42h “Isolated Memory Region 0 Read Mask (IMR0RM)—Offset 42h” on page 187 BFFFFFFFh
43h “Isolated Memory Region 0 Write Mask (IMR0WM)—Offset 43h” on page 189 FFFFFFFFh
4Ah “Isolated Memory Region 2 Read Mask (IMR2RM)—Offset 4Ah” on page 196 BFFFFFFFh
4Bh “Isolated Memory Region 2 Write Mask (IMR2WM)—Offset 4Bh” on page 198 FFFFFFFFh
4Ch “Isolated Memory Region 3 Low Address (IMR3L)—Offset 4Ch” on page 200 00000000h


Default
Value
4Dh “Isolated Memory Region 3 High Address (IMR3H)—Offset 4Dh” on page 200 00000000h
4Eh “Isolated Memory Region 3 Read Mask (IMR3RM)—Offset 4Eh” on page 201 BFFFFFFFh
4Fh “Isolated Memory Region 3 Write Mask (IMR3WM)—Offset 4Fh” on page 203 FFFFFFFFh
5Ah “Isolated Memory Region 6 Read Mask (IMR6RM)—Offset 5Ah” on page 215 BFFFFFFFh
5Bh “Isolated Memory Region 6 Write Mask (IMR6WM)—Offset 5Bh” on page 217 FFFFFFFFh
5Ch “Isolated Memory Region 7 Low Address (IMR7L)—Offset 5Ch” on page 218 00000000h
5Dh “Isolated Memory Region 7 High Address (IMR7H)—Offset 5Dh” on page 219 00000000h
5Eh “Isolated Memory Region 7 Read Mask (IMR7RM)—Offset 5Eh” on page 219 BFFFFFFFh
5Fh “Isolated Memory Region 7 Write Mask (IMR7WM)—Offset 5Fh” on page 221 FFFFFFFFh
81h “eSRAM Control (ESRAMCTRL)—Offset 81h” on page 223 047F3F91h
82h “eSRAM Block Page Control (ESRAMPGCTRL_BLOCK)—Offset 82h” on page 224 850000FFh
83h “eSRAM Correctable Error (ESRAMCERR)—Offset 83h” on page 226 00000000h
84h “eSRAM Uncorrectable Error (ESRAMUERR)—Offset 84h” on page 226 00000000h
88h “eSRAM ECC Error Syndrome (ESRAMSDROME)—Offset 88h” on page 227 00000000h
12.7.4.1 Control (BCTRL)—Offset 1h

Access Method
Type: Message Bus Register BCTRL: [Port: 0x05] + 1h
(Size: 32 bits)
Op Codes:
Default: 00000800h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
MissValidEntries
RSV2
RSV1
RSV0

Bit Default &

Description
Range Access
0000h
RO
0b
RO
1b
RO
0b
RO
0b
RO
0b Miss Valid Entries (MissValidEntries): This mode causes reads to clean valid
8 Memory Manager buffer entries that have zero reference counts so that they look like
RW misses instead of hits. It is mostly present for test purposes,
0b
RO
0h
RO
0b
RO
0b
RO
0b
RO
0b
RO
0b
RO
12.7.4.2 Write Flush Policy (BWFLUSH)—Offset 2h

Access Method
Type: Message Bus Register BWFLUSH: [Port: 0x05] + 2h
(Size: 32 bits)
Op Codes:
Default: 0C070408h
31 28 24 20 16 12 8 4 0
0 0 0 0 1 1 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0
dram_dirty_hwm
RSV0
esram_dirty_lwm
esram_dirty_hwm
DRAMAllEntriesFlushed
dram_dirty_lwm
DRAMAllEntriesIdle
ESRAMAllEntriesIdle
ESRAMAllEntriesFlushed

Bit Default &

Description
Range Access
0h
RO
1b eSRAM All Entries Idle (ESRAMAllEntriesIdle): All entries in the Memory Manager
27 eCACHE tag store have 0 reference counts and are unlocked, indicating that there are
RO no transactions in progress in this cache
1b eSRAM All Entries Flushed (ESRAMAllEntriesFlushed): All entries in the Memory

26
RO Manager eCACHE have been flushed to the eSRAM
00h eSRAM Low Water Mark (esram_dirty_lwm): Low Water Mark for Dirty Entries
25:22
RW retained by eCACHE in the Memory Manager
01h eSRAM High Water Mark (esram_dirty_hwm): High Water Mark for Dirty Entries
21:18
RW retained by eCACHE in the Memory Manager
1b DRAM All Entries Idle (DRAMAllEntriesIdle): All entries in the Memory Manager
17 DCACHE entries have 0 reference counts and are unlocked, indicating that the Memory
RO Manager has no DRAM transactions in progress
1b DRAM All Entries Flushed (DRAMAllEntriesFlushed): All entries in the Memory

16
RO Manager DCACHE have been flushed
04h DRAM Low Water Mark (dram_dirty_lwm): Low water mark for dirty entries
15:8
RW retained by the Memory Manager
08h DRAM High Water Mark (dram_dirty_hwm): High water mark for dirty entries
7:0
RW retained by the Memory Manager
12.7.4.3 Isolated Memory Region Violation Control (BIMRVCTL)—Offset 19h

This register is used to configure the interrupt and to capture status when an IMR is
violated. Note that the Enable Interrupt on IMR Violation field is reset on an IMR
violation event and this register captures the first violation only. It can be set back to 1,
and if a new violation occurs any status previously captured is overridden with the new
violation details. This register is not secured or locked because the violation interrupt is
not part of securing the region and is only intended to help in debugging IMR
configuration.
Access Method
Type: Message Bus Register BIMRVCTL: [Port: 0x05] + 19h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RSV1
RSV0
EnableIMRInt
IMRViolationRegion
IMRViolationAgent
IMRViolationSubAgent

Bit Default &

Description
Range Access
Enable Interrupt on IMR Violation (EnableIMRInt): When set, the Memory

0b Manager will latch violation information into this register and send an interrupt request
31 to the Remote Management Unit. This bit is cleared upon triggering and must be reset
RW by software in order to trigger again. Memory protection is maintained even while the
interrupt/capture mechanism is disabled
0b
RO
IMR Violation Region (IMRViolationRegion): This 14-bit value indicates which

region the last IMR violation occurred on if the IMR interrupt is enabled. A bit of this field
will be asserted for every IMR region, HMBOUND or SMM region violated by the
transaction that caused EnableIMRInt to deassert.
[29]: HMBOUND Violation
[28]: SMM Bound Violation
0000h [27:24]: Reserved
29:16 [23]: IMR7 Violation
RO [22]: IMR6 Violation
[21]: IMR5 Violation
0h
RO
IMR Violation Agent (IMRViolationAgent): This 4-bit value indicates which agent
caused the last IMR violation if the IMR interrupt is enabled:
0000b : CPU
0001b : Host Bridge Arbiter VC0
0h 0010b : Host Bridge Arbiter VC1
11:8 0011b : Reserved
RO 0100b : Reserved
0101b : Reserved
0110b : Reserved
0111b : eSRAM Flush/Init
1000b : Remote Management Unit
00h
RO
IMR Violation Sub Agent (IMRViolationSubAgent): This 3-bit value indicates which
sub-agent caused the last IMR violation, if the IMR interrupt is enabled
0h 000b : Host Bridge Arbiter Sub-Channel 0 (Anonymous)
2:0
RO 001b : Host Bridge Arbiter Sub-Channel 1
010b : Host Bridge Arbiter Sub-Channel 2
100b : Host Bridge Arbiter Sub-Channel 3
12.7.4.4 Debug 1 (DEBUG1)—Offset 31h

Access Method
Type: Message Bus Register DEBUG1: [Port: 0x05] + 31h
(Size: 32 bits)
Op Codes:
Default: 4F08C20Ch

31 28 24 20 16 12 8 4 0
0 1 0 0 1 1 1 1 0 0 0 0 1 0 0 0 1 1 0 0 0 0 1 0 0 0 0 0 1 1 0 0
EnDCACHEPartFill
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access
0b
RO
1b
RO
0b
RO
0h
RO
Fh
RO
0h
RO
1b
RO
0b
RO
0b
RO
0b
RO
1b
RO
1b
RO
0h
RO
0h
RO
2h
RO
0h
RO
1b
RO
Enable DCACHE Partial Entries (EnDCACHEPartFill): If the Memory Manager

1b admitted Request or CPU Snoop Response that has not all Byte Enables set the Entry is
2 considered Partial. Setting this bit will cause the Memory Manager to fill the partial entry
RW from DRAM, before it is being flushed to DRAM. This bit needs to be set if the Memory
Controller enables ECC mode, where partial writes are forbidden

Bit Default &

Description
Range Access
0b
RO
0b
RO
12.7.4.5 Isolated Memory Region 0 Low Address (IMR0L)—Offset 40h

Access Method
Type: Message Bus Register IMR0L: [Port: 0x05] + 40h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
IMRL
RSV0
IMR_LOCK
Bit Default &

Description
Range Access
0b IMR Lock (IMR_LOCK): Setting this bit to “1” locks the IMRX registers, preventing
31
RW/O further updates.
00h
RO
000000h IMR Low Address (IMRL): These bits are compared with bits 31:10 of the incoming
23:2
RW/L address to determine the lower 1KB aligned value of the protected range
00h
RO
12.7.4.6 Isolated Memory Region 0 High Address (IMR0H)—Offset 41h

Access Method
Type: Message Bus Register IMR0H: [Port: 0x05] + 41h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMRH
RSV1
RSV0

Bit Default &

Description
Range Access
00h
RO
000000h IMR High Address (IMRH): These bits are compared with bits 31:10 of the incoming
23:2
RW/L address to determine the upper 1KB aligned value of the protected range
0h
RO
12.7.4.7 Isolated Memory Region 0 Read Mask (IMR0RM)—Offset 42h

Access Method
Type: Message Bus Register IMR0RM: [Port: 0x05] + 42h
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh
31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
RSVD
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access
1b eSRAM Flush/Init (ESRAM_FLUSH_INIT): eSRAM Flush/Init Read Access Allowed to

31
RW/L memory delineated by IMRxL and IMRxH
0b
RO
1b Remote Management Unit (PUNIT): Remote Management Unit Read Access Allowed
29
RW/L to memory delineated by IMRxL and IMRxH
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b Host Bridge Arbiter VC1 Sub-Channel 3 (VC1_SAI_ID3): Host Bridge Arbiter VC1
15
RW/L Sub-Channel 3 Read Accesses Allowed to memory delineated by IMRxL and IMRxH.
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
CPU (CPU_0): CPU Read Access Allowed to memory region delineated by IMRxL and
1b IMRxH.
1
RW/L Note: Bit[0] and bit [1] of the IMR Read Mask register must always be programmed to
the same value
CPU (CPU0): CPU Read Access Allowed to memory region delineated by IMRxL and
1b IMRxH.
0
the same value

12.7.4.8 Isolated Memory Region 0 Write Mask (IMR0WM)—Offset 43h

Access Method
Type: Message Bus Register IMR0WM: [Port: 0x05] + 43h
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
CPU_SNOOP
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access
1b eSRAM Flush/Init (ESRAM_FLUSH_INIT): eSRAM Flush/Init Write Access Allowed

31
1b CPU Snoop (CPU_SNOOP): Dirty CPU Snoop Response Write Allowed to memory
30 delineated by IMRxL and IMRxH. Clean snoop responses are always allowed and does
RW/L not cause an IMR violation
1b Remote Management Unit (PUNIT): Remote Management Unit Write Access Allowed
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
15
RW/L Sub-Channel 3 Write Accesses Allowed to memory delineated by IMRxL and IMRxH.
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
CPU (CPU_0): CPU Write Access Allowed to memory region delineated by IMRxL and
1b IMRxH.
1
the same value
CPU (CPU0): CPU Write Access Allowed to memory region delineated by IMRxL and
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMRL
IMR_LOCK
RSV1
RSV0
Bit Default &
Description
Range Access
0b IMR Lock (IMR_LOCK): Setting this bit to “1” locks the IMRX registers, preventing
31
RW/O further updates.
00h
RO
23:2
00h
RO

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
IMRH
Bit Default &

Description
Range Access
00h
RO
23:2
0h
RO

Access Method
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh

31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
PUNIT
ESRAM_FLUSH_INIT
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_0
CPU0
Bit Default &
Description
Range Access

31
0b
RO
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13

Bit Default &

Description
Range Access
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_SNOOP
CPU_0
CPU0
ESRAM_FLUSH_INIT
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD

Bit Default &

Description
Range Access

31
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9

Bit Default &

Description
Range Access
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMR_LOCK
RSV1
IMRL
RSV0
Bit Default &

Description
Range Access
0b
31 IMR Lock (IMR_LOCK): This bit locks the IMRX registers, preventing further updates
RW/O
00h
RO
23:2
00h
RO


Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
IMRH
Bit Default &
Description
Range Access
00h
RO
23:2
0h
RO
12.7.4.15 Isolated Memory Region 2 Read Mask (IMR2RM)—Offset 4Ah

Access Method
Type: Message Bus Register IMR2RM: [Port: 0x05] + 4Ah
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh
31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
RSVD
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &

Description
Range Access

31
0b
RO
29
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value
12.7.4.16 Isolated Memory Region 2 Write Mask (IMR2WM)—Offset 4Bh

Access Method
Type: Message Bus Register IMR2WM: [Port: 0x05] + 4Bh
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
ESRAM_FLUSH_INIT
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_SNOOP
CPU_0
CPU0
Bit Default &
Description
Range Access

31
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value

Bit Default &

Description
Range Access
1b IMRxH.
0
the same value
12.7.4.17 Isolated Memory Region 3 Low Address (IMR3L)—Offset 4Ch

Access Method
Type: Message Bus Register IMR3L: [Port: 0x05] + 4Ch
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
IMRL
IMR_LOCK
Bit Default &

Description
Range Access
0b
RW/O
00h
RO
23:2
00h
RO
12.7.4.18 Isolated Memory Region 3 High Address (IMR3H)—Offset 4Dh

Access Method
Type: Message Bus Register IMR3H: [Port: 0x05] + 4Dh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMRH
RSV1
RSV0
Bit Default &

Description
Range Access
00h
RO

Bit Default &

Description
Range Access
23:2
0h
RO
12.7.4.19 Isolated Memory Region 3 Read Mask (IMR3RM)—Offset 4Eh

Access Method
Type: Message Bus Register IMR3RM: [Port: 0x05] + 4Eh
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh
31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
RSVD
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access

31
0b
RO
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

12.7.4.20 Isolated Memory Region 3 Write Mask (IMR3WM)—Offset 4Fh

Access Method
Type: Message Bus Register IMR3WM: [Port: 0x05] + 4Fh
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
CPU_SNOOP
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access

31
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMRL
IMR_LOCK
RSV1
RSV0
Bit Default &
Description
Range Access
0b
RW/O
00h
RO
23:2
00h
RO

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
IMRH
Bit Default &

Description
Range Access
00h
RO
23:2
0h
RO

Access Method
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh

31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
PUNIT
ESRAM_FLUSH_INIT
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_0
CPU0
Bit Default &
Description
Range Access

31
0b
RO
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13

Bit Default &

Description
Range Access
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_SNOOP
CPU_0
CPU0
ESRAM_FLUSH_INIT
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD

Bit Default &

Description
Range Access

31
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9

Bit Default &

Description
Range Access
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMR_LOCK
RSV1
IMRL
RSV0
Bit Default &

Description
Range Access
0b
RW/O
00h
RO
23:2
00h
RO


Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
IMRH
Bit Default &
Description
Range Access
00h
RO
23:2
0h
RO

Access Method
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh
31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
RSVD
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &

Description
Range Access

31
0b
RO
29
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
ESRAM_FLUSH_INIT
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_SNOOP
CPU_0
CPU0
Bit Default &
Description
Range Access

31
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value

Bit Default &

Description
Range Access
1b IMRxH.
0
the same value

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
IMRL
IMR_LOCK
Bit Default &

Description
Range Access
0b
RW/O
00h
RO
23:2
00h
RO

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMRH
RSV1
RSV0
Bit Default &

Description
Range Access
00h
RO

Bit Default &

Description
Range Access
23:2
0h
RO
12.7.4.31 Isolated Memory Region 6 Read Mask (IMR6RM)—Offset 5Ah

Access Method
Type: Message Bus Register IMR6RM: [Port: 0x05] + 5Ah
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh
31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
RSVD
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access

31
0b
RO
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value

12.7.4.32 Isolated Memory Region 6 Write Mask (IMR6WM)—Offset 5Bh

Access Method
Type: Message Bus Register IMR6WM: [Port: 0x05] + 5Bh
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
CPU_SNOOP
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
ESRAM_FLUSH_INIT
CPU_0
CPU0
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access

31
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO

Bit Default &

Description
Range Access
1b
RO
15
14
13
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value
12.7.4.33 Isolated Memory Region 7 Low Address (IMR7L)—Offset 5Ch

Access Method
Type: Message Bus Register IMR7L: [Port: 0x05] + 5Ch
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IMRL
IMR_LOCK
RSV1
RSV0
Bit Default &
Description
Range Access
0b
RW/O
00h
RO
23:2
00h
RO
12.7.4.34 Isolated Memory Region 7 High Address (IMR7H)—Offset 5Dh

Access Method
Type: Message Bus Register IMR7H: [Port: 0x05] + 5Dh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
IMRH
Bit Default &

Description
Range Access
00h
RO
23:2
0h
RO
12.7.4.35 Isolated Memory Region 7 Read Mask (IMR7RM)—Offset 5Eh

Access Method
Type: Message Bus Register IMR7RM: [Port: 0x05] + 5Eh
(Size: 32 bits)
Op Codes:
Default: BFFFFFFFh

31 28 24 20 16 12 8 4 0
1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
PUNIT
ESRAM_FLUSH_INIT
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_0
CPU0
Bit Default &
Description
Range Access

31
0b
RO
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13

Bit Default &

Description
Range Access
12
11
10
9
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value
12.7.4.36 Isolated Memory Region 7 Write Mask (IMR7WM)—Offset 5Fh

Access Method
Type: Message Bus Register IMR7WM: [Port: 0x05] + 5Fh
(Size: 32 bits)
Op Codes:
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
VC1_SAI_ID3
VC1_SAI_ID2
VC1_SAI_ID1
VC1_SAI_ID0
VC0_SAI_ID3
VC0_SAI_ID2
VC0_SAI_ID1
VC0_SAI_ID0
CPU_SNOOP
CPU_0
CPU0
ESRAM_FLUSH_INIT
PUNIT
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD

Bit Default &

Description
Range Access

31
29
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
15
14
13
12
11
10
9

Bit Default &

Description
Range Access
8
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b
RO
1b IMRxH.
1
the same value
1b IMRxH.
0
the same value
12.7.4.37 eSRAM Control (ESRAMCTRL)—Offset 81h

Provides control of attributes which affect all eSRAM pages.
Access Method
Type: Message Bus Register ESRAMCTRL: [Port: 0x05] + 81h
(Size: 32 bits)
Op Codes:
Default: 047F3F91h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 1 0 0 0 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 0 0 1 0 0 0 1
eSRAM_AVAILABLE
eSRAM_ENABLE_ALL
RSVD
eSRAM_SIZE
RSVD
eSRAM_GLOBAL_CSR_LOCK
SECDED_ENABLE
RSV2
ECC_THRESH
ECC_THRESH_SB_MSG_EN
RSV1
RSV0

Bit Default &

Description
Range Access
0h
RO
2h
RO
07Fh
24:16 eSRAM Size (eSRAM_SIZE): eSRAM size in 4k pages ( 0 means 1)
RO
ECC Threshold (ECC_THRESH): Total correctable ECC threshold until the eSRAM
Correctable Error Threshold Reached message (opcode 0xD8) is sent to Remote
Management Unit. Valid values are 0x1-0xFF. 0x0 may be used as a test mode to send
3Fh the message immediately without waiting for occurrence of any correctable ECC errors.
15:8 Once the message has been sent in this way, ESRAMCERR.
RW/L CORRECTABLE_ERR_CNT_RST must be written with 1, before another message can be
generated using this test mode. Note that the test mode generation of an eSRAM
Correctable Error Threshold Reached message is not dependent on the value of
ESRAMCTRL.SECDED_ENABLE.
ECC Threshold Message Enable (ECC_THRESH_SB_MSG_EN): Set to 1 to enable

1h the message to the Remote Management Unit which is generated when ECC_THRESH
7
RW/L correctable ECC errors have been generated. If this field is 0, no ECC threshold message
will be generated.
0h
RO
0h
RO
1h eSRAM Available (eSRAM_AVAILABLE): Indicates eSRAM is available. Used as

4
RO qualifier for eSRAM_SIZE.
eSRAM Enable All Ranges (eSRAM_ENABLE_ALL): Used during the early BIOS
stage to enable eSRAM mapping into the system address space. Forces all eSRAM pages
0h
3 which are not already enabled, and are unlocked to be ECC initialized and enabled, and
RW/C stays 0x1 until all such pages have been initialized.
NOTE: This is a locking field, locks on ESRAM_GLOBAL_CSR_LOCK.
eSRAM Global CSR Lock (eSRAM_GLOBAL_CSR_LOCK): When set to 1, all eSRAM

0h global and page (4KB and 512KB) registers are locked. A locked page can still be flushed
2
RW if FLUSH_PG_ENABLE/BLOCK_FLUSH_PG_ENABLE is set to 1. Once set, this field can
only be cleared by a warm reset. This is a locking field, it is locked by being set to 1.
0h
RO
1h SECDED Enable (SECDED_ENABLE): SECDED ECC enable for the eSRAM memory
0
RW/L array.
12.7.4.38 eSRAM Block Page Control (ESRAMPGCTRL_BLOCK)—Offset 82h

This register allows all eSRAM pages to be mapped and controlled as a single 512KB
block page. If this page is enabled, no 4KB pages may be individually mapped.
Access Method
Type: Message Bus Register ESRAMPGCTRL_BLOCK: [Port: 0x05] + 82h
(Size: 32 bits)
Op Codes:
Default: 850000FFh

31 28 24 20 16 12 8 4 0
1 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1
BLOCK_PG_SYSTEM_ADDRESS_16MB
BLOCK_FLUSH_PG_ENABLE
RSVD
BLOCK_PG_BUSY
BLOCK_ENABLE_PG
BLOCK_DISABLE_PG
BLOCK_PAGE_CSR_LOCK
RSV1
RSV0
BLOCK_INIT_IN_PROG
Bit Default &

Description
Range Access
1h Block Flush Page Enable (BLOCK_FLUSH_PG_ENABLE): This field is used to enable

31 or disable flushing of the block page to DRAM by the S3 entry firmware code. The block
RW/L page may be flushed only by the S3 entry firmware code.
0h
RO
Block Disable Page (BLOCK_DISABLE_PG): When written with 0x1 disables block
0h page decoding by eSRAM. This bit stays 0x1 until the block page has been disabled and
29 ECC initialized. Note that this is a locking field, which locks on
RW/C BLOCK_PAGE_CSR_LOCK=1 and ESRAMCTRL.eSRAM_GLOBAL_CSR_LOCK. This field
should only be used by BIOS code.
Block Enable Page (BLOCK_ENABLE_PG): When written with 0x1 enables block
0h page mapping of the eSRAM. When the block page is enabled, address mapping for all
28 pages will be controlled by the block page address, instead of the 4KB page address
RW/C fields. Cleared when the page flush/disable completes. Note that this is a locking field,
locks on BLOCK_PAGE_CSR_LOCK=1 and ESRAMCTRL.eSRAM_GLOBAL_CSR_LOCK.
Block Page Register Lock (BLOCK_PAGE_CSR_LOCK): When set to 1, the block

0h page register (ESRAMPGCTRL_BLOCK) is locked. When locked, the block page may still
27
RW be flushed to DRAM by the firmware S3 entry code by setting
BLOCK_FLUSH_PG_ENABLE to 1.
Block Page Initialization in Progress (BLOCK_INIT_IN_PROG): Reads 0x1 as

1h long as the block page is being re-initialized following the disable. Note that while page
26 is being flushed or re-initialized the eSRAM will block the access to the page stalling any
RO/V requestor trying to access it. It also stays high until the ECC initialization completes
after the reset.
0h
RO
Block Page Busy (BLOCK_PG_BUSY): Reads 0x1 when the block page is enabled and
1h stays 0x1 until the block page has been flushed (if flush was to be performed) and
24
RO/V reinitialized following disable. It also stays high until the ECC initialization completes
after the reset.
0000h
RO
Block Page Base Address (BLOCK_PG_SYSTEM_ADDRESS_16MB): Base address

FFh of the 512KB eSRAM block page (bits [31:24] of the system memory address). The
7:0 eSRAM block page may only be placed on 16MB boundaries. Writes to this register will
RW/L only update the contents when ESRAMPGCTRL_BLOCK.BLOCK_ENABLE_PG is 0. Note
that the value in this register field is locked until the page has been disabled and is free.

12.7.4.39 eSRAM Correctable Error (ESRAMCERR)—Offset 83h

Provides status information for correctable ECC errors.
Access Method
Type: Message Bus Register ESRAMCERR: [Port: 0x05] + 83h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
CORRECTABLE_ERR_PG_NUM
CORRECTABLE_ERR_CNT
CORRECTABLE_ERR_PG_DW_OFFSET
CORRECTABLE_ERR_CNT_RST
Bit Default &

Description
Range Access
0h
RO
0h Correctable Error Counter Reset (CORRECTABLE_ERR_CNT_RST): Resets the

25
RW/C correctable ECC error counter.
0h Correctable Error Counter (CORRECTABLE_ERR_CNT): Correctable ECC error

24:17
RO/P count. Saturates at 8'hFF
0h
RO
0h Correctable Error Page DW Offset (CORRECTABLE_ERR_PG_DW_OFFSET): Page

15:9
RO/P DW offset for the last Correctable ECC Error
0h Correctable Error Page Number (CORRECTABLE_ERR_PG_NUM): Page number

8:0
RO/P for the last Correctable ECC Error
12.7.4.40 eSRAM Uncorrectable Error (ESRAMUERR)—Offset 84h

Provides status information for the uncorrectable ECC error.
Access Method
Type: Message Bus Register ESRAMUERR: [Port: 0x05] + 84h
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
UNCORRECTABLE_ERR_PG_DW_OFFSET
UNCORRECTABLE_ERR_OCCURED
RSV0
UNCORRECTABLE_ERR_OCCURED_FLUSH
UNCORRECTABLE_ERR_PG_NUM
Bit Default &
Description
Range Access
0000h
RO
Uncorrectable Error Occurred during Flush

0h (UNCORRECTABLE_ERR_OCCURED_FLUSH): Sticky register field which asserts
17
RW/1C/P when an uncorrectable ECC error has occurred during an eSRAM flush to DRAM. Set by
hardware, cleared by software writing a 1.
0h Uncorrectable Error Occurred (UNCORRECTABLE_ERR_OCCURED): Sticky

16 register field which asserts when an uncorrectable ECC error has occurred. Set by
RW/1C/P hardware, cleared by software writing a 1.
0h Uncorrectable Error Page DW Offset

15:9 (UNCORRECTABLE_ERR_PG_DW_OFFSET): Page DW offset for the uncorrectable
RO/P ECC Error
0h Uncorrectable Error Page Number (UNCORRECTABLE_ERR_PG_NUM): Page

8:0
RO/P number for the uncorrectable ECC Error
12.7.4.41 eSRAM ECC Error Syndrome (ESRAMSDROME)—Offset 88h

Access Method
Type: Message Bus Register ESRAMSDROME: [Port: 0x05] + 88h
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SYNDROME

Bit Default &

Description
Range Access
Syndrome (SYNDROME): Syndrome for the last ECC (SECDED) error. For single bit
errors, can be used to decode which bit of the eSRAM read data or ECC code was
0h incorrect. Bits [7:0] relate to esram_ecc_out[7:0] and esram_data_out[63:0], Bits
31:0
RO/P [15:8] relate to esram_ecc_out[15:8] and esram_data_out[127:64], Bits [23:16] relate
to esram_ecc_out[23:16] and esram_data_out[191:128], Bits [31:24] relate to
esram_ecc_out[31:24] and esram_data_out[255:192].
12.7.5 Memory Manager eSRAM (Port 0x05)

Default
Value
0h + [0- “eSRAM Page Control Register[0-127] (ESRAMPGCTRL[0-127])—Offset 0h, Count 128, Stride 4h” on
850FFFFFh
127]*4h page 228
12.7.5.1 eSRAM Page Control Register[0-127] (ESRAMPGCTRL[0-127])—Offset

0h, Count 128, Stride 4h
This register provides individual per page control and status information.
Access Method
Type: Message Bus Register Offset[0-127]: [Port: 0x05] + 0h + [0-127]*4h
(Size: 32 bits)
Op Codes:
Default: 850FFFFFh
31 28 24 20 16 12 8 4 0
1 0 0 0 0 1 0 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
DISABLE_PG
INIT_IN_PROG
FLUSH_PG
ENABLE_PG
PG_SYSTEM_ADDRESS_4K
RSV1
RSV0
PAGE_CSR_LOCK
PG_BUSY
FLUSH_PG_ENABLE
Bit Default &

Range Access
Flush Page Enable (FLUSH_PG_ENABLE): When PAGE_CSR_LOCK is set to 0,

setting this bit means the page has a corresponding DRAM overlay and will be flushed to
the DDR and disabled when the ESRAMCTRL.eSRAM_Flush_and_Disable is set. When
1h PAGE_CSR_LOCK is set to 1, setting this bit to 0 will prevent flushes via the FLUSH_PG
31 field as well as prevent flushing and disabling via
RW/L ESRAMCTRL.eSRAM_Flush_and_Disable. Since locked pages which s/w does not want to
flush to DRAM for security reasons, will now remain enabled after a global flush and
disable, software must ensure that traffic to those locked pages is quiesced before
entrance to S3 low power.

Bit Default &

Range Access
Flush Page (FLUSH_PG): Initiates flushing the page to DRAM, the page has to have a
0h DRAM overlay and can't be mapped on top of physical memory. After being set, reads as
30 0x1 until the page has been flushed to DRAM. Note that while the page is being flushed,
RW/C the eSRAM will block accesses to the page stalling any other requestor trying to access
it.
Disable Page (DISABLE_PG): When written with 0x1 disables page decoding by
0h eSRAM. When set in the same cycle as Flush Page, the flushing takes place prior to
29
RW/C disabling and reinitializing. This bit stays 0x1 until the page has been flushed, disabled
and ECC initialized. Note that this is a locking field, locks on PAGE_CSR_LOCK.
Enable Page (ENABLE_PG): When written with 0x1 enables page decoding by eSRAM.
0h Same effect will be achieved on all eSRAM page CSRs when the ESRAMCTRL.
28
RW/C eSRAM_Enable_All is set. Cleared when the page flush/disable completes. Note that this
is a locking field, locks on PAGE_CSR_LOCK.
Lock Page (PAGE_CSR_LOCK): When set to 1, the per page (ESRAMPGCTRLX)

0h register for this page is locked. While the page is locked, it may still be flushed via the
27
RW FLUSH_PG field of this register or via ESRAMCTRL.eSRAM_Flush_and_Disable if
FLUSH_PG_ENABLE is set to 1.
Initialisation In Progress (INIT_IN_PROG): Reads 0x1 as long as the page is being

1h re-initialized following the disable. Note that while page is being flushed or re-initialized
26
RO/V the eSRAM will block the access to the page stalling any requestor trying to access it. It
also stays high until the ECC initialization completes after the reset.
0h
RO
1h Page Busy (PG_BUSY): Reads 0x1 when the page is enabled and stays 0x1 until the
24 page has been flushed (if flush was to be performed) and reinitialized following disable.
RO/V It also stays high until the ECC initialization completes after the reset.
0h
RO
Page 4K Address (PG_SYSTEM_ADDRESS_4K): 20b of base address for the 4K

page. Needs to be stable before the page is enabled and must stay untouched until the
1FFFFFh page has been disabled and flushed. Software may pool the PG_BUSY bit to find out the
19:0 status of the page allocation, and attempt to allocate address and enable new page only
RW/L when the previous deallocation has successfully completed. Another less read intensive
method is using the dynamic pool mechanism as per the global ESRAMPGPOOL register.
Note that the value in this CSR is locked until the page has been disabled and is free.
12.7.6 SoC Unit (Port 0x31)

Default
Value
34h “Thermal Sensor Configuration 4 (SCU_TSCFG4_Config)—Offset 34h” on page 229 00057801h
50h “Sticky Write Once (CFGSTICKY_W1)—Offset 50h” on page 230 00000000h
51h “Sticky Read/Write (CFGSTICKY_RW)—Offset 51h” on page 231 00000000h
52h “Non-Sticky Read/Write Once (CFGNONSTICKY_W1)—Offset 52h” on page 231 00000000h
12.7.6.1 Thermal Sensor Configuration 4 (SCU_TSCFG4_Config)—Offset 34h

Access Method


(Size: 32 bits)
Op Codes:
Default: 00057801h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1
ts_itsrst
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Range Access
0h
RO
0h
RO
AFh
RO
0h
RO
0h
RO
0h
RO
0h
RO
0h
RO
1h
0 Thermal Sensor Reset (ts_itsrst): Resets all Thermal Sensor registers.
RW
12.7.6.2 Sticky Write Once (CFGSTICKY_W1)—Offset 50h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
STICKY_W1_STRATCH
Bit Default &
Range Access
0000h Sticky Write Once Scratchpad (STICKY_W1_STRATCH): Assigned by Software,

31:0
RW/O write once requires a S0 exit to clear
12.7.6.3 Sticky Read/Write (CFGSTICKY_RW)—Offset 51h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
STRICKY_RW_STRATCH
Bit Default &

Range Access
0000h Sticky Read/Write Scratchpad (STRICKY_RW_STRATCH): Assigned by Software,

31:0
RW/P reset by a S0 exit
12.7.6.4 Non-Sticky Read/Write Once (CFGNONSTICKY_W1)—Offset 52h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
NONSTICKY_W1_STRATCH
Bit Default &
Range Access
0000h Non-Sticky Write Once Scratchpad (NONSTICKY_W1_STRATCH): Assigned by

31:0
RW/O Software, reset by a warm reset while in S0
§§

System Memory Controller—Intel® Quark™ SoC X1000
13.0 System Memory Controller
The system memory controller supports DDR3 protocol with one 16-bit wide data
channel and up to 2 ranks of memory, allowing for population of up to 2Gbyte of
system memory using 1, 2 or 4 Gbit standard DDR3 devices. It is capable of data rates
up to 800 MT/s.


Table 79. Memory Signals (Sheet 1 of 2)

Direction
Type
DRAM Differential Clock Pair: (1 pair per Rank)

DDR3_CK[1:0] O
The differential clock pair is used to latch the command into
DDR3_CKB[1:0] DDR3
DRAM. Each pair corresponds to a rank on the DRAM side.
O Chip Select: (1 per Rank). Used to qualify the command on the

DDR3_CSB[1:0]
DDR3 command bus for a particular rank.
O Clock Enable: (power management - 1 per Rank)

DDR3_CKE[1:0]
DDR3 It is used during DRAM power up/power down and self refresh.
Memory Address: Multiplexed Memory address bus (Row, Column)

O for writing data to memory and reading data from memory. These
DDR3_MA[15:0]
DDR3 signals follow common clock protocol w.r.t. DDR3_CK, DDR3_CKB
pairs.
O Bank Select: These signals define which banks are selected within
DDR3_BS[2:0]
DDR3 each DRAM rank.
O Row Address Select: Used with DDR3_CASB and DDR3_WEB

DDR3_RASB
DDR3 (along with DDR3_CSB) to define the DRAM Commands.
O Column Address Select: Used with DDR3_RASB and DDR3_WEB

DDR3_CASB
DDR3 (along with DDR3_CSB) to define the DRAM Commands.
O Write Enable Control Signal: Used with DDR3_WEB and

DDR3_WEB DDR3_CASB (along with control signal, DDR3_CSB) to define the
DDR3 DRAM Commands.
I/O
DDR3_DQ[15:0] Bidirectional Data Lines
DDR3

Intel® Quark™ SoC X1000—System Memory Controller
Table 79. Memory Signals (Sheet 2 of 2)

Direction
Type
Data Mask: DM is an output mask signal for write data. Output

O data is masked when DM is sampled HIGH coincident with that
DDR_DM[1:0]
DDR3 output data during a Write access. DM is sampled on both edges of
DQS.
Data Strobes: DDR3_DQSB[1:0] and its complement signal group

make up a differential strobe pair for each 8 data bits - DQ. The
DDR3_DQS[1:0] I/O data is captured at the crossing point of DDR3_DQS[1:0] and its
DDR3_DQSB[1:0] during read and write transactions. For Read,
DDR3_DQSB[1:0] DDR3 the Strobe crossover and data are edge aligned, whereas in the
Write command, the strobe crossing is in the centre of the data
window.
O ODT signal (One per rank) going to DRAM in order to turn ON the
DDR3_ODT[1:0]
DDR3 DRAM ODT during Write.
This signal must be terminated to VSS on board (refer to the

O Platform Design Guide for resistor value). This external resistor
DDR3_ODTPU
Analog termination scheme is used for Resistor compensation of DRAM
ODT strength.

DDR3_DQPU
Analog termination scheme is used for Resistor compensation of DQ
buffers

DDR3_CMDPU
Analog termination scheme is used for Resistor compensation of CMD
buffers.
DRAM Interface Reference Voltage: This signal voltage level is

used for qualifying logical levels on the DQ bits on reads. The
I Memory interface can also use internally generated reference
DDR3_VREF
Analog voltage to qualify the crossing point between logical levels on Data
bits. Internal Vref is a default setting for Memory Interface and
this interface signal and can be tied to VSS on board.
I
DDR3_ISYSPWRGOOD Asynchronous This signal indicates the status of the DRAM Core power supply.
CMOS
I This signal indicates the status of the DRAM S3 power supply.

DDR3_IDRAM_PWROK Asynchronous Used primarily in the DRAM PHY to determine ACPI S3 power
CMOS state.
DDR3_DRAMRSTB O This signal is used to reset DRAM devices.
13.2 Features
13.2.1 System Memory Technology Supported

The system memory controller supports the following DDR3 Data Transfer Rates and
DRAM Device Technologies:
• DDR3 Data Transfer Rate: 800 MT/s
• DDR3 (1.5V DRAM interface I/Os)
• DDR3 x8 memory modules
Note: x8 means that each DRAM component has 8 data lines. Standard 1Gbit, 2Gbit, and
4Gbit technologies and addressing are supported for x8 devices.

Table 80. Supported DDR3 DRAM Devices

DRAM Bank Row Column
Data Width Banks Page Size
Density Address Address Address
1 Gbit x8 8 BA[2:0] A[13:0] A[9:0] 1 Kbyte
2 Gbit x8 8 BA[2:0] A[14:0] A[9:0] 1 Kbyte
4 Gbit x8 8 BA[2:0] A[15:0] A[9:0] 1 Kbyte
Table 81. Supported DDR3 Memory Configurations

DRAM Chip DRAM Device # of DRAM Devices
Rank Size # of Ranks Total Memory Size
Density Width per Rank
1 Gbit x8 2 256 Mbyte 1 256 Mbyte
2 Gbit x8 2 512 Mbyte 2 1 Gbyte
4 Gbit x8 2 1 Gbyte 1 1 Gbyte
4 Gbit x8 2 1 Gbyte 2 2 Gbyte
13.2.2 Rules for Populating Memory Down Ranks

The devices density and width for both ranks must be the same. Rank0 must be always
populated and Rank1 is optional.
13.2.3 DRAM Error Detection & Correction (EDC)

For high reliability applications the system memory controller supports inclusion of
Error Correction Codes (ECC) in DRAM transactions. In ECC mode, the 8th bank of each
rank is allocated for ECC data storage, thus reducing the total available physical
memory for the system by 12.5% or 1/8. Additionally each transaction to memory
requires an associated ECC transaction reducing the useful memory bandwidth.
The algorithm used allows for on-the-fly correction of single bit errors and detection of
double bit errors.
A separate bank is used for ECC data storage to avoid the page-miss (row pre-charge)
time penalty on ECC and next data fetch that would have been introduced in majority
of cases if the ECC was interleaved across banks. The ECC bank can be configured to
always issue RD/WR with Auto Precharge or dynamic page close policy similar to other
data banks. Since the configuration is on a per bank basis, the ECC bank is not required
to have the same policy as other banks.
When ECC is enabled, the Address Map field in the DRAM Rank Population (DRP)—
Offset 0h register MUST be set to 2. In a two ranks system, the size of both ranks can
be the same or different. Therefore, each rank can have different DRAM device width
and density, and thus different rank size. Setting the address map to 2 applies to both
ranks.

13.2.4 DRAM Data Scrambling

The Memory Controller supports data scrambling. This feature helps in lowering the
MTBF (Mean Time Between Failures) by reducing the probability of occurrence of the
specific bit patterns on the DRAM I/O interface that could cause unpredictable behavior
at the platform design stage, i.e., signal integrity issues.
13.2.5 Power Management

The System Memory Controller Power Management features are detailed in Section 8.5.
13.3 Register Map

Figure 26. Register Map
CPU
Core
Host
Bridge Message Bus Space
B:0,D:0,F0
Port: 0x00 Port: 0x01 Port: 0x02 Port: 0x04 Port: 0x05
Memory Host Remote Memory

Memory Management
Arbiter Controller Bridge Unit Manager
13.4 Message Bus Registers

Default
Value
0h “DRAM Rank Population (DRP)—Offset 0h” on page 237 00000000h
1h “DRAM Timing Register 0 (DTR0)—Offset 1h” on page 238 43001110h
6h “DRAM Power Management Control 0 (DPMC0)—Offset 6h” on page 245 03000000h
8h “DRAM Refresh Control (DRFC)—Offset 8h” on page 247 00012CA7h
9h “DRAM Scheduler Control (DSCH)—Offset 9h” on page 248 00071108h


Default
Value
Ah “DRAM Calibration Control (DCAL)—Offset Ah” on page 249 00001300h
Bh “DRAM Reset Management Control (DRMC)—Offset Bh” on page 250 00000000h
Ch “Power Management Status (PMSTS)—Offset Ch” on page 251 00000000h
Fh “DRAM Control Operation (DCO)—Offset Fh” on page 252 00000000h
4Ah “Sticky Scratchpad 0 (SSKPD0)—Offset 4Ah” on page 252 00000000h
4Bh “Sticky Scratchpad 1 (SSKPD1)—Offset 4Bh” on page 253 00000000h
60h “DRAM ECC Control Register (DECCCTRL)—Offset 60h” on page 253 00000000h
61h “DRAM ECC Status (DECCSTAT)—Offset 61h” on page 254 00000000h
62h “DRAM ECC Single Bit Error Count (DECCSBECNT)—Offset 62h” on page 254 00000000h
68h “DRAM Single Bit ECC Error Captured Address (DECCSBECA)—Offset 68h” on page 255 00000000h
69h “DRAM Single Bit ECC Error Captured Syndrome (DECCSBECS)—Offset 69h” on page 256 00000000h
6Ah “DRAM Double Bit ECC Error Captured Address (DECCDBECA)—Offset 6Ah” on page 256 00000000h
6Bh “DRAM Double Bit ECC Error Captured Syndrome (DECCDBECS)—Offset 6Bh” on page 257 00000000h
70h “Memory Controller Fuse Status (DFUSESTAT)—Offset 70h” on page 257 00000000h
80h “Scrambler Seed (DSCRMSEED)—Offset 80h” on page 258 00000000h
13.4.1 DRAM Rank Population (DRP)—Offset 0h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ADDRMAP
PRI64BSPLITEN
MODE32
DIMMDDEN1
DIMMDWID1
DIMMDDEN0
DIMMDWID0
Rsvd_4
Rsvd_3
Rsvd_1
Rsvd_0
RKEN1
RKEN0
Bit Default &

Range Access
0h
31 Rsvd_4: Reserved
RO
0h 16-bit/32-bit Mode Select (MODE32): 0 - Selects 16-bit DRAM Data Interface.

30
RW/P/L 1 - Selects 32-bit DRAM Data Interface.
0000h
29:16 Rsvd_3: Reserved
RO

Bit Default &

Range Access
Address Map Select (ADDRMAP): See Address Mapping section for full description.
0 Map 0
0h 1 Map 1
15:14
RW/P/L 2 Map 2
Note: The address map select should be set the same for both the Memory Controller
and the Memory Manager.
0h 64B Split Enable (PRI64BSPLITEN): Setting this bit to '1' enables logic to split 64B
13
RW PRI transactions to two 32B transactions. This bit must be set if ECC mode is enabled.
Rank 1 Device Density (DIMMDDEN1): This sets the density of the DRAM devices
populated in Rank 1.
0h 00 1Gbit
12:11
RW/P/L 01 2Gbit
10 4Gbit
11 Reserved
Rank 1 Device Width (DIMMDWID1): Indicates the width of the DRAM devices
0h 00 x8
10:9
RW/P/L 01 Future Support
10 Reserved
11 Reserved
0h
8 Rsvd_1: Reserved
RO
Rank 0 Device Density (DIMMDDEN0): This sets the density of the DRAM devices
0h 00 1Gbit
7:6
RW/P/L 01 2Gbit
10 4Gbit
11 Reserved
Rank 0 Device Width (DIMMDWID0): Indicates the width of the DRAM devices
0h 00 x8
5:4
RW/P/L 01 Future Support
10 Reserved
11 Reserved
0h
RO
0h Rank Enable 1 (RKEN1): Should be set to 1 when device has 2 ranks to enable the
1
RW/P/L use of second rank. Otherwise, must be set to 0.
0h Rank Enable 0 (RKEN0): Should be set to 1 when Rank 0 is populated to enable the
0
RW/P/L use of this rank. Otherwise, must be set to 0.
13.4.2 DRAM Timing Register 0 (DTR0)—Offset 1h

Access Method
(Size: 32 bits)
Op Codes:
Default: 43001110h
31 28 24 20 16 12 8 4 0
0 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0
tRP
Rsvd_13
Rsvd_12
tXSDLL
Rsvd_9
Rsvd_8
Rsvd_5
tCL
CKEDLY
PMEDLY
tZQoper
Rsvd_11
Rsvd_10
tZQCS
tXS
tRCD
DFREQ

Bit Default &

Range Access
Clock Valid To Self-Refresh Exit Delay (CKEDLY): Additional delay between CK/CKB
4h start and SRX command. This delay is needed for clock to stabilize to meet JEDEC
31:28 requirements. Delay is CKEDLY multiples of 256 DRAM Clocks.
RW 0ns to 9,600ns (DDR3-800)
0ns to 7,200ns (Future DDR3-1066)
0h
RO
Power Mode Entry Delay (PMEDLY): The delay, in DRAM clocks, between SR Entry
command and Power-Mode message to DDRIO.
3h 0h - 6 DRAM Clocks.
25:24
RW 1h - 8 DRAM Clocks.
2h - 10 DRAM Clocks.
0h
23 Rsvd_12: Reserved
RO
ZQCal Long Delay (tZQoper): The delay, in DRAM clocks, between ZQC-Long
command to any command.
Note: ZQCL command during DRAM Init flow requires longer latency which is controlled
be BIOS.
22 1h - 384 DRAM Clocks.
RW Note: This field defines the ZQ Calibration Long delay during normal operation. It is not
the same as tZQinit, which uses the same ZQCL command but the delay is longer.
tZQinit applies only during power-on initialization of the DRAM devices, and tZQoper
applies during normal operation. BIOS executes the DRAM initialization sequence, so it
has to ensure tZQinit is met, and not the Memory Controller.
0h
RO
ZQCal Short Delay (tZQCS): The delay, in DRAM clocks, between a ZQC-Short
0h command to any command.
20
0h
RO
Self-Refresh Exit To DLL Delay (tXSDLL): The delay, in DRAM clocks, between SRX
0h command to any command requiring locked DLL. Only ZQCL can be sent before tXSDLL
18 is done.
RW 0h - tXS + 256 DRAM Clocks.
1h - tXS + 384 DRAM Clocks.
0h
17 Rsvd_9: Reserved
RO
Self-Refresh Exit Delay (tXS): The delay, in DRAM clocks, between SRX command to
0h command not requiring locked DLL. The Memory Controller can send a ZQCL command
16 after tXS. JEDEC defines MAX(5CK, tRFC(min)+10ns) so both values take safety margin.
0h
15 Rsvd_8: Reserved
RO
CAS Latency (tCL): Specifies the delay, in DRAM clocks, between the issue of a RD
command and the return of valid data on the DQ bus.
0h - 5 DRAM Clocks (DDR3-800)
1h 2h - Reserved
14:12
RW 3h - Reserved
4h - Reserved
5h - Reserved
6h - Reserved
7h - Reserved

Bit Default &

Range Access
Activate (RAS) to CAS Delay (tRCD): Specifies the delay, in DRAM clocks, between
an ACT command and a RD/WR command to the same bank.
1h - 6 DRAM Clocks (DDR3-800
1h 2h - Reserved
11:8
RW 3h - Reserved
4h - Reserved
5h - Reserved
6h - Reserved
7h - Reserved
Precharge to Activate Delay (tRP): Specifies the delay, in DRAM clocks, between a
PRE command and an ACT command to the same bank.
1h - 6 DRAM Clocks (DDR3-800
1h 2h - Reserved
7:4
RW 3h - Reserved
4h - Reserved
5h - Reserved
6h - Reserved
7h - Reserved
0h
RO
DRAM Frequency (DFREQ): Specifies the DDR3 frequency used by the Memory
Controller for computing proper cycle to cycle timings. Note this configuration has no
impact on the actual DRAM clock.
0h 0h - DDR3-800
1:0
RW 1h - Reserved
2h - Reserved
3h - Reserved
Note: This configuration has no impact on the actual DRAM clock frequency.

Access Method
(Size: 32 bits)
Op Codes:
Default: 02690320h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 1 0 0 1 1 0 1 0 0 1 0 0 0 0 0 0 1 1 0 0 1 0 0 0 0 0
Rsvd_18
Rsvd_17
Rsvd_16
Rsvd_15
Rsvd_14
tWCL
tRRD
tRAS
tCCD
tCMD
tRTP
tFAW
tWTP
Bit Default &

Range Access
0h
RO
Read to Precharge Delay (tRTP): The minimal delay between RD command and PRE
command to same bank.
0h 001 - 4 DRAM Clocks (DDR3-800)
30:28
RW 010 - 5 DRAM Clocks)
011 - 6 DRAM Clocks
100 - 7 DRAM Clocks
0h
RO

Bit Default &

Range Access
Row Activation to Row Activation Delay (tRRD): The minimal time interval
between 2 ACT commands to any bank in the same DRAM device. Limits peak current
profile.
2h 00 - 4 DRAM Clocks (1KB page DDR3-800), (2KB page DDR3-800)
25:24 01 - 5 DRAM Clocks
RW 10 - 6 DRAM Clocks
11 - 7 DRAM Clocks
Note: This timing parameter applies to both Ranks, so set it based on the rank with the
large tRRD value.
Row Activation Period (tRAS): The minimal delay, in DRAM clocks, between ACT
command and PRE command to same bank. At least equal to tRCD + tCWL + tCCD +
tWR 0h -14 DRAM Clocks.
1h -15 DRAM Clocks (DDR3-800)
2h -16 DRAM Clocks.
3h -17 DRAM Clocks.
6h 4h -18 DRAM Clocks.
23:20
RW 5h -19 DRAM Clocks.
6h -20 DRAM Clocks
7h -21 DRAM Clocks.
8h -22 DRAM Clocks.
9h -23 DRAM Clocks.
Ah -24 DRAM Clocks
Others - Reserved
Four Bank Activation Window (tFAW): A rolling time-frame, in which a maximum of

4 ACT commands (per rank) can be sent. Limits peak current profile.
0h - Reserved.
1h - Reserved.
3h - 16 DRAM Clocks (1KB page DDR3-800).
5h - 20 DRAM Clocks (2KB page DDR3-800.
9h 7h - 24 DRAM Clocks .
19:16
Ah - 30 DRAM Clocks.
Bh - 32 DRAM Clocks
Ch - Reserved.
Dh - Reserved.
Eh - Reserved.
Fh - Reserved.
Note: This timing parameter applies to both Ranks, so set it based on the rank with the
large tFAW value.
0h
RO
CAS to CAS delay (tCCD): The minimum delay, in DRAM clocks, between 2 RD/WR
commands.
0h 0h - 4 DRAM Clocks. Functional mode. (DDR3-800).
13:12
RW 1h - 12 DRAM Clocks. DFX stretch mode (x2).
2h - 18 DRAM Clocks. DFX stretch mode (x4).
3h - Reserved
Write To Prechange Delay (tWTP): The minimum delay, in DRAM clocks, between a
WR command and a PRE command to the same bank. Value should be computed as 4 +
tWCL + tWR.
1h - 15 DRAM Clocks DDR3-800).
3h 4h - 18 DRAM Clocks
11:8
7h - 21 DRAM Clocks
Others - Reserved
Note: This is not a JEDEC timing parameter. It is derived from other JEDEC timing
parameters.

Bit Default &

Range Access
0h
RO
Command Transport Duration (tCMD): The time period, in DRAM clocks, that a
command occupies the DRAM command bus. 1N is the DDR3 basic requirement. 2N and
2h 3N are extended modes for board signal-integrity.
5:4 0h - 1 DRAM Clock (1N).
RW 1h - 2 DRAM Clocks (2N).
2h - 3 DRAM Clocks (3N).
Note: This is a board design timing parameter and not part of JEDEC spec.
0h
3 Rsvd_14: Reserved
RO
CAS Write Latency (tWCL): The delay, in DRAM clocks, between the internal write
command and the availability of the first bit of DRAM input data.
0h 0h - 5 DRAM Clocks (DDR3-800)
2:0
RW 1h - 6 DRAM Clocks
2h - 7 DRAM Clocks
3h - 8 DRAM Clocks

Access Method
(Size: 32 bits)
Op Codes:
Default: 00040504h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0
Rsvd4_DTR2
tRWDR
Rsvd2_DTR2
tWWDR
Rsvd0_DTR2
tRRDR
Bit Default &
Range Access
0h
31:20 Rsvd4_DTR2: Reserved
RO
Read to Write Delay (tRWDR): Read to Write DQ delay, different ranks.

1h - 6 DRAM Clocks.
19:16
4h - 9 DRAM Clocks.
00h
RO
Write to Write Delay (tWWDR): Write to Write DQ delay, different ranks.

0h - Reserved
1h - Reserved
2h - 6 DRAM Clocks.
10:8
5h - 9 DRAM Clocks.
6h - Reserved
7h - Reserved

Bit Default &

Range Access
00h
RO
Read to Read Delay (tRRDR): Read to Read DQ delay, different ranks.

0h - Reserved
1h - 6 DRAM Clocks.
2:0
4h - 9 DRAM Clocks.
Others - Reserved

Access Method
(Size: 32 bits)
Op Codes:
Default: 06406205h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 1 1 0 0 0 1 0 0 0 0 0 0 1 0 1
PWDDLY
tXP
Rsvd4_DTR3
Rsvd3_DTR3
tWRSR
Rsvd2_DTR3
tRWSR
Rsvd0_DTR3
tWRDR
Bit Default &
Range Access
0h
RO
6h RD/WR command to Power-down Delay (PWDDLY): Non-JEDEC delay for

27:24
RW performance enhancement. Delay = PWDDLY x 4 DRAM Clocks.
CKR to Command Delay (tXP): Delay from CKE asserted high to any DRAM command
1h 0h - 2 DRAM Clocks (DDR3-800 2N).
23:22 1h - 3 DRAM Clocks (DDR3-800 1N).
RW 2h - 4 DRAM Clocks (Future DDR3-1066 1N).
3h - 5 DRAM Clocks.
0h
RO
Write to Read Command Delay (tWRSR): Write to Read same rank command delay.
Should be set to 4 + tWCL + tWTR
2h - 13 DRAM Clocks (DDR3-800).
16:13
Others - Reserved
0h
12 Rsvd2_DTR3: Reserved
RO

Bit Default &

Range Access
Read to Write Command Delay (tRWSR): Read to Write same rank command delay.
Should be set to tCL - tWCL + 6 + board delay if needed.
0h - 6 DRAM Clocks.
1h - 7 DRAM Clocks.
11:8
Others - Reserved
0h
RO
Write to Read Delay (tWRDR): Write to Read DQ delay, different ranks.

0h - Reserved
1h - 6 DRAM Clocks.
2:0 3h - 8 DRAM Clocks.
Others - Reserved

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000022h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0
TRGSTRDIS
Rsvd_23
Rsvd_22
ODTDIS
WRODTSTRT
RDODTDIS
WRODTSTOP
Bit Default &

Range Access
0h
31:17 Rsvr_24 (RDODTDIS): Reserved
RW
Disable Write ODT Stretching (TRGSTRDIS): Write target rank is not stretched.
When set, stretched ODT as defined above is not applied to the write target rank and
0h ODT command is asserted for 6 DRAM clocks. Should not be used when ODT is pulled-in.
16
RW Note: This bit should be set to 0 for normal operation.
Note: This bit should not be set to 1 when ODT is configured to assert earlier than the
Write command.
0h
15 ODT Disable (ODTDIS): 0 - ODT is enabled. 1 - ODT is disabled
RW
0h
RO

Bit Default &

Range Access
Write command to ODT de-assert delay (WRODTSTOP): For 1N Command Mode

0h WR+6
1h WR+7
2h WR+8
3h N/A
4h N/A
Others Reserved
For 2N Command Mode
0h N/A
2h 1h WR+6
6:4 2h WR+7
RW 3h WR+8
4h N/A
Others Reserved
For 3N Command Mode
0h N/A
1h N/A
2h WR+6
3h WR+7
4h WR+8
Others Reserved
0h
RO
Write command to ODT assert delay (WRODTSTRT): JEDEC requires ODT to be

asserted on the same clock with the WR command. The Memory Controller allows to
pull-in by 1 clock in 2N mode and by 1-2 clocks in 3N mode. For most DIMM
configurations, this register should be programmed to same value as tCMD. A value of
tCMD - ODT_PULLIN can be used according to the table below which shows the ODT
command assertion with respect to the WR command assertion.
For 1N Command Mode
0h WR
1h N/A
2h 2h N/A
1:0 3h Reserved
RW For 2N Command Mode
0h WR-1
1h WR
2h N/A
3h Reserved
For 3N Command Mode
0h WR-2
1h WR-1
2h WR
3h Reserved
13.4.7 DRAM Power Management Control 0 (DPMC0)—Offset 6h

Access Method
(Size: 32 bits)
Op Codes:
Default: 03000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DISPWRDN
PREAPWDEN
Rsvd_31
Rsvd27
DYNSREN
Rsvd_30
PCLSWKOK
Rsvd_29
Rsvd_28
ENPHYCLKGATE
CLKGTDIS
RSVD
PCLSTO
SREDLY
REUTCLKGTDIS

Bit Default &

Range Access
0h
RO
0h Enable PHY Clock Gate Disable During SR (ENPHYCLKGATE): When set to 1, the
29 Memory Controller will turn off the 1x and 2x clock trees to the DDRIO PHY during Self
RW Refresh. The Memory Controller will re-enable the clocks upon Self Refresh exit.
MTE Clock Gate Disable (REUTCLKGTDIS): 0h MTE clock is gated when DCO.PMICTL
0h is set to 0.
28
RW 1h MTE clock is ungated, overriding the DCO.PMICTL config bit.
Note: The DCO.CPGCLOCK bit overrides this bit.
0h
27:26 Rsvd27: Reserved
RO
Disable Power Down (DISPWRDN): Setting this bit to 1 will block CKE high-)low
transitions. May be used by BIOS during init flow and should be set to 0 for functional
1h mode.
25
RW 0 - The Memory Controller dynamically controls the CKE pins to place the DRAM device
in power down mode.
1 - The Memory Controller constantly drives the CKE pins high.
Clock Gating Disabled (CLKGTDIS): Setting this bit to 0 allows a large number of
internal Memory Controller clocks to be gated when there is no activity in order to save
1h power. When set to 1, internal clock-gating is disabled.
24
RW 0 - Enable.
1 - Disable.
Note: This bit should be set to 0 for normal operation.
Dynamic Self-Refresh Enable (DYNSREN): Setting this bit to 1, enables automatic

0h SR command to DRAM and PM message to DDRIO when the PRI bus is idle, all pending
23 requests have been served and the and PRI status is less than 2, SREDLY has timed-out,
RW and all JEDEC requirements are satisfied. This register may be changed by BIOS/FW on-
the-fly.
0h
RO
Close All Pages before Power-Down (PREAPWDEN): Send Precharge All Command
0h to a Rank before PD-Enter. Setting this bit to 1 will allow sending a PREA command
21 before PDE command.
RW 0 - Disable.
1 - Enable.
Wake Allowed for Page Close Timeout (PCLSWKOK): Setting this bit to 1 indicates
the Memory Controller can send DRAM devices a PD-Exit command in order to close
single bank if the page timer expired. Note this bit applies only to cases where at least
0h one other bank in the same rank is open but not timed-out. If all banks in the rank
20
RW timed-out, a PD-Exit command will be sent regardless of this bit. Must be set to 0 during
init/training mode.
0 - Disable.
1 - Enable.
0h
RO
Page Close Timeout Period (PCLSTO): Specifies the time frame, in ns, from last
access to a DRAM page until that page may be scheduled for closing (by sending a PRE
command).
0h - Disable page close timer (init/training).
0h 1h - Immediate page close.
18:16 2h - 30-60 ns to page close.
RW 3h - 60-120 ns to page close.
4h - 120-240 ns to page close.
7h - 1-2 s to page close.
0h
RO
0h
RO

Bit Default &

Range Access
0h Self-Refresh Entry delay (SREDLY): The delay, in core-clocks, between PRI idle (no
7:0 pending requests and PRI status is less than 2) and SR Entry when the Memory
RW Controller is in Dynamic SR mode.
13.4.8 DRAM Refresh Control (DRFC)—Offset 8h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00012CA7h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 1 1 0 0 1 0 1 0 0 1 1 1
REFSKWDIS
REFCNTMAX
REFWMPNC
REFWMLO
Rsvd_36
REFDBTCLR
Rsvd_35
Rsvd_34
tREFI
REFWMHI
Bit Default &
Range Access
0h
RO
Clear Refresh Debit before Self Refresh Entry (REFDBTCLR): To ensure that the
Memory Controller sends enough REF commands to the DRAM, it calculates tREFI period
with 2% less than the JEDEC tREFI value. So instead of tREFI equaling 7.8us, it's 7.6us,
which means over 1000 x tREFI interval, the Memory Controller would have sent 20 REF
0h commands more than required by the JEDEC spec. When this bit is set to 1 and if the
21 Memory Controller was awake for at least 1000 x tREFI period and then enters Self
RW Refresh, the Memory Controller clears the refresh counter and enters Self Refresh
without having to send the accumulated REF commands, since it has already issued 20
more REF commands than required by JEDEC.
0h - Disabled.
1h - Enabled.
Disable Skewing of Refresh Counting between Ranks (REFSKWDIS): Each rank

has its own refresh counter. By default, incrementing these refresh counters are skewed
by 1/4 the tREFI period. Setting this bit to a 1 disables this feature and all refresh
0h counters will increment at the same time per tREFI period. Skewing the tREFI counters
20
RW can improve performance, since traffic to all ranks does not have to be block to perform
refresh.
0h - counters are updated per rank every tREFI.
1h - all counters are updated every tREFI.
0h
RO
Refresh Max tREFI Interval (REFCNTMAX): The maximum interval between ant two
REF commands per rank. JEDEC allows a maximum of 9 x tREFI intervals.
1h 0h - 6 x tREFI.
17:16 1h - 7 x tREFI.
RW 2h - 8 x tREFI.
3h - Reserved.
Should not be changed after initial setting.
0h
RO

Bit Default &

Range Access
Refresh Period (tREFI): Specifies the average time between sending REF commands
to DRAM. The Memory Controller will guarantee that the average time is met, but
maintains a certain degree of flexibility in the exact REF scheduling in order to increase
2h overall performance.
14:12
RW 0h - Refresh disabled
1h - Reserved for pre-silicon simulation.
2h - 3.9 s (Extended Temperature Range, 85-95 C)
3h - 7.8 s (Normal Temperature Range, 0-85 C)
Refresh Panic Watermark (REFWMPNC): When the refresh debit counter, per rank,
is greater than this value, the Memory Controller will send a REF command even if there
are some pending requests and regardless of the PRI status level. See DDR3 spec for
Refresh Postponing/Pulling-In flexibility. May be changed to functional value after init
sequence. Value should be greater than, or equal, to REFWMHI.
0-6h - Reserved
Ch 7h - Postpone 2 REF commands.
11:8
RW 8h - Postpone 3 REF commands.
9h - Postpone 4 REF commands.
Ah - Postpone 5 REF commands.
Bh - Postpone 6 REF commands.
Ch - Postpone 7 REF commands.
Dh - Postpone 8 REF commands.
E-Fh - Reserved.
Refresh High Watermark (REFWMHI): When the refresh debit counter, per rank, is
greater than this value, the Memory Controller will send a REF command even if there
are some pending requests to the rank but not if the PRI status is equal to 3. See DDR3
spec for Refresh Postponing/Pulling-In flexibility. May be changed to functional value
after init sequence. Value should be greater than, or equal, to REFWMLO.
0-6h - Reserved
Ah 7h - Postpone 2 REF commands.
7:4
E-Fh - Reserved.
Refresh Low Watermark (REFWMLO): When the refresh debit counter, per rank, is
greater than this value, the Memory Controller will send a REF command only if there
are no pending requests to the rank and the PRI status is less than 3. See DDR3 spec for
Refresh Postponing/Pulling-In flexibility. May be changed to functional value after init
sequence.
0-6h - Reserved
7h 7h - Postpone 2 REF commands.
3:0
E-Fh - Reserved.
13.4.9 DRAM Scheduler Control (DSCH)—Offset 9h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00071108h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0
IPREQMAX
NEWBYPDIS
OOOST3DIS
OOODIS
OOOAGETRH
Rsvd_40
Rsvd_39
Rsvd_38
Rsvd_37
Bit Default &
Range Access
0h
RO
In-Progress Request Queue Depth (IPREQMAX): Maximal number of In-Progress

7h Requests stored in the Memory Controller.
18:16
RW Number of pending requests = IPREQMAX+1; Value may be changed after init/training
when PRI is idle.
0h
RO
Disable New Request Bypass (NEWBYPDIS): Setting this bit to 0 will allow a new
1h request to bypass the normal Memory Controller internal arbiter when there are no
12 pending commands.
RW 0h - Enable New Request Bypass.
1h - Disable New Request Bypass.
0h
RO
Out-of-Order Disabled when PRI status is 3 (OOOST3DIS): Valid only if OOODIS

0h is 0;
9 0 - Remain OOO if status goes up to 3
RW 1 - Disable OOO if status goes to 3
May be changed after init/training flow.
1h Disable Out-of-Order (OOODIS): 0h - OOO enabled.

8 1h - OOO disabled.
RW Should be disabled during init/training and can be enabled for functional mode.
0h
RO
Out-of-Order Aging Threshold (OOOAGETRH): Specifies the number of requests

that can be processed ahead of another request sitting in the In-Progress request
08h (IPreq) queue before OOO is disabled. Once this threshold is met for any request sitting
4:0 in the IPreq queue, OOO is disabled until the aged request is processed. This
RW mechanism prevents starvation of pending requests in the IPreq queue. Each request
sitting in the IPreq queue has its own age timer. OOOAGETRH sets the default value of
an age timer when a request is loaded into the IPreq queue.
13.4.10 DRAM Calibration Control (DCAL)—Offset Ah

Access Method
Type: Message Bus Register Offset: [Port: 0x01] + Ah
(Size: 32 bits)
Op Codes:
Default: 00001300h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0
SRXZQCL
ZQCINT
Rsvd_43
Rsvd_42
Rsvd_41
Bit Default &
Range Access
0h
RO
ZQ Calibration Long (SRXZQCL): Issued After SR Exit Control

1h 0h - ZQCL commands after SRX are sent in parallel.
13:12 1h - ZQCL commands are sent serially to ranks.
RW/P 2h - No ZQCL is sent after SR Exit. (Debug only).
3h Reserved.
0h
RO
ZQ Calibration Short Interval (ZQCINT): The time interval, in ms, between ZQCS
commands to a DRAM device. ZQCS commands are sent to a single DRAM device and
commands are distributed and non-overlapping in the interval.
0h - Disabled.
3h 1h - 62s (for pre-silicon simulation only)
10:8
RW/P 2h - 31ms.
3h - 63ms.
4h - 126ms.
5-7h - Reserved.
May be changed on-the-fly in response to thermal events.
0h
RO
13.4.11 DRAM Reset Management Control (DRMC)—Offset Bh

Access Method
Type: Message Bus Register Offset: [Port: 0x01] + Bh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ODTVAL
CKEVAL
COLDWAKE
ODTMODE
CKEMODE
Rsvd1_DRMC
Rsvd0_DRMC
Rsvd_46
Rsvd_45
Rsvd_44
Bit Default &

Range Access
0h
RO
0h Cold Wake (COLDWAKE): BIOS should set this bit to 1 before sending WAKE
16 command to Memory Controller after Cold Reset. For S3 Exit, or any other mode in
RW which the DRAM is in SR, this bit must be set to 0.

Bit Default &

Range Access
0h
RO
0h ODT Control Mode (ODTMODE): 0 - Memory Controller auto controls the ODT pins
12 based on DRAM Write transactions.
RW 1 - The value of ODTVAL above directly controls the ODT pins.
0h
11:10 Rsvd1_DRMC: Reserved
RO
0h ODT Control Value (ODTVAL): When ODTMODE is set to 1, ODT pins to DRAM are
9:8
RW overridden by ODTVAL. Used only during init flow by BIOS.
0h
RO
0h CKE Control Mode (CKEMODE): 0 - Memory Controller auto controls the CKE pins
4 based on Power-Down and Self Refresh entry and exit
RW . 1 - The value of CKEVAL directly controls the CKE pins
0h
3:2 Rsvd0_DRMC: Reserved
RO
0h CKE Control Value (CKEVAL): When CKEMODE is set to 1, CKE pins to DRAM are
1:0
RW overridden by CKEVAL. Used only during init flow by BIOS.
13.4.12 Power Management Status (PMSTS)—Offset Ch

Access Method
Type: Message Bus Register Offset: [Port: 0x01] + Ch
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DISR
Rsvd_48
Rsvd_47
WRO
Bit Default &

Range Access
0h
RO
0h Warm Reset Occurred (WRO): The Remote Management Unit writes a 1 to this bit to
8 indicate to BIOS a warm reset has just occurred. Can write a 0 to clear it. This is also
RW/P cleared when powergood = 0. This bit will not clear with reset
0h
RO
DRAM In Self-Refresh Status (DISR): The Memory Controller sets this bit to a 1
after it has placed the DRAM devices in Self Refresh mode. The Memory Controller clears
0h this bit when it brings the DRAM devices out of Self Refresh mode. Writing a 1 to this bit,
0 when the COLDWAKE bit is set to 0, will also clear it. This will not clear with system
RW/P reset, but will clear when powergood = 0.
0 - DRAM not guaranteed to be in Self-Refresh.
1 - DRAM in Self-Refresh

13.4.13 DRAM Control Operation (DCO)—Offset Fh

Access Method
Type: Message Bus Register Offset: [Port: 0x01] + Fh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IC
DIOIC
PMICTL
PMIDIS
Rsvd_50
Rsvd_49
CPGCLOCK
DRPLOCK
Bit Default &
Range Access
Memory Controller Initialization Complete (IC): Indicates that initialization of the

Memory Controller has completed. Memory accesses are permitted and maintenance
0h operation begins. Until this bit is set to a 1, the memory controller will not accept DRAM
31
RW requests from the Memory Manager or the MTE.
Note: Set this bit to 1 only when all other Memory Controller registers has been
configured. Usually set at the last configuration step.
0h DDRIO PHY initialization complete (DIOIC): Status indication that DDRIO

30
RO initialization is complete.
Disable PRI interface (PMIDIS): When this bit is set to 1, the Memory Controller will
not respond to requests from either the Memory Manager or the MTE.
0h Note: This bit should be set to 1, when issuing DRAM Read and Write commands
29 through message 68h. It prevents the Memory Controller trying to pull data from the
RW Memory Manager or the MTE for Writes through message 68h, and also prevents the
Memory Manager and the MTE from taking read data returned from Reads through
message 68h.
PRI Control Select (PMICTL): 0 - Memory Controller PRI is connected to Memory

0h Manager.
28 1 - Memory Controller PRI is connected to MTE.
RW When this bit is toggled, the Memory Controller will flush all pending DRAM requests
from the previous unit before taking requests from the new unit
0h
RO
0h MTE Lock (CPGCLOCK): After this bit is set to 1, the MTE is clock gated and locked and
8 cannot be used. CPGCLOCK can be set only once, and will only reset when powergood =
RW/P/L 0.
0h
RO
DRP Register Lock (DRPLOCK): Write a 1 to this bit to lock the DRP and DTRC
0h registers, and to disable the ability to issues the DRAM Read and Write commands
0
RW/P/L through message 68h. Once locked, the DRP and DTRC registers cannot be written
again. Once set to 1, this bit can only be cleared when powergood = 0.
13.4.14 Sticky Scratchpad 0 (SSKPD0)—Offset 4Ah

Access Method
Type: Message Bus Register Offset: [Port: 0x01] + 4Ah
(Size: 32 bits)
Op Codes:

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAL
Bit Default &
Range Access
0h General Purpose Scratchpad (VAL): May be used for BIOS for data storage. Value is
31:0
RW/P preserved in warm-reset.
13.4.15 Sticky Scratchpad 1 (SSKPD1)—Offset 4Bh

Access Method
Type: Message Bus Register Offset: [Port: 0x01] + 4Bh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 VAL 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Bit Default &

Range Access
0h General Purpose Scratchpad (VAL): May be used for BIOS for data storage. Value is
31:0
RW/P preserved in warm-reset.
13.4.16 DRAM ECC Control Register (DECCCTRL)—Offset 60h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Rsvd_58
Rsvd_57
CLRSBECNT
DBEEN
SBEEN
RSVD
ENCBGEN
RSVD
SYNSEL
RSVD
Bit Default &

Range Access
0h
RO
0h
RO

Bit Default &

Range Access
0h Enable Generation of ECC Check Bits (ENCBGEN): 0 - Disable check bit generation.
17
RW 1 - Enable check bit generation.
0b
RO
00h
RO
00h Clear Single Bit Error Count (CLRSBECNT): Clear ECC Single Bit Error Count:
7 0 - allow single bit error count to increment.
RW 1 - clear single bit error count.
Syndrome Select (SYNSEL): ECC Syndrome Bits Select for Observation: The
Syndrome Bits are generated from read data returned from DRAM and used to detect
ECC errors. Each 64 bits of read data is used to generate 8 Syndrome Bits. SYNSEL
00h selects which set of Syndrome Bits to mux to the DECCSTAT register for observation.
6:5
RW 00 - Selects Syndrome Bits from read data [63:0].
01 - Selects Syndrome Bits from read data [127:64].
000h
RO
0h Double Bit Enable (DBEEN): Enable Double Bit Error Detect

1 0: disable double bit error detect.
RW 1: enable double bit error detect.
0h Single Bit Enable (SBEEN): Enable Single Bit Error Detect and Correct
0 0: disable single bit error detect and correct.
RW 1: enable single bit error detect and correct.
13.4.17 DRAM ECC Status (DECCSTAT)—Offset 61h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Rsvd_59
ECCSYN
Bit Default &

Range Access
0h
RO
0h ECC Syndrome Bits (ECCSYN): This is the 8 ECC Syndrome Bits selected for
7:0
RO observation. Selection is made through the DECCCTRL register.
13.4.18 DRAM ECC Single Bit Error Count (DECCSBECNT)—Offset 62h

Access Method


(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ECCSBECNT
Bit Default &
Range Access
0h ECC Single Bit Error Count (ECCSBECNT): Write a 1 to the CLRSBECNT bit in the
31:0
RO DECCCTRL register to clear this register.
13.4.19 DRAM Single Bit ECC Error Captured Address (DECCSBECA)—

Offset 68h
Note: Write any value to this register to clear both this register and the DRAM Single Bit
ECC Error Captured Syndrome register
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SBE_ROW
RSVD
SBE_VLD
SBE_COL
SBE_RANK
SBE_BANK
Bit Default &

Range Access
0h
RO
0h Single Bit ECC Error Valid (SBE_VLD): 0 - No Single Bit ECC Error was detected.
30
RO/P 1 - A Single Bit ECC Error was detected.
0h Captured Rank Address (SBE_RANK): Captured rank address of a read with a Single
29
RO/P Bit ECC Error
0h Captured Bank Address (SBE_BANK): Captured bank address of a read with a Single
28:26
RO/P Bit ECC Error
0h Captured Row Address (SBE_ROW): Captured row address of a read with a Single
25:10
RO/P Bit ECC Error

Bit Default &

Range Access
0h Captured Column Address (SBE_COL): Captured column address of a read with a

9:0
RO/P Single Bit ECC Error
13.4.20 DRAM Single Bit ECC Error Captured Syndrome (DECCSBECS)—

Offset 69h
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SBE_SDROME
Bit Default &

Range Access
0h Captured Syndrome (SBE_SDROME): Captured syndrome of a read with a Single Bit

31:0
RO/P ECC Error
13.4.21 DRAM Double Bit ECC Error Captured Address (DECCDBECA)—

Offset 6Ah
Note: Write any value to this register to clear both this register and the DRAM Double
Bit ECC Error Captured Syndrome register
Access Method
Type: Message Bus Register Offset: [Port: 0x01] + 6Ah
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DBE_VLD
DBE_ROW
DBE_COL
RSVD
DBE_RANK
DBE_BANK
Bit Default &

Range Access
0h
RO

Bit Default &

Range Access
0h Double Bit ECC Error Valid (DBE_VLD): 0 - No Double Bit ECC Error was detected.
30
RO/P 1 - A Double Bit ECC Error was detected.
0h Captured Rank Address (DBE_RANK): Captured rank address of a read with a

29
RO/P Double Bit ECC Error
0h Captured Bank Address (DBE_BANK): Captured bank address of a read with a

28:26
0h Captured Row Address (DBE_ROW): Captured row address of a read with a Double
25:10
RO/P Bit ECC Error
0h Captured Column Address (DBE_COL): Captured column address of a read with a

9:0
13.4.22 DRAM Double Bit ECC Error Captured Syndrome (DECCDBECS)—

Offset 6Bh
Access Method
Type: Message Bus Register Offset: [Port: 0x01] + 6Bh
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DBE_SDROME
Bit Default &

Range Access
0h Captured Syndrome (DBE_SDROME): Captured syndrome of a read with a Double

31:0
RO/P Bit ECC Error
13.4.23 Memory Controller Fuse Status (DFUSESTAT)—Offset 70h

Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FUSESTAT

Bit Default &

Range Access
0h Fuse Status (FUSESTAT): Memory Controller fuse bits are captured in this register.
31:0 [0] ECC Disable
RO [31:1] Reserved
13.4.24 Scrambler Seed (DSCRMSEED)—Offset 80h

Dynamic data scrambler seed register
Access Method
(Size: 32 bits)
Op Codes:
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Rsvd_71
SCRMSEED
Bit Default &
Range Access
0h
RO
0h Scrambler Seed (SCRMSEED): Holds 18 bit scrambler seed value used to feed into
17:0
RW/P LFSR array matrix.

13.5 Message Bus Commands
Table 78. Message Opcode Definition

Opcode Operation Type Description
00h NOP Msg No operation
Wakes the memory from Self-Refresh mode or puts the

Memory Controller in working mode after cold-boot.
CAh Wake Msg
An ACK is sent after the Wake has completed and
memory is accessible.
This message enables accessing the DRAM internal

registers.
Message Data Payload Bits:
31:23 Reserved (set to 0)
22 Rank Address
21:6 Multiplexed Address: Determines the value of
MA[15:0] when the initialization command is sent.
5:3 Bank Address: Determines the value of BA[2:0]
when the initialization command is sent.
2:0 Command (DDR3_RAS_B, DDR3_CAS_B,
DDR3_WE_B): Determines the value driven on
68h DRAM Init MsgD theDDR3_RAS_B, DDR3_CAS_B, and DDR3_WE_B
signals respectively when the initialization command is
sent. The supported commands are listed below:
000 – MRS (Extended) Mode Register Set
001 – Refresh
010 – Precharge (single or all as specified by MA[10])
011 – Bank Activate
100 – Reserved
101 – Reserved
110 – ZQ Calibration (Long/Short as specified by MA[10])
111 – NOP
§§


PCI Express* 2.0—Intel® Quark™ SoC X1000
14.0 PCI Express* 2.0
There are two lanes and two PCI Express* root ports, each supporting the PCI Express*
Base Specification, Rev. 2.0 at a maximum 2.5 GT/s signaling rate.

Please see Chapter 2.0, “Physical Interfaces” for additional details.

Table 83. PCI Express* 2.0 Signals

Direction/
Type
PCI Express* Transmit

PCIE_PETP[1:0] O
PCIE_PETN[1:0] PCI Express* Transmit pair (P and N) signals. Each pair makes up the
PCIe*
transmit half of a lane.
PCI Express* Receive

PCIE_PERP[1:0] I
PCIE_PERN[1:0] PCI Express* Receive pair (P and N) signals. Each pair makes up the
PCIe*
receive half of lane.
IO Note: Please check the Platform Design Guide for connection details
PCIE_IRCOMP
Analog for this COMP pin.
I Note: Please check the Platform Design Guide for connection details
PCIE_RBIAS
Analog for this BIAS pin.
Note: PCIe reference clocks are supplied by the following:

REF[0/1]_OUTCLKP
REF[0/1]_OUTCLKN
14.2 Features
• Conforms to PCI Express* Base Specification, Rev. 2.0
• 2.5 GT/s operation per root port (limited for power saving)
• Virtual Channel support for VC0
• x1 widths
• Supports 2 x1 Root port configurations
• Interrupts and Events
— Legacy (INTx) and MSI Interrupts
— General Purpose Events

Intel® Quark™ SoC X1000—PCI Express* 2.0
— Express Card Hot Plug Events

— System Error Events
• Power Management
— Link State support for L0s, L1, and L2
— Powered down in ACPI S3 state - L3
14.2.1 Interrupts and Events

A root port is capable of handling interrupts and events from an end point device. A
root port can also generate its own interrupts for some events, including power
management and hot plug events, and also including error events.
There are two interrupt types a root port receives from an end point device: INTx
(legacy), and MSI. MSIs are automatically passed upstream by the root port, just as
other memory writes would be. INTx messages are delivered to the Legacy Bridge’s
interrupt decoding and routing logic by the root port.
Events and interrupts that are handled by the root port are shown in Table 84, with the
possible interrupts they can deliver to the interrupt decoder/router.
Table 84. Possible Interrupts Generated From Events/Packets

Packet/Event Type INTx MSI SERR SCI SMI GPE
INTx Packet X X
PM_PME Packet X X
Power Management (PM) Event X X X X
Hot Plug (HP) Event X X X X
ERR_CORR Packet X
ERR_NONFATAL Packet X
ERR_FATAL Packet X
Internal Error Event X
VDM Packet X
Note: Table 84 lists the possible interrupts and events generated based on packets received,
or events generated in the root port. Configuration is performed by the software to
enable the different interrupts as applicable.
Note: GPE is reported as SCI by the root port.
14.2.1.1 Express Card Hot Plug Events

Express Card Hot Plug is available based on Presence Detection for each root port.
Note: A full Hot Plug Controller is not implemented.
Presence detection occurs when a PCI Express* device is plugged in and power is
supplied. The physical layer detects the presence of the device, and the root port sets
the SLCTL_SLSTS.PDS and SLCTL_SLSTS.PDC bits.
When a device is removed and detected by the physical layer, the root port clears the
SLCTL_SLSTS.PDS bit, and sets the SLCTL_SLSTS.PDC bit.

Interrupts can be generated by the root port when a hot plug event occurs. A hot plug
event is defined as the transition of the SLCTL_SLSTS.PDC bit from 0 to 1. Software
can set the SLCTL_SLSTS.PDE and SLTCTL_SLSTS.HPE bits to allow hot plug events to
generate an interrupt.
If SLCTL_SLSTS.PDE and SLTCTL_SLSTS.HPE are both set, and SLCTL_STSTS.PDC

transitions from 0 to 1, an interrupt is generated.
14.2.1.2 System Error (SERR)

System Error events are supported by both internal and external sources. See the PCI
Express* Base Specification, Rev. 2.0 for details.
14.2.2 Power Management

Each root port’s link supports L0s, L1, and L2/3 link states per PCI Express* Base
Specification, Rev. 2.0. L2/3 is entered on entry to S3.
14.3 References
PCI Express* Base Specification, Rev. 2.0
14.4 Register Map

Each root port supports it’s own extended PCI bridge header in PCI configuration space.
These headers are located on PCI bus 0, device 23, functions 0-1 as shown below.
There are no other registers implemented by the root ports or their controller.
See Chapter 5.0, “Register Access Methods” for details on accessing different register
types.

Figure 27. PCI Express Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI
CAM
(I/O)
Bus 0
PCI PCI Express*
ECAM PCI Bridge
(Mem)
Headers
PCIe*
D:23
RP0 F:0
D:23,F:0-1
SPI0 F:0 RP0 F:1
IO Fabric
D:21
SPI1 F:1
I2C*/GPIOF:2
Legacy Bridge
D:31, F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7

Registers listed are for Function 0 (Root Port 0). Function 1 (Root Port 1) contains the
same registers. Differences between Root Ports are noted in the individual registers.

Offset Default
Offset End Register Name (Register Symbol)
Start Value
0h 3h “Identifiers (ID)—Offset 0h” on page 266 11C38086h
4h 7h “Primary Status (CMD_PSTS)—Offset 4h” on page 266 00100000h
8h Bh “Class Code (RID_CC)—Offset 8h” on page 268 06040000h
Ch Fh “Header Type (CLS_PLT_HTYPE)—Offset Ch” on page 268 00810000h
18h 1Bh “Secondary Latency Timer (BNUM_SLT)—Offset 18h” on page 269 00000000h
1Ch 1Fh “Secondary Status (IOBL_SSTS)—Offset 1Ch” on page 269 00000000h
20h 23h “Memory Base and Limit (MBL)—Offset 20h” on page 270 00000000h


Offset Default
Start Value
24h 27h “Prefetchable Memory Base and Limit (PMBL)—Offset 24h” on page 271 00010001h
28h 2Bh “Prefetchable Memory Base Upper 32 Bits (PMBU32)—Offset 28h” on page 271 00000000h
2Ch 2Fh “Prefetchable Memory Limit Upper 32 Bits (PMLU32)—Offset 2Ch” on page 272 00000000h
34h 37h “Capabilities List Pointer (CAPP)—Offset 34h” on page 272 00000040h
3Ch 3Fh “Bridge Control (INTR_BCTRL)—Offset 3Ch” on page 273 00000000h
40h 43h “PCI Express Capabilities (CLIST_XCAP)—Offset 40h” on page 274 00428010h
44h 47h “Device Capabilities (DCAP)—Offset 44h” on page 275 00008000h
48h 4Bh “Device Status (DCTL_DSTS)—Offset 48h” on page 276 00100000h
4Ch 4Fh “Link Capabilities (LCAP)—Offset 4Ch” on page 277 00110C01h
50h 53h “Link Status (LCTL_LSTS)—Offset 50h” on page 279 10010000h
54h 57h “Slot Capabilities (SLCAP)—Offset 54h” on page 280 00040060h
58h 5Bh “Slot Status (SLCTL_SLSTS)—Offset 58h” on page 281 00000000h
5Ch 5Fh “Root Control (RCTL)—Offset 5Ch” on page 283 00000000h
60h 63h “Root Status (RSTS)—Offset 60h” on page 283 00000000h
64h 67h “Device Capabilities 2 (DCAP2)—Offset 64h” on page 284 00000016h
68h 6Bh “Device Status 2 (DCTL2_DSTS2)—Offset 68h” on page 285 00000000h
6Ch 6Fh “Link Capability 2 (LCAP2)—Offset 6Ch” on page 286 00000000h
70h 73h “Link Status 2 (LCTL2_LSTS2)—Offset 70h” on page 286 00000001h
74h 77h “Slot Capabilities 2 (SLCAP2)—Offset 74h” on page 288 00000000h
78h 7Bh “Slot Status 2 (SLCTL2_SLSTS2)—Offset 78h” on page 288 00000000h
80h 83h “Message Signaled Interrupt Message Control (MID_MC)—Offset 80h” on page 289 00009005h
84h 87h “Message Signaled Interrupt Message Address (MA)—Offset 84h” on page 289 00000000h
88h 8Bh “Message Signaled Interrupt Message Data (MD)—Offset 88h” on page 290 00000000h
90h 93h “Subsystem Vendor Capability (SVCAP)—Offset 90h” on page 290 0000A00Dh
94h 97h “Subsystem Vendor IDs (SVID)—Offset 94h” on page 291 00000000h
A0h A3h “PCI Power Management Capabilities (PMCAP_PMC)—Offset A0h” on page 291 C8020001h
A4h A7h “PCI Power Management Control And Status (PMCS)—Offset A4h” on page 292 00000000h
D0h D3h “Channel Configuration (CCFG)—Offset D0h” on page 293 01000000h
D4h D7h “Miscellaneous Port Configuration 2 (MPC2)—Offset D4h” on page 294 00000000h
D8h DBh “Miscellaneous Port Configuration (MPC)—Offset D8h” on page 295 01110000h
DCh DFh “SMI / SCI Status (SMSCS)—Offset DCh” on page 296 00000000h
F4h F7h “Message Bus Control (PHYCTL_PHYCTL2_IOSFSBCTL)—Offset F4h” on page 297 000C3043h
100h 103h “Advanced Error Reporting Capability Header (AECH)—Offset 100h” on page 298 00000000h
104h 107h “Uncorrectable Error Status (UES)—Offset 104h” on page 299 00000000h
108h 10Bh “Uncorrectable Error Mask (UEM)—Offset 108h” on page 300 00000000h
10Ch 10Fh “Uncorrectable Error Severity (UEV)—Offset 10Ch” on page 301 00060011h
110h 113h “Correctable Error Status (CES)—Offset 110h” on page 302 00000000h
114h 117h “Correctable Error Mask (CEM)—Offset 114h” on page 303 00002000h
118h 11Bh “Advanced Error Capabilities and Control (AECC)—Offset 118h” on page 304 00000000h
11Ch 11Fh “Header Log (HL_DW1)—Offset 11Ch” on page 304 00000000h


Offset Default
Start Value
120h 123h “Header Log (HL_DW2)—Offset 120h” on page 305 00000000h
124h 127h “Header Log (HL_DW3)—Offset 124h” on page 305 00000000h
128h 12Bh “Header Log (HL_DW4)—Offset 128h” on page 305 00000000h
12Ch 12Fh “Root Error Command (REC)—Offset 12Ch” on page 306 00000000h
130h 133h “Root Error Status (RES)—Offset 130h” on page 306 00000000h
134h 137h “Error Source Identification (ESID)—Offset 134h” on page 307 00000000h
14.5.1 Identifiers (ID)—Offset 0h

Access Method
(Size: 32 bits) Offset: [B:0, D:23, F:0] + 0h
Power Well: Core
Default: 11C38086h
31 28 24 20 16 12 8 4 0
0 0 0 1 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
DID
VID
Bit Default &
Range Access
11C3h
31:16 Device Identification (DID): PCI Device ID
RO/V
8086h
15:0 Vendor Identification (VID): PCI Vendor ID
RO
14.5.2 Primary Status (CMD_PSTS)—Offset 4h

Access Method
Type: PCI Configuration Register Offset: [B:0, D:23, F:0] + 4h
(Size: 32 bits)
Power Well: Core
Default: 00100000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VGA_PSE
DPE
SSE
RMA
RTA
STA
PDTS
PFBC
IS
FBE
SEE
PERE
MWIE
SCE
BME
MSE
IOSE
RSVD_1
PC66
RSVD_2
WCC
DPD
CLIST
RSVD
ID
Bit Default &

Range Access
0b DPE Detected Parity Error (DPE): Set when the root port receives a command or
31
RWC data from the backbone with a parity error. This is set even if PCMD.PERE is not set.

Bit Default &

Range Access
0b Signaled System Error (SSE): Set when the root port signals a system error to the
30
RWC internal SERR# logic.
0b Received Master Abort (RMA): Set when the root port receives a completion with
29
RWC unsupported request status from the backbone.
0b Received Target Abort (RTA): Set when the root port receives a completion with
28
RWC completer abort from the backbone.
0b Signaled Target Abort (STA): Set whenever the root port forwards a target abort
27
RWC received from the downstream device onto the backbone.
00b
26:25 Primary DEVSEL# Timing Status (PDTS): Reserved per PCI-Express spec.
RO
0b Master Data Parity Error Detected (DPD): Set when the root port receives a
24
RWC completion with a data parity error on the backbone and PCMD.PERE is set.
0b
23 Primary Fast Back to Back Capable (PFBC): Reserved per PCI-Express spec.
RO
0b
22 Reserved (RSVD_1): Reserved.
RO
0b
21 Primary 66 MHz Capable (PC66): Reserved per PCI-Express spec.
RO
1b
20 Capabilities List (CLIST): Indicates the presence of a capabilities list.
RO
0b Interrupt Status (IS): Indicates status of hot plug and power management interrupts
19 on the root port that result in INTx# message generation. This bit is not set if MSI is
RO/V enabled. If MSI is not enabled, this bit is set regardless of the state of CMD.ID.
000b
18:16 Reserved (RSVD_2): Reserved.
RO
00h
RO
Interrupt Disable (ID): This disables pin-based INTx# interrupts on enabled hot plug
and power management events. This bit has no effect on MSI operation. When set,
0b internal INTx# messages will not be generated. When cleared, internal INTx# messages
10 are generated if there is an interrupt for hot plug or power management and MSI is not
RW/RO enabled. This bit does not affect interrupt forwarding from devices connected to the root
port. Assert_INTx and Deassert_INTx messages will still be forwarded to the internal
interrupt controllers if this bit is set.
0b
9 Fast Back to Back Enable (FBE): Reserved per PCI-Express spec.
RO
0b SERR# Enable (SEE): When set, enables the root port to generate an SERR# message
8
RW when PSTS.SSE is set.
0b
7 Wait Cycle Control (WCC): Reserved per PCI-Express spec.
RO
0b Parity Error Response Enable (PERE): Indicates that the device is capable of
6
RW reporting parity errors as a master on the backbone.
0b
5 VGA Palette Snoop (VGA_PSE): Reserved per PCI-Express spec.
RO
0b
4 Memory Write and Invalidate Enable (MWIE): Reserved per PCI-Express spec.
RO
0b
3 Special Cycle Enable (SCE): Reserved per PCI-Express and PCI bridge spec.
RO

Bit Default &

Range Access
Bus Master Enable (BME): When set, allows the root port to forward Memory and I/O
Read/Write cycles onto the backbone from a PCI-Express device. When this bit is 0b,
0b Memory and I/O requests received at a Root Port must be handled as Unsupported
2
RW Requests (UR). This bit does not affect forwarding of Completions in either the Upstream
or Downstream direction. The forwarding of Requests other than Memory or I/O
requests is not controlled by this bit.
0b Memory Space Enable (MSE): When set, memory cycles within the range specified by
1 the memory base and limit registers can be forwarded to the PCI-Express device. When
RW cleared, these memory cycles are master aborted on the backbone.
0b I/O Space Enable (IOSE): When set, I/O cycles within the range specified by the I/O
0 base and limit registers can be forwarded to the PCI-Express device. When cleared,
RW these cycles are master aborted on the backbone.
14.5.3 Class Code (RID_CC)—Offset 8h

Access Method
Power Well: Core
Default: 06040000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RID
BCC
SCC
PI
Bit Default &

Range Access
06h
31:24 Base Class Code (BCC): Indicates the device is a bridge device.
RO
04h
23:16 Sub-Class Code (SCC): The default indicates the device is a PCI-to-PCI bridge.
RO/V
00h
15:8 Programming Interface (PI): This is a read only register.
RO/V
00h
7:0 Revision ID (RID): Indicates the revision of the bridge.
RO/V
14.5.4 Header Type (CLS_PLT_HTYPE)—Offset Ch

Access Method
(Size: 32 bits) Offset: [B:0, D:23, F:0] + Ch
Default: 00810000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MFD
CT
RSVD
LS
HTYPE
RSVD0

Bit Default &

Range Access
0b
31:24 RSVD0: Reserved
RO
1b
23 Multi-function Device (MFD): This bit is '1' to indicate a multi-function device.
RO
01h Header Type (HTYPE): The default mode identifies the header layout of the
22:16
RO/V configuration space, which is a PCI-to-PCI bridge.
00h
15:11 Latency Count (CT): Reserved per PCI-Express spec.
RO
000b
RO
00h
7:0 Line Size (LS): This is read/write but contains no functionality, per PCI-Express spec.
RW
14.5.5 Secondary Latency Timer (BNUM_SLT)—Offset 18h

This register is reserved for a root port per PCI-Express spec.
Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SBBN
SCBN
PBN
SLT
Bit Default &

Range Access
00h Secondary Latency Timer (SLT): This register is RO and returns 0. This register does
31:24
RW/RO not affect the behavior of any HW logic.
00h Subordinate Bus Number (SBBN): Indicates the highest PCI bus number below the
23:16
RW bridge.
00h
15:8 Secondary Bus Number (SCBN): Indicates the bus number the port.
RW
00h
7:0 Primary Bus Number (PBN): Indicates the bus number of the backbone.
RW
14.5.6 Secondary Status (IOBL_SSTS)—Offset 1Ch

Access Method
(Size: 32 bits) Offset: [B:0, D:23, F:0] + 1Ch
Power Well: Core
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DPE
RSE
RMA
RTA
STA
SDTS
DPD
RSVD
IOLA
IOBA
IOBC
SFBC
SC66
RSVD_1
IOLC
Bit Default &
Range Access
0b
31 Detected Parity Error (DPE): Set when the port receives a poisoned TLP.
RWC
0b Received System Error (RSE): Set when the port receives an ERR_FATAL or
30
RWC ERR_NONFATAL message from the device.
0b Received Master Abort (RMA): Set when the port receives a completion with
29
RWC Unsupported Request status from the device.
0b Received Target Abort (RTA): Set when the port receives a completion with
28
RWC Completion Abort status from the device.
0b Signaled Target Abort (STA): Set when the port generates a completion with
27
RWC Completion Abort status to the device.
00b
26:25 Secondary DEVSEL# Timing Status (SDTS): Reserved per PCI-Express spec.
RO/V
Data Parity Error Detected (DPD): Set when the BCTRL.PERE, and either of the
0b following two conditions occurs:
24
RWC Port receives completion marked poisoned.
Port poisons a write request to the secondary side.
0b
23 Secondary Fast Back to Back Capable (SFBC): Reserved per PCI-Express spec.
RO/V
0b
RO
0b
21 Secondary 66 MHz Capable (SC66): Reserved per PCI Express spec.
RO
00h
RO
0h I/O Address Limit (IOLA): I/O Base bits corresponding to address lines 15:12 for
15:12
RW 4KB alignment. Bits 11:0 are assumed to be padded to FFFh.
0h I/O Limit Address Capability (IOLC): Indicates that the bridge does not support 32-
11:8
RO bit I/O addressing.
0h I/O Base Address (IOBA): I/O Base bits corresponding to address lines 15:12 for
7:4
RW 4KB alignment. Bits 11:0 are assumed to be padded to 000h.
0h I/O Base Address Capability (IOBC): Indicates that the bridge does not support 32-
3:0
RO bit I/O addressing.
14.5.7 Memory Base and Limit (MBL)—Offset 20h

Accesses that are within the ranges specified in this register will be sent to the attached
device if CMD.MSE is set. Accesses from the attached device that are outside the
ranges specified will be forwarded to the backbone if CMD.BME is set.
Access Method
(Size: 32 bits)
Power Well: Core

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ML
RSVD
MB
RSVD_1
Bit Default &
Range Access
000h Memory Limit (ML): These bits are compared with bits 31:20 of the incoming address
31:20
RW to determine the upper 1MB aligned value of the range.
0h
RO
000h Memory Base (MB): These bits are compared with bits 31:20 of the incoming address
15:4
RW to determine the lower 1MB aligned value of the range.
0h
RO
14.5.8 Prefetchable Memory Base and Limit (PMBL)—Offset 24h

Accesses that are within the ranges specified in this register will be sent to the device if
CMD.MSE is set. Accesses from the device that are outside the ranges specified will be
forwarded to the backbone if CMD.BME is set.
Access Method
Power Well: Core
Default: 00010001h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
PML
I64L
PMB
Bit Default &

Field Name (ID): Description I64B
Range Access
000h Prefetchable Memory Limit (PML): These bits are compared with bits 31:20 of the
31:20
RW incoming address to determine the upper 1MB aligned value of the range.
1h
19:16 64-bit Indicator (I64L): Indicates support for 64-bit addressing.
RO
000h Prefetchable Memory Base (PMB): These bits are compared with bits 31:20 of the
15:4
RW incoming address to determine the lower 1MB aligned value of the range.
1h
3:0 64-bit Indicator (I64B): Indicates support for 64-bit addressing.
RO
14.5.9 Prefetchable Memory Base Upper 32 Bits (PMBU32)—Offset 28h

Access Method


(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PMBU
Bit Default &
Range Access
00000000h Prefetchable Memory Base Upper Portion (PMBU): Upper 32-bits of the
31:0
RW prefetchable address base.
14.5.10 Prefetchable Memory Limit Upper 32 Bits (PMLU32)—Offset 2Ch

Access Method
Type: PCI Configuration Register Offset: [B:0, D:23, F:0] + 2Ch
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PMLU
Bit Default &

Range Access
00000000h Prefetchable Memory Limit Upper Portion (PMLU): Upper 32-bits of the
31:0
RW prefetchable address limit.
14.5.11 Capabilities List Pointer (CAPP)—Offset 34h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000040h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
RSVD1
PTR

Bit Default &

Range Access
0000000h
31:8 Reserved (RSVD1): Reserved
RO
Capabilities Pointer (PTR): Indicates that the pointer for the first entry in the
capabilities list.
BIOS can determine which capabilities will be exposed by including or removing them
from the capability linked list.
As this register is RWO, BIOS must write a value to this register, even if it is to re-write
the default value.
40h Capability Linked List (Default Settings)
7:0 Offset Capability Next Pointer
RWO 40h PCI Express 80h
80h Message Signaled Interrupt (MSI) 90h
90h Subsystem Vendor A0h
A0h PCI Power Management 00h
Extended PCIe Capability Linked List
Offset Capability Next Pointer
100h Advanced Error Reporting 000h
14.5.12 Bridge Control (INTR_BCTRL)—Offset 3Ch

Access Method
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
DTSE
SDT
DTS
PDT
FBE
VE
IE
SE
PERE
ILINE
SBR
MAM
V16
IPIN
Bit Default &

Range Access
0h
RO
0b
27 Discard Timer SERR# Enable (DTSE): Reserved per PCI-Express spec.
RO/RW
0b
26 Discard Timer Status (DTS): Reserved per PCI-Express spec.
RO
0b
25 Secondary Discard Timer (SDT): Reserved per PCI-Express spec.
RO/RW
0b
24 Primary Discard Timer (PDT): Reserved per PCI-Express spec.
RO/RW
0b
23 Fast Back to Back Enable (FBE): Reserved per Express spec.
RO
0b
22 Secondary Bus Reset (SBR): Triggers a Hot Reset on the PCI-Express port.
RW
0b
21 Master Abort Mode (MAM): Reserved per PCI-Express spec.
RO/RW

Bit Default &

Range Access
VGA 16-Bit Decode (V16): When set, indicates that the I/O aliases of the VGA range
0b (see BCTRL.VE definition below), are not enabled, and only the base I/O ranges can be
20 decoded.
RW 0: Execute 10-bit address decode on VGA I/O accesses.
1: Execute 16-bit address decode on VGA I/O accesses.
VGA Enable (VE): When set, the following ranges will be claimed off the backbone by
0b the root port:
19 Memory ranges A0000h-BFFFFh
RW I/O ranges 3B0h-3BBh and 3C0h-3DFh, and all aliases of bits 15:10 in any combination
of 1's
ISA Enable (IE): This bit only applies to I/O addresses that are enabled by the I/O
0b Base and I/O Limit registers and are in the first 64KB of PCI I/O space. If this bit is set,
18
RW the root port will block any forwarding from the backbone to the device of I/O
transactions addressing the last 768 bytes in each 1KB block (offsets 100h to 3FFh).
0b SERR# Enable (SE): When set, ERR_COR, ERR_NONFATAL, and ERR_FATAL messages
17
RW received are forwarded to the backbone. When cleared, they are not.
0b Parity Error Response Enable (PERE): When set, poisoned write TLPs and
16
RW completions indicating poisoned TLPs will set the SSTS.DPD.
Interrupt Pin (IPIN): Indicates the interrupt pin driven by the root port. At reset, this
register takes on the following values, which reflect the reset state of the D28IP register
in chipset config space:
Port Bits(15:12) Bits(11:08)
1 0h D28IP.P1IP
00h 2 0h D28IP.P2IP
15:8 3 0h D28IP.P3IP
RO/V 4 0h D28IP.P4IP
5 0h D28IP.P5IP
6 0h D28IP.P6IP
7 0h D28IP.P7IP
8 0h D28IP.P8IP
The value that is programmed into D28IP is always reflected in this register.
00h Interrupt Line (ILINE): Software written value to indicate which interrupt line
7:0
RW (vector) the interrupt is connected to. No hardware action is taken on this register.
14.5.13 PCI Express Capabilities (CLIST_XCAP)—Offset 40h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00428010h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
CV
RSVD_1
IMN
SI
RSVD
DT
NEXT
CID
Bit Default &

Range Access
0b
RO
0b Reserved (RSVD_1): This register at one time was for TCS Routing but that was later
30
RO removed from the PCIe 2.0 spec.

Bit Default &

Range Access
00h Interrupt Message Number (IMN): The PCH does not have multiple MSI interrupt
29:25
RO numbers.
0b Slot Implemented (SI): Indicates whether the root port is connected to a slot. Slot
24 support is platform specific. BIOS programs this field, and it is maintained until a
RWO platform reset.
4h
23:20 Device / Port Type (DT): Indicates this is a PCI-Express root port.
RO
2h Capability Version (CV): Version 2.0 indicates devices compliant to the PCI Express
19:16
RO 2.0 specification which incorporates the Register Expansion ECN.
Next Capability (NEXT): Indicates the location of the next capability.

80h The default value of this register is 80h which points to the MSI Capability structure.
15:8 BIOS can determine which capabilities will be exposed by including or removing them
RWO from the capability linked list. As this register is RWO, BIOS must write a value to this
register, even if it is to re-write the default value.
10h
7:0 Capability ID (CID): Indicates this is a PCI Express capability.
RO
14.5.14 Device Capabilities (DCAP)—Offset 44h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00008000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD_1
RSVD_2
RSVD_3
RSVD_4
RBER
E1AL
E0AL
RSVD
CSPS
CSPV
ETFS
PFS
MPS
FLRC
Bit Default &

Range Access
000b
RO
0b
28 Function Level Reset Capable (FLRC): Not supported in Root Ports.
RO
00b
27:26 Captured Slot Power Limit Scale (CSPS): Not supported.
RO
00h
25:18 Captured Slot Power Limit Value (CSPV): Not supported.
RO
00b
RO
1b Role Based Error Reporting (RBER): Indicates that this device implements the
15
RO functionality defined in the Error Reporting ECN as required by the PCI Express 1.1 spec.
0b Reserved (RSVD_2): On previous version of the specification this was Power Indicator
14
RO Present (PIP).
0b Reserved (RSVD_3): On previous version of the specification this was Attention

13
RO Indicator Present (AIP).

Bit Default &

Range Access
0b Reserved (RSVD_4): On previous version of the specification this was Attention

12
RO Button Present (ABP).
000b
11:9 Endpoint L1 Acceptable Latency (E1AL): Reserved for Root port.
RO
000b
8:6 Endpoint L0 Acceptable Latency (E0AL): Reserved for Root port.
RO
Extended Tag Field Supported (ETFS): The PCH root port never needs to initiate a
0b transaction as a Requester with the Extended Tag bits being set. This bit does not affect
5
RO the root port's ability to forward requests as a bridge as the root port always supports
forwarding requests with extended tags.
00b
4:3 Phantom Functions Supported (PFS): No phantom functions supported.
RO
000b Max Payload Size Supported (MPS): Indicates the maximum payload size supported
2:0
RO is 128B.
14.5.15 Device Status (DCTL_DSTS)—Offset 48h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00100000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TDP
RSVD_1
APD
URD
FED
NFED
CED
RSVD
MRRS
ENS
APME
PFE
ETFE
MPS
ERO
URE
FEE
NFE
CEE
Bit Default &
Range Access
000h
RO
0b Transactions Pending (TDP): This bit has no meaning for the root port since it never
21
RO initiates a non-posted request with its own RequesterID.
1b
20 AUX Power Detected (APD): The root port contains AUX power for wakeup.
RO
0b Unsupported Request Detected (URD): Indicates an unsupported request was

19
RWC detected.
0b Fatal Error Detected (FED): Indicates a fatal error was detected. Set when a fatal
18
RWC error occurred from a data link protocol error, buffer overflow, or malformed TLP.
0b Non-Fatal Error Detected (NFED): Indicates a non-fatal error was detected. Set
17 when received a non-fatal error occurred from a poisoned TLP, unexpected completions,
RWC unsupported requests, completer abort, or completer timeout.
0b Correctable Error Detected (CED): Indicates a correctable error was detected. Set
16 when received an internal correctable error from receiver errors / framing errors, TLP
RWC CRC error, DLLP CRC error, replay number rollover, or replay timeout.
0b
RO

Bit Default &

Range Access
000b
14:12 Max Read Request Size (MRRS): Hardwired to 0
RO
0b Enable No Snoop (ENS): Not supported. The root port will never issue non-snoop
11
RO requests.
0b Aux Power PM Enable (APME): Must be RW for OS testing. The OS will set this bit to
10 '1' if the device connected has detected aux power. It has no effect on the root port
RW/P otherwise. This registers is in the resume well.
0b
9 Phantom Functions Enable (PFE): Not supported.
RO
0b
8 Extended Tag Field Enable (ETFE): Not supported.
RO
000b
7:5 Max Payload Size (MPS): The root port only supports 128B payloads.
RO
0b
4 Enable Relaxed Ordering (ERO): Not supported.
RO
Unsupported Request Reporting Enable (URE): When set, allows signaling

ERR_NONFATAL, ERR_FATAL, or ERR_COR to the Root Control register when detecting
0b an unmasked Unsupported Request (UR). An ERR_COR is signaled when a unmasked
3
RW Advisory Non-Fatal UR is received. An ERR_FATAL, ERR_or NONFATAL, is sent to the
Root Control Register when an uncorrectable non-Advisory UR is received with the
severity set by the Uncorrectable Error Severity register.
0b Fatal Error Reporting Enable (FEE): Enables signaling of ERR_FATAL to the Root
2 Control register due to internally detected errors or error messages received across the
RW link. Other bits also control the full scope of related error reporting.
Non-Fatal Error Reporting Enable (NFE): When set, enables signaling of

0b ERR_NONFATAL to the Root Control register due to internally detected errors or error
1
RW messages received across the link. Other bits also control the full scope of related error
reporting.
Correctable Error Reporting Enable (CEE): When set, enables signaling of

0b ERR_CORR to the Root Control register due to internally detected errors or error
0
RW messages received across the link. Other bits also control the full scope of related error
reporting.
14.5.16 Link Capabilities (LCAP)—Offset 4Ch

Access Method
Power Well: Core
Default: 00110C01h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1
RSVD
APMS
SLS
PN
RSVD_1
LARC
SDERC
CPM
EL1
EL0
MLW

Bit Default &

Range Access
Port Number (PN): Indicates the port number for the root port. This value is different
for each implemented port:
Port # Value of PN field
1 01h
00h 2 02h
31:24 3 03h
RO/V 4 04h
5 05h
6 06h
7 07h
8 08h
00b
RO
0b Reserved (RSVD_1): This port does not support Link Bandwidth Notification
21
RO Capability.
1b Link Active Reporting Capable (LARC): This port supports the optional capability of
20
RO reporting the DL_Active state of the Data Link Control and Management State Machine.
0b Surprise Down Error Reporting Capable (SDERC): Set to '0' to indicate the PCH
19
RO does not support Surprise Down Error Reporting.
0b Clock Power Management (CPM): '0' Indicates that PCH root ports do not support
18
RO the CLKREQ# mechanism.
L1 Exit Latency (EL1): Indicates an exit latency of 2 s to 4 s.

000b Less than 1 s
001b 1 s to less than 2 s
010b 2 s to less than 4 s
010b 011b 4 s to less than 8 s
17:15 100b 8 s to less than 16 s
RWO 101b 16 s to less than 32 s
110b 32 s to 64 s
111b More than 64 s
Note: If PXP PLL shutdown is enabled, BIOS should program this latency to comprehend
PLL lock latency.
L0s Exit Latency (EL0): Indicates an exit latency based upon common-clock
000b configuration:
14:12 LCAP.CCC Value
RO/V 0 MPC.UCEL
1 MPC.CCEL
Active State Link PM Support (APMS): Indicates the level of active state power
management on this link:
11b Bits Definition
11:10 00 (Reserved)
RWO 01 L0s Entry supported
10 Reserved
11 Both L0s and L1 supported
Maximum Link Width (MLW): For the root ports, several values can be taken, based
upon the value of the chipset configuration register field RPC.PC1 for ports 1-4 and
RPC.PC2 for ports 5-6:
RPC.PC1 00 01 10 11
1 01h 02h 02h 04h
000000b 2 01h 01h 01h 01h
9:4 3 01h 01h 02h 01h
RO/V 4 01h 01h 01h 01h
RPC.PC2 00 01 10 11
5 01h 02h 02h 04h
6 01h 01h 01h 01h
7 01h 01h 02h 01h
8 01h 01h 01h 01h
1h Supported Link Speeds (SLS): Indicates the supported link speeds of the Root Port.
3:0 0001b 2.5 GT/s Link speed supported
RO/V 0010b 5.0 GT/s and 2.5GT/s Link speeds supported

14.5.17 Link Status (LCTL_LSTS)—Offset 50h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 10010000h
31 28 24 20 16 12 8 4 0
0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LABIE
LBMIE
SCC
RCBC
NLW
CCC
ECPM
ASPM
RL
RSVD_1
LBMS
LA
LT
RSVD_2
LABS
CLS
RSVD
HAWD
ES
LD
Bit Default &
Range Access
Link Autonomous Bandwidth Status (LABS): This bit is Set by hardware to indicate
that hardware has autonomously changed Link speed or width, without the Port
0b transitioning through DL_Down status, for reasons other than to attempt to correct
31 unreliable Link operation.
RWC This bit must be set if the Physical Layer reports a speed or width change was initiated
by the Downstream component that was indicated as an autonomous change.
The default value of this bit is 0b.
Link Bandwidth Management Status (LBMS): This bit is Set by hardware to indicate
that either of the following has occurred without the Port transitioning through DL_Down
status:
* A Link retraining has completed following a write of 1b to the Retrain Link bit
0b Note: This bit is Set following any write of 1b to the Retrain Link bit, including when the
30 Link is in the process of retraining for some other reason.
RWC * Hardware has changed Link speed or width to attempt to correct unreliable Link
operation, either through an LTSSM timeout or a higher level process
This bit must be set if the Physical Layer reports a speed or width change was initiated
by the Downstream component that was not indicated as an autonomous change.
0b Link Active (LA): Set to 1b when the Data Link Control and Management State
29
RO/V Machine is in the DL_Active state, 0b otherwise.
1b Slot Clock Configuration (SCC): PCH uses the same reference clock as on the
28
RO platform and does not generate its own clock.
0b Link Training (LT): The root port sets this bit whenever link training is occurring, or
27 that 1b was written to the Retrain Link bit but Link training has not yet begun. It clears
RO/V the bit upon completion of link training.
0b Reserved (RSVD_2): Previously this was defined as Link Training Error (LTE) but
26 support for this bit was removed from subsequent versions of the PCI Express
RO specification.
Negotiated Link Width (NLW): For the root ports, this register could take on several
values:
RPC.PC1 00 01 10 11
1 01h 02h 02h 04h
2 01h 01h 01h 01h
000000b 3 01h 01h 02h 01h
25:20 4 01h 01h 01h 01h
RO/V Port # Value of PN field
RPC.PC2 00 01 10 11
5 01h 02h 02h 04h
6 01h 01h 01h 01h
7 01h 01h 02h 01h
8 01h 01h 01h 01h
The value of this register is undefined if the link has not successfully trained.

Bit Default &

Range Access
1h Current Link Speed (CLS): 0001b Link is 2.5 GT/s Link

19:16 0010b Link is 5.0 GT/s Link
RO/V The value of this field is undefined if the link is not up.
0h
RO
0b Link Autonomous Bandwidth Interrupt Enable (LABIE): When Set, this bit enables
11 the generation of an interrupt to indicate that the Link Autonomous Bandwidth Status
RW bit has been Set.
Link Bandwidth Management Interrupt Enable (LBMIE): When Set, this bit
enables the generation of an interrupt to indicate that the Link Bandwidth Management
0b Status bit has been Set. This bit is not applicable and is reserved for Endpoints, PCI
10
RW Express-to-PCI/PCI-X bridges, and Upstream Ports of Switches. Functions that do not
implement the Link Bandwidth Notification Capability must hardwire this bit to 0b.
Default value of this bit is 0b.
0b Hardware Autonomous Width Disable (HAWD): When Set, this bit disables
9 hardware from changing the Link width for reasons other than attempting to correct
RW unreliable Link operation by reducing Link width. Default value of this bit is 0b.
0b Enable Clock Power Management (ECPM): Reserved. Not supported on PCH Root
8
RO Ports.
0b Extended Synch (ES): When set, forces extended transmission of FTS ordered sets in
7
RW FTS and extra TS2 at exit from L1 prior to entering L0.
0b Common Clock Configuration (CCC): When set, indicates that the PCH and device
6
RW are operating with a distributed common reference clock.
Retrain Link (RL): When set, the root port will train its downstream link. This bit
always returns '0' when read. Software uses LSTS.LT and LSTS.LTE to check the status
0b of training. It is permitted to write 1b to this bit while simultaneously writing modified
5 values to other fields in this register. If the LTSSM is not already in Recovery or
WO Configuration, the resulting Link training must use the modified values. If the LTSSM is
already in Recovery or Configuration, the modified values are not required to affect the
Link training that's already in progress.
0b Link Disable (LD): When set, the root port will disable the link by directing the LTSSM
4
RW to the Disabled state.
0b Read Completion Boundary Control (RCBC): Indicates the read completion

3
RO boundary is 64 bytes.
0b
2 Reserved (RSVD_1): Reserved.
RO
Active State Link PM Control (ASPM): Indicates whether the root port should enter
L0s or L1 or both.
Bits Definition
00b 00 Disabled
1:0 01 L0s Entry Enabled
RW 10 L1 Entry Enabled
11 L0s and L1 Entry Enabled
The value of this register is used unless the Root Port ASPM Control Override Enable
register is set, in which case the Root Port ASPM Control Override value is used.
14.5.18 Slot Capabilities (SLCAP)—Offset 54h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00040060h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0
NCCS
SLS
SLV__54_14_8
SLV__54_7_7
HPS
PSN__54_31_24
PSN__54_23_19
EMIP
HPC
PIP
AIP
MSP
PCP
ABP
Bit Default &
Range Access
00h Physical Slot Number (PSN__54_31_24): This is a value that is unique to the slot
31:24
RWO number. BIOS sets this field and it remains set until a platform reset.
00h Physical Slot Number (PSN__54_23_19): This is a value that is unique to the slot
23:19
RWO number. BIOS sets this field and it remains set until a platform reset.
1b No Command Completed Support (NCCS): Set to '1' as this port does not implement
18 a Hot Plug controller and can handle back-2-back writes to all fields of the slot control
RO register without delay between successive writes.
0b Electromechanical Interlock Present (EMIP): Set to 0 to indicate that no electro-

17
RO mechanical interlock is implemented.
00b Slot Power Limit Scale (SLS): Specifies the scale used for the slot power limit value.
16:15
RWO BIOS sets this field and it remains set until a platform reset.
Slot Power Limit Value (SLV__54_14_8): Specifies the upper limit (in conjunction
00h with SLS value), on the upper limit on power supplied by the slot. The two values
14:8
RWO together indicate the amount of power in watts allowed for the slot. BIOS sets this field
and it remains set until a platform reset.
Slot Power Limit Value (SLV__54_7_7): Specifies the upper limit (in conjunction
0b with SLS value), on the upper limit on power supplied by the slot. The two values
7
RWO together indicate the amount of power in watts allowed for the slot. BIOS sets this field
and it remains set until a platform reset.
1b
6 Hot Plug Capable (HPC): When set, Indicates that hot plug is supported.
RWO
1b Hot Plug Surprise (HPS): When set, indicates the device may be removed from the
5
RWO slot without prior notification.
0b Power Indicator Present (PIP): Indicates that a power indicator LED is not present
4
RO for this slot.
0b Attention Indicator Present (AIP): Indicates that an attention indicator LED is not
3
RO present for this slot.
0b
2 MRL Sensor Present (MSP): Indicates that an MRL sensor is not present.
RO
0b Power Controller Present (PCP): Indicates that a power controller is not

1
RO implemented for this slot.
0b Attention Button Present (ABP): Indicates that an attention button is not

0
RO implemented for this slot.
14.5.19 Slot Status (SLCTL_SLSTS)—Offset 58h

Access Method
Power Well: Core
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PDS
PFD
RSVD
PFE
ABE
DLLSC
EMIS
MS
DLLSCE
HPE
CCE
PDE
MSE
RSVD_1
CC
PDC
MSC
ABP
EMIC
PCC
PIC
AIC
Bit Default &
Range Access
00h
RO
Data Link Layer State Changed (DLLSC): This bit is set when the value reported in
0b Data Link Layer Link Active field of the Link Status register is changed. In response to a
24 Data Link Layer State Changed event, software must read Data Link Layer Link Active
RWC field of the Link Status register to determine if the link is active before initiating
configuration cycles to the hot plugged device.
0b Electromechanical Interlock Status (EMIS): Reserved as this port does not support
23
RO and electromechanical interlock.
0b Presence Detect State (PDS): If XCAP.SI is set (indicating that this root port spawns
22 a slot), then this bit indicates whether a device is connected ('1') or empty ('0'). If
RO/V XCAP.SI is cleared, this bit is a '1'.
0b
21 MRL Sensor State (MS): Reserved as the MRL sensor is not implemented.
RO
0b Command Completed (CC): This register is RO as this port does not implement a Hot
20
RO Plug Controller.
0b Presence Detect Changed (PDC): This bit is set by the root port when the PD bit
19
RWC changes state.
0b
18 MRL Sensor Changed (MSC): Reserved as the MRL sensor is not implemented.
RO
0b
17 Power Fault Detected (PFD): Reserved as a power controller is not implemented.
RO
0b Attention Button Pressed (ABP): This register is RO as this port does not implement
16
RO an attention button.
000b
RO
0b Data Link Layer State Changed Enable (DLLSCE): When set, this field enables
12
RW generation of a hot plug interrupt when the Data Link Layer Link Active field is changed.
0b Electromechanical Interlock Control (EMIC): Reserved as this port does not

11
RO support an Electromechanical Interlock.
0b
10 Power Controller Control (PCC): This bit has no meaning for module based hot plug.
RO
00b Power Indicator Control (PIC): This register is RO as this port does not implement a
9:8
RO Hot Plug Controller.
00b Attention Indicator Control (AIC): This register is RO as this port does not
7:6
RO implement a Hot Plug Controller.
0b Hot Plug Interrupt Enable (HPE): When set, enables generation of a hot plug
5
RW interrupt on enabled hot plug events.
0b Command Completed Interrupt Enable (CCE): This register is RO as this port does
4
RO not implement a Hot Plug Controller.
0b Presence Detect Changed Enable (PDE): When set, enables the generation of a hot
3
RW plug interrupt or wake message when the presence detect logic changes state.

Bit Default &

Range Access
0b MRL Sensor Changed Enable (MSE): This register is RO as this port does not
2
0b Power Fault Detected Enable (PFE): This register is RO as this port does not
1
0b Attention Button Pressed Enable (ABE): This register is RO as this port does not
0
14.5.20 Root Control (RCTL)—Offset 5Ch

Access Method
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
PIE
SFE
SNE
SCE
RSVD1
Bit Default &

Range Access
00000h
RO
000h
RO
0b PME Interrupt Enable (PIE): When set, enables interrupt generation when RSTS.PS
3 is in a set state (either due to a '0' to '1' transition, or due to this bit being set with
RW RSTS.PS already set).
0b System Error on Fatal Error Enable (SFE): When set, an SERR# will be generated if
2 a fatal error is reported by any of the devices in the hierarchy of this root port, including
RW fatal errors in this root port. This register is not dependent on CMD.SEE being set.
System Error on Non-Fatal Error Enable (SNE): When set, an SERR# will be
0b generated if a non-fatal error is reported by any of the devices in the hierarchy of this
1
RW root port, including non-fatal errors in this root port. This register is not dependent on
CMD.SEE being set.
System Error on Correctable Error Enable (SCE): When set, an SERR# will be
0b generated if a correctable error is reported by any of the devices in the hierarchy of this
0
RW root port, including correctable errors in this root port. This register is not dependent on
CMD.SEE being set.
14.5.21 Root Status (RSTS)—Offset 60h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
PS
RID
PP
Bit Default &
Range Access
0000h
RO
PME Pending (PP): Indicates another PME is pending when the PME status bit is set.
0b When the original PME is cleared by software, it will be set again, the requestor ID will
17
RO/V be updated, and this bit will be cleared. PCH root ports have a one deep PME pending
queue.
0b PME Status (PS): Indicates that PME was asserted by the requestor ID in RID.
16
RWC Subsequent PMEs are kept pending until this bit is cleared.
PME Requestor ID (RID): Indicates the PCI requestor ID of the last PME requestor.
0000h Valid only when PS is set. PCH root ports are capable of storing the requester ID for two
15:0
RO/V PM_PME messages, with one active (this register) and a one deep pending queue.
Subsequent PM_PME messages will be dropped.
14.5.22 Device Capabilities 2 (DCAP2)—Offset 64h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000016h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0
RSVD_1
RSVD
LTRMS
CTDS
CTRS
Bit Default &
Range Access
00000h
RO
LTR Mechanism Supported (LTRMS): A value of 1b indicates support for the optional
0b Latency Tolerance Reporting (LTR) mechanism capability.
11
RWO BIOS must write to this register with either a '1' or a '0' to enable/disable the root port
from declaring support for the LTR capability.
00h
RO
1b Completion Timeout Disable Supported (CTDS): A value of 1b indicates support for

4
RO the Completion Timeout Disable mechanism.

Bit Default &

Range Access
Completion Timeout Ranges Supported (CTRS): This field indicates device support
for the optional Completion Timeout programmability mechanism. This mechanism
allows system software to modify the Completion Timeout value.
This field is applicable only to Root Ports, Endpoints that issue requests on their own
behalf, and PCI Express to PCI/PCI-X Bridges that take ownership of requests issued on
PCI Express.
For all other devices this field is reserved and must be hardwired to 0000b.
Four time value ranges are defined:
Range A: 50us to 10ms
Range B: 10ms to 250ms
6h Range C: 250ms to 4s
3:0
RO Range D: 4s to 64s
Bits are set according to the table below to show timeout value ranges supported.
0000b Completion Timeout programming not supported.
0001b Range A
0010b Range B
0011b Ranges A and B
0110b Ranges B and C
0111b Ranges A, B and C
1110b Ranges B, C and D
1111b Ranges A, B, C and D
All other values are reserved.
14.5.23 Device Status 2 (DCTL2_DSTS2)—Offset 68h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD_2
LTREN
RSVD_1
RSVD
CTD
CTV
Bit Default &
Range Access
0000h
RO
00h
RO
0b LTR Mechanism Enable (LTREN): When Set to 1b, this bit enables the Latency
10
RW Tolerance Reporting (LTR) mechanism.
00h
RO
Completion Timeout Disable (CTD): When set to 1b, this bit disables the Completion
Timeout mechanism.
This field is required for all devices that support the Completion Timeout Disable
Capability.
0b Software is permitted to set or clear this bit at any time. When set, the Completion
4 Timeout detection mechanism is disabled.
RW If there are outstanding requests when the bit is cleared, it is permitted but not required
for hardware to apply the completion timeout mechanism to the outstanding requests. If
this is done, it is permitted to base the start time for each request on either the time
this bit was cleared or the time each request was issued.
Only the value from Port 1 (for ports 1-4) or Port 5 (for ports 5-8) is used.

Bit Default &

Range Access
Completion Timeout Value (CTV): In Devices that support Completion Timeout

programmability, this field allows system software to modify the Completion Timeout
value. This field is applicable to Root Ports, Endpoints that issue requests on their own
behalf, and PCI Express to PCI/PCI-X Bridges that take ownership of requests issued on
PCI Express. For all other devices this field is reserved and must be hardwired to 0000b.
A Device that does not support this optional capability must hardwire this field to 0000b
and is required to implement a timeout value in the range 50us to 50ms. Devices that
support Completion Timeout programmability must support the values given below
corresponding to the programmability ranges indicated in the Completion Timeout
Values Supported field.
The PCH targeted configurable ranges are listed below, along with the range allowed by
the PCI Express 2.0 specification.
0h Defined encodings:
3:0
RW 0000b Default range: 40-50ms (spec range 50us to 50ms)
Values available if Range B (10ms to 250ms) programmability range is supported:
0101b 40-50ms (spec range is 16ms to 55ms)
Values available if Range C (250ms to 4s) programmability range is supported:
1010b 1.6-1.7s (spec range is 1s to 3.5s)
Values not defined above are Reserved.
Software is permitted to change the value in this field at any time. For requests already
pending when the Completion Timeout Value is changed, hardware is permitted to use
either the new or the old value for the outstanding requests, and is permitted to base
the start time for each request either on when this value was changed or on when each
request was issued.
14.5.24 Link Capability 2 (LCAP2)—Offset 6Ch

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
Bit Default &

Range Access
00000000h
RO
14.5.25 Link Status 2 (LCTL2_LSTS2)—Offset 70h

Access Method
Power Well: Core
Default: 00000001h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
CDL
RSVD
CD
CSOS
SD
HASD
TLS
RSVD_1
EMC
TM
EC
Bit Default &
Range Access
0000h
RO
Current De-emphasis Level (CDL): When the Link is operating at 5 GT/s speed, this
bit reflects the level of de-emphasis.
0b Encodings:
16
RO/V 1b -3.5 dB
0b -6 dB
The value in this bit is undefined when the Link is operating at 2.5 GT/s speed.
000b
RO
Compliance De-emphasis (CD): This bit sets the de-emphasis level in

Polling.Compliance state if the entry occurred due to the Enter Compliance bit being 1b.
Encodings:
0b 1b -3.5 dB
12 0b -6 dB
RW/P When the Link is operating at 2.5 GT/s, the setting of this bit has no effect.
This bit is intended for debug, compliance testing purposes. System firmware and
software is allowed to modify this bit only during debug or compliance testing.
0b Compliance SOS (CSOS): When set to 1b, the LTSSM is required to send SKP Ordered
11 Sets periodically in between the (modified) compliance patterns.
RW/P The default value of this bit is 0b.
Enter Modified Compliance (EMC): When this bit is set to 1b, the device transmits
Modified Compliance Pattern if the LTSSM enters Polling.Compliance substate.
0b Default value of this bit is 0b.
10
RW/P This register is intended for debug, compliance testing purposes only. System firmware
and software is allowed to modify this register only during debug or compliance testing.
In all other cases, the system must ensure that this register is set to the default value.
Transmit Margin (TM): This field controls the value of the nondeemphasized voltage
level at the Transmitter pins. This field is reset to 000b on entry to the LTSSM
Polling.Configuration substate (see PCI Express Chapter 4 for details of how the
Transmitter voltage level is determined in various states).
Encodings:
000b Normal operating range
001b 800-1200 mV for full swing and 400-700 mV for half-swing
010b-(n-1) Values must be monotonic with a non-zero slope. The value of n must be
greater than 3 and less than 7. At least two of these must be below the normal
000b operating range of n : 200-400 mV for full-swing and 100-200 mV for half-swing
9:7
RW/P n-111b reserved
For a Multi-Function device associated with an Upstream Port, the field in Function 0 is
of type RWS, and only Function 0 controls the component's Link behavior. In all other
Functions of that device, this field is of type RsvdP.
Default value of this field is 000b.
Components that support only the 2.5 GT/s speed are permitted to hardwire this bit to
000b.
This register is intended for debug, compliance testing purposes only. System firmware
and software is allowed to modify this register only during debug or compliance testing.
In all other cases, the system must ensure that this register is set to the default value.
Selectable De-emphasis (SD): When the Link is operating at 5.0 GT/s speed, this bit
selects the level of de-emphasis for an Upstream component.
0b Encodings:
6
RW/P 1b -3.5 dB
0b -6 dB
When the Link is operating at 2.5 GT/s speed, the setting of this bit has no effect.

Bit Default &

Range Access
0b Hardware Autonomous Speed Disable (HASD): This port cannot autonomously

5
RO change speeds.
Enter Compliance (EC): Software is permitted to force a Link to enter Compliance

0b mode at the speed indicated in the Target Link Speed field by setting this bit to 1b in
4
RW/P both components on a Link and then initiating a hot reset on the Link.
Default value of this bit following Fundamental Reset is 0b.
1h Target Link Speed (TLS): This field sets an upper limit on Link operational speed by
3:0
RW/F/P restricting the values advertised by the upstream component in its training sequences.
14.5.26 Slot Capabilities 2 (SLCAP2)—Offset 74h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
Bit Default &

Range Access
00000000h
RO
14.5.27 Slot Status 2 (SLCTL2_SLSTS2)—Offset 78h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD_1
RSVD
Bit Default &

Range Access
0000h
RO
0000h
RO

14.5.28 Message Signaled Interrupt Message Control (MID_MC)—Offset

80h
Access Method
(Size: 32 bits)
Power Well: Core
Default: 00009005h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1
MME
MSIE
MMC
C64
RSVD
NEXT
CID
Bit Default &
Range Access
00h
RO
0b
23 64-Bit Address Capable (C64): Capable of generating a 32-bit message only.
RO
000b Multiple Message Enable (MME): These bits are RW for software compatibility, but
22:20
RW only one message is ever sent by the root port.
000b
19:17 Multiple Message Capable (MMC): Only one message is required.
RO
0b MSI Enable (MSIE): If set, MSI is enabled and traditional interrupt pins are not used
16 to generate interrupts. CMD.BME must be set for an MSI to be generated. If CMD.BME is
RW cleared, and this bit is set, no interrupts (not even pin based) are generated.
Next Pointer (NEXT): Indicates the location of the next capability in the list.
The default value of this register is 90h which points to the Subsystem Vendor capability
90h structure.
RWO from the capability linked list.
the default value.
05h
7:0 Capability ID (CID): Capabilities ID indicates MSI.
RO
14.5.29 Message Signaled Interrupt Message Address (MA)—Offset 84h

Access Method
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ADDR
RSVD

Bit Default &

Range Access
00000000h Address (ADDR): Lower 32 bits of the system specified message address, always DW
31:2
RW aligned.
00b
RO
14.5.30 Message Signaled Interrupt Message Data (MD)—Offset 88h

Access Method
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DATA
RSVD1
Bit Default &

Range Access
00000h
RO
0000h Data (DATA): This 16-bit field is programmed by system software if MSI is enabled. Its
15:0 content is driven onto the lower word (PCI AD[15:0]) during the data phase of the MSI
RW memory write transaction.
14.5.31 Subsystem Vendor Capability (SVCAP)—Offset 90h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 0000A00Dh
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 1 1 0 1
RSVD1
NEXT
CID
Bit Default &

Range Access
00000h
RO

Bit Default &

Range Access
Next Capability (NEXT): Indicates the location of the next capability in the list.
The default value of this register is A0h which points to the PCI Power Management
A0h capability structure.
RWO from the capability linked list.
the default value.
0Dh Capability Identifier (CID): Value of 0Dh indicates this is a PCI bridge subsystem
7:0
RO vendor capability.
14.5.32 Subsystem Vendor IDs (SVID)—Offset 94h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SID
SVID
Bit Default &
Range Access
0000h Subsystem Identifier (SID): Indicates the subsystem as identified by the vendor.
31:16 This field is write once and is locked down until a bridge reset occurs (not the PCI bus
RWO reset).
0000h Subsystem Vendor Identifier (SVID): Indicates the manufacturer of the subsystem.
15:0 This field is write once and is locked down until a bridge reset occurs (not the PCI bus
RWO reset).
14.5.33 PCI Power Management Capabilities (PMCAP_PMC)—Offset A0h

Access Method
Type: PCI Configuration Register Offset: [B:0, D:23, F:0] + A0h
(Size: 32 bits)
Power Well: Core
Default: C8020001h
31 28 24 20 16 12 8 4 0
1 1 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
PMES
D2S
D1S
VS
AC
DSI
PMEC
RSVD
NEXT
CID
Bit Default &

Range Access
PME Support (PMES): Indicates PME# is supported for states D0, D3HOT and
11001b D3COLD. The root port does not generate PME#, but reporting that it does is necessary
31:27
RO for legacy Microsoft operating systems to enable PME# in devices connected behind this
root port.

Bit Default &

Range Access
0b
26 D2 Support (D2S): The D2 state is not supported.
RO
0b
25 D1 Support (D1S): The D1 state is not supported.
RO
000b Aux Current (AC): Reports 375mA maximum suspend well current required when in
24:22
RO the D3COLD state.
0b Device Specific Initialization (DSI): Indicates that no device-specific initialization is

21
RO required.
0b
RO
0b
19 PME Clock (PMEC): Indicates that PCI clock is not required to generate PME#.
RO
010b Version (VS): Indicates support for Revision 1.1 of the PCI Power Management
18:16
RO Specification.
00h
15:8 Next Capability (NEXT): Indicates this is the last item in the list.
RO
01h Capability Identifier (CID): Value of 01h indicates this is a PCI power management
7:0
RO capability.
14.5.34 PCI Power Management Control And Status (PMCS)—Offset A4h

Access Method
(Size: 32 bits) Offset: [B:0, D:23, F:0] + A4h
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
BPCE
DTA
B23S
PMES
PMEE
PS
DSC
DSEL
RSVD_1
RSVD
Bit Default &

Range Access
00h
31:24 Data (DTA): Reserved
RO
0b
23 Bus Power / Clock Control Enable (BPCE): Reserved per PCI Express specification.
RO
0b
22 B2/B3 Support (B23S): Reserved per PCI Express specification.
RO
00h
RO
0b
15 PME Status (PMES): Indicates a PME was received on the downstream link.
RO
00b
14:13 Data Scale (DSC): Reserved
RO

Bit Default &

Range Access
0h
12:9 Data Select (DSEL): Reserved
RO
PME Enable (PMEE): Indicates PME is enabled. The root port takes no action on this
0b bit, but it must be RW for legacy Microsoft operating systems to enable PME# on devices
8 connected to this root port. This register resides in the resume well and is not reset on a
RW/W resume from S3/S4/S5.
The reset for this register is RSMRST# which is not asserted during a Warm Reset.
00h
RO
Power State (PS): This field is used both to determine the current power state of the
root port and to set a new power state. The values are:
00 D0 state
00b 11 D3HOT state
1:0 When in the D3HOT state, the controller's configuration space is available, but the I/O
RW and memory spaces are not. Type 1 configuration cycles are also not accepted.
Interrupts are not required to be blocked as software will disable interrupts prior to
placing the port into D3HOT.
If software attempts to write a '10' or '01' to these bits, the write will be ignored.
14.5.35 Channel Configuration (CCFG)—Offset D0h

Access Method
Type: PCI Configuration Register Offset: [B:0, D:23, F:0] + D0h
(Size: 32 bits)
Power Well: Core
Default: 01000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
RSVD
RSVD
UPSD
UNSD
RSVD
RSVD
RSVD
UNRS
UPRS
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &
Range Access
0b
RO
0b
RO
00h
RO
Upstream Posted Split Disable (UPSD): When '0', upstream posted memory
requests will be split on boundaries defined by the UPRS bit in this register.
1b When '1', upstream posted memory requests will not be split and will be presented to
24 the backbone as received from the link.
RW This register has no effect on posted messages which are never split.
BIOS must program this bit to '0'.
Upstream Non-Posted Split Disable (UNSD): When '0', upstream non-posted

requests will be split on boundaries defined by the UNRS bit in this register.
0b When '1', upstream non-posted requests will not be split and will be presented to the
23
RW backbone as received from the link.

Bit Default &

Range Access
00h
RO
0b
RO
0b
RO
Upstream Non-Posted Request Size (UNRS): Sets the size for splitting upstream
memory read requests. Requests will be split on naturally aligned addresses.
0b When '0', requests are split at 128 byte boundaries.
15 When '1', requests are split at 64 byte boundaries. This field is only used if the UNSD bit
RW is '0'.
Upstream Posted Request Size (UPRS): Sets the size for splitting upstream memory
write requests. Requests will be split on naturally aligned addresses.
When '0', requests are split at 128 byte boundaries.
0b When '1', requests are split at 64 byte boundaries. This field is only used if the UPSD bit
14
RW is '0'.
This register has no effect on posted messages which are never split.
00b
RO
0b
RO
0b
RO
0b
RO
0b
RO
00h
RO
14.5.36 Miscellaneous Port Configuration 2 (MPC2)—Offset D4h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IPF
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &

Range Access
00000h
RO

Bit Default &

Range Access
Packet Fast Transmit Mode (IPF): When set, the PCIe transmit block will move the
packet header from the Tx buffer to the retry buffer without waiting for the
0b corresponding data to be available.
11 When cleared, the packet transfer to the retry buffer will not occur until the Tx buffer
RW has the entire data phase available.
BIOS must program this bit to 1.
0b
RO
0b
RO
0b
RO
0b
RO
00b
RO
0b
RO
00b
RO
0b
RO
0b
RO
14.5.37 Miscellaneous Port Configuration (MPC)—Offset D8h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 01110000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
UCEL
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
CCEL
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
PMCE
HPCE
HPME
PMME
Bit Default &

Range Access
0b Power Management SCI Enable (PMCE): Enables the root port to generate SCI
31
RW whenever a power management event is detected.
0b Hot Plug SCI Enable (HPCE): Enables the root port to generate SCI whenever a hot
30
RW plug event is detected.
0b
RO
0b
RO

Bit Default &

Range Access
0b
RO
0b
RO
0b
RO
1b
RO
0b
RO
0b
RO
0b
RO
100b Unique Clock Exit Latency (UCEL): This value represents the L0s Exit Latency for
20:18 unique-clock configurations (LCAP.CCC = '0'). It defaults to 512ns to less than 1us, but
RW may be overridden by BIOS.
010b Common Clock Exit Latency (CCEL): This value represents the L0s Exit Latency for
17:15 common-clock configurations (LCAP.CCC = '1'). It defaults to 128ns to less than 256ns,
RW but may be overridden by BIOS.
0b
RO
0b
RO
0b
RO
0h
RO
0b
RO
000b
RO
0b
RO
0b
RO
0b Hot Plug SMI Enable (HPME): Enables the root port to generate SMI whenever a hot
1
RW plug event is detected.
0b Power Management SMI Enable (PMME): Enables the root port to generate SMI
0
RW whenever a power management event is detected.
14.5.38 SMI / SCI Status (SMSCS)—Offset DCh

Access Method
Type: PCI Configuration Register Offset: [B:0, D:23, F:0] + DCh
(Size: 32 bits)
Power Well: Core
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
RSVD
RSVD
PMCS
PMMS
HPCS
HPLAS
HPPDM
Bit Default &
Range Access
0b Power Management SCI Status (PMCS): This bit is set if the root port PME control
31 logic needs to generate an interrupt, and this interrupt has been routed to generate an
RWC SCI.
0b Hot Plug SCI Status (HPCS): This bit is set if the hot plug controller needs to
30
RWC generate an interrupt, and has this interrupt been routed to generate an SCI.
0000000h
RO
0b Hot Plug Link Active State Changed SMI Status (HPLAS): This bit is set when
4 SLSTS.LASC transitions from '0' to '1', and MPC.HPME is set. When this bit is set, an
RWC SMI# will be generated.
0b
RO
0b
RO
0b Hot Plug Presence Detect SMI Status (HPPDM): This bit is set when SLSTS.PDC
1 transitions from '0' to '1', and MPC.HPME is set. When this bit is set, an SMI# will be
RWC generated.
0b Power Management SMI Status (PMMS): This bit is set when RSTS.PS transitions
0
RWC from '0' to '1', and MPC.PMME is set.
14.5.39 Message Bus Control (PHYCTL_PHYCTL2_IOSFSBCTL)—Offset

F4h
Access Method
(Size: 32 bits) Offset: [B:0, D:23, F:0] + F4h
Default: 000C3043h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 1 0 0 0 0 1 1
RSVD0
SBIC
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
RSVD
Bit Default &

Range Access
0b
RO
0h
RO
11b
RO

Bit Default &

Range Access
Message Bus Idle Counter (SBIC): This register provides configuration flexibility to
00b govern when the Message Bus interface transitions to IDLE.
17:16
RW BIOS must program this field to 11b to prevent transitions to IDLE on the Message Bus
interface.
00b
RO
11b
RO
000b
RO
0b
RO
0b
RO
10b
RO
0b
RO
00b
RO
11b
RO
14.5.40 Advanced Error Reporting Capability Header (AECH)—Offset

100h
The AER capability can optionally be included or excluded from the capabilities list. The
full AER is supported.
Access Method
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CID
NCO
CV
Bit Default &

Range Access
000h
31:20 Next Capability Offset (NCO): Set to 000h as this is the last capability in the list.
RWO
0h Capability Version (CV): For systems that support AER, BIOS should write a 1h to this
19:16
RWO register else it should write 0

Bit Default &

Range Access
0000h Capability ID (CID): For systems that support AER, BIOS should write a 0001h to this
15:0
RWO register else it should write 0
14.5.41 Uncorrectable Error Status (UES)—Offset 104h

This register must maintain its state through a platform reset. It loses its state upon
loss of core power.
Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
UC
RSVD_1
RSVD_2
RSVD
AVS
MT
URE
EE
RO
CA
CT
PT
SDE
FCPE
DLPE
TE
Bit Default &
Range Access
000h
RO
0b
21 ACS Violation Status (AVS): Reserved. Access Control Services are not supported.
RO
0b Unsupported Request Error Status (URE): Indicates an unsupported request was

20
RWC/P received.
0b
19 ECRC Error Status (EE): ECRC is not supported.
RO
0b
18 Malformed TLP Status (MT): Indicates a malformed TLP was received.
RWC/P
0b
17 Receiver Overflow Status (RO): Indicates a receiver overflow occurred.
RWC/P
0b Unexpected Completion Status (UC): Indicates an unexpected completion was

16
RWC/P received.
0b
15 Completer Abort Status (CA): Indicates a completer abort was received.
RWC/P
0b Completion Timeout Status (CT): Indicates a completion timed out. This is signaled if
14 Completion Timeout is enabled and a completion fails to return within the amount of
RWC/P time specified by the Completion Timeout Value.
0b
13 Flow Control Protocol Error Status (FCPE): Not supported.
RO
0b
12 Poisoned TLP Status (PT): Indicates a poisoned TLP was received.
RWC/P
00h
RO
0b
5 Surprise Down Error Status (SDE): Surprise Down is not supported.
RO

Bit Default &

Range Access
0b Data Link Protocol Error Status (DLPE): Indicates a data link protocol error
4
RWC/P occurred.
000b
RO
0b
0 Training Error Status (TE): Not supported.
RO
14.5.42 Uncorrectable Error Mask (UEM)—Offset 108h

When set, the corresponding error in the UES register is masked, and the logged error
will cause no action. When cleared, the corresponding error is enabled. This register is
only reset by a loss of core power.
Access Method
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
AVS
MT
URE
EE
RO
CT
FCPE
PT
SDE
DLPE
TE
UC
CM
RSVD_1
RSVD_2
Bit Default &
Range Access
000h
RO
0b
21 ACS Violation Status (AVS): Reserved. Access Control Services are not supported.
RO
0b
20 Unsupported Request Error Mask (URE): Mask for uncorrectable errors.
RW/P
0b
19 ECRC Error Mask (EE): ECRC is not supported.
RO
0b
18 Malformed TLP Mask (MT): Mask for malformed TLPs.
RW/P
0b
17 Receiver Overflow Mask (RO): Mask for receiver overflows.
RW/P
0b
16 Unexpected Completion Mask (UC): Mask for unexpected completions.
RW/P
0b
15 Completer Abort Mask (CM): Mask for completer abort.
RW/P
0b
14 Completion Timeout Mask (CT): Mask for completion timeouts.
RW/P
0b
13 Flow Control Protocol Error Mask (FCPE): Not supported.
RO

Bit Default &

Range Access
0b
12 Poisoned TLP Mask (PT): Mask for poisoned TLPs.
RW/P
00h
RO
0b
5 Surprise Down Error Mask (SDE): Surprise Down is not supported.
RO
0b
4 Data Link Protocol Error Mask (DLPE): Mask for data link protocol errors.
RW/P
000b
RO
0b
0 Training Error Mask (TE): Not supported.
RO
14.5.43 Uncorrectable Error Severity (UEV)—Offset 10Ch

This register gives the option to make an uncorrectable error fatal or non-fatal. An
error is fatal if the bit is set. An error is non-fatal if the bit is cleared. This register is
Access Method
Power Well: Core
Default: 00060011h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1
AVS
MT
URE
SDE
EE
RO
CA
CT
FCPE
PT
DLPE
TE
UC
RSVD_1
RSVD_2
RSVD
Bit Default &

Range Access
000h
RO
0b
21 ACS Violation Severity (AVS): Reserved. Access Control Services are not supported.
RO
0b Unsupported Request Error Severity (URE): Severity for unsupported request

20
RW/P reception.
0b
19 ECRC Error Severity (EE): ECRC is not supported.
RO
1b
18 Malformed TLP Severity (MT): Severity for malformed TLP reception.
RW/P
1b
17 Receiver Overflow Severity (RO): Severity for receiver overflow occurrences.
RW/P
0b Unexpected Completion Severity (UC): Severity for unexpected completion

16
RW/P reception.

Bit Default &

Range Access
0b
15 Completer Abort Severity (CA): Severity for completer abort.
RW/P
0b
14 Completion Timeout Severity (CT): Severity for completion timeout.
RW/P
0b
13 Flow Control Protocol Error Severity (FCPE): Not supported.
RO
0b
12 Poisoned TLP Severity (PT): Severity for poisoned TLP reception.
RW/P
00h
RO
0b
5 Surprise Down Error Severity (SDE): Surprise Down is not supported.
RO
1b
4 Data Link Protocol Error Severity (DLPE): Severity for data link protocol errors.
RW/P
000b
RO
1b Training Error Severity (TE): TE not supported. This bit is left as RO='1' for ease of
0
RO implementation.
14.5.44 Correctable Error Status (CES)—Offset 110h

This register is only reset by a loss of core power.
Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD_1
RSVD_2
RSVD
RNR
ANFES
RTT
BD
BT
RE
Bit Default &

Range Access
00000h
RO
0b Advisory Non-Fatal Error Status (ANFES): When set, indicates that a Advisory Non-
13
RWC/P Fatal Error occurred.
0b
12 Replay Timer Timeout Status (RTT): Indicates the replay timer timed out.
RWC/P
000b
RO
0b
8 Replay Number Rollover Status (RNR): Indicates the replay number rolled over.
RWC/P

Bit Default &

Range Access
0b
7 Bad DLLP Status (BD): Indicates a bad DLLP was received.
RWC/P
0b
6 Bad TLP Status (BT): Indicates a bad TLP was received.
RWC/P
00h
RO
0b
0 Receiver Error Status (RE): Indicates a receiver error occurred.
RWC/P
14.5.45 Correctable Error Mask (CEM)—Offset 114h

When set, the corresponding error in the CES register is masked, and the logged error
will cause no action. When cleared, the corresponding error is enabled. This register is
Access Method
Power Well: Core
Default: 00002000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
RTT
BD
BT
RE
ANFEM
RSVD_1
RSVD_2
RNR
Bit Default &
Range Access
00000h
RO
Advisory Non-Fatal Error Mask (ANFEM): When set, masks Advisory Non-Fatal
1b errors from (a) signaling ERR_COR to the device control register and (b) updating the
13 Uncorrectable Error Status register.
RW/P This register is set by default to enable compatibility with software that does not
comprehend Role-Based Error Reporting.
0b
12 Replay Timer Timeout Mask (RTT): Mask for replay timer timeout.
RW/P
000b
RO
0b
8 Replay Number Rollover Mask (RNR): Mask for replay number rollover.
RW/P
0b
7 Bad DLLP Mask (BD): Mask for bad DLLP reception.
RW/P
0b
6 Bad TLP Mask (BT): Mask for bad TLP reception.
RW/P
00h
RO

Bit Default &

Range Access
0b
0 Receiver Error Mask (RE): Mask for receiver errors.
RW/P
14.5.46 Advanced Error Capabilities and Control (AECC)—Offset 118h

This register is only reset by a loss of core power.
Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
ECE
EGE
ECC
EGC
FEP
Bit Default &
Range Access
000000h
RO
0b
8 ECRC Check Enable (ECE): ECRC is not supported.
RO
0b
7 ECRC Check Capable (ECC): ECRC is not supported.
RO
0b
6 ECRC Generation Enable (EGE): ECRC is not supported.
RO
0b
5 ECRC Generation Capable (EGC): ECRC is not supported.
RO
00000b First Error Pointer (FEP): Identifies the bit position of the first error reported in the
4:0
RO/V/P Uncorrectable Error Status Register.
14.5.47 Header Log (HL_DW1)—Offset 11Ch

These registers report the header for the TLP corresponding to a detected error. This
register is only reset by a loss of core power.
Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DW1

Bit Default &

Range Access
00000000h
31:0 4th DWord of TLP (DW1): Byte12 and Byte13 and Byte14 and Byte15
RO/V/P
14.5.48 Header Log (HL_DW2)—Offset 120h

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DW2
Bit Default &
Range Access
00000000h
31:0 3rd DWord of TLP (DW2): Byte8 and Byte9 and Byte10 and Byte11
RO/V/P

Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DW2
Bit Default &

Range Access
00000000h
31:0 2nd DWord of TLP (DW2): Byte4 and Byte5 and Byte6 and Byte7
RO/V/P


Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DW1
Bit Default &
Range Access
00000000h
31:0 1st DWord of TLP (DW1): Byte0 and Byte1 and Byte2 and Byte3
RO/V/P
14.5.51 Root Error Command (REC)—Offset 12Ch

This register allows errors to generate interrupts.
Access Method
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
FERE
NERE
CERE
Bit Default &
Range Access
00000000h
RO
0b Fatal Error Reporting Enable (FERE): When set, the root port will generate an
2
RW interrupt when a fatal error is reported by the attached device.
0b Non-fatal Error Reporting Enable (NERE): When set, the root port will generate an
1
RW interrupt when a non-fatal error is reported by the attached device.
0b Correctable Error Reporting Enable (CERE): When set, the root port will generate
0
RW an interrupt when a correctable error is reported by the attached device.
14.5.52 Root Error Status (RES)—Offset 130h

This register can track more than one error and set the multiple bits if a second or
subsequent error occurs and the first has not been serviced. This register is only reset
by a loss of core power.
Access Method


(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0
AEMN 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
NFEMR
FUF
FEMR
MENR
ENR
MCR
CR
Bit Default &
Range Access
00h Advanced Error Interrupt Message Number (AEMN): Reserved. There is only one
31:27
RO error interrupt allocated.
00000h
RO
0b Fatal Error Message Received (FEMR): Set when one or more Fatal Uncorrectable
6
RWC/P Error Messages have been received.
0b Non-Fatal Error Messages Received (NFEMR): Set when one or more Non-Fatal
5
RWC/P Uncorrectable error messages have been received.
0b First Uncorrectable Fatal (FUF): Set when the first Uncorrectable Error message
4
RWC/P received is for a fatal error.
0b Multiple ERR_FATAL/NONFATAL Received (MENR): Set when either a fatal or a

3
RWC/P non-fatal error is received and the ENR bit is already set.
0b ERR_FATAL/NONFATAL Received (ENR): Set when either a fatal or a non-fatal error

2
RWC/P message is received.
0b Multiple ERR_COR Received (MCR): Set when a correctable error message is

1
RWC/P received and the CR bit is already set.
0b
0 ERR_COR Received (CR): Set when a correctable error message is received.
RWC/P
14.5.53 Error Source Identification (ESID)—Offset 134h

Identifies the source (Requester ID) of the first correctable and uncorrectable (Non-
Fatal / Fatal) errors reported in the Root Error Status register. This register is only reset
by a loss of core power.
Access Method
(Size: 32 bits)
Power Well: Core
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EFNFSID
ECSID

Bit Default &

Range Access
0000h ERR_FATAL/NONFATAL Source Identification (EFNFSID): Loaded with the

31:16 Requester ID indicated in the received ERR_FATAL or ERR_NONFATAL Message with the
RO/V/P ERR_FATAL/NONFATAL Received register is not already set.
0000h ERR_COR Source Identification (ECSID): Loaded with the Requester ID indicated in
15:0
RO/V/P the received ERR_COR Message with the ERR_COR Received register is not already set.
§§

10/100 Mbps Ethernet—Intel® Quark™ SoC X1000
15.0 10/100 Mbps Ethernet
The Intel® Quark™ SoC X1000 provides two 10/100 Mbps Ethernet controllers. Each
controller includes a MAC but not a PHY. The integrated controller is compatible with an
industry standard, RMII based Ethernet PHY.


Table 86. 10/100 Ethernet Interface Signals

Direction/
Type
RMII_REF_CLK I 50 MHz reference clock for the RMII interface
MAC[0/1]_TXDATA[1:0] O RMII Transmit data
MAC[0/1]_TXEN O RMII Transmit data enable
MAC[0/1]_RXDATA[1:0] I RMII Receive data
MAC[0/1]_RXDV I RMII Receive data valid
MAC[0/1]_MDC O Management data clock
MAC[0/1]_MDIO I/O Management data
15.2 Features:
• 10 and 100 Mbps data transfer rates with RMII interface to communicate with an
external Fast Ethernet PHY
• Compliant with RMII specification version 1.2 from RMII consortium
• Full-duplex operation:
— IEEE* 802.3x flow control automatic transmission of zero-quanta pause frame
on flow control input de-assertion
— Optional forwarding of received pause control frames to the user application
• Half-duplex operation:
— CSMA/CD Protocol support
• Preamble and start-of-frame data (SFD) insertion in transmit path
• Preamble and SFD deletion in the receive path

Intel® Quark™ SoC X1000—10/100 Mbps Ethernet
• Automatic CRC and pad generation controllable on a per-frame basis

• Automatic Pad and CRC Stripping options for receive frames
• Flexible address filtering modes:
— 64-bit hash filter for multicast and unicast (DA) addresses
— Option to pass all multicast addressed frames
— Promiscuous mode to pass all frames without any filtering for network
monitoring
— Pass all incoming packets (as per filter) with a status report
• Programmable frame length.
• Programmable Interframe Gap (IFG) (40-96 bit times in steps of 8)
• Option to transmit frames with reduced preamble size
• Separate 32-bit status for transmit and receive packets
• IEEE 802.1Q VLAN tag detection for reception frames
• Additional frame filtering:
— VLAN tag-based: hash-based filtering
• Separate transmission, reception, and control interfaces to the application
• Little-endian configuration for transmit and receive paths
• 32-bit data transfer interface on system-side
• Network statistics with RMON/MIB Counters (RFC2819/RFC2665)
• Enhanced receive module for checking IPv4 header checksum and TCP, UDP, or
ICMP checksum encapsulated in IPv4 or IPv6 datagrams (Type 2)
• Support Ethernet frame time stamping as described in IEEE 1588-2002 and
IEEE 1588-2008 The 64-bit timestamps are given in the transmit or receive status
of each frame
• MDIO master interface for PHY device configuration and management
• CRC replacement, source address field insertion or replacement, and VLAN
insertion, replacement, and deletion in transmitted frames with per-frame control
15.3 References
• IEEE 802.3TM Ethernet: http://standards.ieee.org/about/get/802/802.3.html
• Alert Standard Format Specification, Version 1.03: http://www.dmtf.org/standards/
asf

15.4 Register Map
Figure 28. Ethernet Register Map
PCI Space
CPU
Core
Memory
Host Bridge Ethernet Space
D:0,F:0 PCI Header
PCI
D:20,F:6, F7
CAM
(I/O)
Bus 0
PCI MBAR
ECAM
(Mem)
RP0 F:0
PCIe*
D:23
SPI0 F:0 Ethernet

RP0 F:1
IO Fabric
Mem
D:21
SPI1 F:1
Registers
I2C*/GPIOF:2
Legacy Bridge
D:31, F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7

Registers listed are for Function 6 (MAC 0). Function 7 (MAC 1) contain the same
registers. Differences between MACs are noted in individual registers.

Default
Value
0h 1h “Vendor ID (VENDOR_ID)—Offset 0h” on page 312 8086h
2h 3h “Device ID (DEVICE_ID)—Offset 2h” on page 313 0937h
4h 5h “Command Register (COMMAND_REGISTER)—Offset 4h” on page 313 0000h
6h 7h “Status Register (STATUS)—Offset 6h” on page 314 0010h


Default
Value
8h Bh “Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h” on page 315 02000010h
Ch Ch “Cache Line Size (CACHE_LINE_SIZE)—Offset Ch” on page 315 00h
Dh Dh “Latency Timer (LATENCY_TIMER)—Offset Dh” on page 315 00h
Eh Eh “Header Type (HEADER_TYPE)—Offset Eh” on page 316 80h
Fh Fh “BIST (BIST)—Offset Fh” on page 316 00h
10h 13h “Base Address Register (BAR0)—Offset 10h” on page 317 00000000h
28h 2Bh “Cardbus CIS Pointer (CARDBUS_CIS_POINTER)—Offset 28h” on page 317 00000000h
2Ch 2Dh “Subsystem Vendor ID (SUB_SYS_VENDOR_ID)—Offset 2Ch” on page 318 0000h
2Eh 2Fh “Subsystem ID (SUB_SYS_ID)—Offset 2Eh” on page 318 0000h
30h 33h “Expansion ROM Base Address (EXP_ROM_BASE_ADR)—Offset 30h” on page 318 00000000h
34h 37h “Capabilities Pointer (CAP_POINTER)—Offset 34h” on page 319 00000080h
3Ch 3Ch “Interrupt Line Register (INTR_LINE)—Offset 3Ch” on page 319 00h
3Dh 3Dh “Interrupt Pin Register (INTR_PIN)—Offset 3Dh” on page 320 00h
3Eh 3Eh “MIN_GNT (MIN_GNT)—Offset 3Eh” on page 320 00h
3Fh 3Fh “MAX_LAT (MAX_LAT)—Offset 3Fh” on page 320 00h
80h 80h “Capability ID (PM_CAP_ID)—Offset 80h” on page 321 01h
81h 81h “Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h” on page 321 A0h
82h 83h “Power Management Capabilities (PMC)—Offset 82h” on page 321 4803h
84h 85h “Power Management Control/Status Register (PMCSR)—Offset 84h” on page 322 0008h
“PM CSR PCI-to-PCI Bridge Support Extension (PMCSR_BSE)—Offset 86h” on

86h 86h 00h
page 323
87h 87h “Power Management Data Register (DATA_REGISTER)—Offset 87h” on page 323 00h
A0h A0h “Capability ID (MSI_CAP_ID)—Offset A0h” on page 324 05h
A1h A1h “Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h” on page 324 00h
A2h A3h “Message Control (MESSAGE_CTRL)—Offset A2h” on page 324 0100h
A4h A7h “Message Address (MESSAGE_ADDR)—Offset A4h” on page 325 00000000h
A8h A9h “Message Data (MESSAGE_DATA)—Offset A8h” on page 325 0000h
ACh AFh “Mask Bits for MSI (PER_VEC_MASK)—Offset ACh” on page 326 00000000h
B0h B3h “Pending Bits for MSI (PER_VEC_PEND)—Offset B0h” on page 326 00000000h
15.5.1 Vendor ID (VENDOR_ID)—Offset 0h

Access Method
(Size: 16 bits) VENDOR_ID: [B:0, D:20, F:6] + 0h
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value

Bit Default &

Description
Range Access
8086h
15: 0 Vendor ID (value): PCI Vendor ID for Intel
RO
15.5.2 Device ID (DEVICE_ID)—Offset 2h

Access Method
(Size: 16 bits) DEVICE_ID: [B:0, D:20, F:6] + 2h
Default: 0937h
15 12 8 4 0
0 0 0 0 1 0 0 1 0 0 1 1 0 1 1 1
value
Bit Default &
Description
Range Access
0937h
15: 0 Device ID (value): PCI Device ID
RO
15.5.3 Command Register (COMMAND_REGISTER)—Offset 4h

Access Method
Type: PCI Configuration Register COMMAND_REGISTER: [B:0, D:20, F:6] + 4h
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
IntrDis
RSVD
RSVD
MasEn
MEMen
RSVD
SERREn
Bit Default &

Description
Range Access
0h
15: 11 RSVD0 (RSVD0): Reserved
RO
0b Interrupt Disable (IntrDis): Interrupt disable. Disables generation of interrupt

10
RW messages in the PCI Express function. 1 =) disabled, 0 =) not disabled
0h
RO
0b SERR Enable (SERREn): When set, this bit enables the non-fatal and fatal errors
8
RW detected by the function to be reported to the root complex.
00h
7: 3 Reserved (RSVD): Reserved.
RO
0b Bus Master Enable (MasEn): 0=)disables upstream requests 1=)enables upstream

2
RW requests.

Bit Default &

Description
Range Access
0b Memory Space Enable (MEMen): Device support for Memory transactions. 0 =) not
1
RW supported. 1 =) supported.
0h
RO
15.5.4 Status Register (STATUS)—Offset 6h

Access Method
(Size: 16 bits) STATUS: [B:0, D:20, F:6] + 6h
Default: 0010h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
FastB2B
hasCapList
IntrStatus
SigSysErr
RcdMasAb
RSVD
DEVSEL
RSVD
RSVD
capable_66Mhz
RSVD0
RSVD1
Bit Default &
Description
Range Access
0h
15 RSVD0 (RSVD0): Reserved
RO
0b Signaled System Error (SigSysErr): Set when a function detects a system error and
14
RW the SERR Enable bit is set
0b Received master abort (RcdMasAb): Set when requester receives a completion with
13
RW Unsupported Request completion status
0h
RO
0b
10: 9 DEVSEL Timing (DEVSEL): Deprecated: Hardwired to 0
RO
0h
RO
0b
7 Fast Back-to-Back Capable (FastB2B): Deprecated: Hardwired to 0
RO
0h
RO
0b
5 66MHz-Capable (capable_66Mhz): Deprecated: Hardwired to 0
RO
1h Capabilities List (hasCapList): Indicates the presence of one or more capability

4
RO register sets.
0b Interrupt Status (IntrStatus): Indicates that the function has a legacy interrupt
3 request outstanding. This bit has no meaning if Message Signaled Interrupts are being
RO used
0h
RO

15.5.5 Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h

Access Method
Type: PCI Configuration Register REV_ID_CLASS_CODE: [B:0, D:20, F:6] + 8h
(Size: 32 bits)
Default: 02000010h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
rev_id
classCode
subClassCode
progIntf
Bit Default &
Description
Range Access
02h Class Code (classCode): Broadly classifies the type of function that the device
31: 24
RO performs.
00h Sub-Class Code (subClassCode): Identifies more specifically (than the class_code
23: 16
RO byte) the function of the device.
00h Programming Interface (progIntf): Used to define the register set variation within a
15: 8
RO particular sub-class.
10h Revision ID (rev_id): Assigned by the function manufacturer and identifies the
7: 0
RO revision number of the function.
15.5.6 Cache Line Size (CACHE_LINE_SIZE)—Offset Ch

Access Method
Type: PCI Configuration Register CACHE_LINE_SIZE: [B:0, D:20, F:6] + Ch
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h Cache Line Size (value): Implemented as a R/W register for legacy purposes but has
7: 0
RW no effect on device functionality.
15.5.7 Latency Timer (LATENCY_TIMER)—Offset Dh

Access Method
Type: PCI Configuration Register LATENCY_TIMER: [B:0, D:20, F:6] + Dh
(Size: 8 bits)
Default: 00h

7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
7: 0 Latency Timer (value): Deprecated. Hardwire to 0.
RO
15.5.8 Header Type (HEADER_TYPE)—Offset Eh

Access Method
Type: PCI Configuration Register HEADER_TYPE: [B:0, D:20, F:6] + Eh
(Size: 8 bits)
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
cfgHdrFormat
multiFnDev
Bit Default &

Description
Range Access
1h Multi-Function Device (multiFnDev): Hard-wired to 1 to indicate that this is a multi-

7
RO function device
0h Configuration Header Format (cfgHdrFormat): Hard-wired to 0 to indicate that this

6: 0
RO configuration header is a Type 0 header, i.e. it is an endpoint rather than a bridge.
15.5.9 BIST (BIST)—Offset Fh

Access Method
Type: PCI Configuration Register BIST: [B:0, D:20, F:6] + Fh
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
BIST_capable
comp_code
start_bist
RSVD
Bit Default &

Description
Range Access
0h BIST_capable (BIST_capable): Hard-wired to 0. (Returns 1 if the function

7
RO implements a BIST)

Bit Default &

Description
Range Access
0h
6 Start (start_bist): Set to start the functions BIST if BIST is supported.
RO
0h
RO
0h Completion Code (comp_code): Completion code having run BIST if BIST is

3: 0
RO supported. 0=)success. non-zero=)failure
15.5.10 Base Address Register (BAR0)—Offset 10h

Access Method
(Size: 32 bits) BAR0: [B:0, D:20, F:6] + 10h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD
prefetchable
memType
isIO
Bit Default &
Description
Range Access
0h address (address): Used to determine the size of memory required by the device and
31: 12
RW to assign a start address for this required amount of memory.
00h
RO
Prefetchable (prefetchable): Defines the block of memory as prefetchable or not. A

0b block of memory is prefetchable if it fulfils the following 3 conditions (1) no side effects
3 on reads, (2) the device returns all bytes on reads regardless of the byte enables, and
RO (3) host bridges can merge processor writes into this range without causing errors.
Hardwired to 0
00b
2: 1 Type (memType): Hardwired to 0 to indicate a 32-bit decoder
RO
0b Memory Space Indicator (isIO): Hardwired to 0 to indicate the register is a memory

0
RO address decoder
15.5.11 Cardbus CIS Pointer (CARDBUS_CIS_POINTER)—Offset 28h

Access Method
Type: PCI Configuration Register CARDBUS_CIS_POINTER: [B:0, D:20, F:6] + 28h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
31: 0 Cardbus CIS Pointer (value): Reserved. Hardwire to 0.
RO
15.5.12 Subsystem Vendor ID (SUB_SYS_VENDOR_ID)—Offset 2Ch

Access Method
(Size: 16 bits) SUB_SYS_VENDOR_ID: [B:0, D:20, F:6] + 2Ch
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
15: 0 Subsystem Vendor ID (value): PCI Subsystem Vendor ID
RO
15.5.13 Subsystem ID (SUB_SYS_ID)—Offset 2Eh

Access Method
Type: PCI Configuration Register SUB_SYS_ID: [B:0, D:20, F:6] + 2Eh
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
15: 0 Subsystem ID (value): PCI Subsystem ID
RO
15.5.14 Expansion ROM Base Address (EXP_ROM_BASE_ADR)—Offset

30h
Access Method
(Size: 32 bits) EXP_ROM_BASE_ADR: [B:0, D:20, F:6] + 30h
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
AddrDecodeEn
ROM_base_addr
Bit Default &
Description
Range Access
0h ROM Start Address (ROM_base_addr): Used to determine the size of memory

31: 11
RW required by the ROM and to assign a start address for this required amount of memory.
000h
RO
0h Address Decode Enable (AddrDecodeEn): A 1 in this field enables the function's

0 ROM address decoder assuming that the Memory Space bit in the Command Register is
RW also set to 1
15.5.15 Capabilities Pointer (CAP_POINTER)—Offset 34h

Access Method
Type: PCI Configuration Register CAP_POINTER: [B:0, D:20, F:6] + 34h
(Size: 32 bits)
Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
value
RSVD0
Bit Default &

Description
Range Access
0h
RO
80h Capabilities Pointer (value): Pointer to memory location of first entry of linked list of
7: 0 configuration register sets each of which supports a feature. Points to PM (power
RO management) register set at location 0x80
15.5.16 Interrupt Line Register (INTR_LINE)—Offset 3Ch

Access Method
Type: PCI Configuration Register INTR_LINE: [B:0, D:20, F:6] + 3Ch
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
Interrupt Line Register (value): The value in this register tells which input of the
0h system interrupt controller(s) the device's interrupt pin is connected to. The device itself
7: 0 does not use this value, rather it is used by device drivers and operating systems.
RW Device drivers and operating systems can use this information to determine priority and
vector information.
15.5.17 Interrupt Pin Register (INTR_PIN)—Offset 3Dh

Access Method
Type: PCI Configuration Register INTR_PIN: [B:0, D:20, F:6] + 3Dh
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
Interrupt Pin Register (value): The Interrupt Pin register tells which interrupt pin the
device (or device function) uses. A value of 1 corresponds to INTA#. A value of 2
03h corresponds to INTB#. A value of 3 corresponds to INTC#. A value of 4 corresponds to
7: 0
RO INTD#. Devices (or device functions) that do not use an interrupt pin must put a 0 in
this register. The values 05h through FFh are reserved. For this system function 0 is
connected to INTA, 1 to INTB, 2 to INTC 3 to INTD, 4 to INTA, 5 to INTB etc.
15.5.18 MIN_GNT (MIN_GNT)—Offset 3Eh

Access Method
(Size: 8 bits) MIN_GNT: [B:0, D:20, F:6] + 3Eh
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
7: 0 MIN_GNT (value): Hardwired to 0
RO
15.5.19 MAX_LAT (MAX_LAT)—Offset 3Fh

Access Method
Type: PCI Configuration Register MAX_LAT: [B:0, D:20, F:6] + 3Fh
(Size: 8 bits)
Default: 00h

7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
7: 0 MAX_LAT (value): Hardwired to 0
RO
15.5.20 Capability ID (PM_CAP_ID)—Offset 80h

Access Method
(Size: 8 bits) PM_CAP_ID: [B:0, D:20, F:6] + 80h
Default: 01h
7 4 0
0 0 0 0 0 0 0 1
value
Bit Default &

Description
Range Access
01h Capability ID (value): Identifies the feature associated with this register set.
7: 0
RO Hardwired value as per PCI SIG assigned capability ID
15.5.21 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h

Access Method
Type: PCI Configuration Register PM_NXT_CAP_PTR: [B:0, D:20, F:6] + 81h
(Size: 8 bits)
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
a0h Next Capability Pointer (value): Pointer to the next register set of feature specific
7: 0
RO configuration registers. Hardwired to 0xA0 to point to the MSI Capability Structure
15.5.22 Power Management Capabilities (PMC)—Offset 82h

Access Method
(Size: 16 bits) PMC: [B:0, D:20, F:6] + 82h

Default: 4803h
15 12 8 4 0
0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1
PME_support
D2_support
D1_support
aux_curr
RSVD
PME_clock
version
DSI
Bit Default &
Description
Range Access
PME Support (PME_support): PME_Support field Indicates the PM states within which
09h the function is capable of sending a PME (Power Management Event) message. 0 in a bit
15: 11
RO =) PME is not supported in the corresponding PM state, where bit indexes
11,12,13,14,15 correspond to PM states D0, D1, D2, D3hot, D3cold respectively.
0h
10 D2 Support (D2_support): Hardwired to 0 as the D2 state is not supported
RO
0h
RO
0h
8: 6 Aux Current (aux_curr): Hardwired to 0 as the D3hot state is not supported
RO
0h Device Specific Initialisation (DSI): Hardwired to 0 to indicate that the device does
5 not require a device specific initialisation sequence following transition to the D0
RO uninitialised state
0h
RO
0h
3 PME Clock (PME_clock): Deprecated. Hardwired to 0
RO
011b Version (version): This function complies with revision 1.2 of the PCI Power
2: 0
RO Management Interface Specification
15.5.23 Power Management Control/Status Register (PMCSR)—Offset

84h
Access Method
(Size: 16 bits) PMCSR: [B:0, D:20, F:6] + 84h
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
PME_status
no_soft_reset
Data_select
PME_en
RSVD
RSVD
Data_scale
power_state
Bit Default &

Description
Range Access
0h PME Status (PME_status): Set if function has experienced a PME (even if PME_en (bit
15
RW 8 of PMCSR register) is not set).
0h
14: 13 Data Scale (Data_scale): Hardwired to 0 as the data register is not supported
RO

Bit Default &

Description
Range Access
0h
12: 9 Data Select (Data_select): Hardwired to 0 as the data register is not supported
RO
0b PME Enable (PME_en): Enable device function to send PME messages when an event
8
RW occurs. 1=)enabled. 0=)disabled
0h
RO
1b No Soft Reset (no_soft_reset): Devices do perform an internal reset when

3
RO transitioning from D3hot to D0
0h
RO
00b Power State (power_state): Allows software to read current PM state or transition
1: 0
RW device to a new PM state, where 2'b00 = D0, 2'b01=D1, 2'b10=D2, 2'b11=D3hot
15.5.24 PM CSR PCI-to-PCI Bridge Support Extension (PMCSR_BSE)—

Offset 86h
Access Method
(Size: 8 bits) PMCSR_BSE: [B:0, D:20, F:6] + 86h
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h PM CSR PCI-to-PCI Bridge Support Extension (value): Not Supported. Hardwired

7: 0
RO to 0.
15.5.25 Power Management Data Register (DATA_REGISTER)—Offset

87h
Access Method
Type: PCI Configuration Register DATA_REGISTER: [B:0, D:20, F:6] + 87h
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
7: 0 Power Management Data Register (value): Not Supported. Hardwired to 0
RO

15.5.26 Capability ID (MSI_CAP_ID)—Offset A0h

Access Method
Type: PCI Configuration Register MSI_CAP_ID: [B:0, D:20, F:6] + A0h
(Size: 8 bits)
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value
Bit Default &
Description
Range Access
7: 0
15.5.27 Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h

Access Method
Type: PCI Configuration Register MSI_NXT_CAP_PTR: [B:0, D:20, F:6] + A1h
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
00h Next Capability Pointer (value): Hardwired to 0 as this is the last capability structure
7: 0
RO in the chain
15.5.28 Message Control (MESSAGE_CTRL)—Offset A2h

Access Method
Type: PCI Configuration Register MESSAGE_CTRL: [B:0, D:20, F:6] + A2h
(Size: 16 bits)
Default: 0100h
15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
perVecMskCap
bit64Cap
multiMsgCap
MSIEnable
RSVD0
multiMsgEn

Bit Default &

Description
Range Access
0h
RO
1h Per Vector Masking Capable (perVecMskCap): Hardwired to 1 to indicate the

8
RO function supports PVM
0h 64 bit Address Capabale (bit64Cap): This bit is hardwired to 0 to indicate that the
7
RO function is not capable of sending a 64-bit message address.
0h Multi-Message Enable (multiMsgEn): As only one vector is supported per function,

6: 4
RW software should only write a value of 0x0 to this field
0h Multiple Message Enable (multiMsgCap): This field is hardwired to 0x0 to indicate

3: 1
RO that the function is requesting a single vector
0h MSI Enable (MSIEnable): Set to enable MSI to request service. If set then it's
0 prohibited to use the INTx pin. System configuration software sets this bit to enable
RW MSI.
15.5.29 Message Address (MESSAGE_ADDR)—Offset A4h

Access Method
(Size: 32 bits) MESSAGE_ADDR: [B:0, D:20, F:6] + A4h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD0
Bit Default &
Description
Range Access
Message Address (address): If the Message Enable bit (bit 0 of the Message Control
0h register) is set, the contents of this register specify the DWORD-aligned address
31: 2
RW (AD[31:2]) for the MSI memory write transaction. AD[1:0] are driven to zero during the
address phase. This field is read/write
0h
RO
15.5.30 Message Data (MESSAGE_DATA)—Offset A8h

Access Method
(Size: 16 bits) MESSAGE_DATA: [B:0, D:20, F:6] + A8h
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData

Bit Default &

Description
Range Access
Data Field (MsgData): System-specified message data. If the Message Enable bit (bit
0 of the Message Control register) is set, the message data is driven onto the lower
0h word (AD[15:0]) of the memory write transactions data phase. AD[31:16] are driven to
15: 0
RW zero during the memory write transactions data phase. C/BE[3::0]# are asserted during
the data phase of the memory write transaction. None of the message bits will be
changed by hardware
15.5.31 Mask Bits for MSI (PER_VEC_MASK)—Offset ACh

Access Method
(Size: 32 bits) PER_VEC_MASK: [B:0, D:20, F:6] + ACh
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
MSIMask
Bit Default &
Description
Range Access
0h
RO
0h Vector 0 Mask (MSIMask): Mask Bit for Vector 0. If this bit is set, the function will not
0
RW send MSI messages
15.5.32 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h

Access Method
(Size: 32 bits) PER_VEC_PEND: [B:0, D:20, F:6] + B0h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
RSVD0
Bit Default &

Description
Range Access
0h
RO
0h
0 Vector 0 Pending (value): Pending Bit for Vector 0.
RO

15.6 Memory Mapped Registers
Table 88. Summary of Memory Mapped I/O Registers—BAR0

Offset Default
Start Value
0h 3h “MAC Configuration Register (Register 0) (GMAC_REG_0)—Offset 0h” on page 331 00008000h
4h 7h “MAC Frame Filter (Register 1) (GMAC_REG_1)—Offset 4h” on page 334 00000000h
8h Bh “Hash Table High Register (Register 2) (GMAC_REG_2)—Offset 8h” on page 336 00000000h
Ch Fh “Hash Table Low Register (Register 3) (GMAC_REG_3)—Offset Ch” on page 336 00000000h
10h 13h “GMII Address Register (Register 4) (GMAC_REG_4)—Offset 10h” on page 337 00000000h
14h 17h “GMII Data Register (Register 5) (GMAC_REG_5)—Offset 14h” on page 338 00000000h
18h 1Bh “Flow Control Register (Register 6) (GMAC_REG_6)—Offset 18h” on page 339 00000000h
1Ch 1Fh “VLAN Tag Register (Register 7) (GMAC_REG_7)—Offset 1Ch” on page 340 00000000h
20h 23h “Version Register (Register 8) (GMAC_REG_8)—Offset 20h” on page 341 00001037h
24h 27h “Debug Register (Register 9) (GMAC_REG_9)—Offset 24h” on page 342 00000000h
38h 3Bh “Interrupt Register (Register 14) (GMAC_REG_14)—Offset 38h” on page 343 00000000h
3Ch 3Fh “Interrupt Mask Register (Register 15) (GMAC_REG_15)—Offset 3Ch” on page 344 00000000h
40h 43h “MAC Address0 High Register (Register 16) (GMAC_REG_16)—Offset 40h” on page 345 8000FFFFh
44h 47h “MAC Address0 Low Register (Register 17) (GMAC_REG_17)—Offset 44h” on page 345 FFFFFFFFh
100h 103h “MMC Control Register (Register 64) (GMAC_REG_64)—Offset 100h” on page 346 00000000h
104h 107h “MMC Receive Interrupt Register (MMC_INTR_RX)—Offset 104h” on page 347 00000000h
108h 10Bh “MMC Transmit Interrupt Register (MMC_INTR_TX)—Offset 108h” on page 349 00000000h
“MMC Receive Interrupt Mask Register (MMC_INTR_MASK_RX)—Offset 10Ch” on

10Ch 10Fh 00000000h
page 351
“MMC Transmit Interrupt Mask Register (MMC_INTR_MASK_TX)—Offset 110h” on

110h 113h 00000000h
page 353
“MMC Transmit Good Bad Octet Counter Register (TXOCTETCOUNT_GB)—Offset 114h”

114h 117h 00000000h
on page 355
“MMC Transmit Good Bad Frame Counter Register (TXFRAMECOUNT_GB)—Offset 118h”

118h 11Bh 00000000h
on page 355
“MMC Transmit Broadcast Good Frame Counter Register (TXBROADCASTFRAMES_G)—

11Ch 11Fh 00000000h
Offset 11Ch” on page 356
“MMC Transmit Multicast Good Frame Counter Register (TXMULTICASTFRAMES_G)—

120h 123h 00000000h
Offset 120h” on page 356
“MMC Transmit 64 Octet Good Bad Frame Counter Register (TX64OCTETS_GB)—Offset

124h 127h 00000000h
124h” on page 357
“MMC Transmit 65 to 127 Octet Good Bad Frame Counter Register

128h 12Bh 00000000h
(TX65TO127OCTETS_GB)—Offset 128h” on page 357
12Ch 12Fh 00000000h
(TX128TO255OCTETS_GB)—Offset 12Ch” on page 358
130h 133h 00000000h
134h 137h 00000000h
“MMC Transmit 1024 to Maximum Octet Good Bad Frame Counter Register
138h 13Bh 00000000h
(TX1024TOMAXOCTETS_GB)—Offset 138h” on page 359
“MMC Transmit Unicast Good Bad Frame Counter Register (TXUNICASTFRAMES_GB)—

13Ch 13Fh 00000000h

Table 88. Summary of Memory Mapped I/O Registers—BAR0 (Continued)

Offset Default
Start Value
“MMC Transmit Multicast Good Bad Frame Counter Register

140h 143h 00000000h
(TXMULTICASTFRAMES_GB)—Offset 140h” on page 360
“MMC Transmit Broadcast Good Bad Frame Counter Register

144h 147h 00000000h
(TXBROADCASTFRAMES_GB)—Offset 144h” on page 360
“MMC Transmit Underflow Error Frame Counter Register (TXUNDERFLOWERROR)—

148h 14Bh 00000000h
“MMC Transmit Single Collision Good Frame Counter Register (TXSINGLECOL_G)—

14Ch 14Fh 00000000h
“MMC Transmit Multiple Collision Good Frame Counter Register (TXMULTICOL_G)—

150h 153h 00000000h
“MMC Transmit Deferred Frame Counter Register (TXDEFERRED)—Offset 154h” on

154h 157h 00000000h
page 362
“MMC Transmit Late Collision Frame Counter Register (TXLATECOL)—Offset 158h” on

158h 15Bh 00000000h
page 362
“MMC Transmit Excessive Collision Frame Counter Register (TXEXESSCOL)—Offset

15Ch 15Fh 00000000h
15Ch” on page 363
“MMC Transmit Carrier Error Frame Counter Register (TXCARRIERERROR)—Offset 160h”

160h 163h 00000000h
on page 363
“MMC Transmit Good Bad Octet Counter Register (TXOCTETCOUNT_GB)—Offset 114h”

164h 167h 00000000h
on page 355
“MMC Transmit Good Bad Frame Counter Register (TXFRAMECOUNT_GB)—Offset 118h”

168h 16Bh 00000000h
on page 355
“MMC Transmit Excessive Deferral Frame Counter Register (TXEXCESSDEF)—Offset

16Ch 16Fh 00000000h
16Ch” on page 365
“MMC Transmit Pause Frame Counter Register (TXPAUSEFRAMES)—Offset 170h” on

170h 173h 00000000h
page 365
“MMC Transmit VLAN Good Frame Counter Register (TXVLANFRAMES_G)—Offset 174h”

174h 177h 00000000h
on page 366
“MMC Transmit Oversize Good Frame Counter Register (TXOVERSIZE_G)—Offset 178h”

178h 17Bh 00000000h
on page 366
“MMC Receive Good Bad Frame Counter Register (RXFRAMECOUNT_GB)—Offset 180h”

180h 183h 00000000h
on page 366
“MMC Receive Good Bad Octet Counter Register (RXOCTETCOUNT_GB)—Offset 184h”

184h 187h 00000000h
on page 367
“MMC Receive Good Bad Octet Counter Register (RXOCTETCOUNT_GB)—Offset 184h”

188h 18Bh 00000000h
on page 367
“MMC Receive Broadcast Good Frame Counter Register (RXBROADCASTFRAMES_G)—

18Ch 18Fh 00000000h
“MMC Receive Multicast Good Frame Counter Register (RXMULTICASTFRAMES_G)—

190h 193h 00000000h
“MMC Receive CRC Error Frame Counter Register (RXCRCERROR)—Offset 194h” on

194h 197h 00000000h
page 369
“MMC Receive Alignment Error Frame Counter Register (RXALIGNMENTERROR)—Offset

198h 19Bh 00000000h
198h” on page 369
“MMC Receive Runt Frame Counter Register (RXRUNTERROR)—Offset 19Ch” on

19Ch 19Fh 00000000h
page 370
“MMC Receive Jabber Error Frame Counter Register (RXJABBERERROR)—Offset 1A0h”

1A0h 1A3h 00000000h
on page 370
“MMC Receive Undersize Good Frame Counter Register (RXUNDERSIZE_G)—Offset

1A4h 1A7h 00000000h
1A4h” on page 370


Offset Default
Start Value
“MMC Receive Oversize Good Frame Counter Register (RXOVERSIZE_G)—Offset 1A8h”

1A8h 1ABh 00000000h
on page 371
“MMC Receive 64 Octet Good Bad Frame Counter Register (RX64OCTETS_GB)—Offset

1ACh 1AFh 00000000h
1ACh” on page 371
“MMC Receive 65 to 127 Octet Good Bad Frame Counter Register

1B0h 1B3h 00000000h
(RX65TO127OCTETS_GB)—Offset 1B0h” on page 372
1B4h 1B7h 00000000h
1B8h 1BBh 00000000h
1BCh 1BFh 00000000h
(RX512TO1023OCTETS_GB)—Offset 1BCh” on page 373
“MMC Receive 1024 to Maximum Octet Good Bad Frame Counter Register
1C0h 1C3h 00000000h
(RX1024TOMAXOCTETS_GB)—Offset 1C0h” on page 374
“MMC Receive Unicast Good Frame Counter Register (RXUNICASTFRAMES_G)—Offset

1C4h 1C7h 00000000h
1C4h” on page 374
“MMC Receive Length Error Frame Counter Register (RXLENGTHERROR)—Offset 1C8h”

1C8h 1CBh 00000000h
on page 374
“MMC Receive Out Of Range Error Frame Counter Register (RXOUTOFRANGETYPE)—

1CCh 1CFh 00000000h
Offset 1CCh” on page 375
“MMC Receive Pause Frame Counter Register (RXPAUSEFRAMES)—Offset 1D0h” on

1D0h 1D3h 00000000h
page 375
“MMC Receive FIFO Overflow Frame Counter Register (RXFIFOOVERFLOW)—Offset

1D4h 1D7h 00000000h
1D4h” on page 376
“MMC Receive VLAN Good Bad Frame Counter Register (RXVLANFRAMES_GB)—Offset

1D8h 1DBh 00000000h
1D8h” on page 376
“MMC Receive Watchdog Error Frame Counter Register (RXWATCHDOGERROR)—Offset

1DCh 1DFh 00000000h
1DCh” on page 377
1E0h 1E3h “MMC Receive Error Frame Counter Register (RXRCVERROR)—Offset 1E0h” on page 377 00000000h
“MMC Receive Control Frame Counter Register (RXCTRLFRAMES_G)—Offset 1E4h” on

1E4h 1E7h 00000000h
page 378
“MMC IPC Receive Checksum Offload Interrupt Mask Register

200h 203h 00000000h
(MMC_IPC_INTR_MASK_RX)—Offset 200h” on page 378
“MMC Receive Checksum Offload Interrupt Register (MMC_IPC_INTR_RX)—Offset 208h”

208h 20Bh 00000000h
on page 380
“MMC Receive IPV4 Good Frame Counter Register (RXIPV4_GD_FRMS)—Offset 210h”

210h 213h 00000000h
on page 382
“MMC Receive IPV4 Header Error Frame Counter Register (RXIPV4_HDRERR_FRMS)—

214h 217h 00000000h
“MMC Receive IPV4 No Payload Frame Counter Register (RXIPV4_NOPAY_FRMS)—Offset

218h 21Bh 00000000h
218h” on page 383
“MMC Receive IPV4 Fragmented Frame Counter Register (RXIPV4_FRAG_FRMS)—Offset

21Ch 21Fh 00000000h
21Ch” on page 384
“MMC Receive IPV4 UDP Checksum Disabled Frame Counter Register

220h 223h 00000000h
(RXIPV4_UDSBL_FRMS)—Offset 220h” on page 384
“MMC Receive IPV6 Good Frame Counter Register (RXIPV6_GD_FRMS)—Offset 224h”

224h 227h 00000000h
on page 384
“MMC Receive IPV6 Header Error Frame Counter Register (RXIPV6_HDRERR_FRMS)—

228h 22Bh 00000000h
“MMC Receive IPV6 No Payload Frame Counter Register (RXIPV6_NOPAY_FRMS)—Offset

22Ch 22Fh 00000000h
22Ch” on page 385


Offset Default
Start Value
“MMC Receive UDP Good Frame Counter Register (RXUDP_GD_FRMS)—Offset 230h” on

230h 233h 00000000h
page 386
“MMC Receive UDP Error Frame Counter Register (RXUDP_ERR_FRMS)—Offset 234h” on

234h 237h 00000000h
page 386
“MMC Receive TCP Good Frame Counter Register (RXTCP_GD_FRMS)—Offset 238h” on

238h 23Bh 00000000h
page 387
“MMC Receive TCP Error Frame Counter Register (RXTCP_ERR_FRMS)—Offset 23Ch” on

23Ch 23Fh 00000000h
page 387
“MMC Receive ICMP Good Frame Counter Register (RXICMP_GD_FRMS)—Offset 240h”

240h 243h 00000000h
on page 388
“MMC Receive ICMP Error Frame Counter Register (RXICMP_ERR_FRMS)—Offset 244h”

244h 247h 00000000h
on page 388
“MMC Receive IPV4 Good Octet Counter Register (RXIPV4_GD_OCTETS)—Offset 250h”

250h 253h 00000000h
on page 388
“MMC Receive IPV4 Header Error Octet Counter Register (RXIPV4_HDRERR_OCTETS)—

254h 257h 00000000h
“MMC Receive IPV4 No Payload Octet Counter Register (RXIPV4_NOPAY_OCTETS)—

258h 25Bh 00000000h
“MMC Receive IPV4 Fragmented Octet Counter Register (RXIPV4_FRAG_OCTETS)—

25Ch 25Fh 00000000h
“MMC Receive IPV4 UDP Checksum Disabled Octet Counter Register

260h 263h 00000000h
(RXIPV4_UDSBL_OCTETS)—Offset 260h” on page 390
“MMC Receive IPV6 Good Octet Counter Register (RXIPV6_GD_OCTETS)—Offset 264h”

264h 267h 00000000h
on page 391
“MMC Receive IPV6 Good Octet Counter Register (RXIPV6_HDRERR_OCTETS)—Offset

268h 26Bh 00000000h
268h” on page 391
“MMC Receive IPV6 Header Error Octet Counter Register (RXIPV6_NOPAY_OCTETS)—

26Ch 26Fh 00000000h
“MMC Receive IPV6 No Payload Octet Counter Register (RXUDP_GD_OCTETS)—Offset

270h 273h 00000000h
270h” on page 392
“MMC Receive UDP Good Octet Counter Register (RXUDP_ERR_OCTETS)—Offset 274h”

274h 277h 00000000h
on page 392
“MMC Receive TCP Good Octet Counter Register (RXTCP_GD_OCTETS)—Offset 278h” on

278h 27Bh 00000000h
page 393
“MMC Receive TCP Error Octet Counter Register (RXTCP_ERR_OCTETS)—Offset 27Ch”

27Ch 27Fh 00000000h
on page 393
“MMC Receive ICMP Good Octet Counter Register (RXICMP_GD_OCTETS)—Offset 280h”

280h 283h 00000000h
on page 394
“MMC Receive ICMP Error Octet Counter Register (RXICMP_ERR_OCTETS)—Offset

284h 287h 00000000h
284h” on page 394
“VLAN Tag Inclusion or Replacement Register (Register 353) (GMAC_REG_353)—Offset

584h 587h 00000000h
584h” on page 395
“VLAN Hash Table Register (Register 354) (GMAC_REG_354)—Offset 588h” on

588h 58Bh 00000000h
page 396
“Timestamp Control Register (Register 448) (GMAC_REG_448)—Offset 700h” on

700h 703h 00002000h
page 396
“Sub-Second Increment Register (Register 449) (GMAC_REG_449)—Offset 704h” on

704h 707h 00000000h
page 398
“System Time - Seconds Register (Register 450) (GMAC_REG_450)—Offset 708h” on

708h 70Bh 00000000h
page 398


Offset Default
Start Value
“System Time - Nanoseconds Register (Register 451) (GMAC_REG_451)—Offset 70Ch”

70Ch 70Fh 00000000h
on page 399
“System Time - Seconds Update Register (Register 452) (GMAC_REG_452)—Offset

710h 713h 00000000h
710h” on page 399
“System Time - Nanoseconds Update Register (Register 453) (GMAC_REG_453)—Offset

714h 717h 00000000h
714h” on page 400
“Timestamp Addend Register (Register 454) (GMAC_REG_454)—Offset 718h” on

718h 71Bh 00000000h
page 400
“Target Time Seconds Register (Register 455) (GMAC_REG_455)—Offset 71Ch” on

71Ch 71Fh 00000000h
page 401
“Target Time Nanoseconds Register (Register 456) (GMAC_REG_456)—Offset 720h” on

720h 723h 00000000h
page 401
“System Time - Higher Word Seconds Register (Register 457) (GMAC_REG_457)—

724h 727h 00000000h
“Timestamp Status Register (Register 458) (GMAC_REG_458)—Offset 728h” on

728h 72Bh 00000000h
page 403
1000h 1003h “Bus Mode Register (Register 0) (DMA_REG_0)—Offset 1000h” on page 404 00020101h
1004h 1007h “Transmit Poll Demand Register (Register 1) (DMA_REG_1)—Offset 1004h” on page 406 00000000h
1008h 100Bh “Receive Poll Demand Register (Register 2) (DMA_REG_2)—Offset 1008h” on page 406 00000000h
“Receive Descriptor List Address Register (Register 3) (DMA_REG_3)—Offset 100Ch” on

100Ch 100Fh 00000000h
page 407
“Transmit Descriptor List Address Register (Register 4) (DMA_REG_4)—Offset 1010h”

1010h 1013h 00000000h
on page 407
1014h 1017h “Status Register (Register 5) (DMA_REG_5)—Offset 1014h” on page 408 00000000h
1018h 101Bh “Operation Mode Register (Register 6) (DMA_REG_6)—Offset 1018h” on page 411 00000000h
101Ch 101Fh “Interrupt Enable Register (Register 7) (DMA_REG_7)—Offset 101Ch” on page 414 00000000h
“Missed Frame and Buffer Overflow Counter Register (Register 8) (DMA_REG_8)—Offset

1020h 1023h 00000000h
1020h” on page 415
“Receive Interrupt Watchdog Timer Register (Register 9) (DMA_REG_9)—Offset 1024h”

1024h 1027h 00000000h
on page 416
102Ch 102Fh “AHB Status Register (Register 11) (DMA_REG_11)—Offset 102Ch” on page 416 00000000h
“Current Host Transmit Descriptor Register (Register 18) (DMA_REG_18)—Offset

1048h 104Bh 00000000h
1048h” on page 417
“Current Host Receive Descriptor Register (Register 19) (DMA_REG_19)—Offset 104Ch”

104Ch 104Fh 00000000h
on page 417
“Current Host Transmit Buffer Address Register (Register 20) (DMA_REG_20)—Offset

1050h 1053h 00000000h
1050h” on page 418
“Current Host Receive Buffer Address Register (Register 21) (DMA_REG_21)—Offset

1054h 1057h 00000000h
1054h” on page 418
1058h 105Bh “HW Feature Register (Register 22) (DMA_REG_22)—Offset 1058h” on page 419 4B0F3915h
15.6.1 MAC Configuration Register (Register 0) (GMAC_REG_0)—

Offset 0h
The MAC Configuration register establishes receive and transmit operating modes.
Access Method

Type: Memory Mapped I/O Register Offset: [BAR0] + 0h

(Size: 32 bits)
BAR0 Type: PCI Configuration Register (Size: 32 bits)

BAR0 Reference: [B:0, D:20, F:6] + 10h
Default: 00008000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV3
SARC
RSV1
IFG
RSV0
DC
LM
DM
IPC
DR
BL
PRELEN
PE2K
WD
JD
BE
JE
DCRS
FES
DO
LUD
ACS
TE
RE
Bit Default &
Range Access
0b
RO
Source Address Insertion or Replacement Control (SARC): This field controls the
source address insertion or replacement for all transmitted frames.
When Bit 30 is set to 0, the MAC replaces the content of the MAC Address 0 registers
(registers 16 and 17) in the SA field of all transmitted frames based on the values of Bits
[29:28]:
000b - 2b0x: SA insertion is controlled by the internal signal from the MTL layer.
30:28 - 2b10: if Bit 30 is set to 0, the MAC inserts the content of the MAC Address 0 registers
RW (registers 16 and 17) in the SA field of all transmitted frames.
- 2b11: if Bit 30 is set to 0, the MAC replaces the content of the MAC Address 0 registers
(registers 16 and 17) in the SA field of all transmitted frames.
NOTE:Changes to this field take effect only on the start of a frame. If you write this
register field when a frame is being transmitted, only the subsequent frame can use the
updated value, that is, the current frame does not use the updated value.
IEEE 802.3as support for 2K packets Enable (PE2K): When set, the MAC considers
all frames, with up to 2,000 bytes length, as normal packets. When Bit 20 (Jumbo
0b Enable) is not set, the MAC considers all received frames of size more than 2K bytes as
27 Giant frames. When this bit is reset and Bit 20 (Jumbo Enable) is not set, the MAC
RW considers all received frames of size more than 1,518 bytes (1,522 bytes for tagged) as
Giant frames. When Bit 20 (Jumbo Enable) is set, setting this bit has no effect on Giant
Frame status.
000b
RO
Watchdog Disable (WD): When this bit is set, the MAC disables the watchdog timer
0b on the receiver. The MAC can receive frames of up to 16,384 bytes.
23 When this bit is reset, the MAC does not allow more than 2,048 bytes (10,240 if JE is set
RW high) of the frame being received. The MAC cuts off any bytes received after 2,048
bytes.
Jabber Disable (JD): When this bit is set, the MAC disables the jabber timer on the
0b transmitter. The MAC can transfer frames of up to 16,384 bytes. When this bit is reset,
22
RW the MAC cuts off the transmitter if the application sends out more than 2,048 bytes of
data (10,240 if JE is set high) during transmission.
0b
21 Reserved (BE): Reserved.
RW
0b Jumbo Frame Enable (JE): When this bit is set, the MAC allows Jumbo frames of
20 9,018 bytes (9,022 bytes for VLAN tagged frames) without reporting a giant frame error
RW in the receive frame status.
Inter-Frame Gap (IFG): These bits control the minimum IFG between frames during
transmission.
000: 96 bit times
000b 001: 88 bit times
19:17
RW 010: 80 bit times
111: 40 bit times
In the half-duplex mode, the minimum IFG can be configured only for 64 bit times (IFG
= 100). Lower values are not considered.

Bit Default &

Range Access
Disable Carrier Sense During Transmission (DCRS): When set high, this bit makes
0b the MAC transmitter ignore the MII CRS signal during frame transmission in the half-
16 duplex mode. This request results in no errors generated because of Loss of Carrier or
RW No Carrier during such transmission. When this bit is low, the MAC transmitter
generates such errors because of Carrier Sense and can even abort the transmissions.
1b GMII/MII configuration (RSV0): This bit identifies the supported interface:

15 0: GMII (1000 Mbps)
RO 1: MII (10/100 Mbps)
0b RMII Speed (FES): This bit selects the speed in the RMII interface:
14 0: 10 Mbps
RW 1: 100 Mbps
Disable Receive Own (DO): When this bit is set, the MAC disables the reception of
0b frames when the gmii_txen_o is asserted in the half-duplex mode. When this bit is
13
RW reset, the MAC receives all packets that are given by the PHY while transmitting.
This bit is not applicable if the MAC is operating in the full-duplex mode.
0b Loopback Mode (LM): When this bit is set, the MAC operates in the loopback mode at
12
RW MII.
0b Duplex Mode (DM): When this bit is set, the MAC operates in the full-duplex mode
11
RW where it can transmit and receive simultaneously.
Checksum Offload (IPC): When this bit is set, the MAC calculates the 16-bit ones
complement of the ones complement sum of all received Ethernet frame payloads. It
also checks whether the IPv4 Header checksum (assumed to be bytes 2526 or 2930
(VLAN-tagged) of the received Ethernet frame) is correct for the received frame and
gives the status in the receive status word. The MAC also appends the 16-bit checksum
0b calculated for the IP header datagram payload (bytes after the IPv4 header) and
10 appends it to the Ethernet frame transferred to the application (when Type 2 COE is
RW deselected).
When this bit is reset, this function is disabled.
As Type 2 COE (Checksum Offload Engine) is supported, this bit, when set, enables the
IPv4 header checksum checking and IPv4 or IPv6 TCP, UDP, or ICMP payload checksum
checking. When this bit is reset, the COE function in the receiver is disabled and the
corresponding PCE and IP HCE status bits are always cleared.
Disable Retry (DR): When this bit is set, the MAC attempts only one transmission.
When a collision occurs on the MII interface, the MAC ignores the current frame
0b transmission and reports a Frame Abort with excessive collision error in the transmit
9
RW frame status.
When this bit is reset, the MAC attempts retries based on the settings of the BL field
(Bits [6:5]). This bit is applicable only in the half-duplex mode.
0b
8 Reserved (LUD): Reserved.
RO
Automatic Pad or CRC Stripping (ACS): When this bit is set, the MAC strips the Pad
or FCS (Frame Check Sequence) field on the incoming frames only if the value of the
0b length field is less than 1,536 bytes. All received frames with length field greater than or
7
RW equal to 1,536 bytes are passed to the application without stripping the Pad or FCS field.
When this bit is reset, the MAC passes all incoming frames, without modifying them, to
the Host.
Back-Off Limit (BL): The Back-Off limit determines the random integer number (r) of
slot time delays (512 bit times for 10/100 Mbps) for which the MAC waits before
rescheduling a transmission attempt during retries after a collision. This bit is applicable
only in the half-duplex mode.
00b 00: k = min (n, 10)
6:5
RW 01: k = min (n, 8)
10: k = min (n, 4)
11: k = min (n, 1)
where n = retransmission attempt. The random integer r takes the value in the range 0
(= r ( kth power of 2

Bit Default &

Range Access
Deferral Check (DC): When this bit is set, the deferral check function is enabled in the
MAC. The MAC issues a Frame Abort status, along with the excessive deferral error bit
set in the transmit frame status, when the transmit state machine is deferred for more
than 24,288 bit times in the 10 or 100 Mbps mode. If the Jumbo frame mode is enabled
0b in the 10 or 100 Mbps mode, the threshold for deferral is 155,680 bits times. Deferral
4 begins when the transmitter is ready to transmit, but is prevented because of an active
RW carrier sense signal (CRS) on MII. Defer time is not cumulative. When the transmitter
defers for 10,000 bit times, it transmits, collides, backs off, and then defers again after
completion of back-off. The deferral timer resets to 0 and restarts.
When this bit is reset, the deferral check function is disabled and the MAC defers until
the CRS signal goes inactive. This bit is applicable only in the half-duplex mode.
Transmitter Enable (TE): When this bit is set, the transmit state machine of the MAC
0b is enabled for transmission on the MII. When this bit is reset, the MAC transmit state
3
RW machine is disabled after the completion of the transmission of the current frame, and
does not transmit any further frames.
Receiver Enable (RE): When this bit is set, the receiver state machine of the MAC is
0b enabled for receiving frames from the MII. When this bit is reset, the MAC receive state
2
RW machine is disabled after the completion of the reception of the current frame, and does
not receive any further frames from the MII.
Preamble Length for Transmit Frames (PRELEN): These bits control the number of
preamble bytes that are added to the beginning of every Transmit frame. The preamble
00b reduction occurs only when the MAC is operating in the full-duplex mode.
1:0 2'b00: 7 bytes of preamble
RW 2'b01: 5 byte of preamble
2'b10: 3 bytes of preamble
2'b11: reserved
15.6.2 MAC Frame Filter (Register 1) (GMAC_REG_1)—Offset 4h

The MAC Frame Filter register contains the filter controls for receiving frames. Some of
the controls from this register go to the address check block of the MAC, which
performs the first level of address filtering. The second level of filtering is performed on
the incoming frame, based on other controls such as Pass Bad Frames and Pass Control
Frames.
Access Method
(Size: 32 bits) Offset: [BAR0] + 4h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RA
VTFE
RSV1
RSV0
SAF
SAIF
DBF
HMC
HUC
HPF
PCF
PM
DAIF
PR
Bit Default &

Range Access
Receive All (RA): When this bit is set, the MAC Receiver module passes all received
frames, irrespective of whether they pass the address filter or not, to the Application.
0b The result of the SA or DA filtering is updated (pass or fail) in the corresponding bits in
31
RW the Receive Status Word.
When this bit is reset, the Receiver module passes only those frames to the Application
that pass the SA or DA address filter.
0b
RO

Bit Default &

Range Access
VLAN Tag Filter Enable (VTFE): When set, this bit enables the MAC to drop VLAN
0b tagged frames that do not match the VLAN Tag comparison.
16
RW When reset, the MAC forwards all frames irrespective of the match status of the VLAN
Tag.
00000b
RO
Hash or Perfect Filter (HPF): When this bit is set, it configures the address filter to
0b pass a frame if it matches either the perfect filtering or the hash filtering as set by the
10 HMC or HUC bits.
RW When this bit is low and the HUC or HMC bit is set, the frame is passed only if it matches
the Hash filter.
Source Address Filter Enable (SAF): When this bit is set, the MAC compares the SA
field of the received frames with the values programmed in the enabled SA registers. If
0b the comparison matches, then the SA Match bit of RxStatus Word is set high. When this
9
RW bit is set high and the SA filter fails, the MAC drops the frame.
When this bit is reset, the MAC forwards the received frame to the application and with
the updated SA Match bit of the RxStatus depending on the SA address comparison.
SA Inverse Filtering (SAIF): When this bit is set, the Address Check block operates in
0b inverse filtering mode for the SA address comparison. The frames whose SA matches
8 the SA registers are marked as failing the SA Address filter.
RW When this bit is reset, frames whose SA does not match the SA registers are marked as
failing the SA Address filter.
Pass Control Frames (PCF): These bits control the forwarding of all control frames
(including unicast and multicast PAUSE frames).
00: MAC filters all control frames from reaching the application.
01: MAC forwards all control frames except PAUSE control frames to application even if
they fail the Address filter.
10: MAC forwards all control frames to application even if they fail the Address Filter.
11: MAC forwards control frames that pass the Address Filter.
The following conditions should be true for the PAUSE control frames processing:
Condition 1: The MAC is in the full-duplex mode and flow control is enabled by setting
Bit 2 (RFE) of Register 6 (Flow Control Register) to 1.
00b Condition 2: The destination address (DA) of the received frame matches the special
7:6
RW multicast address or the MAC Address 0 when Bit 3 (UP) of the Register 6 (Flow Control
Register) is set.
Condition 3: The Type field of the received frame is 0x8808 and the OPCODE field is
0x0001.
NOTE:
This field should be set to 01 only when the Condition 1 is true, that is, the MAC is
programmed to operate in the full-duplex mode and the RFE bit is enabled. Otherwise,
the PAUSE frame filtering may be inconsistent. When Condition 1 is false, the PAUSE
frames are considered as generic control frames. Therefore, to pass all control frames
(including PAUSE control frames) when the full-duplex mode and flow control is not
enabled, you should set the PCF field to 10 or 11 (as required by the application).
0b Disable Broadcast Frames (DBF): When this bit is set, the AFM module filters all
5 incoming broadcast frames. In addition, it overrides all other filter settings.
RW When this bit is reset, the AFM module passes all received broadcast frames.
0b Pass All Multicast (PM): When set, this bit indicates that all received frames with a
4 multicast destination address (first bit in the destination address field is '1') are passed.
RW When reset, filtering of multicast frame depends on HMC bit.
DA Inverse Filtering (DAIF): When this bit is set, the Address Check block operates
0b in inverse filtering mode for the DA address comparison for both unicast and multicast
3
RW frames.
When reset, normal filtering of frames is performed.
Hash Multicast (HMC): When set, MAC performs destination address filtering of
0b received multicast frames according to the hash table.
2
RW When reset, the MAC performs a perfect destination address filtering for multicast
frames, that is, it compares the DA field with the values programmed in DA registers.
Hash Unicast (HUC): When set, MAC performs destination address filtering of unicast
0b frames according to the hash table.
1
RW When reset, the MAC performs a perfect destination address filtering for unicast frames,
that is, it compares the DA field with the values programmed in DA registers.

Bit Default &

Range Access
0b Promiscuous Mode (PR): When this bit is set, the Address Filter module passes all
0 incoming frames regardless of its destination or source address. The SA or DA Filter Fails
RW status bits of the Receive Status Word are always cleared when PR is set.
15.6.3 Hash Table High Register (Register 2) (GMAC_REG_2)—Offset

8h
The Hash Table High register contains the higher 32 bits of the Hash table. The 64-bit
Hash table is used for group address filtering. For hash filtering, the contents of the
destination address in the incoming frame is passed through the CRC logic, and the
upper 6 bits of the CRC register are used to index the contents of the Hash table. The
most significant bit determines the register to be used (Hash Table High or Hash Table
Low), and the other 5 bits determine which bit within the register. A hash value of
5b'00000 selects Bit 0 of the selected register, and a value of 5b'11111 selects Bit 31 of
the selected register. The hash value of the destination address is calculated in the
following way: 1. Calculate the 32-bit CRC for the DA (See IEEE 802.3, Section 3.2.8
for the steps to calculate CRC32). 2. Perform bitwise reversal for the value obtained in
Step 1. 3. Take the upper 6 bits from the value obtained in Step 2. For example, if the
DA of the incoming frame is received as 0x1F52419CB6AF (0x1F is the first byte
received on MII interface), then the internally calculated 6-bit Hash value is 0x2C and
Bit 12 of Hash Table High register is checked for filtering. If the DA of the incoming
frame is received as 0xA00A98000045, then the calculated 6-bit Hash value is 0x07
and Bit 7 of Hash Table Low register is checked for filtering. If the corresponding bit
value of the register is 1'b1, the frame is accepted. Otherwise, it is rejected. If the PM
(Pass All Multicast) bit is set in Register 1, then all multicast frames are accepted
regardless of the multicast hash values.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
HTH
Bit Default &

Range Access
00000000h
31:0 Hash Table High (HTH): This field contains the upper 32 bits of the Hash table.
RW
15.6.4 Hash Table Low Register (Register 3) (GMAC_REG_3)—Offset

Ch
The Hash Table Low register contains the lower 32 bits of the Hash table.
Access Method

Type: Memory Mapped I/O Register Offset: [BAR0] + Ch

(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
HTL
Bit Default &
Range Access
00000000h
31:0 Hash Table Low (HTL): This field contains the lower 32 bits of the Hash table.
RW
15.6.5 GMII Address Register (Register 4) (GMAC_REG_4)—Offset 10h

The GMII Address register controls the management cycles to the external PHY through
the management interface.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PA
RSV0
GR
CR
GW
GB
Bit Default &
Range Access
0000h
RO
00000b Physical Layer Address (PA): This field indicates which of the 32 possible PHY
15:11
RW devices are being accessed.
00000b GMII Register (GR): These bits select the desired GMII register in the selected PHY
10:6
RW device.

Bit Default &

Range Access
CSR Clock Range (CR): The CSR Clock Range selection determines the frequency of
the serial management clock (MDC) according to the system clock (clk_csr_i) frequency
used in your design, which is 133MHz. When Bit[5] = 0 allowed values are:
- 0001: The frequency of the clk_csr_i clock is 100-150 MHz and the MDC clock is
clk_csr_i/62.
clk_csr_i/16.
clk_csr_i/26.
clk_csr_i/102.
0000b clk_csr_i/124.
5:2 - 0110 and 0111: Reserved
RW Based on a system clock of 133MHz, the CR value that ensures the MDC clock is
approximately between the frequency range 1.0 MHz - 2.5 MHz is 0010
When Bit 5 is set, you can achieve MDC clock of frequency higher than the IEEE 802.3
specified frequency limit of 2.5 MHz and program a clock divider of lower value.
Program the following values only if the interfacing chips support faster MDC clocks:
- 1000: clk_csr_i/4
- 1001: clk_csr_i/6
- 1010: clk_csr_i/8
- 1011: clk_csr_i/10
- 1100: clk_csr_i/12
- 1101: clk_csr_i/14
- 1110: clk_csr_i/16
- 1111: clk_csr_i/18
0b GMII Write (GW): When set, this bit indicates to the PHY that this is a Write operation
1 using the GMII Data register. If this bit is not set, it indicates that this is a Read
RW operation, that is, placing the data in the GMII Data register.
GMII Busy (GB): This bit should read logic 0 before writing to Register 4 and Register
5. During a PHY register access, the software sets this bit to 1'b1 to indicate that a Read
or Write access is in progress.
The Register 5 is invalid until this bit is cleared by the MAC. Therefore, Register 5 (GMII
0b Data) should be kept valid until the MAC clears this bit during a PHY Write operation.
0 Similarly for a read operation, the contents of Register 5 are not valid until this bit is
RW cleared.
The subsequent read or write operation should happen only after the previous operation
is complete. Because there is no acknowledgment from the PHY to MAC after a read or
write operation is completed, there is no change in the functionality of this bit even
when the PHY is not present.
15.6.6 GMII Data Register (Register 5) (GMAC_REG_5)—Offset 14h

The GMII Data register stores Write data to be written to the PHY register located at
the address specified in Register 4 (GMII Address Register). This register also stores
the Read data from the PHY register located at the address specified by Register 4.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
GD
RSV0

Bit Default &

Range Access
0000h
RO
0000h GMII Data (GD): This field contains the 16-bit data value read from the PHY after a
15:0 Management Read operation or the 16-bit data value to be written to the PHY before a
RW Management Write operation.
15.6.7 Flow Control Register (Register 6) (GMAC_REG_6)—Offset 18h

The Flow Control register controls the generation and reception of the Control (Pause
Command) frames by the MAC's Flow control module. A Write to a register with the
Busy bit set to '1' triggers the Flow Control block to generate a Pause Control frame.
The fields of the control frame are selected as specified in the 802.3x specification, and
the Pause Time value from this register is used in the Pause Time field of the control
frame. The Busy bit remains set until the control frame is transferred onto the cable.
The Host must make sure that the Busy bit is cleared before writing to the register.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0RSV1 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
PT
DZPQ
PLT
RFE
TFE
FCB_BPA
UP
Bit Default &
Range Access
Pause Time (PT): This field holds the value to be used in the Pause Time field in the
0000h transmit control frame. If the Pause Time bits is configured to be double-synchronized to
31:16
RW the MII clock domain, then consecutive writes to this register should be performed only
after at least four clock cycles in the destination clock domain.
00h
RO
Disable Zero-Quanta Pause (DZPQ): When this bit is set, it disables the automatic
0b generation of the Zero-Quanta Pause Control frames on the de-assertion of the flow-
7
RW control signal from the FIFO layer. When this bit is reset, normal operation with
automatic Zero-Quanta Pause Control frame generation is enabled.
0b
RO

Bit Default &

Range Access
Pause Low Threshold (PLT): This field configures the threshold of the PAUSE timer at
which the flow-control signal from the FIFO layer is checked for automatic
retransmission of PAUSE Frame.
The threshold values should be always less than the Pause Time configured in
Bits[31:16]. For example, if PT = 100H (256 slot-times), and PLT = 01, then a second
PAUSE frame is automatically transmitted if the flow-control signal from the FIFO layer
00b is asserted at 228 (256 - 28) slot times after the first PAUSE frame is transmitted.
5:4
RW The following list provides the threshold values for different values:
00: The threshold is Pause time minus 4 slot times (PT - 4 slot times).
The slot time is defined as the time taken to transmit 512 bits (64 bytes) on the MII
interface.
Unicast Pause Frame Detect (UP): When this bit is set, then in addition to the
0b detecting Pause frames with the unique multicast address, the MAC detects the Pause
3 frames with the station's unicast address specified in the MAC Address0 High Register
RW and MAC Address0 Low Register. When this bit is reset, the MAC detects only a Pause
frame with the unique multicast address specified in the 802.3x standard.
0b Receive Flow Control Enable (RFE): When this bit is set, the MAC decodes the
2 received Pause frame and disables its transmitter for a specified (Pause) time. When this
RW bit is reset, the decode function of the Pause frame is disabled.
Transmit Flow Control Enable (TFE): In the full-duplex mode, when this bit is set,
the MAC enables the flow control operation to transmit Pause frames. When this bit is
0b reset, the flow control operation in the MAC is disabled, and the MAC does not transmit
1
RW any Pause frames.
In half-duplex mode, when this bit is set, the MAC enables the back-pressure operation.
When this bit is reset, the back-pressure feature is disabled.
Flow Control Busy or Backpressure Activate (FCB_BPA): This bit initiates a Pause
Control frame in the full-duplex mode and activates the backpressure function in the
half-duplex mode if the TFE bit is set.
In the full-duplex mode, this bit should be read as 1'b0 before writing to the Flow
Control register. To initiate a Pause control frame, the Application must set this bit to
1'b1. During a transfer of the Control Frame, this bit continues to be set to signify that a
0b frame transmission is in progress. After the completion of Pause control frame
0 transmission, the MAC resets this bit to 1'b0. The Flow Control register should not be
RW written to until this bit is cleared.
In the half-duplex mode, when this bit is set (and TFE is set), then backpressure is
asserted by the MAC. During backpressure, when the MAC receives a new frame, the
transmitter starts sending a JAM pattern resulting in a collision. This control register bit
is logically ORed with the flow-control signal from the FIFO layer for the backpressure
function. When the MAC is configured for the full-duplex mode, the BPA is automatically
disabled.
15.6.8 VLAN Tag Register (Register 7) (GMAC_REG_7)—Offset 1Ch

The VLAN Tag register contains the IEEE 802.1Q VLAN Tag to identify the VLAN frames.
The MAC compares the 13th and 14th bytes of the receiving frame (Length/Type) with
16'h8100, and the following two bytes are compared with the VLAN tag. If a match
occurs, the MAC sets the received VLAN bit in the receive frame status. The legal length
of the frame is increased from 1,518 bytes to 1,522 Bytes.
Access Method
(Size: 32 bits) Offset: [BAR0] + 1Ch

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ESVL
VL
ETV
RSV0
VTHM
VTIM
Bit Default &
Range Access
000h
RO
VLAN Tag Hash Table Match Enable (VTHM): When set, the most significant four
bits of the VLAN tags CRC are used to index the content of Register 354 (VLAN Hash
0b Table Register). A value of 1 in the VLAN Hash Table register, corresponding to the
19 index, indicates that the frame matched the VLAN hash table. When Bit 16 (ETV) is set,
RW the CRC of the 12-bit VLAN Identifier (VID) is used for comparison whereas when ETV is
reset, the CRC of the 16-bit VLAN tag is used for comparison.
When reset, the VLAN Hash Match operation is not performed.
0b Enable S-VLAN (ESVL): When this bit is set, the MAC transmitter and receiver also
18
RW consider the S-VLAN (Type = 0x88A8) frames as valid VLAN tagged frames.
VLAN Tag Inverse Match Enable (VTIM): When set, this bit enables the VLAN Tag
0b inverse matching. The frames that do not have matching VLAN Tag are marked as
17 matched.
RW When reset, this bit enables the VLAN Tag perfect matching. The frames with matched
VLAN Tag are marked as matched.
Enable 12-Bit VLAN Tag Comparison (ETV): When this bit is set, a 12-bit VLAN
identifier is used for comparing and filtering instead of the complete 16-bit VLAN tag.
0b Bits [11:0] of VLAN tag are compared with the corresponding field in the received VLAN-
16 tagged frame. Similarly, when enabled, only 12 bits of the VLAN tag in the received
RW frame are used for hash-based VLAN filtering.
When this bit is reset, all 16 bits of the 15th and 16th bytes of the received VLAN frame
are used for comparison and VLAN hash filtering.
VLAN Tag Identifier for Receive Frames (VL): This field contains the 802.1Q VLAN
tag to identify the VLAN frames and is compared to the 15th and 16th bytes of the
frames being received for VLAN frames. The following list describes the bits of this field:
Bits [15:13]: User Priority
0000h Bit 12: Canonical Format Indicator (CFI) or Drop Eligible Indicator (DEI)
15:0
RW Bits[11:0]: VLAN tag's VLAN Identifier (VID) field
When the ETV bit is set, only the VID (Bits[11:0]) is used for comparison. If VL
(VL[11:0] if ETV is set) is all zeros, the MAC does not check the fifteenth and 16th bytes
for VLAN tag comparison, and declares all frames with a Type field value of 0x8100 or
0x88a8 as VLAN frames.
15.6.9 Version Register (Register 8) (GMAC_REG_8)—Offset 20h

The Version registers identifies the version of the MAC. This register contains two
bytes: one identifies the core IP release number, and the other that identifies the user
release.
Access Method
(Size: 32 bits)

Default: 00001037h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 0 1 1 1
USERVER
SNPSVER
RSV0
Bit Default &
Range Access
0000h
31:16 RSV0: Reserved
RO
10h
15:8 User-defined Version (1.0) (USERVER): Reserved.
RO
37h
7:0 Synopsys-defined Version (3.7) (SNPSVER): Reserved.
RO
15.6.10 Debug Register (Register 9) (GMAC_REG_9)—Offset 24h

The Debug register gives the status of all main modules of the transmit and receive
data-paths and the FIFOs. An all-zero status indicates that the MAC is in idle state (and
FIFOs are empty) and no activity is going on in the data-paths.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV4
RSV3
RSV2
RSV1
RSV0
TWCSTS
TXPAUSED
RPESTS
TXFSTS
TRCSTS
TFCSTS
TPESTS
RXFSTS
RRCSTS
RWCSTS
RFCFCSTS
Bit Default &
Range Access
0000000b
RO
0b MTL Tx FIFO Not Empty Status (TXFSTS): When high, this bit indicates that the MTL
24
RO Tx FIFO is not empty and some data is left for transmission.
0b
RO
0b MTL Tx FIFO Write Controller Active Status (TWCSTS): When high, this bit
22 indicates that the MTL Tx FIFO Write Controller is active and transferring data to the Tx
RO FIFO.
MTL Tx FIFO Read Controller Status (TRCSTS): This field indicates the state of the
Tx FIFO Read Controller:
00b 00: IDLE state
21:20
RO 01: READ state (transferring data to MAC transmitter)
10: Waiting for TxStatus from MAC transmitter
11: Writing the received TxStatus or flushing the Tx FIFO

Bit Default &

Range Access
0b MAC transmitter in PAUSE (TXPAUSED): When high, this bit indicates that the MAC
19 transmitter is in the PAUSE condition (in the full-duplex only mode) and hence does not
RO schedule any frame for transmission.
MAC Transmit Frame Controller Status (TFCSTS): This field indicates the state of
the MAC Transmit Frame Controller module:
00b 00: IDLE state
18:17
RO 01: Waiting for Status of previous frame or IFG or backoff period to be over
10: Generating and transmitting a PAUSE control frame (in the full-duplex mode)
11: Transferring input frame for transmission
0b MAC MII Transmit Protocol Engine Status (TPESTS): When high, this bit indicates
16 that the MAC MII transmit protocol engine is actively transmitting data and is not in the
RO IDLE state.
000000b
RO
MTL Rx FIFO Fill-level Status (RXFSTS): This field gives the status of the fill-level of
the Rx FIFO:
00b 00: Rx FIFO Empty
9:8
RO 01: Rx FIFO fill level is below the flow-control deactivate threshold
10: Rx FIFO fill level is above the flow-control activate threshold
11: Rx FIFO Full
0b
RO
MTL Rx FIFO Read Controller State (RRCSTS): This field gives the state of the Rx
FIFO read Controller:
00b 00: IDLE state
6:5
RO 01: Reading frame data
10: Reading frame status (or timestamp)
11: Flushing the frame data and status
0b MTL Rx FIFO Write Controller Active Status (RWCSTS): When high, this bit
4 indicates that the MTL Rx FIFO Write Controller is active and is transferring a received
RO frame to the FIFO.
0b
RO
00b MAC Receive Frame Controller FIFO Status (RFCFCSTS): When high, this field
2:1 indicates the active state of the small FIFO Read and Write controllers of the MAC
RO Receive Frame Controller Module.
0b MAC MII Receive Protocol Engine Status (RPESTS): When high, this bit indicates
0
RO that the MAC MII receive protocol engine is actively receiving data and not in IDLE state.
15.6.11 Interrupt Register (Register 14) (GMAC_REG_14)—Offset 38h

The Interrupt Status register identifies the events in the MAC that can generate
interrupt.
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TSIS
MMCRXIPIS
MMCIS
MMCTXIS
MMCRXIS
RSV4
RSV2
RSV0
Bit Default &
Range Access
000000h
RO
Timestamp Interrupt Status (TSIS): When the Advanced Timestamp feature is

enabled, this bit is set when any of the following conditions is true:
The system time value equals or exceeds the value specified in the Target Time High and
Low registers.
0b There is an overflow in the seconds register.
9 The Auxiliary snapshot trigger is asserted.
RO This bit is cleared on reading Bit 0 of the Register 458 (Timestamp Status Register).
If default Timestamping is enabled, when set, this bit indicates that the system time
value is equal to or exceeds the value specified in the Target Time registers. In this
mode, this bit is cleared after the completion of the read of this bit. In all other modes,
this bit is reserved.
0b
RO
0b MMC Receive Checksum Offload Interrupt Status (MMCRXIPIS): This bit is set
7 high when an interrupt is generated in the MMC Receive Checksum Offload Interrupt
RO Register. This bit is cleared when all the bits in this interrupt register are cleared.
0b MMC Transmit Interrupt Status (MMCTXIS): This bit is set high when an interrupt is
6 generated in the MMC Transmit Interrupt Register. This bit is cleared when all the bits in
RO this interrupt register are cleared.
0b MMC Receive Interrupt Status (MMCRXIS): This bit is set high when an interrupt is
5 generated in the MMC Receive Interrupt Register. This bit is cleared when all the bits in
RO this interrupt register are cleared.
0b MMC Interrupt Status (MMCIS): This bit is set high when any of the Bits [7:5] is set
4
RO high and cleared only when all of these bits are low.
0000b
RO
15.6.12 Interrupt Mask Register (Register 15) (GMAC_REG_15)—Offset

3Ch
The Interrupt Mask Register bits enable you to mask the interrupt signal because of the
corresponding event in the Interrupt Status Register.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV3
TSIM
RSV0

Bit Default &

Range Access
000000h
RO
Timestamp Interrupt Mask (TSIM): When set, this bit disables the assertion of the
0b interrupt signal because of the setting of Timestamp Interrupt Status bit in Register 14
9
RW (Interrupt Status Register). This bit is valid only when IEEE1588 timestamping is
enabled. In all other modes, this bit is reserved.
0b
RO
15.6.13 MAC Address0 High Register (Register 16) (GMAC_REG_16)—

Offset 40h
The MAC Address0 High register holds the upper 16 bits of the first 6-byte MAC address
of the station. The first DA byte that is received on the MII interface corresponds to the
LS byte (Bits [7:0]) of the MAC Address Low register. For example, if 0x112233445566
is received (0x11 in lane 0 of the first column) on the MII as the destination address,
then the MacAddress0 Register [47:0] is compared with 0x665544332211. Using the
standard IEEE 802 format for printing MAC-48 addresses this corresponds to
11:22:33:44:55:66 where 0x11 is the LS byte (Bits [7:0]) of the MAC Address Low
register.
Access Method
(Size: 32 bits)

Default: 8000FFFFh
31 28 24 20 16 12 8 4 0
1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
ADDRHI
RSV0
AE
Bit Default &

Range Access
1b
31 Address Enable (AE): This bit is always set to 1.
RO
0000h
RO
ffffh MAC Address0 High (ADDRHI): This field contains the upper 16 bits (47:32) of the
15:0 first 6-byte MAC address. The MAC uses this field for filtering the received frames and
RW inserting the MAC address in the Transmit Flow Control (PAUSE) Frames.
15.6.14 MAC Address0 Low Register (Register 17) (GMAC_REG_17)—

Offset 44h
The MAC Address0 Low register holds the lower 32 bits of the first 6-byte MAC address
of the station.
Access Method


(Size: 32 bits)

Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
ADDRLO
Bit Default &
Range Access
ffffffffh MAC Address0 Low (ADDRLO): This field contains the lower 32 bits of the first 6-byte
31:0 MAC address. This is used by the MAC for filtering the received frames and inserting the
RW MAC address in the Transmit Flow Control (PAUSE) Frames.
15.6.15 MMC Control Register (Register 64) (GMAC_REG_64)—Offset

100h
The MMC Control register establishes the operating mode of the management counters.
NOTE: The bit 0 (Counters Reset) has higher priority than bit 4 (Counter Preset).
Therefore, when the Software tries to set both bits in the same write cycle, all counters
are cleared and the bit 4 is not set.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNTPRST
CNTRST
UCDBC
RSV1
RSV0
CNTSTOPRO
CNTPRSTLVL
CNTFREEZ
RSTONRD
Bit Default &

Range Access
000000h
RO
Update MMC Counters for Dropped Broadcast Frames (UCDBC): When set, this
0b bit enables MAC to update all the related MMC Counters for Broadcast frames dropped
8
RW due to setting of DBF bit (Disable Broadcast Frames) of MAC Filter Register at offset
0x0004. When reset, MMC Counters are not updated for dropped Broadcast frames.
00b
RO

Bit Default &

Range Access
Full-Half Preset (CNTPRSTLVL): When low and bit 4 is set, all MMC counters get
preset to almost-half value. All octet counters get preset to 0x7FFF_F800 (half -
2KBytes) and all frame-counters gets preset to 0x7FFF_FFF0 (half - 16). When this bit is
0b high and bit 4 is set, all MMC counters get preset to almost-full value. All octet counters
5
RW get preset to 0xFFFF_F800 (full - 2KBytes) and all frame-counters gets preset to
0xFFFF_FFF0 (full - 16). For 16-bit counters, the almost-half preset values are 0x7800
and 0x7FF0 for the respective octet and frame counters. Similarly, the almost-full preset
values for the 16-bit counters are 0xF800 and 0xFFF0.
Counters Preset (CNTPRST): When this bit is set, all counters are initialized or preset
0b to almost full or almost half according to bit 5. This bit is cleared automatically after 1
4
RW clock cycle. This bit, along with bit 5, is useful for debugging and testing the assertion of
interrupts because of MMC counter becoming half-full or full.
MMC Counter Freeze (CNTFREEZ): When this bit is set, it freezes all MMC counters to
0b their current value. Until this bit is reset to 0, no MMC counter is updated because of any
3
RW transmitted or received frame. If any MMC counter is read with the Reset on Read bit
set, then that counter is also cleared in this mode
0b Reset on Read (RSTONRD): When this bit is set, the MMC counters are reset to zero
2 after Read (self-clearing after reset). The counters are cleared when the least significant
RW byte lane (bits[7:0]) is read.
0b Counters Stop Rollover (CNTSTOPRO): When this bit is set, after reaching
1
RW maximum value, the counter does not roll over to zero.
0b Counters Reset (CNTRST): When this bit is set, all counters are reset. This bit is
0
RW cleared automatically after one clock cycle.
15.6.16 MMC Receive Interrupt Register (MMC_INTR_RX)—Offset 104h

The MMC Receive Interrupt Register maintains the interrupt generated from all of the
receive statistic counters. An interrupt bit is cleared when the respective MMC counter
that caused the interrupt is read.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
RX512T1023OCTGBFIS
RX256T511OCTGBFIS
RX128T255OCTGBFIS
RXCTRLFIS
RXRCVERRFIS
RXWDOGFIS
RXVLANGBFIS
RXPAUSFIS
RX65T127OCTGBFIS
RXALGNERFIS
RXFOVFIS
RXORANGEFIS
RXUCGFIS
RX1024TMAXOCTGBFIS
RX64OCTGBFIS
RXJABERFIS
RXMCGFIS
RXGBFRMIS
RXLENERFIS
RXOSIZEGFIS
RXUSIZEGFIS
RXRUNTFIS
RXCRCERFIS
RXBCGFIS
RXGOCTIS
RXGBOCTIS
Bit Default &

Range Access
000000b
RO

Bit Default &

Range Access
0b MMC Receive Control Frame Counter Interrupt Status (RXCTRLFIS): This bit is
25 set when the rxctrlframes_g counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive Error Frame Counter Interrupt Status (RXRCVERRFIS): This bit is
24 set when the rxrcverror counter reaches half of the maximum value or the maximum
RO value.
0b MMC Receive Watchdog Error Frame Counter Interrupt Status (RXWDOGFIS):

23 This bit is set when the rxwatchdog error counter reaches half of the maximum value or
RO the maximum value.
0b MMC Receive VLAN Good Bad Frame Counter Interrupt Status (RXVLANGBFIS):
22 This bit is set when the rxvlanframes_gb counter reaches half of the maximum value or
0b MMC Receive FIFO Overflow Frame Counter Interrupt Status (RXFOVFIS): This
21 bit is set when the rxfifooverflow counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive Pause Frame Counter Interrupt Status (RXPAUSFIS): This bit is set
20 when the rxpauseframes counter reaches half of the maximum value or the maximum
RO value.
0b MMC Receive Out Of Range Error Frame Counter Interrupt Status

19 (RXORANGEFIS): This bit is set when the rxoutofrangetype counter reaches half of the
RO maximum value or the maximum value.
0b MMC Receive Length Error Frame Counter Interrupt Status (RXLENERFIS): This
18 bit is set when the rxlengtherror counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive Unicast Good Frame Counter Interrupt Status (RXUCGFIS): This
17 bit is set when the rxunicastframes_g counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive 1024 to Maximum Octet Good Bad Frame Counter Interrupt
16 Status (RX1024TMAXOCTGBFIS): This bit is set when the rx1024tomaxoctets_gb
RO counter reaches half of the maximum value or the maximum value.
0b MMC Receive 512 to 1023 Octet Good Bad Frame Counter Interrupt Status
15 (RX512T1023OCTGBFIS): This bit is set when the rx512to1023octets_gb counter
RO reaches half of the maximum value or the maximum value.
12 (RX65T127OCTGBFIS): This bit is set when the rx65to127octets_gb counter reaches
RO half of the maximum value or the maximum value.
0b MMC Receive 64 Octet Good Bad Frame Counter Interrupt Status

11 (RX64OCTGBFIS): This bit is set when the rx64octets_gb counter reaches half of the
0b MMC Receive Oversize Good Frame Counter Interrupt Status (RXOSIZEGFIS):

10 This bit is set when the rxoversize_g counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive Undersize Good Frame Counter Interrupt Status (RXUSIZEGFIS):

9 This bit is set when the rxundersize_g counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive Jabber Error Frame Counter Interrupt Status (RXJABERFIS): This
8 bit is set when the rxjabbererror counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive Runt Frame Counter Interrupt Status (RXRUNTFIS): This bit is set
7 when the rxrunterror counter reaches half of the maximum value or the maximum
RO value.

Bit Default &

Range Access
0b MMC Receive Alignment Error Frame Counter Interrupt Status (RXALGNERFIS):

6 This bit is set when the rxalignmenterror counter reaches half of the maximum value or
0b MMC Receive CRC Error Frame Counter Interrupt Status (RXCRCERFIS): This bit
5 is set when the rxcrcerror counter reaches half of the maximum value or the maximum
RO value.
0b MMC Receive Multicast Good Frame Counter Interrupt Status (RXMCGFIS): This
4 bit is set when the rxmulticastframes_g counter reaches half of the maximum value or
0b MMC Receive Broadcast Good Frame Counter Interrupt Status (RXBCGFIS): This
3 bit is set when the rxbroadcastframes_g counter reaches half of the maximum value or
0b MMC Receive Good Octet Counter Interrupt Status (RXGOCTIS): This bit is set
2 when the rxoctetcount_g counter reaches half of the maximum value or the maximum
RO value.
0b MMC Receive Good Bad Octet Counter Interrupt Status (RXGBOCTIS): This bit is
1 set when the rxoctetcount_gb counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive Good Bad Frame Counter Interrupt Status (RXGBFRMIS): This bit
0 is set when the rxframecount_gb counter reaches half of the maximum value or the
RO maximum value.
15.6.17 MMC Transmit Interrupt Register (MMC_INTR_TX)—Offset 108h

The maintains the interrupt generated from all of the transmit statistic counters. An
interrupt bit is cleared when the respective MMC counter that caused the interrupt is
read.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TXGBFRMIS
TXOSIZEGFIS
TXPAUSFIS
TXEXDEFFIS
TXGFRMIS
TXVLANGFIS
TXCARERFIS
TXEXCOLFIS
TXLATCOLFIS
TXDEFFIS
TXSCOLGFIS
TXUFLOWERFIS
TXBCGBFIS
TXMCGBFIS
TXUCGBFIS
TXGOCTIS
TXMCOLGFIS
TX1024TMAXOCTGBFIS
TXBCGFIS
TX512T1023OCTGBFIS
TX256T511OCTGBFIS
TX128T255OCTGBFIS
TX65T127OCTGBFIS
TXGBOCTIS
TX64OCTGBFIS
TXMCGFIS
RSV0
Bit Default &

Range Access
000000b
RO
0b MMC Transmit Oversize Good Frame Counter Interrupt Status (TXOSIZEGFIS):

25 This bit is set when the txoversize_g counter reaches half of the maximum value or the
RO maximum value.

Bit Default &

Range Access
0b MMC Transmit VLAN Good Frame Counter Interrupt Status (TXVLANGFIS): This
24 bit is set when the txvlanframes_g counter reaches half of the maximum value or the
RO maximum value.
0b MMC Transmit Pause Frame Counter Interrupt Status (TXPAUSFIS): This bit is
23 set when the txpauseframeserror counter reaches half of the maximum value or the
RO maximum value.
0b MMC Transmit Excessive Deferral Frame Counter Interrupt Status

22 (TXEXDEFFIS): This bit is set when the txexcessdef counter reaches half of the
0b MMC Transmit Good Frame Counter Interrupt Status (TXGFRMIS): This bit is set
21 when the txframecount_g counter reaches half of the maximum value or the maximum
RO value.
0b MMC Transmit Good Octet Counter Interrupt Status (TXGOCTIS): This bit is set
20 when the txoctetcount_g counter reaches half of the maximum value or the maximum
RO value.
0b MMC Transmit Carrier Error Frame Counter Interrupt Status (TXCARERFIS):

19 This bit is set when the txcarriererror counter reaches half of the maximum value or the
RO maximum value.
0b MMC Transmit Excessive Collision Frame Counter Interrupt Status

18 (TXEXCOLFIS): This bit is set when the txexesscol counter reaches half of the
0b MMC Transmit Late Collision Frame Counter Interrupt Status (TXLATCOLFIS):

17 This bit is set when the txlatecol counter reaches half of the maximum value or the
RO maximum value.
0b MMC Transmit Deferred Frame Counter Interrupt Status (TXDEFFIS): This bit is
16 set when the txdeferred counter reaches half of the maximum value or the maximum
RO value.
0b MMC Transmit Multiple Collision Good Frame Counter Interrupt Status

15 (TXMCOLGFIS): This bit is set when the txmulticol_g counter reaches half of the
0b MMC Transmit Single Collision Good Frame Counter Interrupt Status

14 (TXSCOLGFIS): This bit is set when the txsinglecol_g counter reaches half of the
0b MMC Transmit Underflow Error Frame Counter Interrupt Status

13 (TXUFLOWERFIS): This bit is set when the txunderflowerror counter reaches half of
RO the maximum value or the maximum value.
0b MMC Transmit Broadcast Good Bad Frame Counter Interrupt Status

12 (TXBCGBFIS): This bit is set when the txbroadcastframes_gb counter reaches half of
0b MMC Transmit Multicast Good Bad Frame Counter Interrupt Status

11 (TXMCGBFIS): The bit is set when the txmulticastframes_gb counter reaches half of
0b MMC Transmit Unicast Good Bad Frame Counter Interrupt Status (TXUCGBFIS):
10 This bit is set when the txunicastframes_gb counter reaches half of the maximum value
RO or the maximum value.
0b MMC Transmit 1024 to Maximum Octet Good Bad Frame Counter Interrupt
9 Status (TX1024TMAXOCTGBFIS): This bit is set when the tx1024tomaxoctets_gb
RO counter reaches half of the maximum value or the maximum value.
0b MMC Transmit 512 to 1023 Octet Good Bad Frame Counter Interrupt Status
8 (TX512T1023OCTGBFIS): This bit is set when the tx512to1023octets_gb counter

Bit Default &

Range Access
5 (TX65T127OCTGBFIS): This bit is set when the tx65to127octets_gb counter reaches
RO half the maximum value, and also when it reaches the maximum value.
0b MMC Transmit 64 Octet Good Bad Frame Counter Interrupt Status

4 (TX64OCTGBFIS): This bit is set when the tx64octets_gb counter reaches half of the
0b MMC Transmit Multicast Good Frame Counter Interrupt Status (TXMCGFIS):

3 This bit is set when the txmulticastframes_g counter reaches half of the maximum value
RO or the maximum value.
0b MMC Transmit Broadcast Good Frame Counter Interrupt Status (TXBCGFIS):

2 This bit is set when the txbroadcastframes_g counter reaches half of the maximum
RO value or the maximum value.
0b MMC Transmit Good Bad Frame Counter Interrupt Status (TXGBFRMIS): This bit
1 is set when the txframecount_gb counter reaches half of the maximum value or the
RO maximum value.
0b MMC Transmit Good Bad Octet Counter Interrupt Status (TXGBOCTIS): This bit
0 is set when the txoctetcount_gb counter reaches half of the maximum value or the
RO maximum value.
15.6.18 MMC Receive Interrupt Mask Register (MMC_INTR_MASK_RX)—

Offset 10Ch
The MMC Receive Interrupt Mask Register maintains the mask for the interrupt
generated from all of the receive statistic counters.
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 10Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RXALGNERFIM
RSV0
RXCTRLFIM
RXWDOGFIM
RXVLANGBFIM
RXPAUSFIM
RXRCVERRFIM
RXFOVFIM
RXORANGEFIM
RXOSIZEGFIM
RXUSIZEGFIM
RXBCGFIM
RXGBFRMIM
RXLENERFIM
RXUCGFIM
RX1024TMAXOCTGBFIM
RX512T1023OCTGBFIM
RX256T511OCTGBFIM
RX128T255OCTGBFIM
RX65T127OCTGBFIM
RX64OCTGBFIM
RXJABERFIM
RXRUNTFIM
RXMCGFIM
RXGOCTIM
RXGBOCTIM
RXCRCERFIM
Bit Default &

Range Access
000000b
RO
0b MMC Receive Control Frame Counter Interrupt Mask (RXCTRLFIM): Setting this
25 bit masks the interrupt when the rxctrlframes_g counter reaches half of the maximum
RW value or the maximum value.
0b MMC Receive Error Frame Counter Interrupt Mask (RXRCVERRFIM): Setting this
24 bit masks the interrupt when the rxrcverror counter reaches half of the maximum value
RW or the maximum value.

Bit Default &

Range Access
0b MMC Receive Watchdog Error Frame Counter Interrupt Mask (RXWDOGFIM):

23 Setting this bit masks the interrupt when the rxwatchdog counter reaches half of the
RW maximum value or the maximum value.
0b MMC Receive VLAN Good Bad Frame Counter Interrupt Mask (RXVLANGBFIM):
22 Setting this bit masks the interrupt when the rxvlanframes_gb counter reaches half of
RW the maximum value or the maximum value.
0b MMC Receive FIFO Overflow Frame Counter Interrupt Mask (RXFOVFIM):

21 Setting this bit masks the interrupt when the rxfifooverflow counter reaches half of the
0b MMC Receive Pause Frame Counter Interrupt Mask (RXPAUSFIM): Setting this
20 bit masks the interrupt when the rxpauseframes counter reaches half of the maximum
0b MMC Receive Out Of Range Error Frame Counter Interrupt Mask

19 (RXORANGEFIM): Setting this bit masks the interrupt when the rxoutofrangetype
RW counter reaches half of the maximum value or the maximum value.
0b MMC Receive Length Error Frame Counter Interrupt Mask (RXLENERFIM):

18 Setting this bit masks the interrupt when the rxlengtherror counter reaches half of the
0b MMC Receive Unicast Good Frame Counter Interrupt Mask (RXUCGFIM): Setting
17 this bit masks the interrupt when the rxunicastframes_g counter reaches half of the
MMC Receive 1024 to Maximum Octet Good Bad Frame Counter Interrupt Mask
0b (RX1024TMAXOCTGBFIM): Setting this bit masks the interrupt when the
16
RW rx1024tomaxoctets_gb counter reaches half of the maximum value or the maximum
value.
MMC Receive 512 to 1023 Octet Good Bad Frame Counter Interrupt Mask
0b (RX512T1023OCTGBFIM): Setting this bit masks the interrupt when the
15
RW rx512to1023octets_gb counter reaches half of the maximum value or the maximum
value.
14
value.
13
value.
0b MMC Receive 65 to 127 Octet Good Bad Frame Counter Interrupt Mask
12 (RX65T127OCTGBFIM): Setting this bit masks the interrupt when the
RW rx65to127octets_gb counter reaches half of the maximum value or the maximum value.
0b MMC Receive 64 Octet Good Bad Frame Counter Interrupt Mask

11 (RX64OCTGBFIM): Setting this bit masks the interrupt when the rx64octets_gb
0b MMC Receive Oversize Good Frame Counter Interrupt Mask (RXOSIZEGFIM):

10 Setting this bit masks the interrupt when the rxoversize_g counter reaches half of the
0b MMC Receive Undersize Good Frame Counter Interrupt Mask (RXUSIZEGFIM):

9 Setting this bit masks the interrupt when the rxundersize_g counter reaches half of the
0b MMC Receive Jabber Error Frame Counter Interrupt Mask (RXJABERFIM):

8 Setting this bit masks the interrupt when the rxjabbererror counter reaches half of the
0b MMC Receive Runt Frame Counter Interrupt Mask (RXRUNTFIM): Setting this bit
7 masks the interrupt when the rxrunterror counter reaches half of the maximum value or
RW the maximum value.
0b MMC Receive Alignment Error Frame Counter Interrupt Mask (RXALGNERFIM):

6 Setting this bit masks the interrupt when the rxalignmenterror counter reaches half of

Bit Default &

Range Access
0b MMC Receive CRC Error Frame Counter Interrupt Mask (RXCRCERFIM): Setting
5 this bit masks the interrupt when the rxcrcerror counter reaches half of the maximum
0b MMC Receive Multicast Good Frame Counter Interrupt Mask (RXMCGFIM):

4 Setting this bit masks the interrupt when the rxmulticastframes_g counter reaches half
RW of the maximum value or the maximum value.
0b MMC Receive Broadcast Good Frame Counter Interrupt Mask (RXBCGFIM):

3 Setting this bit masks the interrupt when the rxbroadcastframes_g counter reaches half
0b MMC Receive Good Octet Counter Interrupt Mask (RXGOCTIM): Setting this bit
2 masks the interrupt when the rxoctetcount_g counter reaches half of the maximum
0b MMC Receive Good Bad Octet Counter Interrupt Mask (RXGBOCTIM): Setting
1 this bit masks the interrupt when the rxoctetcount_gb counter reaches half of the
0b MMC Receive Good Bad Frame Counter Interrupt Mask (RXGBFRMIM): Setting
0 this bit masks the interrupt when the rxframecount_gb counter reaches half of the
15.6.19 MMC Transmit Interrupt Mask Register

(MMC_INTR_MASK_TX)—Offset 110h
The MMC Transmit Interrupt Mask Register maintains the mask for the interrupt
generated from all of the transmit statistic counters.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
TXOSIZEGFIM
TXVLANGFIM
TXPAUSFIM
TXEXDEFFIM
TXUCGBFIM
TXGFRMIM
TXGOCTIM
TXCARERFIM
TXEXCOLFIM
TXLATCOLFIM
TXDEFFIM
TXMCOLGFIM
TXSCOLGFIM
TXBCGBFIM
TXMCGBFIM
TXUFLOWERFIM
TX1024TMAXOCTGBFIM
TX512T1023OCTGBFIM
TXBCGFIM
TXGBFRMIM
TXGBOCTIM
TX256T511OCTGBFIM
TX128T255OCTGBFIM
TX65T127OCTGBFIM
TX64OCTGBFIM
TXMCGFIM
Bit Default &

Range Access
000000b
RO
0b MMC Transmit Oversize Good Frame Counter Interrupt Mask (TXOSIZEGFIM):

25 Setting this bit masks the interrupt when the txoversize_g counter reaches half of the
0b MMC Transmit VLAN Good Frame Counter Interrupt Mask (TXVLANGFIM):

24 Setting this bit masks the interrupt when the txvlanframes_g counter reaches half of the

Bit Default &

Range Access
0b MMC Transmit Pause Frame Counter Interrupt Mask (TXPAUSFIM): Setting this
23 bit masks the interrupt when the txpauseframes counter reaches half of the maximum
0b MMC Transmit Excessive Deferral Frame Counter Interrupt Mask

22 (TXEXDEFFIM): Setting this bit masks the interrupt when the txexcessdef counter
RW reaches half of the maximum value or the maximum value.
0b MMC Transmit Good Frame Counter Interrupt Mask (TXGFRMIM): Setting this bit
21 masks the interrupt when the txframecount_g counter reaches half of the maximum
0b MMC Transmit Good Octet Counter Interrupt Mask (TXGOCTIM): Setting this bit
20 masks the interrupt when the txoctetcount_g counter reaches half of the maximum
0b MMC Transmit Carrier Error Frame Counter Interrupt Mask (TXCARERFIM):

19 Setting this bit masks the interrupt when the txcarriererror counter reaches half of the
0b MMC Transmit Excessive Collision Frame Counter Interrupt Mask

18 (TXEXCOLFIM): Setting this bit masks the interrupt when the txexcesscol counter
0b MMC Transmit Late Collision Frame Counter Interrupt Mask (TXLATCOLFIM):

17 Setting this bit masks the interrupt when the txlatecol counter reaches half of the
0b MMC Transmit Deferred Frame Counter Interrupt Mask (TXDEFFIM): Setting this
16 bit masks the interrupt when the txdeferred counter reaches half of the maximum value
RW or the maximum value.
0b MMC Transmit Multiple Collision Good Frame Counter Interrupt Mask

15 (TXMCOLGFIM): Setting this bit masks the interrupt when the txmulticol_g counter
0b MMC Transmit Single Collision Good Frame Counter Interrupt Mask

14 (TXSCOLGFIM): Setting this bit masks the interrupt when the txsinglecol_g counter
0b MMC Transmit Underflow Error Frame Counter Interrupt Mask

13 (TXUFLOWERFIM): Setting this bit masks the interrupt when the txunderflowerror
0b MMC Transmit Broadcast Good Bad Frame Counter Interrupt Mask

12 (TXBCGBFIM): Setting this bit masks the interrupt when the txbroadcastframes_gb
0b MMC Transmit Multicast Good Bad Frame Counter Interrupt Mask

11 (TXMCGBFIM): Setting this bit masks the interrupt when the txmulticastframes_gb
0b MMC Transmit Unicast Good Bad Frame Counter Interrupt Mask (TXUCGBFIM):
10 Setting this bit masks the interrupt when the txunicastframes_gb counter reaches half
MMC Transmit 1024 to Maximum Octet Good Bad Frame Counter Interrupt
0b Mask (TX1024TMAXOCTGBFIM): Setting this bit masks the interrupt when the
9
RW tx1024tomaxoctets_gb counter reaches half of the maximum value or the maximum
value.
MMC Transmit 512 to 1023 Octet Good Bad Frame Counter Interrupt Mask
0b (TX512T1023OCTGBFIM): Setting this bit masks the interrupt when the
8
RW tx512to1023octets_gb counter reaches half of the maximum value or the maximum
value.
7
value.
6
value.

Bit Default &

Range Access
0b MMC Transmit 65 to 127 Octet Good Bad Frame Counter Interrupt Mask
5 (TX65T127OCTGBFIM): Setting this bit masks the interrupt when the
RW tx65to127octets_gb counter reaches half of the maximum value or the maximum value.
0b MMC Transmit 64 Octet Good Bad Frame Counter Interrupt Mask

4 (TX64OCTGBFIM): Setting this bit masks the interrupt when the tx64octets_gb
0b MMC Transmit Multicast Good Frame Counter Interrupt Mask (TXMCGFIM):

3 Setting this bit masks the interrupt when the txmulticastframes_g counter reaches half
0b MMC Transmit Broadcast Good Frame Counter Interrupt Mask (TXBCGFIM):

2 Setting this bit masks the interrupt when the txbroadcastframes_g counter reaches half
0b MMC Transmit Good Bad Frame Counter Interrupt Mask (TXGBFRMIM): Setting
1 this bit masks the interrupt when the txframecount_gb counter reaches half of the
0b MMC Transmit Good Bad Octet Counter Interrupt Mask (TXGBOCTIM): Setting
0 this bit masks the interrupt when the txoctetcount_gb counter reaches half of the
15.6.20 MMC Transmit Good Bad Octet Counter Register

(TXOCTETCOUNT_GB)—Offset 114h
Number of bytes transmitted, exclusive of preamble and retried bytes, in good and bad
frames.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
31:0 Counter value (CNT): Reserved.
RO
15.6.21 MMC Transmit Good Bad Frame Counter Register

(TXFRAMECOUNT_GB)—Offset 118h
Number of good and bad frames transmitted, exclusive of retried frames.
Access Method


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.22 MMC Transmit Broadcast Good Frame Counter Register

(TXBROADCASTFRAMES_G)—Offset 11Ch
Number of good broadcast frames transmitted.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.23 MMC Transmit Multicast Good Frame Counter Register

(TXMULTICASTFRAMES_G)—Offset 120h
Number of good multicast frames transmitted.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
15.6.24 MMC Transmit 64 Octet Good Bad Frame Counter Register

(TX64OCTETS_GB)—Offset 124h
Number of good and bad frames transmitted with length 64 bytes, exclusive of
preamble and retried frames.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.25 MMC Transmit 65 to 127 Octet Good Bad Frame Counter

Register (TX65TO127OCTETS_GB)—Offset 128h
Number of good and bad frames transmitted with length between 65 and 127
(inclusive) bytes, exclusive of preamble and retried frames.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

Register (TX128TO255OCTETS_GB)—Offset 12Ch
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
Number of good and bad frames transmitted with length between 512 and 1,023
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.29 MMC Transmit 1024 to Maximum Octet Good Bad Frame Counter
Register (TX1024TOMAXOCTETS_GB)—Offset 138h
Number of good and bad frames transmitted with length between 1,024 and maxsize
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.30 MMC Transmit Unicast Good Bad Frame Counter Register

(TXUNICASTFRAMES_GB)—Offset 13Ch
Number of good and bad unicast frames transmitted.
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.31 MMC Transmit Multicast Good Bad Frame Counter Register

(TXMULTICASTFRAMES_GB)—Offset 140h
Number of good and bad multicast frames transmitted.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.32 MMC Transmit Broadcast Good Bad Frame Counter Register

(TXBROADCASTFRAMES_GB)—Offset 144h
Number of good and bad broadcast frames transmitted.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
15.6.33 MMC Transmit Underflow Error Frame Counter Register

(TXUNDERFLOWERROR)—Offset 148h
Number of frames aborted because of frame underflow error.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.34 MMC Transmit Single Collision Good Frame Counter Register

(TXSINGLECOL_G)—Offset 14Ch
Number of successfully transmitted frames after a single collision in the half-duplex
mode.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

15.6.35 MMC Transmit Multiple Collision Good Frame Counter Register

(TXMULTICOL_G)—Offset 150h
Number of successfully transmitted frames after multiple collisions in the half-duplex
mode.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.36 MMC Transmit Deferred Frame Counter Register

(TXDEFERRED)—Offset 154h
Number of successfully transmitted frames after a deferral in the half-duplex mode.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.37 MMC Transmit Late Collision Frame Counter Register

(TXLATECOL)—Offset 158h
Number of frames aborted because of late collision error.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.38 MMC Transmit Excessive Collision Frame Counter Register

(TXEXESSCOL)—Offset 15Ch
Number of frames aborted because of excessive (16) collision errors.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.39 MMC Transmit Carrier Error Frame Counter Register

(TXCARRIERERROR)—Offset 160h
Number of frames aborted because of carrier sense error (no carrier or loss of carrier).
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.40 MMC Transmit Good Octet Counter Register

(TXOCTETCOUNT_G)—Offset 164h
Number of bytes transmitted, exclusive of preamble, in good frames only.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.41 MMC Transmit Good Frame Counter Register

(TXFRAMECOUNT_G)—Offset 168h
Number of good frames transmitted.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
15.6.42 MMC Transmit Excessive Deferral Frame Counter Register

(TXEXCESSDEF)—Offset 16Ch
Number of frames aborted because of excessive deferral error (deferred for more than
two max-sized frame times).
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.43 MMC Transmit Pause Frame Counter Register

(TXPAUSEFRAMES)—Offset 170h
Number of good PAUSE frames transmitted.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

15.6.44 MMC Transmit VLAN Good Frame Counter Register

(TXVLANFRAMES_G)—Offset 174h
Number of good VLAN frames transmitted, exclusive of retried frames.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.45 MMC Transmit Oversize Good Frame Counter Register

(TXOVERSIZE_G)—Offset 178h
Number of frames transmitted without errors and with length greater than the maxsize
(1,518 or 1,522 bytes for VLAN tagged frames; 2000 bytes if enabled in Bit 27 of
Register 0 (MAC Configuration Register)).
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.46 MMC Receive Good Bad Frame Counter Register

(RXFRAMECOUNT_GB)—Offset 180h
Number of good and bad frames received.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.47 MMC Receive Good Bad Octet Counter Register

(RXOCTETCOUNT_GB)—Offset 184h
Number of bytes received, exclusive of preamble, in good and bad frames.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.48 MMC Receive Good Octet Counter Register

(RXOCTETCOUNT_G)—Offset 188h
Number of bytes received, exclusive of preamble, only in good frames.
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.49 MMC Receive Broadcast Good Frame Counter Register

(RXBROADCASTFRAMES_G)—Offset 18Ch
Number of good broadcast frames received.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.50 MMC Receive Multicast Good Frame Counter Register

(RXMULTICASTFRAMES_G)—Offset 190h
Number of good multicast frames received.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
15.6.51 MMC Receive CRC Error Frame Counter Register

(RXCRCERROR)—Offset 194h
Number of frames received with CRC error.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.52 MMC Receive Alignment Error Frame Counter Register

(RXALIGNMENTERROR)—Offset 198h
Number of frames received with alignment (dribble) error.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

15.6.53 MMC Receive Runt Frame Counter Register (RXRUNTERROR)—

Offset 19Ch
Number of frames received with runt ((64 bytes and CRC error) error.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.54 MMC Receive Jabber Error Frame Counter Register

(RXJABBERERROR)—Offset 1A0h
Number of giant frames received with length (including CRC) greater than 1,518 bytes
(1,522 bytes for VLAN tagged) and with CRC error. If Jumbo Frame mode is enabled,
then frames of length greater than 9,018 bytes (9,022 for VLAN tagged) are considered
as giant frames.
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1A0h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.55 MMC Receive Undersize Good Frame Counter Register

(RXUNDERSIZE_G)—Offset 1A4h
Number of frames received with length less than 64 bytes, without any errors.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.56 MMC Receive Oversize Good Frame Counter Register

(RXOVERSIZE_G)—Offset 1A8h
Number of frames received without errors, with length greater than the maxsize (1,518
or 1,522 for VLAN tagged frames; 2,000 bytes if enabled in Bit 27 of Register 0 (MAC
Configuration Register)).
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.57 MMC Receive 64 Octet Good Bad Frame Counter Register

(RX64OCTETS_GB)—Offset 1ACh
Number of good and bad frames received with length 64 bytes, exclusive of preamble.
Access Method
(Size: 32 bits) Offset: [BAR0] + 1ACh

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
(RX65TO127OCTETS_GB)—Offset 1B0h
Number of good and bad frames received with length between 65 and 127 (inclusive)
bytes, exclusive of preamble.
Access Method
(Size: 32 bits) Offset: [BAR0] + 1B0h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.59 MMC Receive 128 to 255 Octet Good Bad Frame Counter
Register (RX128TO255OCTETS_GB)—Offset 1B4h
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1B4h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
Register (RX256TO511OCTETS_GB)—Offset 1B8h
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1B8h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
Register (RX512TO1023OCTETS_GB)—Offset 1BCh
Number of good and bad frames received with length between 512 and 1,023
(inclusive) bytes, exclusive of preamble.
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1BCh
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

15.6.62 MMC Receive 1024 to Maximum Octet Good Bad Frame Counter
Register (RX1024TOMAXOCTETS_GB)—Offset 1C0h
Number of good and bad frames received with length between 1,024 and maxsize
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1C0h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.63 MMC Receive Unicast Good Frame Counter Register

(RXUNICASTFRAMES_G)—Offset 1C4h
Number of received good unicast frames.
Access Method
(Size: 32 bits) Offset: [BAR0] + 1C4h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.64 MMC Receive Length Error Frame Counter Register

(RXLENGTHERROR)—Offset 1C8h
Number of frames received with length error (Length type field doesn't match frame
size), for all frames with valid length field.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.65 MMC Receive Out Of Range Error Frame Counter Register

(RXOUTOFRANGETYPE)—Offset 1CCh
Number of frames received with length field not equal to the valid frame size (greater
than 1,500 but less than 1,536).
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1CCh
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.66 MMC Receive Pause Frame Counter Register

(RXPAUSEFRAMES)—Offset 1D0h
Number of good and valid PAUSE frames received.
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1D0h
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.67 MMC Receive FIFO Overflow Frame Counter Register

(RXFIFOOVERFLOW)—Offset 1D4h
Number of missed received frames because of FIFO overflow.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.68 MMC Receive VLAN Good Bad Frame Counter Register

(RXVLANFRAMES_GB)—Offset 1D8h
Number of good and bad VLAN frames received.
Access Method
(Size: 32 bits) Offset: [BAR0] + 1D8h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
15.6.69 MMC Receive Watchdog Error Frame Counter Register

(RXWATCHDOGERROR)—Offset 1DCh
Number of frames received with error because of watchdog timeout error (frames with
a data load larger than 2,048 bytes).
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1DCh
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.70 MMC Receive Error Frame Counter Register (RXRCVERROR)—

Offset 1E0h
Number of frames received with Receive error or Frame Extension error on the MII
interface.
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + 1E0h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

15.6.71 MMC Receive Control Frame Counter Register

(RXCTRLFRAMES_G)—Offset 1E4h
Number of received good control frames.
Access Method
(Size: 32 bits) Offset: [BAR0] + 1E4h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.72 MMC IPC Receive Checksum Offload Interrupt Mask Register

(MMC_IPC_INTR_MASK_RX)—Offset 200h
The MMC IPC Receive Checksum Offload Interrupt Mask maintains the mask for the
interrupt generated from the receive IPC statistic counters.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RXUDPGFIM
RXIPV4UDSBLFIM
RXIPV4FRAGFIM
RXICMPEROIM
RXICMPGOIM
RXUDPGOIM
RXIPV6NOPAYOIM
RXIPV6HEROIM
RXIPV6GOIM
RXIPV4UDSBLOIM
RXIPV4FRAGOIM
RXIPV4NOPAYOIM
RSV0
RXTCPGOIM
RXTCPGFIM
RXUDPEROIM
RXIPV4HEROIM
RXIPV4GOIM
RXICMPERFIM
RXICMPGFIM
RXTCPERFIM
RXUDPERFIM
RXIPV6NOPAYFIM
RXIPV6HERFIM
RXIPV6GFIM
RXIPV4NOPAYFIM
RXIPV4HERFIM
RXIPV4GFIM
RXTCPEROIM
Bit Default &

Range Access
0b
RO
0b MMC Receive ICMP Error Octet Counter Interrupt Mask (RXICMPEROIM):

29 Setting this bit masks the interrupt when the rxicmp_err_octets counter reaches half of
0b MMC Receive ICMP Good Octet Counter Interrupt Mask (RXICMPGOIM): Setting
28 this bit masks the interrupt when the rxicmp_gd_octets counter reaches half of the

Bit Default &

Range Access
0b MMC Receive TCP Error Octet Counter Interrupt Mask (RXTCPEROIM): Setting
27 this bit masks the interrupt when the rxtcp_err_octets counter reaches half of the
0b MMC Receive TCP Good Octet Counter Interrupt Mask (RXTCPGOIM): Setting
26 this bit masks the interrupt when the rxtcp_gd_octets counter reaches half of the
0b MMC Receive UDP Good Octet Counter Interrupt Mask (RXUDPEROIM): Setting
25 this bit masks the interrupt when the rxudp_err_octets counter reaches half of the
0b MMC Receive IPV6 No Payload Octet Counter Interrupt Mask (RXUDPGOIM):

24 Setting this bit masks the interrupt when the rxudp_gd_octets counter reaches half of
MMC Receive IPV6 Header Error Octet Counter Interrupt Mask

0b (RXIPV6NOPAYOIM): Setting this bit masks the interrupt when the
23
RW rxipv6_nopay_octets counter reaches half of the maximum value or the maximum
value.
0b MMC Receive IPV6 Good Octet Counter Interrupt Mask (RXIPV6HEROIM):

22 Setting this bit masks the interrupt when the rxipv6_hdrerr_octets counter reaches half
0b MMC Receive IPV6 Good Octet Counter Interrupt Mask (RXIPV6GOIM): Setting
21 this bit masks the interrupt when the rxipv6_gd_octets counter reaches half of the
0b MMC Receive IPV4 UDP Checksum Disabled Octet Counter Interrupt Mask
20 (RXIPV4UDSBLOIM): Setting this bit masks the interrupt when the
RW rxipv4_udsbl_octets counter reaches half of the maximum value or the maximum value.
0b MMC Receive IPV4 Fragmented Octet Counter Interrupt Mask

19 (RXIPV4FRAGOIM): Setting this bit masks the interrupt when the rxipv4_frag_octets
MMC Receive IPV4 No Payload Octet Counter Interrupt Mask

0b (RXIPV4NOPAYOIM): Setting this bit masks the interrupt when the
18
RW rxipv4_nopay_octets counter reaches half of the maximum value or the maximum
value.
0b MMC Receive IPV4 Header Error Octet Counter Interrupt Mask

17 (RXIPV4HEROIM): Setting this bit masks the interrupt when the rxipv4_hdrerr_octets
0b MMC Receive IPV4 Good Octet Counter Interrupt Mask (RXIPV4GOIM): Setting
16 this bit masks the interrupt when the rxipv4_gd_octets counter reaches half of the
00b
RO
0b MMC Receive ICMP Error Frame Counter Interrupt Mask (RXICMPERFIM):

13 Setting this bit masks the interrupt when the rxicmp_err_frms counter reaches half of
0b MMC Receive ICMP Good Frame Counter Interrupt Mask (RXICMPGFIM): Setting
12 this bit masks the interrupt when the rxicmp_gd_frms counter reaches half of the
0b MMC Receive TCP Error Frame Counter Interrupt Mask (RXTCPERFIM): Setting
11 this bit masks the interrupt when the rxtcp_err_frms counter reaches half of the
0b MMC Receive TCP Good Frame Counter Interrupt Mask (RXTCPGFIM): Setting
10 this bit masks the interrupt when the rxtcp_gd_frms counter reaches half of the
0b MMC Receive UDP Error Frame Counter Interrupt Mask (RXUDPERFIM): Setting
9 this bit masks the interrupt when the rxudp_err_frms counter reaches half of the

Bit Default &

Range Access
0b MMC Receive UDP Good Frame Counter Interrupt Mask (RXUDPGFIM): Setting
8 this bit masks the interrupt when the rxudp_gd_frms counter reaches half of the
0b MMC Receive IPV6 No Payload Frame Counter Interrupt Mask

7 (RXIPV6NOPAYFIM): Setting this bit masks the interrupt when the
RW rxipv6_nopay_frms counter reaches half of the maximum value or the maximum value.
0b MMC Receive IPV6 Header Error Frame Counter Interrupt Mask

6 (RXIPV6HERFIM): Setting this bit masks the interrupt when the rxipv6_hdrerr_frms
0b MMC Receive IPV6 Good Frame Counter Interrupt Mask (RXIPV6GFIM): Setting
5 this bit masks the interrupt when the rxipv6_gd_frms counter reaches half of the
0b MMC Receive IPV4 UDP Checksum Disabled Frame Counter Interrupt Mask
4 (RXIPV4UDSBLFIM): Setting this bit masks the interrupt when the rxipv4_udsbl_frms
0b MMC Receive IPV4 Fragmented Frame Counter Interrupt Mask

3 (RXIPV4FRAGFIM): Setting this bit masks the interrupt when the rxipv4_frag_frms
0b MMC Receive IPV4 No Payload Frame Counter Interrupt Mask

2 (RXIPV4NOPAYFIM): Setting this bit masks the interrupt when the
RW rxipv4_nopay_frms counter reaches half of the maximum value or the maximum value.
0b MMC Receive IPV4 Header Error Frame Counter Interrupt Mask

1 (RXIPV4HERFIM): Setting this bit masks the interrupt when the rxipv4_hdrerr_frms
0b MMC Receive IPV4 Good Frame Counter Interrupt Mask (RXIPV4GFIM): Setting
0 this bit masks the interrupt when the rxipv4_gd_frms counter reaches half of the
15.6.73 MMC Receive Checksum Offload Interrupt Register

(MMC_IPC_INTR_RX)—Offset 208h
The MMC Receive Checksum Offload Interrupt register maintains the interrupts
generated when receive IPC statistic counters reach half their maximum values, and
when they cross their maximum values. When Counter Stop Rollover is set, then
interrupts are set but the counter remains at all-ones. When the MMC IPC counter that
caused the interrupt is read, its corresponding interrupt bit is cleared.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
RXICMPEROIS
RXUDPGOIS
RXIPV4FRAGOIS
RXIPV4HERFIS
RXICMPGOIS
RXIPV6NOPAYOIS
RXIPV6HEROIS
RXUDPGFIS
RXIPV6NOPAYFIS
RXIPV6HERFIS
RXTCPGFIS
RXUDPERFIS
RXIPV6GFIS
RXIPV4UDSBLFIS
RXIPV4NOPAYFIS
RXTCPEROIS
RXTCPGOIS
RXUDPEROIS
RXIPV6GOIS
RXIPV4UDSBLOIS
RXIPV4NOPAYOIS
RXIPV4HEROIS
RXIPV4GOIS
RXICMPERFIS
RXICMPGFIS
RXTCPERFIS
RXIPV4FRAGFIS
RXIPV4GFIS

Bit Default &

Range Access
0b
RO
0b MMC Receive ICMP Error Octet Counter Interrupt Status (RXICMPEROIS): This
29 bit is set when the rxicmp_err_octets counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive ICMP Good Octet Counter Interrupt Status (RXICMPGOIS): This
28 bit is set when the rxicmp_gd_octets counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive TCP Error Octet Counter Interrupt Status (RXTCPEROIS): This bit
27 is set when the rxtcp_err_octets counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive TCP Good Octet Counter Interrupt Status (RXTCPGOIS): This bit is
26 set when the rxtcp_gd_octets counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive UDP Error Octet Counter Interrupt Status (RXUDPEROIS): This bit
25 is set when the rxudp_err_octets counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive UDP Good Octet Counter Interrupt Status (RXUDPGOIS): This bit
24 is set when the rxudp_gd_octets counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive IPV6 No Payload Octet Counter Interrupt Status

23 (RXIPV6NOPAYOIS): This bit is set when the rxipv6_nopay_octets counter reaches
0b MMC Receive IPV6 Header Error Octet Counter Interrupt Status

22 (RXIPV6HEROIS): This bit is set when the rxipv6_hdrerr_octets counter reaches half
RO of the maximum value or the maximum value.
0b MMC Receive IPV6 Good Octet Counter Interrupt Status (RXIPV6GOIS): This bit
21 is set when the rxipv6_gd_octets counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive IPV4 UDP Checksum Disabled Octet Counter Interrupt Status
20 (RXIPV4UDSBLOIS): This bit is set when the rxipv4_udsbl_octets counter reaches
0b MMC Receive IPV4 Fragmented Octet Counter Interrupt Status

19 (RXIPV4FRAGOIS): This bit is set when the rxipv4_frag_octets counter reaches half of
0b MMC Receive IPV4 No Payload Octet Counter Interrupt Status

18 (RXIPV4NOPAYOIS): This bit is set when the rxipv4_nopay_octets counter reaches
0b MMC Receive IPV4 Header Error Octet Counter Interrupt Status

17 (RXIPV4HEROIS): This bit is set when the rxipv4_hdrerr_octets counter reaches half
0b MMC Receive IPV4 Good Octet Counter Interrupt Status (RXIPV4GOIS): This bit
16 is set when the rxipv4_gd_octets counter reaches half of the maximum value or the
RO maximum value.
00b
RO
0b MMC Receive ICMP Error Frame Counter Interrupt Status (RXICMPERFIS): This
13 bit is set when the rxicmp_err_frms counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive ICMP Good Frame Counter Interrupt Status (RXICMPGFIS): This
12 bit is set when the rxicmp_gd_frms counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive TCP Error Frame Counter Interrupt Status (RXTCPERFIS): This bit
11 is set when the rxtcp_err_frms counter reaches half of the maximum value or the
RO maximum value.

Bit Default &

Range Access
0b MMC Receive TCP Good Frame Counter Interrupt Status (RXTCPGFIS): This bit is
10 set when the rxtcp_gd_frms counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive UDP Error Frame Counter Interrupt Status (RXUDPERFIS): This bit
9 is set when the rxudp_err_frms counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive UDP Good Frame Counter Interrupt Status (RXUDPGFIS): This bit
8 is set when the rxudp_gd_frms counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive IPV6 No Payload Frame Counter Interrupt Status

7 (RXIPV6NOPAYFIS): This bit is set when the rxipv6_nopay_frms counter reaches half
0b MMC Receive IPV6 Header Error Frame Counter Interrupt Status

6 (RXIPV6HERFIS): This bit is set when the rxipv6_hdrerr_frms counter reaches half of
0b MMC Receive IPV6 Good Frame Counter Interrupt Status (RXIPV6GFIS): This bit
5 is set when the rxipv6_gd_frms counter reaches half of the maximum value or the
RO maximum value.
0b MMC Receive IPV4 UDP Checksum Disabled Frame Counter Interrupt Status
4 (RXIPV4UDSBLFIS): This bit is set when the rxipv4_udsbl_frms counter reaches half
0b MMC Receive IPV4 Fragmented Frame Counter Interrupt Status

3 (RXIPV4FRAGFIS): This bit is set when the rxipv4_frag_frms counter reaches half of
0b MMC Receive IPV4 No Payload Frame Counter Interrupt Status

2 (RXIPV4NOPAYFIS): This bit is set when the rxipv4_nopay_frms counter reaches half
0b MMC Receive IPV4 Header Error Frame Counter Interrupt Status

1 (RXIPV4HERFIS): This bit is set when the rxipv4_hdrerr_frms counter reaches half of
0b MMC Receive IPV4 Good Frame Counter Interrupt Status (RXIPV4GFIS): This bit
0 is set when the rxipv4_gd_frms counter reaches half of the maximum value or the
RO maximum value.

(RXIPV4_GD_FRMS)—Offset 210h
Number of good IPv4 datagrams received with the TCP, UDP, or ICMP payload.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO

(RXIPV4_HDRERR_FRMS)—Offset 214h
Number of IPv4 datagrams received with header (checksum, length, or version
mismatch) errors.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

(RXIPV4_NOPAY_FRMS)—Offset 218h
Number of IPv4 datagram frames received that did not have a TCP, UDP, or ICMP
payload processed by the Checksum engine.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

15.6.77 MMC Receive IPV4 Fragmented Frame Counter Register

(RXIPV4_FRAG_FRMS)—Offset 21Ch
Number of good IPv4 datagrams with fragmentation.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.78 MMC Receive IPV4 UDP Checksum Disabled Frame Counter

Register (RXIPV4_UDSBL_FRMS)—Offset 220h
Number of good IPv4 datagrams received that had a UDP payload with checksum
disabled.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

(RXIPV6_GD_FRMS)—Offset 224h
Number of good IPv6 datagrams received with TCP, UDP, or ICMP payloads.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO

(RXIPV6_HDRERR_FRMS)—Offset 228h
Number of IPv6 datagrams received with header errors (length or version mismatch).
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

(RXIPV6_NOPAY_FRMS)—Offset 22Ch
Number of IPv6 datagram frames received that did not have a TCP, UDP, or ICMP
payload. This includes all IPv6 datagrams with fragmentation or security extension
headers.
Access Method

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.82 MMC Receive UDP Good Frame Counter Register

(RXUDP_GD_FRMS)—Offset 230h
Number of good IP datagrams with a good UDP payload. This counter is not updated
when the rxipv4_udsbl_frms counter is incremented.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.83 MMC Receive UDP Error Frame Counter Register

(RXUDP_ERR_FRMS)—Offset 234h
Number of good IP datagrams whose UDP payload has a checksum error.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
15.6.84 MMC Receive TCP Good Frame Counter Register

(RXTCP_GD_FRMS)—Offset 238h
Number of good IP datagrams with a good TCP payload.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.85 MMC Receive TCP Error Frame Counter Register

(RXTCP_ERR_FRMS)—Offset 23Ch
Number of good IP datagrams whose TCP payload has a checksum error.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

15.6.86 MMC Receive ICMP Good Frame Counter Register

(RXICMP_GD_FRMS)—Offset 240h
Number of good IP datagrams with a good ICMP payload.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.87 MMC Receive ICMP Error Frame Counter Register

(RXICMP_ERR_FRMS)—Offset 244h
Number of good IP datagrams whose ICMP payload has a checksum error.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

(RXIPV4_GD_OCTETS)—Offset 250h
Number of bytes received in good IPv4 datagrams encapsulating TCP, UDP, or ICMP
data. (Ethernet header, FCS, pad, or IP pad bytes are not included in this counter or in
the octet counters listed below).
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO

(RXIPV4_HDRERR_OCTETS)—Offset 254h
Number of bytes received in IPv4 datagrams with header errors (checksum, length,
version mismatch). The value in the Length field of IPv4 header is used to update this
counter.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

(RXIPV4_NOPAY_OCTETS)—Offset 258h
Number of bytes received in IPv4 datagrams that did not have a TCP, UDP, or ICMP
payload. The value in the IPv4 headers Length field is used to update this counter.
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.91 MMC Receive IPV4 Fragmented Octet Counter Register

(RXIPV4_FRAG_OCTETS)—Offset 25Ch
Number of bytes received in fragmented IPv4 datagrams. The value in the IPv4
headers Length field is used to update this counter.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.92 MMC Receive IPV4 UDP Checksum Disabled Octet Counter

Register (RXIPV4_UDSBL_OCTETS)—Offset 260h
Number of bytes received in a UDP segment that had the UDP checksum disabled. This
counter does not count IP Header bytes.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO

(RXIPV6_GD_OCTETS)—Offset 264h
Number of bytes received in good IPv6 datagrams encapsulating TCP, UDP or ICMPv6
data.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO

(RXIPV6_HDRERR_OCTETS)—Offset 268h
Number of bytes received in IPv6 datagrams with header errors (length, version
mismatch). The value in the IPv6 headers Length field is used to update this counter.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO


(RXIPV6_NOPAY_OCTETS)—Offset 26Ch
Number of bytes received in IPv6 datagrams that did not have a TCP, UDP, or ICMP
payload. The value in the IPv6 headers Length field is used to update this counter.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO

(RXUDP_GD_OCTETS)—Offset 270h
Number of bytes received in a good UDP segment. This counter (and the counters
below) does not count IP header bytes.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.97 MMC Receive UDP Good Octet Counter Register

(RXUDP_ERR_OCTETS)—Offset 274h
Number of bytes received in a UDP segment that had checksum errors.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.98 MMC Receive TCP Good Octet Counter Register

(RXTCP_GD_OCTETS)—Offset 278h
Number of bytes received in a good TCP segment.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.99 MMC Receive TCP Error Octet Counter Register

(RXTCP_ERR_OCTETS)—Offset 27Ch
Number of bytes received in a TCP segment with checksum errors.
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &
Range Access
00000000h
RO
15.6.100 MMC Receive ICMP Good Octet Counter Register

(RXICMP_GD_OCTETS)—Offset 280h
Number of bytes received in a good ICMP segment.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT
Bit Default &

Range Access
00000000h
RO
15.6.101 MMC Receive ICMP Error Octet Counter Register

(RXICMP_ERR_OCTETS)—Offset 284h
Number of bytes received in an ICMP segment with checksum errors.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CNT

Bit Default &

Range Access
00000000h
RO
15.6.102 VLAN Tag Inclusion or Replacement Register (Register 353)

(GMAC_REG_353)—Offset 584h
The VLAN Tag Inclusion or Replacement register contains the VLAN tag for insertion or
replacement in the transmit frames.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VLP
VLC
RSV0
CSVL
VLT
Bit Default &
Range Access
000h
RO
0b C-VLAN or S-VLAN (CSVL): When this bit is set, S-VLAN type (0x88A8) is inserted or
19 replaced in the 13th and 14th bytes of transmitted frames. When this bit is reset, C-
RW VLAN type (0x8100) is inserted or replaced in the transmitted frames.
0b VLAN Priority Control (VLP): When this bit is set, the control Bits [17:16] are used
18 for VLAN deletion, insertion, or replacement. When this bit is reset, the internal control
RW signal from the MTL layer is used, and Bits [17:16] are ignored.
VLAN Tag Control in Transmit Frames (VLC): 2'b00: No VLAN tag deletion,
insertion, or replacement
2'b01: VLAN tag deletion. The MAC removes the VLAN type (bytes 13 and 14) and VLAN
tag (bytes 15 and 16) of all transmitted frames with VLAN tags.
2'b10: VLAN tag insertion. The MAC inserts VLT in bytes 15 and 16 of the frame after
00b inserting the Type value (0x8100/0x88a8) in bytes 13 and 14. This operation is
17:16 performed on all transmitted frames, irrespective of whether they already have a VLAN
RW tag.
2'b11: VLAN tag replacement. The MAC replaces VLT in bytes 15 and 16 of all VLAN-type
transmitted frames (Bytes 13 and 14 are 0x8100/0x88a8).
NOTE: Changes to this field take effect only on the start of a frame. If you write this
register field when a frame is being transmitted, only the subsequent frame can use the
updated value, that is, the current frame does not use the updated value.
VLAN Tag for Transmit Frames (VLT): This field contains the value of the VLAN tag
0000h to be inserted or replaced. The value must only be changed when the transmit lines are
15:0
RW inactive or during the initialization phase. Bits[15:13] are the User Priority, Bit 12 is the
CFI/DEI, and Bits[11:0] are the VLAN tags VID field.

15.6.103 VLAN Hash Table Register (Register 354) (GMAC_REG_354)—

Offset 588h
The 16-bit Hash table is used for group address filtering based on VLAN tag when Bit 18
(VTHM) of Register 7 (VLAN Tag Register) is set. For hash filtering, the content of the
16-bit VLAN tag or 12-bit VLAN ID (based on Bit 16 (ETV) of VLAN Tag Register) in the
incoming frame is passed through the CRC logic and the upper four bits of the
calculated CRC are used to index the contents of the VLAN Hash table. For example, a
hash value of 4b'1000 selects Bit 8 of the VLAN Hash table. The hash value of the
destination address is calculated in the following way: 1. Calculate the 32-bit CRC for
the VLAN tag or ID (See IEEE 802.3, Section 3.2.8 for the steps to calculate CRC32). 2.
Perform bitwise reversal for the value obtained in Step 1. 3. Take the upper four bits
from the value obtained in Step 2. If the corresponding bit value of the register is 1'b1,
the frame is accepted. Otherwise, it is rejected.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
VLHT
Bit Default &
Range Access
0000h
RO
0000h
15:0 VLAN Hash Table (VLHT): This field contains the 16-bit VLAN Hash Table.
RW
15.6.104 Timestamp Control Register (Register 448) (GMAC_REG_448)—

Offset 700h
This register controls the operation of the System Time generator and the processing of
PTP packets for timestamping in the Receiver.
Access Method
(Size: 32 bits)

Default: 00002000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV4
TSENMACADDR
SNAPTYPSEL
RSV0
TSADDREG
TSTRIG
TSENALL
TSMSTRENA
TSIPV4ENA
TSIPV6ENA
TSVER2ENA
TSIPENA
TSINIT
TSCFUPDT
TSENA
TSEVNTENA
TSCTRLSSR
TSUPDT

Bit Default &

Range Access
0b
RO
0b Enable MAC address for PTP Frame Filtering (TSENMACADDR): When set, the DA
18 MAC address (that matches any MAC Address register) is used to filter the PTP frames
RW when PTP is directly sent over Ethernet.
00b Select PTP packets for Taking Snapshots (SNAPTYPSEL): These bits along with
17:16
RW Bits 15 and 14 decide the set of PTP packet types for which snapshot needs to be taken.
0b Enable Snapshot for Messages Relevant to Master (TSMSTRENA): When set, the
15 snapshot is taken only for the messages relevant to the master node. Otherwise, the
RW snapshot is taken for the messages relevant to the slave node.
Enable Timestamp Snapshot for Event Messages (TSEVNTENA): When set, the
0b timestamp snapshot is taken only for event messages (SYNC, Delay_Req, Pdelay_Req,
14
RW or Pdelay_Resp). When reset, the snapshot is taken for all messages except Announce,
Management, and Signaling.
Enable Processing of PTP Frames Sent over IPv4-UDP (TSIPV4ENA): When set,
1b the MAC receiver processes the PTP packets encapsulated in UDP over IPv4 packets.
13
RW When this bit is clear, the MAC ignores the PTP transported over UDP-IPv4 packets. This
bit is set by default.
0b Enable Processing of PTP Frames Sent Over IPv6-UDP (TSIPV6ENA): When set,
12 the MAC receiver processes PTP packets encapsulated in UDP over IPv6 packets. When
RW this bit is clear, the MAC ignores the PTP transported over UDP-IPv6 packets.
0b Enable Processing of PTP over Ethernet Frames (TSIPENA): When set, the MAC
11 receiver processes the PTP packets encapsulated directly in the Ethernet frames. When
RW this bit is clear, the MAC ignores the PTP over Ethernet packets.
0b Enable PTP packet Processing for Version 2 Format (TSVER2ENA): When set, the
10 PTP packets are processed using the 1588 version 2 format. Otherwise, the PTP packets
RW are processed using the version 1 format.
Timestamp Digital or Binary Rollover Control (TSCTRLSSR): When set, the

Timestamp Low register rolls over after 0x3B9A_C9FF value (that is, 1 nanosecond
0b accuracy) and increments the timestamp (High) seconds. When reset, the rollover value
9
RW of sub-second register is 0x7FFF_FFFF. The sub-second increment has to be
programmed correctly depending on the PTP reference clock frequency and the value of
this bit.
0b Enable Timestamp for All Frames (TSENALL): When set, the timestamp snapshot is
8
RW enabled for all frames received by the MAC.
00b
RO
0b Addend Reg Update (TSADDREG): When set, the content of the Timestamp Addend
5 register is updated in the PTP block for fine correction. This is cleared when the update
RW is completed. This register bit should be zero before setting it.
Timestamp Interrupt Trigger Enable (TSTRIG): When set, the timestamp interrupt
0b is generated when the System Time becomes greater than the value written in the
4
RW Target Time register. This bit is reset after the generation of the Timestamp Trigger
Interrupt.
Timestamp Update (TSUPDT): When set, the system time is updated (added or
subtracted) with the value specified in Register 452 (System Time - Seconds Update
0b Register) and Register 453 (System Time - Nanoseconds Update Register).
3
RW This bit should be read zero before updating it. This bit is reset when the update is
completed in hardware. The Timestamp Higher Word register (if enabled during core
configuration) is not updated.
Timestamp Initialize (TSINIT): When set, the system time is initialized (overwritten)
with the value specified in the Register 452 (System Time - Seconds Update Register)
0b and Register 453 (System Time - Nanoseconds Update Register).
2
RW This bit should be read zero before updating it. This bit is reset when the initialization is
complete. The Timestamp Higher Word register (if enabled during core configuration)
can only be initialized.

Bit Default &

Range Access
0b Timestamp Fine or Coarse Update (TSCFUPDT): When set, this bit indicates that
1 the system times update should be done using the fine update method. When reset, it
RW indicates the system timestamp update should be done using the Coarse method.
Timestamp Enable (TSENA): When set, the timestamp is added for the transmit and
0b receive frames. When disabled, timestamp is not added for the transmit and receive
0 frames and the Timestamp Generator is also suspended. You need to initialize the
RW Timestamp (system time) after enabling this mode. On the receive side, the MAC
processes the 1588 frames only if this bit is set.
15.6.105 Sub-Second Increment Register (Register 449)

In the Coarse Update mode (TSCFUPDT bit in Register 448), the value in this register is
added to the system time every clock cycle of the internal 50MHz PTP reference clock.
In the Fine Update mode, the value in this register is added to the system time
whenever the Accumulator gets an overflow.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
SSINC
Bit Default &
Range Access
000000h
RO
Sub-second Increment Value (SSINC): The value programmed in this field is

accumulated every clock cycle (of clk_ptp_i) with the contents of the sub-second
00h register. For example, when PTP clock is 50 MHz (period is 20 ns), you should program
7:0 20 (0x14) when the System Time-Nanoseconds register has an accuracy of 1 ns
RW (TSCTRLSSR bit is set). When TSCTRLSSR is clear, the Nanoseconds register has a
resolution of ~0.465ns. In this case, you should program a value of 43 (0x2B) that is
derived by 20ns/0.465.
15.6.106 System Time - Seconds Register (Register 450)

The System Time -Seconds register, along with System-TimeNanoseconds register,
indicates the current value of the system time maintained by the MAC. Though it is
updated on a continuous basis, there is some delay from the actual time because of
clock domain transfer latencies.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TSS
Bit Default &
Range Access
00000000h Timestamp Second (TSS): The value in this field indicates the current value in
31:0
RO seconds of the System Time maintained by the MAC.
15.6.107 System Time - Nanoseconds Register (Register 451)

(GMAC_REG_451)—Offset 70Ch
The value in this field has the sub second representation of time, with an accuracy of
0.46 ns. When TSCTRLSSR is set, each bit represents 1 ns and the maximum value is
0x3B9A_C9FF, after which it rolls-over to zero.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
TSSS
Bit Default &

Range Access
0b
RO
Timestamp Sub Seconds (TSSS): The value in this field has the sub second
00000000h representation of time, with an accuracy of 0.46 ns. When bit 9 (TSCTRLSSR) is set in
30:0
RO Register 448 (Timestamp Control Register), each bit represents 1 ns and the maximum
value is 0x3B9A_C9FF, after which it rolls-over to zero.
15.6.108 System Time - Seconds Update Register (Register 452)

The System Time - Seconds Update register, along with the System Time -
Nanoseconds Update register, initializes or updates the system time maintained by the
MAC. You must write both of these registers before setting the TSINIT or TSUPDT bits
in the Timestamp Control register.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TSS
Bit Default &
Range Access
00000000h Timestamp Second (TSS): The value in this field indicates the time in seconds to be
31:0
RW initialized or added to the system time.
15.6.109 System Time - Nanoseconds Update Register (Register 453)

The System Time - Nanoseconds Update register, along with the System Time -
Seconds Update register, initializes or updates the system time maintained by the MAC.
You must write both of these registers before setting the TSINIT or TSUPDT bits in the
Timestamp Control register.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ADDSUB
TSSS
Bit Default &

Range Access
0b Add or subtract time (ADDSUB): When this bit is set, the time value is subtracted
31 with the contents of the update register. When this bit is reset, the time value is added
RW with the contents of the update register.
Timestamp Sub Second (TSSS): The value in this field has the sub second
00000000h representation of time, with an accuracy of 0.46 ns. When bit 9 (TSCTRLSSR) is set in
30:0
RW Register 448 (Timestamp Control Register), each bit represents 1 ns and the
programmed value should not exceed 0x3B9A_C9FF.
15.6.110 Timestamp Addend Register (Register 454) (GMAC_REG_454)—

Offset 718h
This register value is used only when the system time is configured for Fine Update
mode (TSCFUPDT bit in Register 448). This register content is added to a 32-bit
accumulator in every clock cycle (of the internal 50MHz PTP reference clock) and the
system time is updated whenever the accumulator overflows.

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TSAR
Bit Default &
Range Access
00000000h Timestamp Addend (TSAR): This field indicates the 32-bit time value to be added to
31:0
RW the Accumulator register to achieve time synchronization.
15.6.111 Target Time Seconds Register (Register 455)

(GMAC_REG_455)—Offset 71Ch
The Target Time Seconds register, along with Target Time Nanoseconds register, is used
to schedule an interrupt event (Register 458[1] when Advanced Timestamping is
enabled; otherwise, TS interrupt bit in Register14[9]) when the system time exceeds
the value programmed in these registers.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TSTR
Bit Default &

Range Access
Target Time Seconds (TSTR): This field stores the time in seconds. When the
00000000h timestamp value matches or exceeds both Target Timestamp registers, then based on
31:0
RW Bits [6:5] of Register 459 (PPS Control Register), the MAC starts or stops the PPS signal
output and generates an interrupt (if enabled).
15.6.112 Target Time Nanoseconds Register (Register 456)

This register is present only when the IEEE 1588 Timestamp feature is selected without
external timestamp input.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TRGTBUSY
TTSLO
Bit Default &
Range Access
Target Time Register Busy (TRGTBUSY): The MAC sets this bit when the PPSCMD
field (Bits[3:0]) in Register 459 (PPS Control Register) is programmed to 010 or 011.
Programming the PPSCMD field to 010 or 011, instructs the MAC to synchronize the
0b Target Time Registers to the PTP clock domain.
31 The MAC clears this bit after synchronizing the Target Time Registers to the PTP clock
RO domain The application must not update the Target Time Registers when this bit is read
as 1. Otherwise, the synchronization of the previous programmed time gets corrupted.
This bit is reserved when the Enable Flexible Pulse-Per-Second Output feature is not
selected.
Target Timestamp Low (TTSLO): The Target Time Nanoseconds register, along with
00000000h Target Time Seconds register, is used to schedule an interrupt event (Register 458[1]
30:0
RW when Advanced Timestamping is enabled; otherwise, TS interrupt bit in Register14[9])
when the system time exceeds the value programmed in these registers.
15.6.113 System Time - Higher Word Seconds Register (Register 457)

Contains the most significant 16-bits of the timestamp seconds value.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TSHWR
RSV0
Bit Default &

Range Access
0000h
RO
Timestamp Higher Word (TSHWR): This field contains the most significant 16-bits of
0000h the timestamp seconds value. The register is directly written to initialize the value. This
15:0
RW register is incremented when there is an overflow from the 32-bits of the System Time -
Seconds register.

15.6.114 Timestamp Status Register (Register 458) (GMAC_REG_458)—

Offset 728h
All non reserved bits are cleared when the host reads this register.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TSTARGT
RSV0
TSTRGTERR3
TSTARGT3
TSTRGTERR2
TSTARGT2
TSTRGTERR1
TSTARGT1
TSTRGTERR
AUXTSTRIG
TSSOVF
Bit Default &
Range Access
0b
RO
0b Timestamp Target Time Error (TSTRGTERR3): This bit is set when the target time,
9 being programmed in Register 496 and Register 497, is already elapsed. This bit is
RO/CR cleared when read by the application.
Timestamp Target Time Reached for Target Time PPS3 (TSTARGT3): When set,
0b this bit indicates that the value of system time is greater than or equal to the value
8
RO/CR specified in Register 496 (PPS3 Target Time High Register) and Register 497 (PPS3
Target Time Low Register).
6
4
0b Timestamp Target Time Error (TSTRGTERR): This bit is set when the target time,
3 being programmed in Target Time Registers, is already elapsed. This bit is cleared when
RO/CR read by the application.
0b
2 Reserved (AUXTSTRIG): Reserved.
RO
0b Timestamp Target Time Reached (TSTARGT): When set, this bit indicates that the
1 value of system time is greater or equal to the value specified in the Register 455
RO/CR (Target Time Seconds Register) and Register 456 (Target Time Nanoseconds Register).
0b Timestamp Seconds Overflow (TSSOVF): When set, this bit indicates that the
0 seconds value of the timestamp (when supporting version 2 format) has overflowed
RO/CR beyond 32'hFFFF_FFFF.

15.6.115 Bus Mode Register (Register 0) (DMA_REG_0)—Offset 1000h

The Bus Mode register establishes the bus operating modes for the DMA.
Access Method

Default: 00020101h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1
MB
AAL
PBL
DSL
RIX
PBL8X
RPBL
FB
ATDS
DA
RSV0
PRWG
USP
TXPR
PR
SWR
Bit Default &
Range Access
RIX: Rebuild INCRx Burst

When this bit is set high and the AHB master gets an EBT (Retry, Split, or Losing
bus grant), the AHB master interface rebuilds the pending beats of any burst
0b transfer initiated with INCRx. The AHB master interface rebuilds the beats with a
31 combination of specified bursts with INCRx and SINGLE. By default, the AHB
RW master interface rebuilds pending beats of an EBT with an unspecified (INCR)
burst.
This bit is valid only in the GMAC-AHB configuration. It is reserved in all other
configuration.
0b
RO
Channel Priority Weights (PRWG): This field sets the priority weights for Channel 0
during the round-robin arbitration between the DMA channels for the system bus.
00b 00: The priority weight is 1.
29:28
RO 01: The priority weight is 2.
10: The priority weight is 3.
11: The priority weight is 4.
0b Transmit Priority (TXPR): When set, this bit indicates that the transmit DMA has
27
RW higher priority than the receive DMA during arbitration for the system-side bus.
Mixed Burst (MB): When this bit is set high and the FB bit is low, the AHB Master
0b interface starts all bursts of length more than 16 with INCR (undefined burst) whereas it
26 reverts to fixed burst transfers (INCRx and SINGLE) for burst length of 16 and less.
RW NOTE: for bandwidth reason, it is recommended to avoid using mixed bursts.
Recommended setting is MB=0, FB=1.
Address Aligned Beats (AAL): When this bit is set high and the FB bit is equal to 1,
0b the AHB interface generates all bursts aligned to the start address LS bits. If the FB bit
25
RW is equal to 0, the first burst (accessing the data buffer's start address) is not aligned,
but subsequent bursts are aligned to the address.
8xPBL Mode (PBL8X): When set high, this bit multiplies the programmed PBL value
0b (Bits[22:17] and Bits[13:8]) eight times. Therefore, the DMA transfers the data in 8,
24 16, 32, 64, 128, and 256 beats depending on the PBL value.
RW NOTE: This bit function is not backward compatible. Before release 3.50a, this bit was
4xPBL.
Use Separate PBL (USP): When set high, this bit configures the Rx DMA to use the
0b value configured in Bits[22:17] as PBL. The PBL value in Bits[13:8] is applicable only to
23
RW the Tx DMA operations.
When reset to low, the PBL value in Bits[13:8] is applicable for both DMA engines.

Bit Default &

Range Access
Rx DMA PBL (RPBL): This field indicates the maximum number of beats to be
transferred in one Rx DMA transaction. This is the maximum value that is used in a
000001b single block Read or Write.
22:17 The Rx DMA always attempts to burst as specified in the RPBL bit each time it starts a
RW Burst transfer on the host bus. You can program RPBL with values of 1, 2, 4, 8, 16, and
32. Any other value results in undefined behavior. This field is valid and applicable only
when USP is set high.
Fixed Burst (FB): This bit controls whether the AHB or AXI Master interface performs
fixed burst transfers or not. When set, the AHB interface uses only SINGLE, INCR4,
0b INCR8, or INCR16 during start of the normal burst transfers. When reset, the AHB or
16
RW AXI interface uses SINGLE and INCR burst transfer operations.
NOTE: for bandwidth reason, it is recommended to avoid using mixed bursts.
Recommended setting is MB=0, FB=1.
Priority Ratio (PR): These bits control the priority ratio in the weighted round-robin
arbitration between the Rx DMA and Tx DMA. These bits are valid only when Bit 1 (DA)
is reset. The priority ratio is Rx:Tx or Tx:Rx depending on whether Bit 27 (TXPR) is reset
00b or set.
15:14
RW 00: The Priority Ratio is 1:1.
01: The Priority Ratio is 2:1.
Programmable Burst Length (PBL): These bits indicate the maximum number of
beats to be transferred in one DMA transaction. This is the maximum value that is used
in a single block Read or Write. The DMA always attempts to burst as specified in PBL
each time it starts a Burst transfer on the host bus. PBL can be programmed with
permissible values of 1, 2, 4, 8, 16, and 32. Any other value results in undefined
behavior. When USP is set high, this PBL value is applicable only for Tx DMA
000001b transactions. If the number of beats to be transferred is more than 32, then perform the
13:8
RW following steps:
1. Set the 8xPBL mode.
2. Set the PBL.
For example, if the maximum number of beats to be transferred is 64, then first set
8xPBL to 1 and then set PBL to 8.
All values up to 256 are allowed using a combination of PBL and 8xPBL. All PBL values
are supported in the full-duplex mode and half-duplex modes.
Alternate (Enhanced) Descriptor Size (ATDS): When set, the size of the alternate
descriptor increases to 32 bytes (8 DWORDS). This is required when the Advanced
Timestamp feature or the IPC Full Offload Engine (Type 2) is enabled in the receiver. The
0b enhanced descriptor is not required if the Advanced Timestamp and IPC Full Checksum
7
RW Offload (Type 2) features are not enabled. In such cases, you can use the 16 bytes
descriptor to save 4 bytes of memory.
When reset, the descriptor size reverts back to 4 DWORDs (16 bytes). This bit preserves
the backward compatibility for the descriptor size.
Descriptor Skip Length (DSL): This bit specifies the number of Word, Dword, or
00000b Lword (depending on the 32-bit, 64-bit, or 128-bit bus) to skip between two unchained
6:2 descriptors. The address skipping starts from the end of current descriptor to the start
RW of next descriptor. When the DSL value is equal to zero, then the descriptor table is
taken as contiguous by the DMA in Ring mode.
DMA Arbitration Scheme (DA): This bit specifies the arbitration scheme between the
transmit and receive paths of Channel 0.
0b 0: Weighted round-robin with Rx:Tx or Tx:Rx. The priority between the paths is
1 according to the priority specified in bits 15:14 (PR) and priority weights specified in Bit
RW 27 (TXPR).
1: Fixed priority. The transmit path has priority over receive path when Bit 27 (TXPR) is
set. Otherwise, receive path has priority over the transmit path.
Software Reset (SWR): When this bit is set, the MAC DMA Controller resets the logic
and all internal registers of the MAC. It is cleared automatically after the reset operation
1b has completed in all of the MAC clock domains. Before reprogramming any register of
0 the MAC, you should read a zero (0) value in this bit .
RW NOTE: The reset operation is completed only when all resets in all active clock domains
are de-asserted. Therefore, it is essential that all the PHY inputs clocks (applicable for
the selected PHY interface) are present for the software reset completion.

15.6.116 Transmit Poll Demand Register (Register 1) (DMA_REG_1)—

Offset 1004h
The Transmit Poll Demand register enables the Tx DMA to check whether or not the
DMA owns the current descriptor. The Transmit Poll Demand command is given to wake
up the Tx DMA if it is in the Suspend mode. The Tx DMA can go into the Suspend mode
because of an Underflow error in a transmitted frame or the unavailability of descriptors
owned by it. You can give this command anytime and the Tx DMA resets this command
when it again starts fetching the current descriptor from host memory.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TPD
Bit Default &
Range Access
Transmit Poll Demand (TPD): When these bits are written with any value, the DMA
00000000b reads the current descriptor pointed to by Register 18 (Current Host Transmit Descriptor
31:0 Register). If that descriptor is not available (owned by the Host), the transmission
RW returns to the Suspend state and the Bit 2 (TU) of Register 5 (Status Register) is
asserted. If the descriptor is available, the transmission resumes.
15.6.117 Receive Poll Demand Register (Register 2) (DMA_REG_2)—

Offset 1008h
The Receive Poll Demand register enables the receive DMA to check for new
descriptors. This command is used to wake up the Rx DMA from the Suspend state. The
RxDMA can go into the Suspend state only because of the unavailability of descriptors it
owns.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RPD

Bit Default &

Range Access
Receive Poll Demand (RPD): When these bits are written with any value, the DMA
00000000b reads the current descriptor pointed to by Register 19 (Current Host Receive Descriptor
31:0 Register). If that descriptor is not available (owned by the Host), the reception returns
RW to the Suspended state and the Bit 7 (RU) of Register 5 (Status Register) is not
asserted. If the descriptor is available, the Rx DMA returns to the active state.
15.6.118 Receive Descriptor List Address Register (Register 3)

(DMA_REG_3)—Offset 100Ch
The Receive Descriptor List Address register points to the start of the Receive
Descriptor List. The descriptor lists reside in the host's physical memory space and
must be Word, Dword, or Lword-aligned (for 32-bit, 64-bit, or 128-bit data bus). The
DMA internally converts it to bus width aligned address by making the corresponding
LS bits low. Writing to this register is permitted only when reception is stopped. When
stopped, this register must be written to before the receive Start command is given.
You can write to this register only when Rx DMA has stopped, that is, Bit 1 (SR) is set
to zero in Register 6 (Operation Mode Register). When stopped, this register can be
written with a new descriptor list address. When you set the SR bit to 1, the DMA takes
the newly programmed descriptor base address. If this register is not changed when
the SR bit is set to 0, then the DMA takes the descriptor address where it was stopped
earlier.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
RDESLA_32BIT
Bit Default &

Range Access
Start of Receive List (RDESLA_32BIT): This field contains the base address of the
00000000h first descriptor in the Receive Descriptor list. The LSB bits (1:0) for 32-bit bus width are
31:2
RW ignored and internally taken as all-zero by the DMA. Therefore, these LSB bits are read-
only (RO).
00b
RO
15.6.119 Transmit Descriptor List Address Register (Register 4)

(DMA_REG_4)—Offset 1010h
The Transmit Descriptor List Address register points to the start of the Transmit
Descriptor List. The descriptor lists reside in the host's physical memory space and
must be Word, Dword, or Lword-aligned (for 32-bit, 64-bit, or 128-bit data bus). The
DMA internally converts it to bus width aligned address by making the corresponding

LSB to low. You can write to this register only when the Tx DMA has stopped, that is, Bit
13 (ST) is set to zero in Register 6 (Operation Mode Register). When stopped, this
register can be written with a new descriptor list address. When you set the ST bit to 1,
the DMA takes the newly programmed descriptor base address. If this register is not
changed when the ST bit is set to 0, then the DMA takes the descriptor address where
it was stopped earlier.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
TDESLA_32BIT
Bit Default &

Range Access
Start of Transmit List (TDESLA_32BIT): This field contains the base address of the
00000000h first descriptor in the Transmit Descriptor list. The LSB bits (1:0) for 32-bit bus width are
31:2
RW ignored and are internally taken as all-zero by the DMA. Therefore, these LSB bits are
read-only (RO).
00b
RO
15.6.120 Status Register (Register 5) (DMA_REG_5)—Offset 1014h

The Status register contains all status bits that the DMA reports to the host. The
Software driver reads this register during an interrupt service routine or polling. Most of
the fields in this register cause the host to be interrupted. The bits of this register are
not cleared when read. Writing 1'b1 to (unreserved) Bits[16:0] of this register clears
these bits and writing 1'b0 has no effect. Each field (Bits[16:0]) can be masked by
masking the appropriate bit in Register 7 (Interrupt Enable Register).
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
UNF
OVF
EB
TS
TJT
RS
NIS
AIS
RWT
RPS
TPS
RSV2
TTI
RSV1
GMI
GLI
FBI
ERI
RSV0
ETI
RU
RI
TU
TI

Bit Default &

Range Access
0b
RO
Timestamp Trigger Interrupt (TTI): This bit indicates an interrupt event in the
0b Timestamp Generator block of MAC. The software must read the corresponding registers
29
RO in the MAC to get the exact cause of interrupt and clear its source to reset this bit to
1'b0. When this bit is high, the interrupt signal from the MAC subsystem is high.
0b
RO
MAC MMC Interrupt (GMI): This bit reflects an interrupt event in the MAC
0b Management Counters (MMC) module. The software must read the corresponding
27 registers in the MAC to get the exact cause of interrupt and clear the source of interrupt
RO to make this bit as 1'b0. The interrupt signal from the MAC subsystem is high when this
bit is high.
0b
26 Reserved (GLI): Reserved.
RO
Error Bits (EB): This field indicates the type of error that caused a Bus Error, for
example, error response on the AHB or AXI interface. This field is valid only when Bit 13
(FBI) is set. This field does not generate an interrupt.
* Bit 23
1'b1: Error during data transfer by the Tx DMA
000b 1'b0: Error during data transfer by the Rx DMA
25:23
RO * Bit 24
1'b1: Error during read transfer
1'b0: Error during write transfer
* Bit 25
1'b1: Error during descriptor access
1'b0: Error during data buffer access
Transmit Process State (TS): This field indicates the Transmit DMA FSM state. This
field does not generate an interrupt.
3'b000: Stopped; Reset or Stop Transmit Command issued
3'b001: Running; Fetching Transmit Transfer Descriptor
000b 3'b010: Running; Waiting for status
22:20 3'b011: Running; Reading Data from host memory buffer and queuing it to transmit
RO buffer (Tx FIFO)
3'b100: TIME_STAMP write state
3'b101: Reserved for future use
3'b110: Suspended; Transmit Descriptor Unavailable or Transmit Buffer Underflow
3'b111: Running; Closing Transmit Descriptor
Received Process State (RS): This field indicates the Receive DMA FSM state. This
field does not generate an interrupt.
3'b000: Stopped: Reset or Stop Receive Command issued
3'b001: Running: Fetching Receive Transfer Descriptor
000b 3'b010: Reserved for future use
19:17 3'b011: Running: Waiting for receive packet
RO 3'b100: Suspended: Receive Descriptor Unavailable
3'b101: Running: Closing Receive Descriptor
3'b110: TIME_STAMP write state
3'b111: Running: Transferring the receive packet data from receive buffer to host
memory
Normal Interrupt Summary (NIS): Normal Interrupt Summary bit value is the logical
OR of the following when the corresponding interrupt bits are enabled in Register 7
(Interrupt Enable Register):
Register 5[0]: Transmit Interrupt
0b Register 5[2]: Transmit Buffer Unavailable
16 Register 5[6]: Receive Interrupt
RW Register 5[14]: Early Receive Interrupt
Only unmasked bits (interrupts for which interrupt enable is set in Register 7) affect the
Normal Interrupt Summary bit.
This is a sticky bit and must be cleared (by writing 1 to this bit) each time a
corresponding bit, which causes NIS to be set, is cleared.

Bit Default &

Range Access
Abnormal Interrupt Summary (AIS): Abnormal Interrupt Summary bit value is the
logical OR of the following when the corresponding interrupt bits are enabled in Register
7 (Interrupt Enable Register):
Register 5[1]: Transmit Process Stopped
Register 5[3]: Transmit Jabber Timeout
Register 5[4]: Receive FIFO Overflow
0b Register 5[5]: Transmit Underflow
15 Register 5[7]: Receive Buffer Unavailable
RW Register 5[8]: Receive Process Stopped
Register 5[9]: Receive Watchdog Timeout
Register 5[10]: Early Transmit Interrupt
Register 5[13]: Fatal Bus Error
Only unmasked bits affect the Abnormal Interrupt Summary bit.
This is a sticky bit and must be cleared each time a corresponding bit, which causes AIS
to be set, is cleared.
0b Early Receive Interrupt (ERI): This bit indicates that the DMA had filled the first data
14
RW buffer of the packet. Bit 6 (RI) of this register automatically clears this bit.
0b Fatal Bus Error Interrupt (FBI): This bit indicates that a bus error occurred, as
13 described in Bits[25:23]. When this bit is set, the corresponding DMA engine disables all
RW of its bus accesses.
00b
RO
0b Early Transmit Interrupt (ETI): This bit indicates that the frame to be transmitted is
10
RW fully transferred to the MTL Transmit FIFO.
0b Receive Watchdog Timeout (RWT): This bit is asserted when a frame with length
9
RW greater than 2,048 bytes is received (10, 240 when Jumbo Frame mode is enabled).
0b Receive Process Stopped (RPS): This bit is asserted when the Receive Process enters
8
RW the Stopped state.
Receive Buffer Unavailable (RU): This bit indicates that the host owns the Next
Descriptor in the Receive List and the DMA cannot acquire it. The Receive Process is
0b suspended. To resume processing Receive descriptors, the host should change the
7 ownership of the descriptor and issue a Receive Poll Demand command. If no Receive
RW Poll Demand is issued, the Receive Process resumes when the next recognized incoming
frame is received. This bit is set only when the previous Receive Descriptor is owned by
the DMA.
Receive Interrupt (RI): This bit indicates that the frame reception is complete. When
0b reception is complete, the Bit 31 of RDES1 (Disable Interrupt on Completion) is reset in
6
RW the last Descriptor, and the specific frame status information is updated in the
descriptor. The reception remains in the Running state.
0b Transmit Underflow (UNF): This bit indicates that the Transmit Buffer had an
5 Underflow during frame transmission. Transmission is suspended and an Underflow
RW Error TDES0[1] is set.
0b Receive Overflow (OVF): This bit indicates that the Receive Buffer had an Overflow
4 during frame reception. If the partial frame is transferred to the application, the
RW overflow status is set in RDES0[11].
Transmit Jabber Timeout (TJT): This bit indicates that the Transmit Jabber Timer
0b expired, which happens when the frame size exceeds 2,048 (10,240 bytes when the
3 Jumbo frame is enabled). When the Jabber Timeout occurs, the transmission process is
RW aborted and placed in the Stopped state. This causes the Transmit Jabber Timeout
TDES0[14] flag to assert.
Transmit Buffer Unavailable (TU): This bit indicates that the host owns the Next
0b Descriptor in the Transmit List and the DMA cannot acquire it. Transmission is
2 suspended. Bits[22:20] explain the Transmit Process state transitions.
RW To resume processing Transmit descriptors, the host should change the ownership of the
descriptor by setting TDES0[31] and then issue a Transmit Poll Demand command.
0b
1 Transmit Process Stopped (TPS): This bit is set when the transmission is stopped.
RW

Bit Default &

Range Access
Transmit Interrupt (TI): This bit indicates that the frame transmission is complete.
0b When transmission is complete, the Bit 31 (Interrupt on Completion) of TDES1 is reset
0
RW in the first descriptor, and the specific frame status information is updated in the
descriptor.
15.6.121 Operation Mode Register (Register 6) (DMA_REG_6)—Offset

1018h
The Operation Mode register establishes the Transmit and Receive operating modes and
commands. This register should be the last CSR to be written as part of the DMA
initialization.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV8
EFC
RTC
RSF
DFF
RFA_2
RFD_2
TSF
FTF
RSV5
TTC
FEF
FUF
RSV1
OSF
RSV0
SR
DT
ST
RFD
RFA
Bit Default &
Range Access
0h
RO
Disable Dropping of TCP/IP Checksum Error Frames (DT): When this bit is set,
0b the MAC does not drop the frames which only have errors detected by the Receive
26 Checksum Offload engine. Such frames do not have any errors (including FCS error) in
RW the Ethernet frame received by the MAC but have errors only in the encapsulated
payload. When this bit is reset, all error frames are dropped if the FEF bit is reset.
Receive Store and Forward (RSF): When this bit is set, the MTL reads a frame from
0b the Rx FIFO only after the complete frame has been written to it, ignoring the RTC bits.
25
RW When this bit is reset, the Rx FIFO operates in the cut-through mode, subject to the
threshold specified by the RTC bits.
0b Disable Flushing of Received Frames (DFF): When this bit is set, the Rx DMA does
24 not flush any frames because of the unavailability of receive descriptors or buffers as it
RW does normally when this bit is reset.
MSB of Threshold for Activating Flow Control (RFA_2): If the Rx FIFO depth is 8
KB or more, this bit (when set) provides additional threshold levels for activating the
flow control in both half-duplex and full-duplex modes. This bit (as Most Significant Bit)
0b along with the RFA (Bits[10:9]) gives the following thresholds for activating flow
23 control:
RW 100: Full minus 5 KB, that is, FULL - 5KB
101: Full minus 6 KB, that is, FULL - 6KB
111: Reserved

Bit Default &

Range Access
MSB of Threshold for Deactivating Flow Control (RFD_2): If the Rx FIFO size is 8
KB or more, this bit (when set) provides additional threshold levels for deactivating the
flow control in both half-duplex and full-duplex modes. This bit (as Most Significant Bit)
0b along with the RFD (Bits[12:11]) gives the following thresholds for deactivating flow
22 control:
111: Reserved
Transmit Store and Forward (TSF): When this bit is set, transmission starts when a
0b full frame resides in the MTL Transmit FIFO. When this bit is set, the TTC values specified
21
RW in Bits[16:14] are ignored. This bit should be changed only when the transmission is
stopped.
Flush Transmit FIFO (FTF): When this bit is set, the transmit FIFO controller logic is
reset to its default values and thus all data in the Tx FIFO is lost or flushed. This bit is
cleared internally when the flushing operation is completed. The Operation Mode
0b register should not be written to until this bit is cleared. The data which is already
20 accepted by the MAC transmitter is not flushed. It is scheduled for transmission and
RW results in underflow and runt frame transmission.
NOTE: The flush operation is complete only when the Tx FIFO is emptied of its contents
and all the pending Transmit Status of the transmitted frames are accepted by the host.
To complete this flush operation, the PHY transmit clock is required to be active.
000b
RO
Transmit Threshold Control (TTC): These bits control the threshold level of the MTL
Transmit FIFO. Transmission starts when the frame size within the MTL Transmit FIFO is
larger than the threshold. In addition, full frames with a length less than the threshold
are also transmitted. These bits are used only when Bit 21 (TSF) is reset.
000: 64
000b 001: 128
16:14
RW 010: 192
011: 256
100: 40
101: 32
110: 24
111: 16
Start or Stop Transmission Command (ST): When this bit is set, transmission is
placed in the Running state, and the DMA checks the Transmit List at the current
position for a frame to be transmitted. Descriptor acquisition is attempted either from
the current position in the list, which is the Transmit List Base Address set by Register 4
(Transmit Descriptor List Address Register), or from the position retained when
transmission was stopped previously. If the DMA does not own the current descriptor,
transmission enters the Suspended state and Bit 2 (Transmit Buffer Unavailable) of
Register 5 (Status Register) is set. The Start Transmission command is effective only
0b when transmission is stopped. If the command is issued before setting Register 4
13
RW (Transmit Descriptor List Address Register), then the DMA behavior is unpredictable.
When this bit is reset, the transmission process is placed in the Stopped state after
completing the transmission of the current frame. The Next Descriptor position in the
Transmit List is saved, and it becomes the current position when transmission is
restarted. To change the list address, you need to program Register 4 (Transmit
Descriptor List Address Register) with a new value when this bit is reset. The new value
is considered when this bit is set again. The stop transmission command is effective only
when the transmission of the current frame is complete or the transmission is in the
Suspended state.
Threshold for Deactivating Flow Control (RFD): These bits control the threshold
(Fill-level of Rx FIFO) at which the flow control is de-asserted after activation (in half-
duplex and full-duplex).
00b 00: Full minus 1 KB, that is, FULL - 1KB
12:11 01: Full minus 2 KB, that is, FULL - 2KB
The de-assertion is effective only after flow control is asserted. If the Rx FIFO is 8 KB or
more, an additional bit (RFD[2]) is used for more threshold levels as described in Bit 22.

Bit Default &

Range Access
Threshold for Activating Flow Control (RFA): These bits control the threshold (Fill
level of Rx FIFO) at which the flow control is activated (in half-duplex and full-duplex).
00b 01: Full minus 2 KB, that is, FULL - 2KB
10:9 10: Full minus 3 KB, that is, FULL - 3KB
These values only apply to Rx FIFOs of 4 KB or more when the EFC bit is set high. If the
Rx FIFO is 8 KB or more, an additional bit (RFA[2]) is used for more threshold levels as
described in Bit 23.
Enable HW Flow Control (EFC): When this bit is set, the flow control signal operation
0b based on the fill-level of Rx FIFO is enabled. When reset, the flow control operation is
8
RW disabled. This bit is not used (reserved and always reset) when the Rx FIFO is less than
4 KB.
Forward Error Frames (FEF): When this bit is reset, the Rx FIFO drops frames with
error status (CRC error, collision error, GMII_ER, giant frame, watchdog timeout, or
overflow). However, if the start byte (write) pointer of a frame is already transferred to
0b the read controller side (in Threshold mode), then the frame is not dropped.
7 When the FEF bit is set, all frames except runt error frames are forwarded to the DMA. If
RW the Bit 25 (RSF) is set and the Rx FIFO overflows when a partial frame is written, then
the frame is dropped irrespective of the FEF bit setting. However, if the Bit 25 (RSF) is
reset and the Rx FIFO overflows when a partial frame is written, then a partial frame
may be forwarded to the DMA.
Forward Undersized Good Frames (FUF): When set, the Rx FIFO forwards
Undersized frames (frames with no Error and length less than 64 bytes) including pad-
0b bytes and CRC.
6
RW When reset, the Rx FIFO drops all frames of less than 64 bytes, unless a frame is
already transferred because of the lower value of Receive Threshold, for example, RTC =
01.
0b
RO
Receive Threshold Control (RTC): These two bits control the threshold level of the
MTL Receive FIFO. Transfer (request) to DMA starts when the frame size within the MTL
Receive FIFO is larger than the threshold. In addition, full frames with length less than
the threshold are transferred automatically.
00b These bits are valid only when the RSF bit is zero, and are ignored when the RSF bit is
4:3
RW set to 1.
00: 64
01: 32
10: 96
11: 128
0b Operate on Second Frame (OSF): When this bit is set, it instructs the DMA to process
2 the second frame of the Transmit data even before the status for the first frame is
RW obtained.
Start or Stop Receive (SR): When this bit is set, the Receive process is placed in the
Running state. The DMA attempts to acquire the descriptor from the Receive list and
processes the incoming frames. The descriptor acquisition is attempted from the current
position in the list, which is the address set by Register 3 (Receive Descriptor List
Address Register) or the position retained when the Receive process was previously
stopped. If the DMA does not own the descriptor, reception is suspended and Bit 7
0b (Receive Buffer Unavailable) of Register 5 (Status Register) is set. The Start Receive
1 command is effective only when the reception has stopped. If the command is issued
RW before setting Register 3 (Receive Descriptor List Address Register), the DMA behavior is
unpredictable.
When this bit is cleared, the Rx DMA operation is stopped after the transfer of the
current frame. The next descriptor position in the Receive list is saved and becomes the
current position after the Receive process is restarted. The Stop Receive command is
effective only when the Receive process is in either the Running (waiting for receive
packet) or in the Suspended state.
0b
RO

15.6.122 Interrupt Enable Register (Register 7) (DMA_REG_7)—Offset

101Ch
The Interrupt Enable register enables the interrupts reported by Register 5 (Status
Register). Setting a bit to 1'b1 enables a corresponding interrupt. After a hardware or
software reset, all interrupts are disabled.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
NIE
AIE
FBE
ERE
ETE
RWE
RSE
RUE
RIE
UNE
OVE
TJE
TUE
TSE
TIE
Bit Default &
Range Access
0000h
RO
Normal Interrupt Summary Enable (NIE): When this bit is set, normal interrupt
summary is enabled. When this bit is reset, normal interrupt summary is disabled. This
0b bit enables the following interrupts in Register 5 (Status Register):
16 Register 5[0]: Transmit Interrupt
RW Register 5[2]: Transmit Buffer Unavailable
Register 5[6]: Receive Interrupt
Register 5[14]: Early Receive Interrupt
Abnormal Interrupt Summary Enable (AIE): When this bit is set, abnormal
interrupt summary is enabled. When this bit is reset, the abnormal interrupt summary is
disabled. This bit enables the following interrupts in Register 5 (Status Register):
Register 5[1]: Transmit Process Stopped
Register 5[3]: Transmit Jabber Timeout
0b Register 5[4]: Receive Overflow
15
RW Register 5[5]: Transmit Underflow
Register 5[7]: Receive Buffer Unavailable
Register 5[8]: Receive Process Stopped
Register 5[9]: Receive Watchdog Timeout
Register 5[10]: Early Transmit Interrupt
Register 5[13]: Fatal Bus Error
0b Early Receive Interrupt Enable (ERE): When this bit is set with Normal Interrupt
14 Summary Enable (Bit 16), the Early Receive Interrupt is enabled. When this bit is reset,
RW the Early Receive Interrupt is disabled.
0b Fatal Bus Error Enable (FBE): When this bit is set with Abnormal Interrupt Summary
13 Enable (Bit 15), the Fatal Bus Error Interrupt is enabled. When this bit is reset, the Fatal
RW Bus Error Enable Interrupt is disabled.
00b
RO
0b Early Transmit Interrupt Enable (ETE): When this bit is set with an Abnormal
10 Interrupt Summary Enable (Bit 15), the Early Transmit Interrupt is enabled. When this
RW bit is reset, the Early Transmit Interrupt is disabled.
0b Receive Watchdog Timeout Enable (RWE): When this bit is set with Abnormal
9 Interrupt Summary Enable (Bit 15), the Receive Watchdog Timeout Interrupt is enabled.
RW When this bit is reset, the Receive Watchdog Timeout Interrupt is disabled.
0b Receive Stopped Enable (RSE): When this bit is set with Abnormal Interrupt
8 Summary Enable (Bit 15), the Receive Stopped Interrupt is enabled. When this bit is
RW reset, the Receive Stopped Interrupt is disabled.

Bit Default &

Range Access
0b Receive Buffer Unavailable Enable (RUE): When this bit is set with Abnormal
7 Interrupt Summary Enable (Bit 15), the Receive Buffer Unavailable Interrupt is enabled.
RW When this bit is reset, the Receive Buffer Unavailable Interrupt is disabled.
0b Receive Interrupt Enable (RIE): When this bit is set with Normal Interrupt Summary
6 Enable (Bit 16), the Receive Interrupt is enabled. When this bit is reset, the Receive
RW Interrupt is disabled.
0b Underflow Interrupt Enable (UNE): When this bit is set with Abnormal Interrupt
5 Summary Enable (Bit 15), the Transmit Underflow Interrupt is enabled. When this bit is
RW reset, the Underflow Interrupt is disabled.
0b Overflow Interrupt Enable (OVE): When this bit is set with Abnormal Interrupt
4 Summary Enable (Bit 15), the Receive Overflow Interrupt is enabled. When this bit is
RW reset, the Overflow Interrupt is disabled.
0b Transmit Jabber Timeout Enable (TJE): When this bit is set with Abnormal Interrupt
3 Summary Enable (Bit 15), the Transmit Jabber Timeout Interrupt is enabled. When this
RW bit is reset, the Transmit Jabber Timeout Interrupt is disabled.
0b Transmit Buffer Unavailable Enable (TUE): When this bit is set with Normal
2 Interrupt Summary Enable (Bit 16), the Transmit Buffer Unavailable Interrupt is
RW enabled. When this bit is reset, the Transmit Buffer Unavailable Interrupt is disabled.
0b Transmit Stopped Enable (TSE): When this bit is set with Abnormal Interrupt
1 Summary Enable (Bit 15), the Transmission Stopped Interrupt is enabled. When this bit
RW is reset, the Transmission Stopped Interrupt is disabled.
0b Transmit Interrupt Enable (TIE): When this bit is set with Normal Interrupt
0 Summary Enable (Bit 16), the Transmit Interrupt is enabled. When this bit is reset, the
RW Transmit Interrupt is disabled.
15.6.123 Missed Frame and Buffer Overflow Counter Register (Register

8) (DMA_REG_8)—Offset 1020h
The DMA maintains two counters to track the number of frames missed during
reception.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV0
OVFCNTOVF
MISCNTOVF
OVFFRMCNT
MISFRMCNT
Bit Default &

Range Access
000b
RO
0b
28 FIFO Overflow Counter Overflow (OVFCNTOVF): Reserved.
RO

Bit Default &

Range Access
FIFO Overflow Counter (OVFFRMCNT): This field indicates the number of frames
000h missed by the application due to buffer overflow conditions and runt frames (good
27:17
RO frames of less than 64 bytes) dropped by the MTL. The counter is cleared when this
register is read with the LS Byte enabled.
0b
16 Missed Frame Counter Overflow (MISCNTOVF): Reserved.
RO
Missed Frame Counter (MISFRMCNT): This field indicates the number of frames
0000h missed by the controller because of the Host Receive Buffer being unavailable. This
15:0
RO counter is incremented each time the DMA discards an incoming frame. The counter is
cleared when this register is read with the LS Byte enabled.
15.6.124 Receive Interrupt Watchdog Timer Register (Register 9)

This register, when written with non-zero value, enables the watchdog timer for the
Receive Interrupt (Bit 6) of Register 5 (Status Register).
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RIWT
RSV0
Bit Default &

Range Access
000000h
RO
RI Watchdog Timer Count (RIWT): This bit indicates the number of system clock
cycles multiplied by 256 for which the watchdog timer is set. The watchdog timer gets
00h triggered with the programmed value after the Rx DMA completes the transfer of a
7:0 frame for which the RI status bit is not set because of the setting in the corresponding
RW descriptor RDES1[31]. When the watchdog timer runs out, the RI bit is set and the timer
is stopped. The watchdog timer is reset when the RI bit is set high because of automatic
setting of RI as per RDES1[31] of any received frame.
15.6.125 AHB Status Register (Register 11) (DMA_REG_11)—Offset

102Ch
This register provides the active status of the AHB master interface interface's read and
write channels. This register is useful for debugging purposes.
Access Method


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
AXWHSTS
RSV0
Bit Default &
Range Access
00000000h
RO
0b AHB Master Status (AXWHSTS): This bit indicates that the AHB master interface
0
RO FSMs are in the non-idle state.
15.6.126 Current Host Transmit Descriptor Register (Register 18)

The Current Host Transmit Descriptor register points to the start address of the current
Transmit Descriptor read by the DMA.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CURTDESAPTR
Bit Default &

Range Access
00000000h Host Transmit Descriptor Address Pointer (CURTDESAPTR): Cleared on Reset.

31:0
RO Pointer updated by the DMA during operation.
15.6.127 Current Host Receive Descriptor Register (Register 19)

(DMA_REG_19)—Offset 104Ch
The Current Host Receive Descriptor register points to the start address of the current
Receive Descriptor read by the DMA.
Access Method


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CURRDESAPTR
Bit Default &
Range Access
00000000h Host Receive Descriptor Address Pointer (CURRDESAPTR): Cleared on Reset.

31:0
RO Pointer updated by the DMA during operation.
15.6.128 Current Host Transmit Buffer Address Register (Register 20)

The Current Host Transmit Buffer Address register points to the current Transmit Buffer
Address being read by the DMA.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CURTBUFAPTR
Bit Default &

Range Access
00000000h Host Transmit Buffer Address Pointer (CURTBUFAPTR): Cleared on Reset. Pointer
31:0
RO updated by the DMA during operation.
15.6.129 Current Host Receive Buffer Address Register (Register 21)

The Current Host Receive Buffer Address register points to the current Receive Buffer
address being read by the DMA.
Access Method
(Size: 32 bits)


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CURRBUFAPTR
Bit Default &
Range Access
00000000h Host Receive Buffer Address Pointer (CURRBUFAPTR): Cleared on Reset. Pointer
31:0
RO updated by the DMA during operation.
15.6.130 HW Feature Register (Register 22) (DMA_REG_22)—Offset

1058h
This register indicates the presence of the optional features or functions of the MAC.
Set field indicates the feature is supported. The software driver can use this register to
dynamically enable or disable the programs related to the optional blocks.
Access Method
(Size: 32 bits)

Default: 4B0F3915h
31 28 24 20 16 12 8 4 0
0 1 0 0 1 0 1 1 0 0 0 0 1 1 1 1 0 0 1 1 1 0 0 1 0 0 0 1 0 1 0 1
FLEXIPPSEN
RSV0
MMCSEL
MGKSEL
L3L4FLTREN
ADDMACADRSEL
GMIISEL
MIISEL
ACTPHYIF
SAVLANINS
INTTSEN
ENHDESSEL
AVSEL
TSVER2SEL
TSVER1SEL
RWKSEL
SMASEL
RXFIFOSIZE
TXCOESEL
EEESEL
PCSSEL
HASHSEL
EXTHASHEN
HDSEL
TXCHCNT
RXCHCNT
RXTYP2COE
RXTYP1COE
Bit Default &

Range Access
0b
RO
Active or Selected PHY interface (ACTPHYIF): This field indicates the supported
PHY interface:
0000: GMII or MII
0001: RGMII
100b 0010: SGMII
30:28 0011: TBI
RO 0100: RMII
0101: RTBI
0110: SMII
0111: RevMII
All Others: Reserved
1b
27 Source Address or VLAN Insertion (SAVLANINS): Reserved.
RO

Bit Default &

Range Access
0b
26 Flexible Pulse-Per-Second Output (FLEXIPPSEN): Reserved.
RO
1b
25 Timestamping with Internal System Time (INTTSEN): Reserved.
RO
1b
24 Alternate (Enhanced Descriptor) (ENHDESSEL): Reserved.
RO
00b
23:22 Number of additional Tx channels (TXCHCNT): Reserved.
RO
00b
21:20 Number of additional Rx channels (RXCHCNT): Reserved.
RO
1b
19 Rx FIFO > 2,048 Bytes (RXFIFOSIZE): Reserved.
RO
1b
18 IP Checksum Offload (Type 2) in Rx (RXTYP2COE): Reserved.
RO
1b
17 IP Checksum Offload (Type 1) in Rx (RXTYP1COE): Reserved.
RO
1b
16 Checksum Offload in Tx (TXCOESEL): Reserved.
RO
0b
15 AV Feature (AVSEL): Reserved.
RO
0b
14 Energy Efficient Ethernet (EEESEL): Reserved.
RO
1b
13 IEEE 1588-2008 Advanced Timestamp (TSVER2SEL): Reserved.
RO
1b
12 Only IEEE 1588-2002 Timestamp (TSVER1SEL): Reserved.
RO
1b
11 RMON Module (MMCSEL): Reserved.
RO
0b
10 PMT Magic Packet (MGKSEL): Reserved.
RO
0b
9 PMT Remote Wakeup (RWKSEL): Reserved.
RO
1b
8 SMA (MDIO) Interface (SMASEL): Reserved.
RO
0b
7 L3L4FLTREN: Reserved.
RO
0b
6 PCS registers (PCSSEL): Reserved.
RO
0b
5 Multiple MAC Address Registers (ADDMACADRSEL): Reserved.
RO
1b
4 HASH Filter (HASHSEL): Reserved.
RO
0b
3 Expanded DA Hash Filter (EXTHASHEN): Reserved.
RO
1b
2 Half-Duplex support (HDSEL): Reserved.
RO
0b
1 1000 Mbps Support (GMIISEL): Reserved.
RO

Bit Default &

Range Access
1b
0 10 and 100 Mbps Support (MIISEL): Reserved.
RO
15.7 MAC Descriptor Details

This section provides bit-field definitions of the current transmit and receive descriptor
registers described in Section 15.6, specifically:
• Current Host Transmit Descriptor Register (Register 18) (DMA_REG_18)—Offset
1048h
• Current Host Receive Descriptor Register (Register 19) (DMA_REG_19)—Offset
104Chh
15.7.1 Descriptor Overview

The descriptor structure has 8 DWORDS (32-bytes). The features of the descriptor
structure are:
• The descriptor structure is implemented to support buffers of up to 8 KB (useful for
Jumbo frames).
• There is a re-assignment of control and status bits in TDES0, TDES1, RDES0
(Advanced timestamp or IPC full offload configuration), and RDES1.
• The transmit descriptor stores the timestamp in TDES6 and TDES7.
• This receive descriptor structure is also used for storing the extended status
(RDES4) and timestamp (RDES6 and RDES7).
• You can select one of the following options for descriptor structure:
— If timestamping is enabled in Register 448 (Timestamp Control Register) or
Checksum Offload is enabled in Register 0 (MAC Configuration Register), the
software needs to allocate 32-bytes (8 DWORDS) of memory for every
descriptor. For this, the software should set Bit 7 (Alternate Descriptor Size) of
Register 0 (Bus Mode Register).
— If timestamping or Checksum Offload is not enabled, the extended descriptors
(DES4 to DES7) are not required. Therefore, the software can use alternate
descriptors with the default size of 16 bytes.
15.7.2 Descriptor Endianness

The descriptor addresses must be aligned to the bus width (Word, DWord, or LWord for
32-bit bus). The data bus is configured for little-endian format.
The structure of the descriptor with respect to the data bus endianness is as follows:
• Data Bus Endianness: Little-endian
• Descriptor Endianness: Same-endian
• Data Bus: 32-bit data bus
15.7.3 Transmit Descriptor

The transmit descriptor structure is shown in Figure 29. The application software must
program the control bits TDES0[31:18] during descriptor initialization. When the DMA
updates the descriptor, it writes back all the control bits except the OWN bit (which it

clears) and updates the status bits[7:0]. The contents of the transmitter descriptor
word 0 (TDES0) through word 3 (TDES3) are given in Table 89 through Table 91,
respectively.
The snapshot of the timestamp to be taken can be enabled for a given frame by setting
Bit 25 (TTSE) of TDES0. When the descriptor is closed (that is, when the OWN bit is
cleared), the timestamp is written into TDES6 and TDES7. This is indicated by the
status Bit 17 (TTSS) of TDES0 shown in Figure 29. The contents of TDES6 and TDES7
are mentioned in Table 93 and Table 94.
Figure 29. Transmit Descriptor Fields
The DMA always reads or fetches four DWORDS of the descriptor from system memory
to obtain the buffer and control information as shown in Figure 30. When the AV
feature is enabled, TDES0 has additional control bits[6:3] for Channel 1 and Channel 2.
For Channel 0, Bits [6:3] are ignored. Bits [6:3] are described in Table 89.

Figure 30. Transmit Descriptor Fetch (Read)
Table 89. Transmit Descriptor Word 0 (TDES0) (Sheet 1 of 3)

Bit Description
OWN: Own Bit

When set, this bit indicates that the descriptor is owned by the DMA. When this bit is reset, it
indicates that the descriptor is owned by the Host. The DMA clears this bit either when it completes
31 the frame transmission or when the buffers allocated in the descriptor are read completely. The
ownership bit of the frame’s first descriptor must be set after all subsequent descriptors belonging
to the same frame have been set. This avoids a possible race condition between fetching a
descriptor and the driver setting an ownership bit.
IC: Interrupt on Completion

30 When set, this bit sets the Transmit Interrupt (Register 5[0]) after the present frame has been
transmitted.
LS: Last Segment

29 When set, this bit indicates that the buffer contains the last segment of the frame. When this bit is
set, the TBS1 or TBS2 field in TDES1 should have a non-zero value.
FS: First Segment

28
When set, this bit indicates that the buffer contains the first segment of a frame.
DC: Disable CRC

27 When this bit is set, the MAC does not append a cyclic redundancy check (CRC) to the end of the
transmitted frame. This is valid only when the first segment (TDES0[28]) is set.
DP: Disable Pad

When set, the MAC does not automatically add padding to a frame shorter than 64 bytes. When this
26 bit is reset, the DMA automatically adds padding and CRC to a frame shorter than 64 bytes, and the
CRC field is added despite the state of the DC (TDES0[27]) bit. This is valid only when the first
segment (TDES0[28]) is set.
TTSE: Transmit Timestamp Enable

25 When set, this bit enables IEEE1588 hardware timestamping for the transmit frame referenced by
the descriptor. This field is valid only when the Enable IEEE1588 Timestamping option is selected
during core configuration and the First Segment control bit (TDES0[28]) is set.
CRCR: CRC Replacement Control

When set, the MAC replaces the last four bytes of the transmitted packet with recalculated CRC
24 bytes. The host should ensure that the CRC bytes are present in the frame being transferred from
the Transmit Buffer. This bit is valid when the Enable SA, VLAN, and CRC Insertion on TX option is
selected during core configuration and the First Segment control bit (TDES0[28]) is set.


Bit Description
CIC: Checksum Insertion Control

These bits control the checksum calculation and insertion. The following list describes the bit
encoding:
• 2’b00: Checksum Insertion Disabled.
• 2’b01: Only IP header checksum calculation and insertion are enabled.
23:22 • 2’b10: IP header checksum and payload checksum calculation and insertion are enabled, but
pseudo-header checksum is not calculated in hardware.
• 2’b11: IP Header checksum and payload checksum calculation and insertion are enabled, and
pseudo-header checksum is calculated in hardware.
This field is valid when the Enable Transmit Full TCP/IP Checksum (Type 2) option is selected during
core configuration and the First Segment control bit (TDES0[28]) is set.
TER: Transmit End of Ring

21 When set, this bit indicates that the descriptor list reached its final descriptor. The DMA returns to
the base address of the list, creating a descriptor ring.
TCH: Second Address Chained

20 When set, this bit indicates that the second address in the descriptor is the Next Descriptor address
rather than the second buffer address. When TDES0[20] is set, TBS2 (TDES1[28:16]) is a “don’t
care” value. TDES0[21] takes precedence over TDES0[20].
VLIC: VLAN Insertion Control When set, these bits request the MAC to perform VLAN tagging or
untagging before transmitting the frames. If the frame is modified for VLAN tags, the MAC
automatically recalculates and replaces the CRC bytes.
The following list describes the values of these bits:
• 2'b00: Do not add a VLAN tag.
• 2'b01: Remove the VLAN tag from the frames before transmission. This option should be used
19:18 only with the VLAN frames.
• 2'b10: Insert a VLAN tag with the tag value programmed in Register 353 (VLAN Tag Inclusion or
Replacement Register).
• 2'b11: Replace the VLAN tag in frames with the Tag value programmed in Register 353 (VLAN
Tag Inclusion or Replacement Register). This option should be used only with the VLAN frames.
These bits are valid when the Enable SA, VLAN, and CRC Insertion on TX option is selected during
core configuration and the First Segment control bit (TDES0[28]) is set.
TTSS: Transmit Timestamp Status

17 This field is used as a status bit to indicate that a timestamp was captured for the described transmit
frame. When this bit is set, TDES2 and TDES3 have a timestamp value captured for the transmit
frame. This field is only valid when the descriptor’s Last Segment control bit (TDES0[29]) is set.
IHE: IP Header Error

When set, this bit indicates that the MAC transmitter detected an error in the IP datagram header.
The transmitter checks the header length in the IPv4 packet against the number of header bytes
16 received from the application and indicates an error status if there is a mismatch. For IPv6 frames, a
header error is reported if the main header length is not 40 bytes. Furthermore, the Ethernet
Length/Type field value for an IPv4 or IPv6 frame must match the IP header version received with
the packet. For IPv4 frames, an error status is also indicated if the Header Length field has a value
less than 0x5.
ES: Error Summary

Indicates the logical OR of the following bits:
• TDES0[14]: Jabber Timeout
• TDES0[13]: Frame Flush
• TDES0[11]: Loss of Carrier
• TDES0[10]: No Carrier
15
• TDES0[9]: Late Collision
• TDES0[8]: Excessive Collision
• TDES0[2]: Excessive Deferral
• TDES0[1]: Underflow Error
• TDES0[16]: IP Header Error
• TDES0[12]: IP Payload Error
JT: Jabber Timeout

14 When set, this bit indicates the MAC transmitter has experienced a jabber time-out. This bit is only
set when Bit 22 (Jabber Disable) of Register 0 (MAC Configuration Register) is not set.


Bit Description
FF: Frame Flushed

13 When set, this bit indicates that the DMA or MTL flushed the frame because of a software Flush
command given by the CPU.
IPE: IP Payload Error

When set, this bit indicates that MAC transmitter detected an error in the TCP, UDP, or ICMP IP
datagram payload.
12
The transmitter checks the payload length received in the IPv4 or IPv6 header against the actual
number of TCP, UDP, or ICMP packet bytes received from the application and issues an error status
in case of a mismatch.
LC: Loss of Carrier

When set, this bit indicates that a loss of carrier occurred during frame transmission (that is, the
11 gmii_crs_i signal was inactive for one or more transmit clock periods during frame transmission).
This is valid only for the frames transmitted without collision when the MAC operates in the half-
duplex mode.
NC: No Carrier
10 When set, this bit indicates that the Carrier Sense signal form the PHY was not asserted during
transmission.
LC: Late Collision

9 When set, this bit indicates that frame transmission is aborted because of a collision occurring after
the collision window (64 byte-times, including preamble, in MII mode and 512 byte-times, including
preamble and carrier extension, in GMII mode). This bit is not valid if the Underflow Error bit is set.
EC: Excessive Collision

When set, this bit indicates that the transmission was aborted after 16 successive collisions while
8 attempting to transmit the current frame. If Bit 9 (Disable Retry) bit in the Register 0 (MAC
Configuration Register) is set, this bit is set after the first collision, and the transmission of the
frame is aborted.
VF: VLAN Frame

7
When set, this bit indicates that the transmitted frame is a VLAN-type frame.
CC: Collision Count (Status field)

These status bits indicate the number of collisions that occurred before the frame was transmitted.
This count is not valid when the Excessive Collisions bit (TDES0[8]) is set. The core updates this
status field only in the half-duplex mode.
-or-
6:3 SLOTNUM: Slot Number Control Bits in AV Mode
These bits indicate the slot interval in which the data should be fetched from the corresponding
buffers addressed by TDES2 or TDES3.
When the transmit descriptor is fetched, the DMA compares the slot number value in this field with
the slot interval maintained in the core (Register 11xx). It fetches the data from the buffers only if
there is a match in values. These bits are valid only for AV channels (not Channel 0).
ED: Excessive Deferral

2 When set, this bit indicates that the transmission has ended because of excessive deferral of over
24,288 bit times (155,680 bits times in 1,000-Mbps mode or if Jumbo Frame is enabled) if Bit 4
(Deferral Check) bit in Register 0 (MAC Configuration Register) is set high.
UF: Underflow Error

When set, this bit indicates that the MAC aborted the frame because the data arrived late from the
1 Host memory. Underflow Error indicates that the DMA encountered an empty transmit buffer while
transmitting the frame. The transmission process enters the Suspended state and sets both
Transmit Underflow (Register 5[5]) and Transmit Interrupt (Register 5[0]).
DB: Deferred Bit

0 When set, this bit indicates that the MAC defers before transmission because of the presence of
carrier. This bit is valid only in the half-duplex mode.

Table 90. Transmit Descriptor Word 1 (TDES1)

Bit Description
SAIC: SA Insertion Control

These bits request the MAC to add or replace the Source Address field in the Ethernet frame with
the value given in the MAC Address 0 register. If the Source Address field is modified in a frame, the
MAC automatically recalculates and replaces the CRC bytes.
The Bit 31 specifies the MAC Address Register (1 or 0) value that is used for Source Address
insertion or replacement. The following list describes the values of Bits[30:29]:
• 2'b00: Do not include the source address.
31:29 • 2'b01: Include or insert the source address. For reliable transmission, the application must
provide frames without source addresses.
• 2'b10: Replace the source address. For reliable transmission, the application must provide
frames with source addresses.
• 2'b11: Reserved
These bits are valid in the GMAC-DMA, GMAC-AXI, and GMAC-AHB configurations when the Enable
SA, VLAN, and CRC Insertion on TX is selected during core configuration and when the First
Segment control bit (TDES0[28]) is set.
TBS2: Transmit Buffer 2 Size

28:16
This field indicates the second data buffer size in bytes. This field is not valid if TDES0[20] is set.
15:13 Reserved
TBS1: Transmit Buffer 1 Size

12:0 These bits indicate the first data buffer byte size, in bytes. If this field is 0, the DMA ignores this
buffer and uses Buffer 2 or the next descriptor, depending on the value of TCH (TDES0[20]).
Table 91. Transmit Descriptor 2 (TDES2)

Bit Description
Buffer 1 Address Pointer

31:0 These bits indicate the physical address of Buffer 1. There is no limitation on the buffer address
alignment.

Bit Description
Buffer 2 Address Pointer (Next Descriptor Address)

Indicates the physical address of Buffer 2 when a descriptor ring structure is used. If the Second
31:0 Address Chained (TDES1[24]) bit is set, this address contains the pointer to the physical memory
where the Next Descriptor is present. The buffer address pointer must be aligned to the bus width
only when TDES1[24] is set. (LSBs are ignored internally.)

Bit Description
TTSL: Transmit Frame Timestamp Low

31:0 This field is updated by DMA with the least significant 32 bits of the timestamp captured for the
corresponding transmit frame. This field has the timestamp only if the Last Segment bit (LS) in the
descriptor is set and Timestamp status (TTSS) bit is set.


Bit Description
TTSH: Transmit Frame Timestamp High

31:0 This field is updated by DMA with the most significant 32 bits of the timestamp captured for the
corresponding receive frame. This field has the timestamp only if the Last Segment bit (LS) in the
descriptor is set and Timestamp status (TTSS) bit is set.
15.7.4 Receive Descriptor

The structure of the received descriptor is shown in Figure 31. It has 32 bytes of
descriptor data (8 DWORDs).
Figure 31. Receive Descriptor Fields
The contents of RDES0 are identified in Table 95. The contents of RDES1 through
RDES3 are identified in Table 96 through Table 98, respectively.
Note: Some of the bit functions of RDES0 are not backward compatible to Release 3.41a and
previous versions. These bits are Bit 7, Bit 0, and Bit 5. The function of Bit 5 is
backward compatible to Release 3.30a and previous versions.

Table 95. Receive Descriptor Fields (RDES0) (Sheet 1 of 2)

Bit Description
OWN: Own Bit

When set, this bit indicates that the descriptor is owned by the DMA of the DWC_gmac. When this
31 bit is reset, this bit indicates that the descriptor is owned by the Host. The DMA clears this bit either
when it completes the frame reception or when the buffers that are associated with this descriptor
are full.
AFM: Destination Address Filter Fail

30
When set, this bit indicates a frame that failed in the DA Filter in the MAC.
FL: Frame Length

These bits indicate the byte length of the received frame that was transferred to host memory
(including CRC). This field is valid when Last Descriptor (RDES0[8]) is set and either the Descriptor
Error (RDES0[14]) or Overflow Error bits are reset. The frame length also includes the two bytes
29:16 appended to the Ethernet frame when IP checksum calculation (Type 1) is enabled and the received
frame is not a MAC control frame.
This field is valid when Last Descriptor (RDES0[8]) is set. When the Last Descriptor and Error
Summary bits are not set, this field indicates the accumulated number of bytes that have been
transferred for the current frame.
ES: Error Summary

Indicates the logical OR of the following bits:
• RDES0[1]: CRC Error
• RDES0[3]: Receive Error
• RDES0[4]: Watchdog Timeout
15 • RDES0[6]: Late Collision
• RDES0[7]: Giant Frame
• RDES4[4:3]: IP Header or Payload Error
• RDES0[11]: Overflow Error
• RDES0[14]: Descriptor Error
This field is valid only when the Last Descriptor (RDES0[8]) is set.
DE: Descriptor Error

14 When set, this bit indicates a frame truncation caused by a frame that does not fit within the current
descriptor buffers, and that the DMA does not own the Next Descriptor. The frame is truncated. This
field is valid only when the Last Descriptor (RDES0[8]) is set.
SAF: Source Address Filter Fail

13
When set, this bit indicates that the SA field of frame failed the SA Filter in the MAC.
LE: Length Error

12 When set, this bit indicates that the actual length of the frame received and that the Length/ Type
field does not match. This bit is valid only when the Frame Type (RDES0[5]) bit is reset.
OE: Overflow Error

When set, this bit indicates that the received frame was damaged because of buffer overflow in MTL.
11 Note: This bit is set only when the DMA transfers a partial frame to the application. This happens
only when the Rx FIFO is operating in the threshold mode. In the store-and-forward mode, all
partial frames are dropped completely in Rx FIFO.
VLAN: VLAN Tag

10 When set, this bit indicates that the frame to which this descriptor is pointing is a VLAN frame
tagged by the MAC. The VLAN tagging depends on checking the VLAN fields of received frame based
on the VLAN Tag Register (Register 7) (GMAC_REG_7)—Offset 1Ch setting.
FS: First Descriptor

9 When set, this bit indicates that this descriptor contains the first buffer of the frame. If the size of
the first buffer is 0, the second buffer contains the beginning of the frame. If the size of the second
buffer is also 0, the next Descriptor contains the beginning of the frame.
LS: Last Descriptor

8 When set, this bit indicates that the buffers pointed to by this descriptor are the last buffers of the
frame.

Table 95. Receive Descriptor Fields (RDES0) (Sheet 2 of 2)

Bit Description
Timestamp Available, IP Checksum Error (Type1), or Giant Frame

When Advanced Timestamp feature is present, when set, this bit indicates that a snapshot of the
Timestamp is written in descriptor words 6 (RDES6) and 7 (RDES7). This is valid only when the Last
Descriptor bit (RDES0[8]) is set.
7 When IP Checksum Engine (Type 1) is selected, this bit, when set, indicates that the 16-bit IPv4
Header checksum calculated by the core did not match the received checksum bytes.
Otherwise, this bit, when set, indicates the Giant Frame Status. Giant frames are larger than 1,518-
byte (or 1,522-byte for VLAN or 2,000-byte when Bit 27 (2KPE) of MAC Configuration register is
set) normal frames and larger than 9,018-byte (9,022-byte for VLAN) frame when Jumbo Frame
processing is enabled.
LC: Late Collision

6 When set, this bit indicates that a late collision has occurred while receiving the frame in the half-
duplex mode.
FT: Frame Type

5 When set, this bit indicates that the Receive Frame is an Ethernet-type frame (the LT field is greater
than or equal to 16’h0600). When this bit is reset, it indicates that the received frame is an
IEEE802.3 frame. This bit is not valid for Runt frames less than 14 bytes.
RWT: Receive Watchdog Timeout

4 When set, this bit indicates that the Receive Watchdog Timer has expired while receiving the current
frame and the current frame is truncated after the Watchdog Timeout.
RE: Receive Error

3 When set, this bit indicates that the gmii_rxer_i signal is asserted while gmii_rxdv_i is asserted
during frame reception. This error also includes carrier extension error in the GMII and half-duplex
mode. Error can be of less or no extension, or error (rxd ≠ 0f) during extension.
DE: Dribble Bit Error

2 When set, this bit indicates that the received frame has a non-integer multiple of bytes (odd
nibbles). This bit is valid only in the MII Mode.
CE: CRC Error

1 When set, this bit indicates that a Cyclic Redundancy Check (CRC) Error occurred on the received
frame. This field is valid only when the Last Descriptor (RDES0[8]) is set.
Extended Status Available/Rx MAC Address

When either Advanced Timestamp or IP Checksum Offload (Type 2) is present, this bit, when set,
indicates that the extended status is available in descriptor word 4 (RDES4). This is valid only when
the Last Descriptor bit (RDES0[8]) is set.
0
When Advance Timestamp Feature or IPC Full Offload is not selected, this bit indicates Rx MAC
Address status. When set, this bit indicates that the Rx MAC Address registers value (1 to 15)
matched the frame’s DA field. When reset, this bit indicates that the Rx MAC Address Register 0
value matched the DA field.
Table 96. Receive Descriptor Fields 1 (RDES1) (Sheet 1 of 2)

Bit Description
DIC: Disable Interrupt on Completion

31 When set, this bit prevents setting the Status Register’s RI bit (CSR5[6]) for the received frame
ending in the buffer indicated by this descriptor. This, in turn, disables the assertion of the interrupt
to Host because of RI for that frame.
30:29 Reserved
RBS2: Receive Buffer 2 Size

These bits indicate the second data buffer size, in bytes. The buffer size must be a multiple of 4, 8,
28:16 or 16, depending on the bus widths (32, 64, or 128, respectively), even if the value of RDES3
(buffer2 address pointer) is not aligned to bus width. If the buffer size is not an appropriate multiple
of 4, 8, or 16, the resulting behavior is undefined. This field is not valid if RDES1[14] is set.
RER: Receive End of Ring

15 When set, this bit indicates that the descriptor list reached its final descriptor. The DMA returns to
the base address of the list, creating a descriptor ring.


Bit Description
RCH: Second Address Chained

14 When set, this bit indicates that the second address in the descriptor is the Next Descriptor address
rather than the second buffer address. When this bit is set, RBS2 (RDES1[28:16]) is a “don’t care”
value. RDES1[15] takes precedence over RDES1[14].
13 Reserved
RBS1: Receive Buffer 1 Size

Indicates the first data buffer size in bytes. The buffer size must be a multiple of 4, 8, or 16,
12:0 depending upon the bus widths (32, 64, or 128), even if the value of RDES2 (buffer1 address
pointer) is not aligned. When the buffer size is not a multiple of 4, 8, or 16, the resulting behavior is
undefined. If this field is 0, the DMA ignores this buffer and uses Buffer 2 or next descriptor
depending on the value of RCH (Bit 14).
Table 97. Receive Descriptor Fields 2 (RDES2)

Bit Description
Buffer 1 Address Pointer

These bits indicate the physical address of Buffer 1. There are no limitations on the buffer address
alignment except for the following condition: The DMA uses the configured value for its address
31:0 generation when the RDES2 value is used to store the start of frame. The DMA performs a write
operation with the RDES2[3:0, 2:0, or 1:0] bits as 0 during the transfer of the start of frame but the
frame data is shifted as per the actual Buffer address pointer. The DMA ignores RDES2[3:0, 2:0, or
1:0] (corresponding to bus width of 128, 64, or 32) if the address pointer is to a buffer where the
middle or last part of the frame is stored.

Bit Description
Buffer 2 Address Pointer (Next Descriptor Address)

These bits indicate the physical address of Buffer 2 when a descriptor ring structure is used. If the
Second Address Chained (RDES1[24]) bit is set, this address contains the pointer to the physical
memory where the Next Descriptor is present.
If RDES1[24] is set, the buffer (Next Descriptor) address pointer must be bus width-aligned
31:0 (RDES3[3, 2, or 1:0] = 0, corresponding to a bus width of 128, 64, or 32. LSBs are ignored
internally.) However, when RDES1[24] is reset, there are no limitations on the RDES3 value, except
for the following condition: The DMA uses the configured value for its buffer address generation
when the RDES3 value is used to store the start of frame. The DMA ignores RDES3 [3, 2, or 1:0]
(corresponding to a bus width of 128, 64, or 32) if the address pointer is to a buffer where the
middle or last part of the frame is stored.
The status written is as shown in Table 99. The status is written only when there is
status related to IPC or timestamp available. The availability of extended status is
indicated by Bit 0 of RDES0. This status is available only when the Advance Timestamp
or IPC Full Offload feature is selected.


Bit Description
31:28 Reserved
Layer 3 and Layer 4 Filter Number Matched

These bits indicate the number of the Layer 3 and Layer 4 Filter that matched the received frame.
• 00: Filter 0
• 01: Filter 1
27:26
• 10: Filter 2
• 11: Filter 3
This field is valid only when Bit 24 or Bit 25 is set high. When more than one filter matches, these
bits give only the lowest filter number.
Layer 4 Filter Match

When set, this bit indicates that the received frame matches one of the enabled Layer 4 Port
Number fields. This status is given only when one of the following conditions is true:
25 • Layer 3 fields are not enabled and all enabled Layer 4 fields match.
• All enabled Layer 3 and Layer 4 filter fields match.
When more than one filter matches, this bit gives the layer 4 filter status of filter indicated by Bits
[27:26].
Layer 3 Filter Match

When set, this bit indicates that the received frame matches one of the enabled Layer 3 IP Address
fields.
This status is given only when one of the following conditions is true:
24
• All enabled Layer 3 fields match and all enabled Layer 4 fields are bypassed.
• All enabled filter fields match.
When more than one filter matches, this bit gives the layer 3 filter status of filter indicated by Bits
[27:26].
23:21 Reserved
VLAN Tag Priority Value

These bits give the VLAN tag’s user value in the received packet. These bits are valid only when the
20:18
RDES4 Bits 16 and 17 are set.
These bits are available only when you select the AV feature.
AV Tagged Packet Received

When set, this bit indicates that an AV tagged packet is received. Otherwise, this bit indicates that
17
an untagged AV packet is received. This bit is valid when Bit 16 is set.
This bit is available only when you select the AV feature.
AV Packet Received
16 When set, this bit indicates that an AV packet is received. This bit is available only when you select
the AV feature.
15 Reserved
Timestamp Dropped
14 When set, this bit indicates that the timestamp was captured for this frame but got dropped in the
MTL Rx FIFO because of overflow. This bit is available only when you select the Advanced
Timestamp feature. Otherwise, this bit is reserved.
PTP Version
13 When set, this bit indicates that the received PTP message is having the IEEE 1588 version 2
format. When reset, it has the version 1 format. This bit is available only when you select the
Advanced Timestamp feature. Otherwise, this bit is reserved.
PTP Frame Type

When set, this bit indicates that the PTP message is sent directly over Ethernet. When this bit is not
12 set and the message type is non-zero, it indicates that the PTP message is sent over UDP-IPv4 or
UDP-IPv6. The information about IPv4 or IPv6 can be obtained from Bits 6 and 7.
This bit is available only when you select the Advanced Timestamp feature.


Bit Description
Message Type
These bits are encoded to give the type of the message received.
• 0000: No PTP message received
• 0001: SYNC (all clock types)
• 0010: Follow_Up (all clock types)
• 0011: Delay_Req (all clock types)
• 0100: Delay_Resp (all clock types)
11:8 • 0101: Pdelay_Req (in peer-to-peer transparent clock)
• 0110: Pdelay_Resp (in peer-to-peer transparent clock)
• 0111: Pdelay_Resp_Follow_Up (in peer-to-peer transparent clock)
• 1000: Announce • 1001: Management
• 1010: Signaling • 1011-1110: Reserved
• 1111: PTP packet with Reserved message type
These bits are available only when you select the Advance Timestamp feature.
Note: Values 1000, 1001, and 1010 are not backward compatible with release 3.50a.
IPv6 Packet Received

When set, this bit indicates that the received packet is an IPv6 packet. This bit is updated only when
7
Bit 10 (IPC) of MAC Configuration Register (Register 0) (GMAC_REG_0)—Offset 0h) is set.
This bit is available when you select the Enable Receive Full TCP/IP Checksum (Type 2) feature.
IPv4 Packet Received

When set, this bit indicates that the received packet is an IPv4 packet. This bit is updated only when
6
Bit 10 (IPC) of MAC Configuration Register (Register 0) (GMAC_REG_0)—Offset 0h) is set.
IP Checksum Bypassed
5 When set, this bit indicates that the checksum offload engine is bypassed. This bit is available when
you select the Enable Receive Full TCP/IP Checksum (Type 2) feature.
IP Payload Error
When set, this bit indicates that the 16-bit IP payload checksum (that is, the TCP, UDP, or ICMP
checksum) that the core calculated does not match the corresponding checksum field in the
4
received segment. It is also set when the TCP, UDP, or ICMP segment length does not match the
payload length value in the IP Header field. This bit is valid when either Bit 7 or Bit 6 is set.
IP Header Error
When set, this bit indicates that either the 16-bit IPv4 header checksum calculated by the core does
3 not match the received checksum bytes, or the IP datagram version is not consistent with the
Ethernet Type value. This bit is valid when either Bit 7 or Bit 6 is set.
IP Payload Type
These bits indicate the type of payload encapsulated in the IP datagram processed by the Receive
Checksum Offload Engine (COE). The COE also sets these bits to 2'b00 if it does not process the IP
datagram’s payload due to an IP header error or fragmented IP.
• 3'b000: Unknown or did not process IP payload
2:0 • 3'b001: UDP
• 3'b010: TCP
• 3'b011: ICMP
• 3’b1xx: Reserved
This bit is valid when either Bit 7 or Bit 6 is set. This bit is available when you select the Enable
Receive Full TCP/IP Checksum (Type 2) feature.
RDES6 and RDES7 contain the snapshot of the timestamp. The availability of the
snapshot of the timestamp in RDES6 and RDES7 is indicated by Bit 7 in the RDES0
descriptor. The contents of RDES6 and RDES7 are identified in Table 100 and Table 101,
respectively.


Bit Description
RTSL: Receive Frame Timestamp Low

31:0 This field is updated by DMA with the least significant 32 bits of the timestamp captured for the
corresponding receive frame. This field is updated by DMA only for the last descriptor of the receive
frame which is indicated by Last Descriptor status bit (RDES0[8]).

Bit Description
RTSH: Receive Frame Timestamp High

31:0 This field is updated by DMA with the most significant 32 bits of the timestamp captured for the
corresponding receive frame. This field is updated by DMA only for the last descriptor of the receive
frame which is indicated by Last Descriptor status bit (RDES0[8]).
§§


USB 2.0—Intel® Quark™ SoC X1000
16.0 USB 2.0
The Intel® Quark™ SoC X1000 USB subsystem provides a two-port USB 2.0 Host
Controller and one USB 2.0 Device port.


Table 102. Signals

Direction/
Type
USBH[0/1]_DP Universal Serial Bus Host Port 0 and Port 1. Differentials: Bus Data/
I/O
USBH[0/1]_DN Address/Command Bus
USBH0_OC_B Over current Indicators: These signals set corresponding bits in the USB
I controller to indicate that an over current condition has occurred.
USBH1_OC_B Overcurrent indicators are provided for both Host ports.
USBH0_PWR_EN
O Power Enable signal to the USB host port
USBH1_PWR_EN
USBD_DP Universal Serial Bus Device Port. Differentials: Bus Data/ Address/
I/O
USBD_DN Command Bus
RCOMP OUT. Note: Please check the Platform Design Guide for connection
OUSBCOMP O
details for this COMP pin.
RCOMP IN. Note: Please check the Platform Design Guide for connection
IUSBCOMP I
details for this COMP pin.
16.2 Features
16.2.1 USB2.0 Host Controller Features

• 2-Port USB 2.0 Host Controller compatible with the following standards:
— Universal Serial Bus Specification (Revision 2.0, April 27, 2000)
— Enhanced Host Controller Interface Specification for Universal Serial Bus
(Revision1.0, March 12, 2002)
— EHCI 1.1 Addendum (Revision v0.6, October 2007)
— OpenHCI: Open Host Controller Interface Specification for USB (Release 1.0a,
September 14, 1999)

Intel® Quark™ SoC X1000—USB 2.0
• EHCI features
Supported:
— 512-byte Packet Buffer depth for in/out data buffering
— Programmable Packet Buffer depth
— Extended capability pointer (EECP = 8’hC0)
— Programmable frame list flag
— 32-bit only addressing capability
— Per port power control
— PCI Power Management
Not supported:
— Descriptor/data prefetching
— Asynchronous schedule park capability
— HSIC functionality
— Link Power Management (LPM) ECN
• OHCI features
Supported:
— One OHCI companion controller
— Per port power control
Not supported:
— Keyboard/Mouse legacy interface
16.2.2 USB2.0 Device Features

• High-speed (480 Mbps) and full-speed (12 Mbps) operation
• 3 logical endpoints in addition to logical endpoint 0
• 1 configuration in addition to configuration 0
• Enables user-configurable endpoint information
• Multiple data packets for each OUT endpoint (Multiple Receive FIFO).
• Both DMA option and Slave-Only modes
• True scatter-gather DMA implementation
• Descriptor-based memory structures in application memory when in DMA mode
16.3 References
• USB 2.0 specification at http://www.usb.org/developers/docs

16.4 Register Map
Figure 32. USB Register Map
EHCI Host Memory Space

PCI Space
PCI Header
D:20,F:3
CPU
Core
MBAR
Host Bridge
D:0,F:0
EHCI Host
PCI Mem
CAM Registers
(I/O)
Bus 0
OHCI Host Memory Space
PCI
ECAM PCI Header
(Mem) D:20,F:4
RP0 F:0 MBAR

PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
D:21
SPI1 F:1
I2C*/GPIO F:2 OHCI Host

Mem
Legacy Bridge Registers
D:31, F:0
SDIO/eMMC F:0 Memory Space
USB Device
HSUART0 F:1 PCI Header
D:20,F:2
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4 MBAR
HSUART1 F:5
MAC0 F:6
MAC1 F:7 USB Device
Mem
Registers
16.5.1 USB Device

Default
Value


Default
Value
8h Bh “Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h” on page 440 0C03FE10h

86h 86h 00h
page 449
16.5.1.1 Vendor ID (VENDOR_ID)—Offset 0h

Access Method
Type: PCI Configuration Register VENDOR_ID: [B:0, D:20, F:2] + 0h
(Size: 16 bits)
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value

Bit Default &

Description
Range Access
8086h
RO
16.5.1.2 Device ID (DEVICE_ID)—Offset 2h

Access Method
Type: PCI Configuration Register DEVICE_ID: [B:0, D:20, F:2] + 2h
(Size: 16 bits)
Default: 0939h
15 12 8 4 0
0 0 0 0 1 0 0 1 0 0 1 1 1 0 0 1
value
Bit Default &
Description
Range Access
0939h
RO
16.5.1.3 Command Register (COMMAND_REGISTER)—Offset 4h

Access Method
(Size: 16 bits) COMMAND_REGISTER: [B:0, D:20, F:2] + 4h
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IntrDis
SERREn
MasEn
MEMen
RSVD0
RSVD
RSVD
RSVD
Bit Default &
Description
Range Access
0h
RO

10
0h
RO
8
00h
RO

2
RW requests.
1

Bit Default &

Description
Range Access
0h
RO
16.5.1.4 Status Register (STATUS)—Offset 6h

Access Method
Type: PCI Configuration Register STATUS: [B:0, D:20, F:2] + 6h
(Size: 16 bits)
Default: 0010h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
capable_66Mhz
RcdMasAb
RSVD0
RSVD1
DEVSEL
FastB2B
hasCapList
SigSysErr
RSVD
RSVD
RSVD
IntrStatus
Bit Default &
Description
Range Access
0h
RO
14
13
0h
RO
0b
RO
0h
RO
0b
RO
0h
RO
0b
RO

4
RO register sets.
RO used
0h
RO
16.5.1.5 Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h

Access Method


(Size: 32 bits)
Default: 0C03FE10h
31 28 24 20 16 12 8 4 0
0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 0 1 0 0 0 0
progIntf
classCode
subClassCode
rev_id
Bit Default &
Description
Range Access
0Ch Class Code (classCode): Broadly classifies the type of function that the device
31: 24
RO performs.
23: 16
FEh Programming Interface (progIntf): Used to define the register set variation within a
15: 8
7: 0
16.5.1.6 Cache Line Size (CACHE_LINE_SIZE)—Offset Ch

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0
16.5.1.7 Latency Timer (LATENCY_TIMER)—Offset Dh

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
RO
16.5.1.8 Header Type (HEADER_TYPE)—Offset Eh

Access Method
(Size: 8 bits)
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
cfgHdrFormat
multiFnDev
Bit Default &

Description
Range Access

7
RO function device

6: 0
16.5.1.9 BIST (BIST)—Offset Fh

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
BIST_capable
comp_code
start_bist
RSVD
Bit Default &

Description
Range Access

7
0h
RO
0h
RO

Bit Default &

Description
Range Access

3: 0
16.5.1.10 Base Address Register (BAR0)—Offset 10h

Access Method
Type: PCI Configuration Register BAR0: [B:0, D:20, F:2] + 10h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
prefetchable
memType
address
RSVD
isIO
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder
16.5.1.11 Cardbus CIS Pointer (CARDBUS_CIS_POINTER)—Offset 28h

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
RO
16.5.1.12 Subsystem Vendor ID (SUB_SYS_VENDOR_ID)—Offset 2Ch

Access Method
Type: PCI Configuration Register SUB_SYS_VENDOR_ID: [B:0, D:20, F:2] + 2Ch
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO
16.5.1.13 Subsystem ID (SUB_SYS_ID)—Offset 2Eh

Access Method
(Size: 16 bits) SUB_SYS_ID: [B:0, D:20, F:2] + 2Eh
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO
16.5.1.14 Expansion ROM Base Address (EXP_ROM_BASE_ADR)—Offset 30h

Access Method
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
AddrDecodeEn
ROM_base_addr
Bit Default &
Description
Range Access

31: 11
000h
RO

RW also set to 1
16.5.1.15 Capabilities Pointer (CAP_POINTER)—Offset 34h

Access Method
(Size: 32 bits)
Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
RSVD0
value
Bit Default &
Description
Range Access
0h
RO
16.5.1.16 Interrupt Line Register (INTR_LINE)—Offset 3Ch

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
vector information.
16.5.1.17 Interrupt Pin Register (INTR_PIN)—Offset 3Dh

Access Method
(Size: 8 bits) INTR_PIN: [B:0, D:20, F:2] + 3Dh
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
7: 0
16.5.1.18 MIN_GNT (MIN_GNT)—Offset 3Eh

Access Method
Type: PCI Configuration Register MIN_GNT: [B:0, D:20, F:2] + 3Eh
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO
16.5.1.19 MAX_LAT (MAX_LAT)—Offset 3Fh

Access Method
(Size: 8 bits)
Default: 00h

7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO
16.5.1.20 Capability ID (PM_CAP_ID)—Offset 80h

Access Method
Type: PCI Configuration Register PM_CAP_ID: [B:0, D:20, F:2] + 80h
(Size: 8 bits)
Default: 01h
7 4 0
0 0 0 0 0 0 0 1
value
Bit Default &

Description
Range Access
7: 0
16.5.1.21 Next Capability Pointer (PM_NXT_CAP_PTR)—Offset 81h

Access Method
(Size: 8 bits)
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0
16.5.1.22 Power Management Capabilities (PMC)—Offset 82h

Access Method
Type: PCI Configuration Register PMC: [B:0, D:20, F:2] + 82h
(Size: 16 bits)

Default: 4803h
15 12 8 4 0
0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1
PME_support
D2_support
D1_support
aux_curr
RSVD
PME_clock
version
DSI
Bit Default &
Description
Range Access
15: 11
0h
RO
0h
RO
0h
RO
0h
RO
0h
RO
2: 0
16.5.1.23 Power Management Control/Status Register (PMCSR)—Offset 84h

Access Method
Type: PCI Configuration Register PMCSR: [B:0, D:20, F:2] + 84h
(Size: 16 bits)
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
PME_status
PME_en
Data_scale
power_state
Data_select
no_soft_reset
RSVD
RSVD
Bit Default &

Description
Range Access
15
0h
RO

Bit Default &

Description
Range Access
0h
RO
8
0h
RO

3
0h
RO
1: 0

86h
Access Method
Type: PCI Configuration Register PMCSR_BSE: [B:0, D:20, F:2] + 86h
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access

7: 0
RO to 0.
16.5.1.25 Power Management Data Register (DATA_REGISTER)—Offset 87h

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

16.5.1.26 Capability ID (MSI_CAP_ID)—Offset A0h

Access Method
(Size: 8 bits)
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value
Bit Default &
Description
Range Access
7: 0
16.5.1.27 Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0
RO in the chain
16.5.1.28 Message Control (MESSAGE_CTRL)—Offset A2h

Access Method
(Size: 16 bits)
Default: 0100h
15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
perVecMskCap
bit64Cap
multiMsgCap
MSIEnable
RSVD0
multiMsgEn

Bit Default &

Description
Range Access
0h
RO

8
7

6: 4

3: 1
RW MSI.
16.5.1.29 Message Address (MESSAGE_ADDR)—Offset A4h

Access Method
Type: PCI Configuration Register MESSAGE_ADDR: [B:0, D:20, F:2] + A4h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD0
Bit Default &
Description
Range Access
31: 2
0h
RO
16.5.1.30 Message Data (MESSAGE_DATA)—Offset A8h

Access Method
Type: PCI Configuration Register MESSAGE_DATA: [B:0, D:20, F:2] + A8h
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData

Bit Default &

Description
Range Access
15: 0
changed by hardware
16.5.1.31 Mask Bits for MSI (PER_VEC_MASK)—Offset ACh

Access Method
Type: PCI Configuration Register PER_VEC_MASK: [B:0, D:20, F:2] + ACh
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
MSIMask
Bit Default &
Description
Range Access
0h
RO
0
16.5.1.32 Pending Bits for MSI (PER_VEC_PEND)—Offset B0h

Access Method
Type: PCI Configuration Register PER_VEC_PEND: [B:0, D:20, F:2] + B0h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
value
Bit Default &

Description
Range Access
0h
RO
0h
RO

16.5.2 USB EHCI

Default
Value
8h Bh “Revision ID and Class Code (REV_ID_CLASS_CODE)—Offset 8h” on page 440 0C032010h
60h 60h “Serial Bus Release Number Register (SBRN)—Offset 60h” on page 462 20h
61h 61h “Frame Length Adjustment Register (FLADJ)—Offset 61h” on page 462 20h
82h 83h “Power Management Capabilities (PMC)—Offset 82h” on page 463 F803h

86h 86h 00h
page 465
A1h A1h “Next Capability Pointer (MSI_NXT_CAP_PTR)—Offset A1h” on page 466 C0h
“USB Legacy Support Extended Capability (USBLEGSUP)—Offset C0h” on

C0h C3h 00000001h
page 468
C4h C7h “USB Legacy Support Control/Status (USBLEGCTLSTS)—Offset C4h” on page 469 00000000h


Access Method
(Size: 16 bits)
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value
Bit Default &
Description
Range Access
8086h
RO

Access Method
(Size: 16 bits)
Default: 0939h
15 12 8 4 0
0 0 0 0 1 0 0 1 0 0 1 1 1 0 0 1
value
Bit Default &

Description
Range Access
0939h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IntrDis
RSVD0
RSVD
RSVD
MasEn
SERREn
MEMen
RSVD
Bit Default &

Description
Range Access
0h
RO

Bit Default &

Description
Range Access

10
0h
RO
8
00h
RO

2
RW requests.
1
0h
RO

Access Method
(Size: 16 bits)
Default: 0010h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0
capable_66Mhz 1 0 0 0 0
RSVD0
hasCapList
SigSysErr
DEVSEL
FastB2B
RSVD1
RSVD
RSVD
RSVD
RcdMasAb
IntrStatus
Bit Default &

Description
Range Access
0h
RO
14
13
0h
RO
0b
RO
0h
RO
0b
RO
0h
RO

Bit Default &

Description
Range Access
0b
RO

4
RO register sets.
RO used
0h
RO

Access Method
(Size: 32 bits)
Default: 0C032010h
31 28 24 20 16 12 8 4 0
0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0
classCode
subClassCode
progIntf
rev_id
Bit Default &
Description
Range Access
31: 24
RO performs.
23: 16
15: 8
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 8 bits) HEADER_TYPE: [B:0, D:20, F:3] + Eh
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
cfgHdrFormat
multiFnDev
Bit Default &

Description
Range Access

7
RO function device

6: 0

Access Method
(Size: 8 bits)
Default: 00h

7 4 0
0 0 0 0 0 0 0 0
start_bist
RSVD
BIST_capable
comp_code
Bit Default &
Description
Range Access

7
0h
RO
0h
RO

3: 0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD
memType
isIO
prefetchable
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder


Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO


Access Method
Type: PCI Configuration Register EXP_ROM_BASE_ADR: [B:0, D:20, F:3] + 30h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ROM_base_addr
RSVD
AddrDecodeEn
Bit Default &
Description
Range Access

31: 11
000h
RO

RW also set to 1

Access Method
(Size: 32 bits)
Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
value 0 0 0 0
RSVD0
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
vector information.

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO


Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO
16.5.2.20 Serial Bus Release Number Register (SBRN)—Offset 60h

Access Method
Type: PCI Configuration Register SBRN: [B:0, D:20, F:3] + 60h
(Size: 8 bits)
Default: 20h
7 4 0
0 0 1 0 0 0 0 0
SBRN
Bit Default &

Description
Range Access
20h Serial Bus Specification Release Number (SBRN): Serial Bus Specification Release
7: 0
RO Number.
16.5.2.21 Frame Length Adjustment Register (FLADJ)—Offset 61h

Access Method
(Size: 8 bits) FLADJ: [B:0, D:20, F:3] + 61h
Default: 20h
7 4 0
0 0 1 0 0 0 0 0
FLADJ
RSVD0
Bit Default &

Description
Range Access
0h
RO

Bit Default &

Description
Range Access
Frame Length Timing Value (FLADJ): Each decimal value change to this register
20h corresponds to 16 highspeed bit times. The SOF cycle time (number of SOF counter
5: 0
RW clock periods to generate a SOF micro-frame length) is equal to 59488 + value in this
field. The default value is decimal 32 (20h), which gives a SOF cycle time of 60000.

Access Method
(Size: 8 bits)
Default: 01h
7 4 0
0 0 0 0 0 0 0 1
value
Bit Default &
Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 16 bits)
Default: F803h

15 12 8 4 0
1 1 1 1 1 0 0 0 0 0 0 0 0 0 1 1
PME_support
D2_support
D1_support
aux_curr
RSVD
PME_clock
version
DSI
Bit Default &
Description
Range Access
1Fh the function is capable of sending a PME (Power Management Event) message. 0 in a bit
15: 11
0h
RO
0h
RO
0h
RO
0h
RO
0h
RO
2: 0

Access Method
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
PME_status
PME_en
Data_scale
power_state
Data_select
no_soft_reset
RSVD
RSVD
Bit Default &

Description
Range Access
15
0h
RO
0h
RO

Bit Default &

Description
Range Access
8
0h
RO

3
0h
RO
1: 0

86h
Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access

7: 0
RO to 0.

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method


(Size: 8 bits)
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value
Bit Default &
Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: C0h
7 4 0
1 1 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
C0h Next Capability Pointer (value): Pointer to the next register set of feature specific
7: 0 configuration registers. Hardwired to 0xC0 to point to the USB Legacy Support Extended
RO Capability Structure

Access Method
(Size: 16 bits)
Default: 0100h
15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
perVecMskCap
bit64Cap
multiMsgCap
MSIEnable
RSVD0
multiMsgEn
Bit Default &

Description
Range Access
0h
RO

Bit Default &

Description
Range Access

8
7

6: 4

3: 1
RW MSI.

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD0
Bit Default &
Description
Range Access
31: 2
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData

Bit Default &

Description
Range Access
15: 0
changed by hardware

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
MSIMask
Bit Default &
Description
Range Access
0h
RO
0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
value
Bit Default &

Description
Range Access
0h
RO
0h
RO
16.5.2.35 USB Legacy Support Extended Capability (USBLEGSUP)—Offset C0h

Access Method

Type: PCI Configuration Register USBLEGSUP: [B:0, D:20, F:3] + C0h

(Size: 32 bits)
Default: 00000001h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
RSVD0
HC_BIOS_Owned_Semaphore
HC_OS_Owned_Semaphore
RSVD
NXT_CAP_PTR
CAP_ID
Bit Default &
Description
Range Access
0h
RO
0b HC OS Owned Semaphore (HC_OS_Owned_Semaphore): System software sets

24 this bit to request ownership of the EHCI controller. Ownership is obtained when this bit
RW reads as one and the HC BIOS Owned Semaphore bit reads as zero.
00h
RO
0b HC BIOS Owned Semaphore (HC_BIOS_Owned_Semaphore): The BIOS sets this

16 bit to establish ownership of the EHCI controller. System BIOS will set this bit to a zero
RW in response to a request for ownership of the EHCI controller by system software.
0h Next EHCI Extended Capability Pointer (NXT_CAP_PTR): This field points to the
15: 8 PCI configuration space offset of the next extended capability pointer. A value of 00h
RO indicates the end of the extended capability list.
Capability ID (CAP_ID): This field identifies the extended capability. A value of 01h
01h identifies the capability as Legacy Support. This extended capability requires one
7: 0
RO additional 32-bit register for control/status information, and this register is located at
offset EECP+04h.
16.5.2.36 USB Legacy Support Control/Status (USBLEGCTLSTS)—Offset C4h

Access Method
(Size: 32 bits) USBLEGCTLSTS: [B:0, D:20, F:3] + C4h
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SMI_ASYNC_ADVANCE
SMI_FRAME_LIST_ROLLOVER
SMI_USB_COMPLETE
SMI_BAR
SMI_HOST_SYSTEM_ERROR
SMI_PCI_CMD
SMI_USB_ERROR
SMI_OS_OWNR_CHANGE
RSVD
RSVD
SMI_HOST_SYSTEM_ERROR_EN
SMI_USB_COMPLETE_EN
SMI_PORT_CHANGE_DETECT
SMI_BAR_EN
SMI_ASYNC_ADVANCE_EN
SMI_FRAME_LIST_ROLLOVER_EN
SMI_PORT_CHANGE_DETECT_EN
SMI_USB_ERROR_EN
SMI_PCI_CMD_EN
SMI_OS_OWNR_CHANGE_EN
Bit Default &
Description
Range Access
0b SMI on BAR (SMI_BAR): This bit is set to one whenever the Base Address Register
31
RW (BAR) is written.
0b SMI on PCI Command (SMI_PCI_CMD): This bit is set to one whenever the PCI
30
RW Command Register is written.
0b SMI on OS Ownership Change (SMI_OS_OWNR_CHANGE): This bit is set to one

29 whenever the HC OS Owned Semaphore bit in the USBLEGSUP register transitions from
RW 1 to a 0 or 0 to a 1
00h
RO
SMI on Async Advance (SMI_ASYNC_ADVANCE): Shadow bit of the Interrupt on

0b Async Advance bit in the USBSTS register see Section 2.3.2 for definition. To set this bit
21
RO to a zero, system software must write a one to the Interrupt on Async Advance bit in the
USBSTS register.
SMI on Host System Error (SMI_HOST_SYSTEM_ERROR): Shadow bit of Host

0b System Error bit in the USBSTS register, see Section 2.3.2 for definition and effects of
20
RO the events associated with this bit being set to a one. To set this bit to a zero, system
software must write a one to the Host System Error bit in the USBSTS register.
SMI on Frame List Rollover (SMI_FRAME_LIST_ROLLOVER): Shadow bit of Frame

0b List Rollover bit in the USBSTS register see Section 2.3.2 for definition. To set this bit to
19
RO a zero, system software must write a one to the Frame List Rollover bit in the USBSTS
register.
SMI on Port Change Detect (SMI_PORT_CHANGE_DETECT): Shadow bit of Port

0b Change Detect bit in the USBSTS register see Section 2.3.2 for definition. To set this bit
18
RO to a zero, system software must write a one to the Port Change Detect bit in the
USBSTS register.
SMI on USB Error (SMI_USB_ERROR): Shadow bit of USB Error Interrupt

0b (USBERRINT) bit in the USBSTS register see Section 2.3.2 for definition. To set this bit
17
RO to a zero, system software must write a one to the USB Error Interrupt bit in the
USBSTS register.
0b SMI on USB Complete (SMI_USB_COMPLETE): Shadow bit of USB Interrupt

16 (USBINT) bit in the USBSTS register see Section 2.3.2 for definition. To set this bit to a
RO zero, system software must write a one to the USB Interrupt bit in the USBSTS register.
0b SMI on BAR Enable (SMI_BAR_EN): When this bit is one and SMI on BAR is one,
15
RW then the host controller will issue an SMI.
0b SMI on PCI Command Enable (SMI_PCI_CMD_EN): When this bit is one and SMI
14
RW on PCI Command is one, then the host controller will issue an SMI.
0b SMI on OS Ownership Enable (SMI_OS_OWNR_CHANGE_EN): When this bit is a

13
RW one AND the OS Ownership Change bit is one, the host controller will issue an SMI.

Bit Default &

Description
Range Access
00h
RO
0b SMI on Async Advance Enable (SMI_ASYNC_ADVANCE_EN): When this bit is a

5 one, and the SMI on Async Advance bit (above) in this register is a one, the host
RW controller will issue an SMI immediately
0b SMI on Host System Error Enable (SMI_HOST_SYSTEM_ERROR_EN): When this

4 bit is a one, and the SMI on Host System Error bit (above) in this register is a one, the
RW host controller will issue an SMI immediately
0b SMI on Frame List Rollover Enable (SMI_FRAME_LIST_ROLLOVER_EN): When

3 this bit is a one, and the SMI on Frame List Rollover bit (above) in this register is a one,
RW the host controller will issue an SMI immediately.
0b SMI on Port Change Enable (SMI_PORT_CHANGE_DETECT_EN): When this bit is

2 a one, and the SMI on Port Change Detect bit (above) in this register is a one, the host
RW controller will issue an SMI immediately.
0b SMI on USB Error Enable (SMI_USB_ERROR_EN): When this bit is a one, and the
1 SMI on USB Error bit (above) in this register is a one, the host controller will issue an
RW SMI immediately.
0b USB SMI Enable (SMI_USB_COMPLETE_EN): When this bit is a one, and the SMI on
0 USB Complete bit (above) in this register is a one, the host controller will issue an SMI
RW immediately.
16.5.3 USB OHCI

Default
Value
2h 3h “Device ID (DEVICE_ID)—Offset 2h” on page 439 093Ah


Default
Value

86h 86h 00h
page 465

Access Method
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value
Bit Default &

Description
Range Access
8086h
RO

Access Method
Default: 093Ah
15 12 8 4 0
0 0 0 0 1 0 0 1 0 0 1 1 1 0 1 0
value

Bit Default &

Description
Range Access
093Ah
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
IntrDis
RSVD
RSVD
MasEn
SERREn
MEMen
RSVD
Bit Default &
Description
Range Access
0h
RO

10
0h
RO
8
00h
RO

2
RW requests.
1
0h
RO

Access Method
(Size: 16 bits)
Default: 0010h

15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
hasCapList
IntrStatus
SigSysErr
RSVD
DEVSEL
RSVD
FastB2B
RSVD
RcdMasAb
capable_66Mhz
RSVD0
RSVD1
Bit Default &
Description
Range Access
0h
RO
14
13
0h
RO
0b
RO
0h
RO
0b
RO
0h
RO
0b
RO

4
RO register sets.
RO used
0h
RO

Access Method
(Size: 32 bits) REV_ID_CLASS_CODE: [B:0, D:20, F:4] + 8h
Default: 0C031010h
31 28 24 20 16 12 8 4 0
0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0
progIntf
classCode
subClassCode
rev_id

Bit Default &

Description
Range Access
31: 24
RO performs.
23: 16
15: 8
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits) LATENCY_TIMER: [B:0, D:20, F:4] + Dh
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method

Default: 80h
7 4 0
1 0 0 0 0 0 0 0
multiFnDev
cfgHdrFormat
Bit Default &
Description
Range Access

7
RO function device

6: 0

Access Method
(Size: 8 bits) BIST: [B:0, D:20, F:4] + Fh
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
RSVD
BIST_capable
comp_code
start_bist
Bit Default &

Description
Range Access

7
0h
RO
0h
RO

3: 0

Access Method
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD
prefetchable
memType
isIO
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder

Access Method
(Size: 32 bits) CARDBUS_CIS_POINTER: [B:0, D:20, F:4] + 28h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
Default: 0000h

15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ROM_base_addr
RSVD
AddrDecodeEn
Bit Default &

Description
Range Access

31: 11
000h
RO

RW also set to 1


Access Method
(Size: 32 bits)
Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
RSVD0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
vector information.

Access Method
(Size: 8 bits) INTR_PIN: [B:0, D:20, F:4] + 3Dh
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
Bit Default & value

Description
Range Access
0h
RO

Access Method
(Size: 8 bits) MAX_LAT: [B:0, D:20, F:4] + 3Fh
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
Default: 01h

7 4 0
0 0 0 0 0 0 0 1
value
Bit Default &
Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 16 bits)
Default: 4803h
15 12 8 4 0
0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1
D2_support
D1_support
PME_support
aux_curr
DSI
RSVD
PME_clock
version
Bit Default &

Description
Range Access
15: 11
0h
RO
0h
RO

Bit Default &

Description
Range Access
0h
RO
0h
RO
0h
RO
2: 0

Access Method
(Size: 16 bits)
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
Data_scale
no_soft_reset
power_state
Data_select
PME_en
RSVD
RSVD
PME_status
Bit Default &

Description
Range Access
15
0h
RO
0h
RO
8
0h
RO

3
0h
RO
1: 0


86h
Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access

7: 0
RO to 0.

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value
Bit Default &

Description
Range Access
7: 0


Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
7: 0
RO in the chain

Access Method
(Size: 16 bits)
Default: 0100h
15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
MSIEnable
RSVD0
perVecMskCap
bit64Cap
multiMsgCap
multiMsgEn
Bit Default &

Description
Range Access
0h
RO

8
7

6: 4

3: 1
RW MSI.

Access Method


(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
address
Bit Default &
Description
Range Access
31: 2
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData
Bit Default &

Description
Range Access
15: 0
changed by hardware

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
MSIMask

Bit Default &

Description
Range Access
0h
RO
0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
RSVD0
Bit Default &
Description
Range Access
0h
RO
0h
RO
16.6.1 USB Device

Offset Default
Start Value
0h 3h “IN Endpoint 0 Control Register (ep0_in_ctrl_udc_reg)—Offset 0h” on page 489 00000000h
4h 7h “IN Endpoint 0 Status Register (ep0_in_sts_udc_reg)—Offset 4h” on page 490 00000000h
8h Bh “IN Endpoint 0 Buffer Size Register (ep0_in_bufsize_udc_reg)—Offset 8h” on page 492 00000000h
“IN Endpoint 0 Maximum Packet Size Register (ep0_in_mpkt_sz_reg)—Offset Ch” on

Ch Fh 00000000h
page 493
“IN Endpoint 0 Data Descriptor Pointer Register (ep0_in_desptr_udc_reg)—Offset 14h”

14h 17h 00000000h
on page 493
“IN Endpoint 0 Write Confirmation register (for Slave-Only mode)

1Ch 1Fh 00000000h
(ep0_wr_cfrm_udc_reg)—Offset 1Ch” on page 494
“IN Endpoint 1 Buffer Size Register (ep1_in_bufsize_udc_reg)—Offset 28h” on

28h 2Bh 00000000h
page 497


Offset Default
Start Value
“IN Endpoint 1 Maximum Packet Size Register (ep1_in_mpkt_sz_reg)—Offset 2Ch” on

2Ch 2Fh 00000000h
page 498

34h 37h 00000000h
on page 498

3Ch 3Fh 00000000h

48h 4Bh 00000000h
page 502

4Ch 4Fh 00000000h
page 503

54h 57h 00000000h
on page 503

5Ch 5Fh 00000000h

68h 6Bh 00000000h
page 507

6Ch 6Fh 00000000h
page 508

74h 77h 00000000h
on page 508

7Ch 7Fh 00000000h
200h 203h “OUT Endpoint 0 Control Register (ep0_out_ctrl_udc_reg)—Offset 200h” on page 509 00000000h
204h 207h “OUT Endpoint 0 Status Register (ep0_out_sts_udc_reg)—Offset 204h” on page 510 00000100h
“OUT Endpoint 0 Receive Packet Frame Number Register (ep0_out_rpf_udc_reg)—

208h 20Bh 00000000h
“OUT Endpoint 0 Buffer Size Register (ep0_out_bufsize_udc_reg)—Offset 20Ch” on

20Ch 20Fh 00000000h
page 513
“OUT Endpoint 0 SETUP Buffer Pointer Register (ep0_subptr_udc_reg)—Offset 210h” on

210h 213h 00000000h
page 513
“OUT Endpoint 0 Data Descriptor Pointer Register (ep0_out_desptr_udc_reg)—Offset

214h 217h 00000000h
214h” on page 514
“OUT Endpoint 0 Read Confirmation Register for zero-length OUT data (for Slave-Only
21Ch 21Fh 00000000h
mode) (ep0_rd_cfrm_udc_reg)—Offset 21Ch” on page 514

228h 22Bh 00000000h

22Ch 22Fh 00000000h
page 519

230h 233h 00000000h
page 519

234h 237h 00000000h
234h” on page 520


Offset Default
Start Value
23Ch 23Fh 00000000h

248h 24Bh 00000000h

24Ch 24Fh 00000000h
page 525

250h 253h 00000000h
page 525

254h 257h 00000000h
254h” on page 526
25Ch 25Fh 00000000h

268h 26Bh 00000000h

26Ch 26Fh 00000000h
page 531

270h 273h 00000000h
page 531

274h 277h 00000000h
274h” on page 532
27Ch 27Fh 00000000h
400h 403h “Device Configuration Register (d_cfg_udc_reg)—Offset 400h” on page 533 00000020h
404h 407h “Device Control Register (d_ctrl_udc_reg)—Offset 404h” on page 534 00000400h
408h 40Bh “Device Status Register (d_sts_udc_reg)—Offset 408h” on page 536 00000000h
40Ch 40Fh “Device Interrupt Register (d_intr_udc_reg)—Offset 40Ch” on page 537 00000000h
410h 413h “Device Interrupt Mask Register (d_intr_msk_udc_reg)—Offset 410h” on page 538 00000000h
414h 417h “Endpoints Interrupt Register (ep_intr_udc_reg)—Offset 414h” on page 539 00000000h
418h 41Bh “Endpoints Interrupt Mask Register (ep_intr_msk_udc_reg)—Offset 418h” on page 539 00000000h
41Ch 41Fh “Test Mode Register (test_mode_udc_reg)—Offset 41Ch” on page 540 00000000h
420h 423h “Product Release Number Register (revision_udc_reg)—Offset 420h” on page 541 3234352Ah
“SETUP command address pointer register (udc_desc_addr_udc_reg)—Offset 500h” on

500h 503h 00000000h
page 541
504h 507h “Physical Endpoint 0 Register (udc_ep_ne_udc_reg_0)—Offset 504h” on page 542 00000000h
508h 50Bh “Physical Endpoint 1 Register (udc_ep_ne_udc_reg_1)—Offset 508h” on page 542 00000000h
50Ch 50Fh “Physical Endpoint 2 Register (udc_ep_ne_udc_reg_2)—Offset 50Ch” on page 543 00000000h
518h 51Bh “Physical Endpoint 5 Register (udc_ep_ne_udc_reg_5)—Offset 518h” on page 546 00000000h
51Ch 51Fh “Physical Endpoint 6 Register (udc_ep_ne_udc_reg_6)—Offset 51Ch” on page 546 00000000h


Offset Default
Start Value
800h + [0- “RxFIFO Array[0-511] (udc_rx_fifo_reg_array[0-511])—Offset 800h, Count 512, Stride

803h 00000000h
511]*4h 4h” on page 547
1000h + “TxFIFO 0 Array[0-255] (udc_tx_fifo_reg_0_array[0-255])—Offset 1000h, Count 256,

1003h 00000000h
[0-255]*4h Stride 4h” on page 548

1403h 00000000h

1803h 00000000h
1C00h + “TxFIFO 3 Array[0-255] (udc_tx_fifo_reg_3_array[0-255])—Offset 1C00h, Count 256,

1C03h 00000000h
16.6.1.1 IN Endpoint 0 Control Register (ep0_in_ctrl_udc_reg)—Offset 0h

This register is used to program the endpoint as required by the application.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rrdy
close_desc
cnak
snak
nak
et
f
reserved
mrx_flush
null_bit
p
sn
s
Bit Default &
Range Access
0x0 reserved: Reserved bits.

31:13
RO These bits are reserved and should be set to zero.
0x0 Receive FIFO Flush for Multiple Receive FIFO (mrx_flush): This is an OUT
12
RO endpoint field only. These bits are reserved and should be set to zero.
0x0 Close Descriptor Channel (close_desc): Close Descriptor Enhancement is not

11
RO supported. These bits are reserved and should be set to zero.
Send Null Packet (null_bit): This bit provides the application with a mechanism to
0x0 instruct the USB Device Controller to send a NULL (zero length) packet when no data is
10
RW available in the particular endpoint TxFIFO. If this bit is set, when no data is available in
the endpoint TxFIFO, the USB Device Controller sends a NULL packet.
0x0 Receive Ready (rrdy): This is an OUT endpoint field only. These bits are reserved and
9
RO should be set to zero.

Bit Default &

Range Access
Clear NAK (cnak): Used by the application to clear the NAK bit (bit 6). After the
subsystem sets bit 6 (NAK), the application must clear it with a write of 1 to the CNAK
bit. (For example, after the application has decoded the SETUP packet and determined it
is not an invalid command, the application must set the CNAK bit of the control endpoint
0x0 to 1 to clear the NAK bit.)The application also must clear the NAK bit whenever the
8 subsystem sets it. (The subsystem sets it due to the application setting the Stall
WO bit.)The application can clear this bit only when the RxFIFO is empty (for single RxFIFO
implementation) or when the RxFIFO corresponding to the same logical endpoint is
empty (Multiple RxFIFO implementation). The application can write CNAK any time
without waiting for a RxFIFO empty condition. The NAK bit is cleared immediately upon
write to CNAK bit. No polling is necessary.
0x0 Set NAK (snak): Used by the application to set the NAK bit (bit 6 of this register). The
7 application must not set the NAK bit for an IN endpoint until it has received an IN token
WO interrupt indicating that the TxFIFO is empty.
NAK Bit (nak): After a SETUP packet, which is decoded by the application, is received
by the core, the core sets the NAK bit for all control IN/OUT endpoints. NAK is also set
after a STALL response for the endpoint (the STALL bit is set in Endpoint Control register
0x0 bit 0).
6 1: The endpoint responds to the USB host with a NAK handshake.
RO 0: The endpoint responds normally.
Note 1: A SETUP packet is sent to the application regardless of whether the NAK bit is
set.
Note 2: A NAK set on ISOC OUT endpoint rejects the ISOC out data packet.
Endpoint Type (et): The possible options are:

0x0 00: Control endpoint
5:4 01: Isochronous endpoint
RW 10: Bulk endpoint
11: Interrupt endpoint
Poll Demand (p): Poll demand from the application. The application can set this bit
0x0 after an IN token is received from the endpoint. The application can also set this bit
3 before an IN token is received for the endpoint, if it has the IN transfer data in advance.
RW Note: After sending a zero-length or short packet, the application must wait for the next
XFERDONE_TXEMPTY interrupt, before setting up the P bit for the next transfer.
0x0 Snoop Mode (sn): This is an OUT endpoint field only. These bits are reserved and
2
Flush TxFIFO (f): The application firmware sets this bit to 1 after it has detected a
0x0 disconnect/connect on the USB cable, then waits for the IN token endpoint interrupt
1
RW before resetting this bit to 0. This flushes the stale data out of the TxFIFO. This bit is
cleared by the core when TDC (Transmit DMA Complete) occurs on this endpoint.
Stall Handshake (s): On successful reception of a SETUP packet (decoded by the

application), the subsystem clears both IN and OUT Stall bits, and sets both the IN and
OUT NAK bits. The application must check for RxFIFO emptiness before setting the IN
and OUT STALL bit.
For non-SETUP packets, the subsystem clears either IN or out STALL bits only if a STALL
handshake is returned to the USB host, then sets the corresponding NAK bit.
0x0 The subsystem returns a STALL handshake for the subsequent transactions of the
0 stalled endpoint until the USB host issues a Clear_Feature command to clear it. Once
RW this bit is set, if the subsystem has already returned a STALL handshake to the USB
host, the application firmware cannot clear the S bit to stop the subsystem from sending
the STALL handshake on the endpoint. The host must instead send one of the following
commands to clear the endpoint Halt status:
- Clear Feature (Halt)
- SetConfiguration
- SetInterface.
16.6.1.2 IN Endpoint 0 Status Register (ep0_in_sts_udc_reg)—Offset 4h

The Endpoint Status register indicates the endpoint status.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo_empty
mrxfifo_empty
reserved_1
xfer_done_txf_empty
set_feature_halt
clr_feature_halt
out_tok
reserved_3
bna
in_tok
cdc
isoc_xfer_done
tdc
he
rx_pkt_size
Bit Default &
Range Access
0x0 reserved_1: Reserved bits.

31:29
0x0 Close Descriptor Channel (cdc): Close Descriptor Enhancement is not supported.
28
Transfer Done/Transmit FIFO Empty (xfer_done_txf_empty): This bit indicates

0x0 that the TXFIFO is empty after the DMA transfer has been completed. The application
27
RW/1C can use this bit to set the poll bit for the next transfer. The application must first clear
this bit after servicing the interrupt.
Received Set Halt (set_feature_halt): This bit indicates that the Set Feature (EP
HALT) command is received for this endpoint. To stall this endpoint, the application can
0x0 set the S bit in Endpoint Control Register. After this, the application clears this RSS bit to
26
RW/1C acknowledge the reception of Set Feature stall command. Once this RSS bit is cleared,
core sends the zero-length packet for the Status IN phase of Set Feature command.
Received Set Stall indication is applicable only for Bulk and Interrupt transactions.
Received Clear Halt (clr_feature_halt): This bit indicates that the Clear Feature (EP
HALT) command is received for this endpoint. To continue to stall the endpoint, the
0x0 application must set the S bit Endpoint Control Register. After this, the application clears
25 this RCS bit to acknowledge the reception of clear stall command. Once this bit is
RW/1C cleared, core sends the zero-length packet for the Status IN phase of clear stall
command. Received Clear Stall indication is applicable only for Bulk and Interrupt
transactions.
Transmit FIFO Empty detected (tx_fifo_empty): This bit indicates that the
0x0 Transmit FIFO Empty condition is triggered. Application can use this information to load
24 the subsequent data into the Transmit FIFO. The application must clear this bit after
RW/1C writing the data into the Transmit FIFO.
Note: This bit is not used for Isochronous endpoints.
Isochronous IN transaction complete (isoc_xfer_done): This bit indicates that the

0x0 isochronous IN transaction for this endpoint is complete. The application can use this
23
RW/1C information to program the isochronous IN data for the next microframe. This bit is used
only in Slave-Only mode.
0x0 Receive Packet Size (rx_pkt_size): This is an OUT endpoint field only. These bits are
22:11
RO reserved and should be set to zero.
0x0 Transmit DMA Completion (tdc): Indicates the transmit DMA has completed
10 transferring a descriptor chain data to the Tx FIFO. After servicing the interrupt, the
RW/1C application must clear this bit.
0x0 Error response on the host bus (he): Error response on the host bus (AHB) when
9 doing a data transfer, descriptor fetch, or descriptor update for this particular endpoint.
RW/1C After servicing the interrupt, the application must clear this bit.
0x0 Receive Address FIFO Empty Status (mrxfifo_empty): This is an OUT endpoint
8
RO field only. These bits are reserved and should be set to zero.

Bit Default &

Range Access
Buffer Not Available (bna): The subsystem sets this bit when the descriptor status is
0x0 either Host Busy or DMA Done to indicate that the descriptor was not ready at the time
7
RW/1C the DMA tried to access it. After servicing the interrupt, the application must clear this
bit.
0x0 IN token (in_tok): An IN token has been received by this endpoint. After servicing the
6
RW/1C interrupt, application must clear this bit.
0x0 OUT token (out_tok): This is an OUT endpoint field only. These bits are reserved and
5:4

3:0
16.6.1.3 IN Endpoint 0 Buffer Size Register (ep0_in_bufsize_udc_reg)—Offset

8h
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size_frame_number
reserved
isoc_data_pid
buff_size_1
Bit Default &

Range Access

31:18
Isochronous IN Transaction PID (isoc_data_pid): Initial data PID to be sent for a

high-bandwidth isochronous IN transaction. This field is used only in Slave-Only mode.
0x0 00: DATA0 PID is sent
17:16
RW 01: DATA0 PID is sent
10: DATA1 PID is sent
0x0 Buffer Size (buff_size_frame_number): These bits are reserved and should be set
15:10
RO to zero.
Buffer Size (buff_size_1): Buffer size required for this endpoint. The application can
0x0 program this field to make each endpoints buffers adaptive, providing flexibility in buffer
9:0 size when the interface or configuration is changed. This value is in 32-bit words, and
RW indicates the number of 32-bit word entries in the Transmit FIFO.
NOTE: the maximum size of the TxFIFO is 1024 32-bit word entries.

16.6.1.4 IN Endpoint 0 Maximum Packet Size Register (ep0_in_mpkt_sz_reg)—

Offset Ch
This register also specifies the maximum packet size an endpoint should support. This
maximum size is used to calculate whether the Receive FIFO has sufficient space to
accept a packet. When changing the maximum packet size for a specific endpoint, the
user must also program the USB Device Controller register space.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
mpkt_size
Bit Default &
Range Access
0x0 Buffer Size (buff_size): This is an OUT endpoint field only. These bits are reserved
31:16
RO and should be set to zero.
Max Packet Size (mpkt_size): Maximum packet size for the endpoint in bytes. This
0x0 maximum size is used to calculate whether the Receive FIFO has sufficient space to
15:0
RO accept a packet. When changing the maximum packet size for a specific endpoint, the
user must also program the corresponding Physical Endpoint Register.

(ep0_in_desptr_udc_reg)—Offset 14h
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
31:0 Descriptor Pointer (desptr): This register contains the data descriptor pointers.
RW


(ep0_wr_cfrm_udc_reg)—Offset 1Ch
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
wr_cfrm
Bit Default &
Range Access
0x0 Write Confirmation (for Slave-Only mode) (wr_cfrm): Writing to this register
31:0
WO confirms the IN data into the TxFIFO.

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rrdy
f
reserved
mrx_flush
close_desc
null_bit
cnak
snak
nak
et
p
sn
Bit Default &

Range Access

31:13
12

11
10
9

Bit Default &

Range Access
0x0 bit 0).
set.

2
1

and OUT STALL bit.
- SetConfiguration
- SetInterface.

Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo_empty
mrxfifo_empty
reserved_1
xfer_done_txf_empty
set_feature_halt
clr_feature_halt
out_tok
reserved_3
bna
in_tok
cdc
isoc_xfer_done
tdc
he
rx_pkt_size
Bit Default &
Range Access

31:29
28

27
26
transactions.

23
22:11
8

Bit Default &

Range Access
7
bit.
6
5:4

3:0

28h
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size_frame_number 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
isoc_data_pid
buff_size_1
Bit Default &

Range Access

31:18

17:16
15:10
RO to zero.


Offset 2Ch
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
mpkt_size
Bit Default &
Range Access
31:16
15:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
RW


Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
wr_cfrm
Bit Default &
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rrdy
f
reserved
mrx_flush
close_desc
null_bit
cnak
snak
nak
et
p
sn
s
Bit Default &
Range Access

31:13
12

11
10
9

Bit Default &

Range Access
0x0 bit 0).
set.

2
1

and OUT STALL bit.
- SetConfiguration
- SetInterface.

Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo_empty
mrxfifo_empty
reserved_1
xfer_done_txf_empty
set_feature_halt
clr_feature_halt
out_tok
reserved_3
bna
in_tok
cdc
isoc_xfer_done
tdc
he
rx_pkt_size
Bit Default &
Range Access

31:29
28

27
26
transactions.

23
22:11
8

Bit Default &

Range Access
7
bit.
6
5:4

3:0

48h
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
isoc_data_pid
buff_size_1
Bit Default &

Range Access

31:18

17:16
15:10
RO to zero.


Offset 4Ch
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
mpkt_size
Bit Default &
Range Access
31:16
15:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
RW


Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
wr_cfrm
Bit Default &
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rrdy
f
reserved
mrx_flush
close_desc
null_bit
cnak
snak
nak
et
p
sn
Bit Default &

Range Access

31:13
12

11
10
9

Bit Default &

Range Access
0x0 bit 0).
set.

2
1

and OUT STALL bit.
- SetConfiguration
- SetInterface.

Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo_empty
mrxfifo_empty
reserved_1
xfer_done_txf_empty
set_feature_halt
clr_feature_halt
out_tok
reserved_3
bna
in_tok
cdc
isoc_xfer_done
tdc
he
rx_pkt_size
Bit Default &
Range Access

31:29
28

27
26
transactions.

23
22:11
8

Bit Default &

Range Access
7
bit.
6
5:4

3:0

68h
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size_frame_number 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
isoc_data_pid
buff_size_1
Bit Default &

Range Access

31:18

17:16
15:10
RO to zero.


Offset 6Ch
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
mpkt_size
Bit Default &
Range Access
31:16
15:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
RW


Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
wr_cfrm
Bit Default &
Range Access
31:0

200h
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
mrx_flush
close_desc
p
sn
s
null_bit
rrdy
cnak
snak
nak
et
Bit Default &

Range Access

31:13
Receive FIFO Flush for Multiple Receive FIFO (mrx_flush): When the application
wants to flush the endpoint receive FIFO, it must first set the SNAK bit in the Endpoint
0x0 Control register. If the receive DMA is in progress, then the core will finish the current
12
WO descriptor, terminate the DMA, and flush the data in the endpoint receive FIFO and clear
this bit. The application must clear this bit after the EP receive FIFO is empty, by
checking the MRXFIFO EMPTY bit in the Endpoint Status register.

11
0x0 Send Null Packet (null_bit): This is an IN endpoint field only. These bits are reserved
10

Bit Default &

Range Access
Receive Ready (rrdy): If this bit is set by the application, on receiving an OUT packet,
the DMA sends the packet to system memory. This bit is deasserted at the end of packet
0x0 if the Descriptor Update bit is set in the Device Control register. This bit is deasserted at
9
RW the end of payload if the Descriptor Update bit is deasserted. This bit can be set by the
application at any time. The application cannot clear this bit if the DMA is busy
transferring the data.
WO bit.)The application can clear this bit only when the
RxFIFO is empty (for single RxFIFO implementation) or when the RxFIFO corresponding
to the same logical endpoint is empty (Multiple RxFIFO implementation). The application
can write CNAK any time without waiting for a RxFIFO empty condition. The NAK bit is
cleared immediately upon write to CNAK bit. No polling is necessary.
0x0 Set NAK. (snak): Used by the application to set the NAK bit (bit 6 of this register). The
NAK Handshake Enable (nak): After a SETUP packet, which is decoded by the
application, is received by the core, the core sets the NAK bit for all control IN/OUT
endpoints. NAK is also set after a STALL response for the endpoint (the STALL bit is set
0x0 in Endpoint Control register bit 0).
set.
Note 2: The NAK set on ISOC OUT endpoint rejects the ISOC out data packet.
Endpoint Type (et): Endpoint Type (2-bit). The possible options are:
0x0 Poll Demand (p): This is an IN endpoint field only. These bits are reserved and should
3
RO be set to zero.
0x0 Snoop Mode (sn): Configures the endpoint for Snoop mode. In this mode, the
2 subsystem does not check the correctness of OUT packets before transferring them to
RW application memory.
0x0 Flush TxFIFO (f): This is an IN endpoint field only. These bits are reserved and should
1
RO be set to zero.

and OUT STALL bit.
handshake is returned to the USB host, then sets the corresponding NAK bit. The
0x0 subsystem returns a STALL handshake for the subsequent transactions of the stalled
0 endpoint until the USB host issues a Clear_Feature command to clear it. Once this bit is
RW set, if the subsystem has already returned a STALL handshake to the USB host, the
application firmware cannot clear the S bit to stop the subsystem from sending the
STALL handshake on the endpoint. The host must instead send one of the following
- SetConfiguration
- SetInterface.
16.6.1.26 OUT Endpoint 0 Status Register (ep0_out_sts_udc_reg)—Offset 204h

Access Method


(Size: 32 bits)

Default: 00000100h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
tx_fifo_empty
mrxfifo_empty
reserved_1
xfer_done_txf_empty
set_feature_halt
clr_feature_halt
out_tok
reserved_3
bna
in_tok
cdc
isoc_xfer_done
tdc
he
rx_pkt_size
Bit Default &
Range Access

31:29
28
0x0 Transfer Done/Transmit FIFO Empty (xfer_done_txf_empty): This is an IN

27
HALT) command is received for this endpoint .To stall this endpoint, the application can
26
RW/1C acknowledge the reception of Set Feature sta command. Once this RSS bit is cleared,
transactions.
0x0 Transmit FIFO Empty detected (tx_fifo_empty): This is an IN endpoint field only.
24
0x0 Isochronous IN transaction complete (isoc_xfer_done): This is an IN endpoint

23
Receive Packet Size (rx_pkt_size): Indicates the number of bytes in the current
receive packet the RxFIFO is receiving. Because the USB host always sends 8 bytes of
SETUP data, these bits do not indicate the receipt of 8 bytes of SETUP data for a SETUP
0x0 packet. Rather, these bits indicate the
22:11
RW configuration status (Configuration number [22:19], Interface number [18:15], and
Alternate
Setting number [14:11]). This field is used in slave mode only. In DMA mode, the
application must check the status from the endpoint data descriptor.
0x0 Transmit DMA Completion (tdc): This is an IN endpoint field only. These bits are
10
0x0 System Host Error (he): Error response on the host bus (AHB) when doing a data
9 transfer, descriptor fetch, or descriptor update for this particular endpoint. After
RW/1C servicing the interrupt, the application must clear this bit.

Bit Default &

Range Access
Receive Address FIFO Empty Status (mrxfifo_empty): This bit indicates the empty
status of the endpoint receive address FIFO. This bit is set by the core after the DMA
1h transfers data to system memory, and there are no new packets received from the USB.
8
RO This bit is cleared by the core after receiving a valid packet from the USB.
1: EP RXFIFO is empty
0: EP RXFIFO is not empty
7
RW/1C the DMA tried to access it.
After servicing the interrupt, the application must clear this bit.
0x0 IN token (in_tok): This is an IN endpoint field only. These bits are reserved and
6
OUT token. (out_tok): An OUT packet has been received by this endpoint. The
encoding of these two bits
indicates the type of data received. The possible options are:
0x0 00: None
5:4
RW/1C 01: Received data
10: Received SETUP data (8 bytes)
11: Reserved
(The application must write the same values to clear these bits.)

3:0

(ep0_out_rpf_udc_reg)—Offset 208h
This register contains the frame number in which the packet is received. This frame
number information is useful when handling isochronous traffic.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
isoc_data_pid
Bit Default &

Range Access

31:18

Bit Default &

Range Access
Isochronous OUT Transaction PID (isoc_data_pid): Data PID received for a high-
bandwidth isochronous OUT transaction. This field indicates that the data PID for the
current packet is available in the Receive FIFO. This field is used only in Slave-Only
0x0 mode.
17:16
RO 00: DATA0 PID is received
01: DATA1 PID is received
11: MDATA PID is received
Buffer Size/Frame Number (buff_size_frame_number): Frame number in which

the packet is received. For high-speed operation:
[15:14] Reserved
0x0 [13:3] Millisecond frame number
15:0
RO [2:0] Microframe number
For full-speed operation:
[15:11] Reserved
[10:0] Millisecond frame number

Offset 20Ch
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
max_pkt_size
Bit Default &

Range Access
Packet Size (buff_size): Buffer size required for this endpoint. The application can
program this field to make each endpoint buffers adaptive, providing flexibility in buffer
0x0 size when the interface or configuration is changed. This value is in 32-bit words, and
31:16
RW indicates the number of 32-bit word entries in the Receive FIFO. This value cannot be
changed dynamically during operation.
NOTE: the maximum size of the RxFIFO is 512 32-bit word entries.
Max Packet Size (max_pkt_size): Maximum packet size for the endpoint. This is the
0x0 value in bytes. This maximum size is used to calculate whether the Transmit FIFO has
15:0 sufficient space to accept a packet. When changing the maximum packet size for a
RW specific endpoint, the user must also program the corresponding Physical Endpoint
Register.

(ep0_subptr_udc_reg)—Offset 210h
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
subptr
Bit Default &
Range Access
SETUP Buffer Pointer (subptr): Endpoint SETUP buffer pointers are used for SETUP
0x0 commands. The Endpoint SETUP Buffer Pointer register is used only in DMA mode.
31:0
RW NOTE: This is applicable only to control endpoints. For all other endpoints this register is
reserved.

(ep0_out_desptr_udc_reg)—Offset 214h
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
31:0 Descriptor Pointer (desptr): This register contains the data descriptor pointer.
RW
16.6.1.31 OUT Endpoint 0 Read Confirmation Register for zero-length OUT data
(for Slave-Only mode) (ep0_rd_cfrm_udc_reg)—Offset 21Ch
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rd_cfrm
Bit Default &
Range Access
0x0 Read Confirmation for zero-length OUT data (for Slave-Only mode) (rd_cfrm):
31:0 For zero-length OUT data, the application must perform a dummy read from this
WO register

220h
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rrdy
close_desc
cnak
snak
nak
et
f
reserved
mrx_flush
null_bit
p
sn
s
Bit Default &
Range Access

31:13
12

11
10
9

Bit Default &

Range Access
set.
3
RO be set to zero.
1
RO be set to zero.

and OUT STALL bit.
- SetConfiguration
- SetInterface.

Access Method
(Size: 32 bits)


Default: 00000100h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
xfer_done_txf_empty
set_feature_halt
tx_fifo_empty
mrxfifo_empty
clr_feature_halt
bna
in_tok
out_tok
cdc
isoc_xfer_done
tdc
rx_pkt_size
he
reserved_1
reserved_3
Bit Default &
Range Access

31:29
28

27
26
transactions.
24

23
22:11
Alternate
10
8

Bit Default &

Range Access
7
6
0x0 00: None
5:4
11: Reserved

3:0

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
isoc_data_pid
Bit Default &

Range Access

31:18
0x0 mode.
17:16

Bit Default &

Range Access

[15:14] Reserved
15:0
[15:11] Reserved

Offset 22Ch
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
max_pkt_size
Bit Default &
Range Access
31:16
Register.

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
subptr
Bit Default &
Range Access
31:0
reserved.

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
RW
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rd_cfrm

Bit Default &

Range Access
WO register

240h
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
mrx_flush
rrdy
snak
nak
et
f
reserved
close_desc
null_bit
cnak
p
sn
s
Bit Default &
Range Access

31:13
12

11
10
9

Bit Default &

Range Access
set.
3
RO be set to zero.
1
RO be set to zero.

and OUT STALL bit.
- SetConfiguration
- SetInterface.

Access Method

Default: 00000100h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
xfer_done_txf_empty
set_feature_halt
tx_fifo_empty
mrxfifo_empty
clr_feature_halt
bna
in_tok
out_tok
cdc
isoc_xfer_done
tdc
rx_pkt_size
he
reserved_1
reserved_3
Bit Default &
Range Access

31:29
28

27
26
transactions.
24

23
22:11
Alternate
10
8
7

Bit Default &

Range Access
6
0x0 00: None
5:4
11: Reserved

3:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
isoc_data_pid
Bit Default &

Range Access

31:18
0x0 mode.
17:16

Bit Default &

Range Access

[15:14] Reserved
15:0
[15:11] Reserved

Offset 24Ch
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
max_pkt_size
Bit Default &
Range Access
31:16
Register.

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
subptr
Bit Default &
Range Access
31:0
reserved.

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
RW
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rd_cfrm

Bit Default &

Range Access
WO register

260h
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
mrx_flush
rrdy
snak
nak
et
f
reserved
close_desc
null_bit
cnak
p
sn
s
Bit Default &
Range Access

31:13
12

11
10
9

Bit Default &

Range Access
set.
3
RO be set to zero.
1
RO be set to zero.

and OUT STALL bit.
- SetConfiguration
- SetInterface.

Access Method

Default: 00000100h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
xfer_done_txf_empty
set_feature_halt
tx_fifo_empty
mrxfifo_empty
clr_feature_halt
bna
in_tok
out_tok
cdc
isoc_xfer_done
tdc
rx_pkt_size
he
reserved_1
reserved_3
Bit Default &
Range Access

31:29
28

27
26
transactions.
24

23
22:11
Alternate
10
8
7

Bit Default &

Range Access
6
0x0 00: None
5:4
11: Reserved

3:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
isoc_data_pid
Bit Default &

Range Access

31:18
0x0 mode.
17:16

Bit Default &

Range Access

[15:14] Reserved
15:0
[15:11] Reserved

Offset 26Ch
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buff_size
max_pkt_size
Bit Default &
Range Access
31:16
Register.

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
subptr
Bit Default &
Range Access
31:0
reserved.

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
desptr
Bit Default &

Range Access
0x0
RW
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rd_cfrm

Bit Default &

Range Access
WO register
16.6.1.53 Device Configuration Register (d_cfg_udc_reg)—Offset 400h

This register configures the device. It is only set during initial configuration or when
there is a change in the configuration.
Access Method
(Size: 32 bits)

Default: 00000020h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0
phy_error_detect
set_desc
status_1
pi
reserved
lpm_en
lpm_auto
ss_ddr
halt_status
hs_timeout_calib
status
csr_prg
dir
ss
fs_timeout_calib
sp
rwkp
spd
Bit Default &
Range Access

31:22
0x0 Link Power Mode Enable (lpm_en): Link Power Mode is not supported. These bits
21
RO are reserved and should be set to zero.
0x0 Link Power Mode Automatic (lpm_auto): Link Power Mode is not supported. These
20
RO bits are reserved and should be set to zero.
0x0 Double Data Rate (ss_ddr): ULPI PHY interface is not supported. These bits are
19
Set Descriptor Request Enable (set_desc): Indicates that the device supports Set
0x0 Descriptor requests.
18
RW 0: The USB Device Controller returns a STALL handshake to the USB host.
1: The SETUP packet for the Set Descriptor request passes to the application.
Dynamic Register Programming (csr_prg): The application can program the USB
Device Controller registers dynamically whenever it has received an interrupt for either
0x0 a Set Configuration or a Set Interface request. If this bit is set to 1, the USB Device
17
RW Controller returns a NAK handshake during the status IN stage of both the Set
Configuration and Set Interface requests until the application has written 1 to the
CSR_DONE bit 13 of the Device Control Register.
Halt Status (halt_status): This bit indicates whether the USB Device Controller must
0x0 respond with a STALL or an ACK handshake when the USB host has issued a
16 Clear_Feature (ENDPOINT_HALT) request for Endpoint 0. Options are:
RW 0: ACK
1: STALL
HS Timeout Counter (hs_timeout_calib): Number of PHY Clocks to the USB Device

0x0 Controller Timeout Counter(for HS).The application uses these bits to increase the
15:13
RW timeout value (736 to 848 bit times in high-speed operation), which depends on the PHY
delay in generating a line state condition. The default timeout value is 736 bit times.

Bit Default &

Range Access
FS Timeout Counter (fs_timeout_calib): Number of PHY Clocks to the USB Device

0x0 Controller Timeout Counter(for FS).These three bits indicate the number of PHY clocks
12:10 to the USB Device Controller timeout counter. The application uses these bits to increase
RW the timeout value (16 to 18 bit times in full-speed operation), which depends on the PHY
delay in generating line state condition. The default timeout value is 16 bit times.
0x0 PHY Error Detect (phy_error_detect): If the application sets this bit, the device
9 detects the phy_rxvalid or phy_rxactive input signal to be continuously asserted for 2
RW ms, indicating PHY error.
Non-zero Control Handshake 1 (status_1): This bit, together with STATUS Bit 7,
0x0 provides an option for the USB Device Controller to respond to the USB host with a
8
RW STALL or ACK handshake if the USB host has issued a non-zero-length data packet
during the STATUS-OUT stage of a CONTROL transfer.
Non-zero Control Handshake (status): This bit, together with STATUS Bit 8,
0x0 provides an option for the USB Device Controller to respond to the USB host with a
7
RW STALL or ACK handshake if the USB host has issued a non-zero-length data packet
during the STATUS-OUT stage of a CONTROL transfer.
UTMI Data Bus Direction (dir): This bit indicates if the UTMI data bus interface has to
0x0 support a unidirectional or bidirectional interface.
6
RO 0: Unidirectional interface
1: Bidirectional interface
PHY Interface Width (pi): Indicates if the UTMI PHY supports an 8-bit or 16-bit
interface.
1h 0: 16-bit
5
RW 1: 8-bit.
NOTE: even if this field is writable, only 8bit interface is supported. Writing 0 will lead to
undefined behavior.
0x0
4 Sync Frame Support (ss): Indicates that the Device Supports Sync Frame.
RW
0x0
3 Self-Powered Device (sp): Indicates that the Device is Self-Powered.
RW
0x0 Remote Wake Up Capable (rwkp): Indicates that the device is remote wake up
2
RW capable.
Device Speed (spd): This is the expected speed the application programs to the
subsystem. The actual speed the subsystem operates depends on the enumeration
0x0 speed (ENUM SPD) of the Device Status register.
1:0 00: HS (PHY clock = 60 MHz)
RW 01: FS (PHY clock = 60 MHz)
10: Reserved
11: Reserved
16.6.1.54 Device Control Register (d_ctrl_udc_reg)—Offset 404h

This register is set at runtime and controls the device after device configuration.
Access Method
(Size: 32 bits)

Default: 00000400h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0
devnak
bf
thlen
brlen
bren
srx_flush
csr_done
scale
sd
du
res
mode
the
be
tde
rde
reserved_1
reserved_2
Bit Default &
Range Access
0x0 Threshold Length (thlen): Indicates the number (THLEN + 1) of 32-bit entries in the
31:24
RW RxFIFO before the DMA can start data transfer.
0x0 Burst Length (brlen): Indicates the length, in 32-bit transfers, of a single burst on the
23:16
RW AHB. The subsystem sends number of 32-bit transfers equal to (BRLEN + 1).

15
0x0 Receive FIFO Flush for Single Receive FIFO (srx_flush): Multiple receive FIFOs are
14
RO implemented. These bits are reserved and should be set to zero.
Dynamic Register Programming Done (csr_done): The application uses this bit to
0x0 notify the USB Device Controller that the application has completed programming all
13
WO required USB Device Controller registers, and the Subsystem can acknowledge the
current Set Configuration or Set Interface command.
Device NAK (devnak): When the application sets this bit, the Subsystem core returns
0x0 a NAK handshake to all OUT endpoints. By writing 1 to this bit, the application does not
12
RW need to write 1 to the
SNAK bit 7 of each Endpoint Control register.
Scale Down (scale): This bit reduces the timer values inside the USB Device Controller
0x0 when running gate-level simulation only. When this bit is set to 1, timer values are
11 scaled down to reduce simulation time. In Scale-Down mode, the USB Device Controller
RW detects a USB reset within 150 PHY clock cycles (60-MHz PHY clock, 8-bit UTMI).
Reset this bit to 0 for normal operation.
1h Soft Disconnect (sd): The application software uses this bit to signal the USB Device
10 Controller to soft-disconnect. When set to 1, this bit causes the device to enter the
RW disconnected state.
0x0 DMA/Slave-Only Mode (mode): Enables the application to dictate the subsystem
9
RW operation in either DMA mode (1) or Slave-Only mode (0) operation.
0x0
8 Burst Enable (bren): When this bit is set, transfers on the AHB are split into bursts.
RW
0x0 Threshold Enable (the): When this bit is set, a number of quadlets equivalent to the
7
RW threshold value is transferred from the RxFIFO to the memory.
0x0 Buffer Fill Mode (bf): The DMA is in Buffer Fill mode and transfers data into
6
RW contiguous locations pointed to by the buffer address.
0x0
5 System Endianness (be): A value of 1 indicates a big endian system.
RW
0x0 Descriptor Update (du): When this bit is set, the DMA updates the descriptor at the
4
RW end of each packet processed.
0x0 Transmit DMA Enable (tde): 0: disabled

3
RW 1: enabled
0x0 Receive DMA Enable (rde): 0: disabled

2
RW 1: enabled

1

Bit Default &

Range Access
Remote Wakeup Resume (res): To perform a remote wakeup resume the application
sets this bit to 1, then resets it to 0 after 1 ms. The USB Device Controller signals the
0x0 USB host to resume the USB bus. However:
0 The application must first set RWKP bit 2 in the Device Configuration Register, indicating
RW that the subsystem supports the Remote Wakeup feature.
The host must already have issued a Set Feature request to enable the device Remote
Wakeup feature.
16.6.1.55 Device Status Register (d_sts_udc_reg)—Offset 408h

This register reflects status information needed to service some of the interrupts.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rmtwkup_feat_sts
phy_error
rxfifo_empty
alt
intf
ts
enum_spd
susp
cfg
Bit Default &
Range Access
Received SOF Frame Number (ts): For high-speed operation:

[31:21]: Millisecond frame number
0x0 [20:18]: Microframe number
31:18
RO For full-speed operation:
[31:29]: Reserved
[28:18]: Millisecond frame number
Remote Wakeup Status (rmtwkup_feat_sts): Status of Remote wakeup feature

due to Set/Clear Feature (Remotewakeup) command from the host. A value of 1
0x0 indicates a Set Feature (Remotewakeup) has been received. A value of 0 indicates Clear
17
RO Feature (Remotewakeup) has been received.
Any change to this bit sets an interrupt in bit 7 of Device Interrupt register, if not
masked.
PHY Error (phy_error): Either the phy_rxvalid or phy_rxactive input signal is detected
to be continuously asserted for 2 ms, indicating PHY error. The USB Device Controller
goes to the Suspend state
0x0 as a result. When the application serves the early suspend interrupt (ES bit 2 of the
16
RO Device
Interrupt register) it also must check this bit to determine if the early suspend interrupt
was
generated due to PHY error detection.
0h Receive Address FIFO Empty Status (rxfifo_empty): Multiple receive FIFOs are
15
RO implemented. These bits are reserved and should be set to zero.

Bit Default &

Range Access
Enumerated Speed (enum_spd): These bits hold the speed at which the subsystem
comes up after the speed enumeration. Possible options are:
00 HS: If the SPD is high speed and the subsystem connects to a 2.0 host controller,
then after Speed Enumeration, these bits indicate that the subsystem is operating in
0x0 high speed mode.
14:13
RO 01 FS: If the expected speed (SPD of the Device Configuration register) is high speed
and the subsystem connects to a 1.1 host controller, then after Speed Enumeration,
these bits indicate that the subsystem is operating in full speed mode.
10 : Reserved
110: Reserved
0x0 Suspend Status (susp): This bit is set as long as a Suspend condition is detected on
12
RO the USB.
0x0 Alternate Setting (alt): This 4-bit field represents the alternate setting to which the
11:8
RO above interface is switched.
0x0 SetInterface Command (intf): This 4-bit field reflects the interface set by the
7:4
RO SetInterface command.
0x0 SetConfiguration Command (cfg): This 4-bit field reflects the configuration set by
3:0
RO the SetConfiguration command.
16.6.1.56 Device Interrupt Register (d_intr_udc_reg)—Offset 40Ch

Device interrupts are set when there are system-level events. Interrupts are used by
the application to make system-level decisions. After checking the register, the
application must clear the interrupt by writing a 1?b1 to the correct bit.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
sof
si
reserved
lpm_tkn
enumc
ur
rmtwkup
us
es
sc
e_slpm
slpm
Bit Default &

Range Access

31:11
0x0 LPM Early Sleep (e_slpm): Link Power Mode is not supported. These bits are reserved
10
0x0 LPM Sleep (slpm): Link Power Mode is not supported. These bits are reserved and
9
0x0 LPM Transaction (lpm_tkn): Link Power Mode is not supported. These bits are
8
0x0 Remote Wakeup (rmtwkup): A Set/Clear Feature (Remote Wakeup) is received by

7 the core. This bit is set by the core whenever bit 17 of the Device Status Register
RW/1C changes: HIGH to LOW or LOW to HIGH.

Bit Default &

Range Access
0x0
6 Speed Enumeration Completed (enumc): Speed enumeration is complete.
RW/1C
0x0
5 SOF Token Detected (sof): An SOF token is detected on the USB.
RW/1C
0x0 Suspend State (us): A suspend state is detected on the USB for a duration of 3
4
RW/1C milliseconds, following the 3 ms Idle State interrupt activity due to an idle state.
0x0 USB Reset Detected (ur): NOTE: If the application has not served this interrupt, the
3 USB Device Controller returns a NAK handshake for all transactions except the 8 SETUP
RW/1C packet bytes from the USB host.
Idle State Detected (es): An idle state is detected on the USB for a duration of 3 ms.
0x0 This interrupt bit is used for the application firmware to finish its job before the
2
RW/1C subsystem generates a true suspend (US) interrupt (another 3 ms after the ES
interrupt)
Set_Interface Command Received (si): The device has received a Set_Interface

0x0 command.
1 NOTE: If the application has not served this interrupt, the USB Device Controller returns
RW/1C a NAK handshake for all transactions except the 8 SETUP packet bytes from the USB
host.
0x0 Set_Configuration Command Received (sc): NOTE: If the application has not
0 served this interrupt, the USB Device Controller returns a NAK handshake for all
RW/1C transactions except the 8 SETUP packet bytes from the USB host.
16.6.1.57 Device Interrupt Mask Register (d_intr_msk_udc_reg)—Offset 410h

The device interrupt mask can be set for system-level interrupts using this register.
Programming 1'b1 in the appropriate bit position in the Interrupt Mask register masks
the designated interrupt. Once masked, an interrupt signal does not reach the
application, nor does its interrupt bit get set.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
lpm_tkn
e_slpm
slpm
mask1
Bit Default &

Range Access

31:11
0x0 LPM Early Sleep (e_slpm): Link Power Mode is not supported. These bits are reserved
10
0x0 LPM Sleep (slpm): Link Power Mode is not supported. These bits are reserved and
9
0x0 LPM Transaction (lpm_tkn): Link Power Mode is not supported. These bits are
8

Bit Default &

Range Access
0x0 Device Interrupt Mask (mask1): Masks equivalent device interrupt bit in the Device
7:0
RW Interrupt Register.
16.6.1.58 Endpoints Interrupt Register (ep_intr_udc_reg)—Offset 414h

The Endpoint Interrupt register is used to set endpoint-level interrupts. Since all 4
endpoints are bidirectional, each endpoint has two interrupt bits (one for each
direction). The application needs to clear the interrupt by writing a 1'b1 to the correct
bit after checking the register.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
out_ep
in_ep
Bit Default &
Range Access
OUT Endpoint Interrupt (out_ep): One bit is associated to one of the 4 supported
OUT endpoint, set when there is an event on that endpoint.
0x0 bit0: OUT ED0
31:16 bit1: OUT ED1
RW/1C bit2: OUT ED2
bit3: OUT ED3
bit4-bit15: Reserved
IN Endpoint Interrupt (in_ep): One bit is associated to one of the 4 supported IN

endpoint, set when there is an event on that endpoint.
0x0 bit0: IN ED0
15:0 bit1: IN ED1
RW/1C bit2: IN ED2
bit3: IN ED3
16.6.1.59 Endpoints Interrupt Mask Register (ep_intr_msk_udc_reg)—Offset

418h
This register is used to mask endpoint interrupts. A write of 1'b1 to any bit in this
register masks the corresponding endpoint for any possible interrupts. Once masked,
an interrupt signal does not reach the application, nor does its interrupt bit get set.
Access Method

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
out_ep_mask
in_ep_mask
Bit Default &
Range Access
OUT Endpoint Interrupt Mask (out_ep_mask): Masks the OUT Endpoint Interrupt
Register bits of the equivalent OUT endpoint.
0h bit0: OUT ED0
31:16 bit1: OUT ED1
RW bit2: OUT ED2
bit3: OUT ED3
IN Endpoint Interrupt Mask (in_ep_mask): Masks the IN Endpoint Interrupt

Register bits of the equivalent IN endpoint.
0h bit0: IN ED0
15:0 bit1: IN ED1
RW bit2: IN ED2
bit3: IN ED3
16.6.1.60 Test Mode Register (test_mode_udc_reg)—Offset 41Ch

In Test mode, the application can use the AHB to read from a TxFIFO or write to an
RxFIFO using AHB read/write cycles. Test mode is supported only in the Slave-Only
operational mode. In Test mode, only single- DWORD transactions are supported: byte
and word transactions on the AHB are not supported. In non-Test modes, reading from
a TxFIFO or writing to an RxFIFO results in an AHB error response. The application
must never read an empty TxFIFO or write a full RxFIFO even though an AHB error
response is not provided. NOTE: Writing to the RxFIFO in Test mode and enabling the
DMA to transfer the data are not supported.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
tstmode
Bit Default &

Range Access

31:1
0x0 Test Mode indicator (tstmode): 0: Normal mode

0
RW 1: Test mode

16.6.1.61 Product Release Number Register (revision_udc_reg)—Offset 420h

Access Method
(Size: 32 bits)

Default: 3234352Ah
31 28 24 20 16 12 8 4 0
0 0 1 1 0 0 1 0 0 0 1 1 0 1 0 0 0 0 1 1 0 1 0 1 0 0 1 0 1 0 1 0
release_id
Bit Default &
Range Access
Product Release Number (release_id): This field indicates the ASCII characters of
3234352Ah the four-digit release number in hexadecimal format. For example, 32_34_35_ 2A
31:0
RO represents 2.45* in ASCII character, where * is an alphabetic character (for example, a,
b, or c) that represents a update to the release, which does not impact the RTL source.
16.6.1.62 SETUP command address pointer register (udc_desc_addr_udc_reg)—

Offset 500h
NOTE: The SETUP command address pointer register is not writable and returns 0 when
read because the SETUP command address pointer is hardcoded to xFFF0.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
dev_desc_addr_ptr
su_cmd_addr_ptr
Bit Default &

Range Access
0x0 Device Descriptor Address Pointer (dev_desc_addr_ptr): Device Descriptor

31:16
RO Address Pointer
0x0 SETUP Command Address Pointer (su_cmd_addr_ptr): SETUP Command Address

15:0
RO Pointer

16.6.1.63 Physical Endpoint 0 Register (udc_ep_ne_udc_reg_0)—Offset 504h

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
alt
intf
reserved
ep_dir
max_pkt_size
cfg
ep_type
ep_num
Bit Default &
Range Access

31:30
RW These bits are reserved and should be set to zero.
0x0 Maximum Packet Size (max_pkt_size): NOTE: In full-speed mode operation,

29:19 application firmware must program the maximum packet size of default endpoint 0 in
RW accordance with the value defined in USB 2.0 specification
0x0
18:15 Alternate Setting (alt): Alternate setting to which this endpoint belongs.
RW
0x0
14:11 Interface Number (intf): Interface number to which this endpoint belongs.
RW
0x0
10:7 Configuration Number (cfg): Configuration number to which this endpoint belongs
RW
Endpoint Type (ep_type): The possible options are:

0x0 00: Control
6:5 01: Isochronous
RW 10: Bulk
11: Interrupt
0x0 Endpoint Direction (ep_dir): 0: OUT

4
RW 1: IN
0x0
3:0 Logical Endpoint Number (ep_num): Logical Endpoint Number
RW

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
alt
intf
reserved
ep_dir
max_pkt_size
cfg
ep_type
ep_num
Bit Default &
Range Access

31:30

0x0
RW
0x0
RW
0x0
RW

0x0 00: Control
6:5 01: Isochronous
RW 10: Bulk
11: Interrupt

4
RW 1: IN
0x0
RW
16.6.1.65 Physical Endpoint 2 Register (udc_ep_ne_udc_reg_2)—Offset 50Ch

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ep_num
alt
intf
reserved
ep_dir
max_pkt_size
cfg
ep_type
Bit Default &

Range Access

31:30

Bit Default &

Range Access

0x0
RW
0x0
RW
0x0
RW

0x0 00: Control
6:5 01: Isochronous
RW 10: Bulk
11: Interrupt

4
RW 1: IN
0x0
RW

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
ep_dir
max_pkt_size
cfg
ep_type
ep_num
intf
alt
Bit Default &

Range Access

31:30

0x0
RW
0x0
RW
0x0
RW

Bit Default &

Range Access

0x0 00: Control
6:5 01: Isochronous
RW 10: Bulk
11: Interrupt

4
RW 1: IN
0x0
RW

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
ep_dir
max_pkt_size
cfg
ep_type
intf
ep_num
alt
Bit Default &

Range Access

31:30

0x0
RW
0x0
RW
0x0
RW

0x0 00: Control
6:5 01: Isochronous
RW 10: Bulk
11: Interrupt

4
RW 1: IN
0x0
RW


Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
alt
intf
reserved
ep_dir
max_pkt_size
cfg
ep_type
ep_num
Bit Default &
Range Access

31:30

0x0
RW
0x0
RW
0x0
RW

0x0 00: Control
6:5 01: Isochronous
RW 10: Bulk
11: Interrupt

4
RW 1: IN
0x0
RW
16.6.1.69 Physical Endpoint 6 Register (udc_ep_ne_udc_reg_6)—Offset 51Ch

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
alt
intf
reserved
ep_dir
max_pkt_size
cfg
ep_type
ep_num
Bit Default &
Range Access

31:30

0x0
RW
0x0
RW
0x0
RW

0x0 00: Control
6:5 01: Isochronous
RW 10: Bulk
11: Interrupt

4
RW 1: IN
0x0
RW
16.6.1.70 RxFIFO Array[0-511] (udc_rx_fifo_reg_array[0-511])—Offset 800h,

Count 512, Stride 4h
Access Method
(Size: 32 bits) Offset[0-511]: [BAR0] + 800h + [0-511]*4h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rx_fifo
Bit Default &

Range Access
0x0
31:0 Receive FIFO (rx_fifo): Receive FIFO
RW


Access Method
(Size: 32 bits) Offset[0-255]: [BAR0] + 1000h + [0-255]*4h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo
Bit Default &
Range Access
0x0
31:0 Transmit FIFO 0 (tx_fifo): Transmit FIFO 0
RW

Access Method
Type: Memory Mapped I/O Register Offset[0-255]: [BAR0] + 1400h + [0-255]*4h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo
Bit Default &

Range Access
0x0
RW

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo
Bit Default &
Range Access
0x0
RW

1C00h, Count 256, Stride 4h
Access Method
Type: Memory Mapped I/O Register Offset[0-255]: [BAR0] + 1C00h + [0-255]*4h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
tx_fifo
Bit Default &

Range Access
0x0
RW
16.6.2 USB EHCI

Offset Default
Start Value
“Host Controller Interface Version Number and Capability Registers Length

0h 3h 01000010h
(HCCAPBASE)—Offset 0h” on page 550
4h 7h “Host Controller Structural Parameters (HCSPARAMS)—Offset 4h” on page 550 00001212h
8h Bh “Host Controller Capability Parameters (HCCPARAMS)—Offset 8h” on page 552 0000C012h
10h 13h “USB Command (USBCMD)—Offset 10h” on page 553 00080000h
14h 17h “USB Status (USBSTS)—Offset 14h” on page 555 00001000h
18h 1Bh “USB Interrupt Enable (USBINTR)—Offset 18h” on page 557 00000000h
1Ch 1Fh “USB Frame Index (FRINDEX)—Offset 1Ch” on page 558 00000000h
20h 23h “4 Gigabyte Memory Segment Selector (CTRLDSSEGMENT)—Offset 20h” on page 559 00000000h
24h 27h “Periodic Frame List Base Address (PERIODICLISTBASE)—Offset 24h” on page 559 00000000h


Offset Default
Start Value
28h 2Bh “Asynchronous List Address (ASYNCLISTADDR)—Offset 28h” on page 560 00000000h
50h 53h “Configure Flag (CONFIGFLAG)—Offset 50h” on page 560 00000000h
54h + [0-
57h “Port Status/Control[0-1] (PORTSC[0-1])—Offset 54h, Count 2, Stride 4h” on page 561 00002000h
1]*4h
90h 93h “Programmable Microframe Base Value (INSNREG00)—Offset 90h” on page 564 00000000h
“Programmable Packet Buffer OUT/IN Thresholds (INSNREG01)—Offset 94h” on

94h 97h 00200020h
page 565
98h 9Bh “Programmable Packet Buffer Depth (INSNREG02)—Offset 98h” on page 565 00000080h
9Ch 9Fh “Programmable Controller Settings (INSNREG03)—Offset 9Ch” on page 566 00002001h
A0h A3h “Programmable Controller Settings (INSNREG04)—Offset A0h” on page 567 00000000h
A4h A7h “UTMI Configuration (INSNREG05)—Offset A4h” on page 568 00001000h
16.6.2.1 Host Controller Interface Version Number and Capability Registers

Length (HCCAPBASE)—Offset 0h
Access Method
(Size: 32 bits)

Default: 01000010h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
reserved_15_8
caplength
hciversion
Bit Default &

Range Access
Host Controller Interface Version Number (hciversion): This is a two-byte register

0100h containing a BCD encoding of the EHCI revision number supported by this host
31:16
RO controller. The most significant byte of this register represents a major revision and the
least significant byte is the minor revision.
0h reserved_15_8: Reserved bits.

15:8
010h Capability Registers Length (caplength): This register is used as an offset to add to
7:0
RO register base to find the beginning of the Operational Register Space.
16.6.2.2 Host Controller Structural Parameters (HCSPARAMS)—Offset 4h

Access Method
(Size: 32 bits)


Default: 00001212h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1 0 0 1 0
debug_port_number
p_indicator
n_cc
reserved_31_24
reserved_19_17
n_pcc
port_route_rules
n_ports
reserved_6_5
ppc
Bit Default &
Range Access

31:24
Debug Port Number (debug_port_number): This register identifies which of the

0h host controller ports is the debug port. The value is the port number (one-based) of the
23:20 debug port. A non-zero value in this field indicates the presence of a debug port. The
RO value in this register must not be greater than N_PORTS.
NOTE: Debug Port is not supported
000b reserved_19_17: Reserved bits.

19:17
Port Indicator (p_indicator): This bit indicates whether the ports support port
0b indicator control. When this bit is a one, the port status and control registers include a
16
RO read/writeable field for controlling the state of the port indicator.
NOTE: Port Indicator is not supported
Number of Companion Controllers (n_cc): This field indicates the number of

companion controllers associated with this USB 2.0 host controller. A zero in this field
1h indicates there are no companion host controllers. Port-ownership hand-off is not
15:12 supported. Only high-speed devices are supported on the host controller root ports.
RO A value larger than zero in this field indicates there are companion USB 1.1 host
controller(s). Port-ownership hand-offs are supported. High, Full- and Low-speed
devices are supported on the host controller root ports.
Number of Ports per Companion Controller (n_pcc): This field indicates the
number of ports supported per companion host controller. It is used to indicate the port
routing configuration to system software.
2h For example, if N_PORTS has a value of 6 and N_CC has a value of 2 then N_PCC could
11:8 have a value of 3. The convention is that the first N_PCC ports are assumed to be routed
RO to companion controller 1, the next N_PCC ports to companion controller 2, etc. In the
previous example, the N_PCC could have been 4, where the first 4 are routed to
companion controller 1 and the last two are routed to companion controller 2. The
number in this field must be consistent with N_PORTS and N_CC.
Port Routing Rules (port_route_rules): This field indicates the method used by this
implementation for how all ports are mapped to companion controllers. The value of this
field has the following interpretation:
0b 0: The first N_PCC ports are routed to the lowest numbered function companion host
7
RO controller, the next N_PCC port are routed to the next lowest function companion
controller, and so on.
1: The port routing is explicitly enumerated by the first N_PORTS elements of the HCSP-
PORTROUTE array.
00b reserved_6_5: Reserved bits.

6:5
Port Power Control (ppc): This field indicates whether the host controller
implementation includes port power control. A one in this bit indicates the ports have
1b port power switches. A zero in this bit indicates the port do not have port power
4
RO switches. The value of this field affects the functionality of the Port Power field in each
port status and control register.
NOTE: Port Power Control is supported.

Bit Default &

Range Access
Number of Physical Downstream Ports (n_ports): This field specifies the number
2h of physical downstream ports implemented on this host controller. The value of this field
3:0
RO determines how many port registers are addressable in the Operational Register Space.
Valid values are in the range of 1 to F. A zero in this field is undefined.
16.6.2.3 Host Controller Capability Parameters (HCCPARAMS)—Offset 8h

Access Method
(Size: 32 bits)

Default: 0000C012h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0
reserved_31_18
reserved_16
reserved_3
link_power_mgmt_cap
eecp
isoc_schedule_threshold
async_schedule_park_cap
frame_list_flag
address_64bit_cap
Bit Default &
Range Access

31:18
0h Link Power Management Capability (link_power_mgmt_cap): NOTE: Link Power

17
RO Management is not supported.
0h reserved_16: Reserved bits.

16
EHCI Extended Capabilities Pointer (eecp): This optional field indicates the
existence of a capabilities list. A value of 00h indicates no extended capabilities are
C0h implemented. A non-zero value in this register indicates the offset in PCI configuration
15:8 space of the first EHCI extended capability. The pointer value must be 40 or greater if
RO implemented to maintain the consistency of the PCI header defined for this class of
device.
NOTE: EHCI Extended Capabilities is supported.
Isochronous Scheduling Threshold (isoc_schedule_threshold): This field

indicates, relative to the current position of the executing host controller, where
software can reliably update the isochronous schedule.
1h When bit [7] is zero, the value of the least significant 3 bits indicates the number of
7:4
RO micro-frames a host controller can hold a set of isochronous data structures (one or
more) before flushing the state.
When bit [7] is a one, then host software assumes the host controller may cache an
isochronous data structure for an entire frame.

3

Bit Default &

Range Access
Async Schedule Park Capability (async_schedule_park_cap): If this bit is set to a

one, then the host controller supports the park feature for high-speed queue heads in
0h the Asynchronous Schedule and he feature can be disabled or enabled and set to a
2
RO specific level by using the Asynchronous Schedule Park Mode Enable and Asynchronous
Schedule Park Mode Count fields in the USBCMD register.
NOTE: Async Schedule Park capability is not supported.
Programmable Frame List Flag (frame_list_flag): If this bit is set to a zero, then
system software must use a frame list length of 1024 elements with this host controller
and the USBCMD register Frame List Size field is a read-only register.
1h If set to a one, then system software can specify and use a smaller frame list and
1
RO configure the host controller via the USBCMD register Frame List Size field. The frame
list must always be aligned on a 4K page boundary. This requirement ensures that the
frame list is always physically contiguous.
NOTE: Programmable Frame List Flag is supported.
64-bit Addressing Capability (address_64bit_cap): This field documents the

addressing range capability of this implementation.
0h The value of this field determines whether software should use the data structures.
0 Values for this field have the following interpretation:
RO 0: data structures using 32-bit address memory pointers
1: data structures using 64-bit address memory pointers
NOTE: 64-bit Addressing is not supported
16.6.2.4 USB Command (USBCMD)—Offset 10h

This register indicates the command to be executed by the serial bus host controller.
Writing to the register causes a command to be executed.
Access Method
(Size: 32 bits)

Default: 00080000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_28
reserved_15_12
async_schedule_park_mode_cnt
host_initiated_resume_duration
async_schedule_park_mode_enable
intr_on_async_advance_drbell
intr_threshold_ctrl
reserved_10
light_hcreset
hcreset
async_schedule_enable
run_stop
periodic_schedule_enable
frame_list_size
Bit Default &

Range Access

31:28

Bit Default &

Range Access
Host Initiated Resume Duration (host_initiated_resume_duration): If the Link

0h Power Management Capability bit in the HCCPARAMS register is set to one then this bit
27:24
RO is R/W; otherwise RO and not functional.
NOTE: Link Power Management is not supported.
Interrupt Threshold Control (intr_threshold_ctrl): This field is used by system

software to select the maximum rate at which the host controller will issue interrupts. If
software writes an invalid value to this register, the results are undefined. Allowed
values are:
00: Reserved
01: 1 micro-frame
08h 02: 2 micro-frames
23:16
RW 04: 4 micro-frames
08: 8 micro-frames (default, equates to 1ms)
10: 16 micro-frames (2ms)
Any other value in this register yields undefined results. Software modifications to this
bit while HCHalted bit is equal to zero results in undefined behavior.

15:12
Asynchronous Schedule Park Mode Enable

(async_schedule_park_mode_enable): If the Asynchronous Park Capability bit in
0h the HCCPARAMS register is a one, then this bit defaults to a 1 and is R/W. Otherwise the
11 bit must be a zero and is RO. Software uses this bit to enable or disable Park mode.
RO When this bit is one, Park mode is enabled. When this bit is a zero, Park mode is
disabled.
NOTE: Asynchronous Park Capability is not supported.

10
Asynchronous Schedule Park Mode Count (async_schedule_park_mode_cnt):

If the Asynchronous Park Capability bit in the HCCPARAMS register is a one, then this
field defaults to 3 and is R/W. Otherwise it defaults to zero and is RO. It contains a count
0h of the number of successive transactions the host controller is allowed to execute from a
9:8
RO high-speed queue head on the Asynchronous schedule before continuing traversal of the
Asynchronous schedule. Valid values are 1h to 3h. Software must not write a zero to this
bit when Park Mode Enable is a one as this will result in undefined behavior.
NOTE: Asynchronous Park Capability is not supported.
Light Host Controller Reset (light_hcreset): This control bit allows the driver to
reset the EHCI controller without affecting the state of the ports or the relationship to
the companion host controllers. For example, the PORSTC registers should not be reset
to their default values and the CF bit setting should not go to zero (retaining port
0h ownership relationships).A host software read of this bit as zero indicates the Light Host
7 Controller Reset has completed and it is safe for host software to re-initialize the host
RW controller. A host software read of this bit as a one indicates the Light Host Controller
Reset has not yet completed. If not implemented a read of this field will always return a
zero.
NOTE: this control bit is supported and resets the EHCI List Processor Master Controller
Unit.
Interrupt on Async Advance Doorbell (intr_on_async_advance_drbell): This bit

is used as a doorbell by software to tell the host controller to issue an interrupt the next
time it advances asynchronous schedule.
Software must write a 1 to this bit to ring the doorbell. When the host controller has
0h evicted all appropriate cached schedule state, it sets the Interrupt on Async Advance
6 status bit in the USBSTS register. If the Interrupt on Async Advance Enable bit in the
RW USBINTR register is a one then the host controller will assert an interrupt at the next
interrupt threshold. The host controller sets this bit to a zero after it has set the
Interrupt on Async Advance status bit in the USBSTS register to a one.
Software should not write a one to this bit when the asynchronous schedule is disabled.
Doing so will yield undefined results.
Asynchronous Schedule Enable (async_schedule_enable): This bit controls

0h whether the host controller skips processing the Asynchronous Schedule. Values mean:
5
RW 0: Do not process the Asynchronous Schedule
1: Use the ASYNCLISTADDR register to access the Asynchronous Schedule

Bit Default &

Range Access
Periodic Schedule Enable (periodic_schedule_enable): This bit controls whether

0h the host controller skips processing the Periodic Schedule. Values mean:
4
RW 0: Do not process the Periodic Schedule
1: Use the PERIODICLISTBASE register to access the Periodic Schedule
Frame List Size (frame_list_size): This field is RW only if Programmable Frame List
Flag in the HCCPARAMS registers is set to a one. This field specifies the size of the frame
list. The size the frame list controls which bits in the Frame Index Register should be
0h used for the Frame List Current index. Values mean:
3:2
RW 00: 1024 elements (4096) Default value
01: 512 elements (2048 )
10: 256 elements (1024 ) for resource-constrained environments
11: Reserved
Host Controller Reset (hcreset): This control bit is used by software to reset host
controller.
The effects of this on Root Hub registers are similar to a Chip Hardware Reset. When
software writes a one to this bit, the Host Controller resets its internal pipelines, timers,
counters, state machines, etc. to their initial value. Any transaction currently in progress
on USB is immediately terminated. A USB reset is not driven on downstream ports.
0h PCI Configuration registers are not affected by this reset.
1
RW All operational registers, including port registers and port state machines are set to their
initial values. Port ownership reverts to the companion host controller(s).
This bit is set to zero by the Host Controller when the reset process is complete.
Software cannot terminate the reset process early by writing a zero to this register.
Software should not set this bit to a one when the HCHalted bit in the USBSTS register
is a zero. Attempting to reset an actively running host controller will result in undefined
behavior.
Run/Stop bit (run_stop): 1: Run

0: Stop
When set to a 1, the Host Controller proceeds with execution of the schedule. The Host
Controller continues execution as long as this bit is set to a 1. When this bit is set to 0,
the Host Controller completes the current and any actively pipelined transactions on the
0h USB and then halts. The Host Controller must halt within 16 micro-frames after software
0
RW clears the Run bit. The HC Halted bit in the status register indicates when the Host
Controller has finished its pending pipelined transactions and has entered the stopped
state.
Software must not write a one to this field unless the host controller is in the Halted
state (i.e. HCHalted in the USBSTS register is a one). Doing so will yield undefined
results.
16.6.2.5 USB Status (USBSTS)—Offset 14h

This register indicates pending interrupts and various states of the Host Controller. The
status resulting from a transaction on the serial bus is not indicated in this register.
Access Method
(Size: 32 bits)

Default: 00001000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0
host_system_error
periodic_schedule_status
port_change_detect
async_schedule_status
usberrint
reclamation
usbint
reserved_31_16
hchalted
reserved_11_6
intr_on_async_advance
frame_list_rollover
Bit Default &
Range Access

31:16
Asynchronous Schedule Status (async_schedule_status): The bit reports the

current real status of the Asynchronous Schedule. If this bit is a zero then the status of
the Asynchronous Schedule is disabled. If this bit is a one then the status of the
0h Asynchronous Schedule is enabled.
15
RO The Host Controller is not required to immediately disable or enable the Asynchronous
Schedule when software transitions the Asynchronous Schedule Enable bit in the
USBCMD register. When this bit and the Asynchronous Schedule Enable bit are the same
value, the Asynchronous Schedule is either enabled (1) or disabled (0).
Periodic Schedule Status (periodic_schedule_status): The bit reports the current

real status of the Periodic Schedule. If this bit is a zero then the status of the Periodic
Schedule is disabled. If this bit is a one then the status of the Periodic Schedule is
0h enabled.
14
RO The Host Controller is not required to immediately disable or enable the Periodic
Schedule when software transitions the Periodic Schedule Enable bit in the USBCMD
register. When this bit and the Periodic Schedule Enable bit are the same value, the
Periodic Schedule is either enabled (1) or disabled (0).
0h Reclamation (reclamation): This bit is used to detect an empty asynchronous

13
RO schedule.
HcHalted (hchalted): This bit is a zero whenever the Run/Stop bit is a one. The Host
1h Controller sets this bit to one after it has stopped executing as a result of the Run/Stop
12
RO bit being set to 0, either by software or by the Host Controller hardware (e.g. internal
error).

11:6
Interrupt on Async Advance (intr_on_async_advance): System software can

0h force the host controller to issue an interrupt the next time the host controller advances
5 the asynchronous schedule by writing a one to the Interrupt on Async Advance Doorbell
RW/1C bit in the USBCMD register. This status bit indicates the assertion of that interrupt
source.
Host System Error (host_system_error): The Host Controller sets this bit to 1 when
a serious error occurs during a host system access involving the Host Controller module.
0h In a PCI system, conditions that set this bit to 1 include PCI Parity error, PCI Master
4
RW/1C Abort, and PCI Target Abort.
When this error occurs, the Host Controller clears the Run/Stop bit in the Command
register to prevent further execution of the scheduled TDs.
Frame List Rollover (frame_list_rollover): The Host Controller sets this bit to a one
when the Frame List Index rolls over from its maximum value to zero. The exact value
0h at which the rollover occurs depends on the frame list size. For example, if the frame list
3
RW/1C size (as programmed in the Frame List Size field of the USBCMD register) is 1024, the
Frame Index Register rolls over every time FRINDEX[13] toggles. Similarly, if the size is
512, the Host Controller sets this bit to a one every time FRINDEX[12] toggles.

Bit Default &

Range Access
Port Change Detect (port_change_detect): The Host Controller sets this bit to a
one when any port for which the Port Owner bit is set to zero has a change bit transition
from a zero to a one or a Force Port Resume bit transition from a zero to a one as a
0h result of a J-K transition detected on a suspended port. This bit will also be set as a
2 result of the Connect Status Change being set to a one after system software has
RW/1C relinquished ownership of a connected port by writing a one to a port's Port Owner bit.
On a D3 to D0 transition of the EHCI HC device, this bit is loaded with the OR of all of
the PORTSC change bits (including: Force port resume, over-current change, enable/
disable change and connect status change).
USB Error Interrupt (usberrint): The Host Controller sets this bit to 1 when
completion of a USB transaction results in an error condition (e.g., error counter
0h underflow). If the TD on which the error interrupt occurred also had its IOC bit set, both
1
RW/1C this bit and USBINT bit are set.
Refer to EHCI Specification for a list of the USB errors that will result in this bit being set
to a one.
USB Interrupt (usbint): The Host Controller sets this bit to 1 on the completion of a
0h USB transaction, which results in the retirement of a Transfer Descriptor that had its IOC
0
RW/1C bit set. The Host Controller also sets this bit to 1 when a short packet is detected (actual
number of bytes received was less than the expected number of bytes).
16.6.2.6 USB Interrupt Enable (USBINTR)—Offset 18h

This register enables and disables reporting of the corresponding interrupt to the
software. When a bit is set and the corresponding interrupt is active, an interrupt is
generated to the host. Interrupt sources that are disabled in this register still appear in
the USBSTS to allow the software to poll for events.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
frame_list_rollover_enable
intr_on_async_advance_enable
host_system_err_enable
usberrint_enable
reserved_31_6
port_change_intr_enable
usbint_enable
Bit Default &

Range Access

31:6
Interrupt on Async Advance Enable (intr_on_async_advance_enable): When

0h this bit is a one, and the Interrupt on Async Advance bit in the USBSTS register is a one,
5
RW the host controller will issue an interrupt at the next interrupt threshold. The interrupt is
acknowledged by software clearing the Interrupt on Async Advance bit.

Bit Default &

Range Access
Host System Error Enable (host_system_err_enable): When this bit is a one, and
0h the Host System Error Status bit in the USBSTS register is a one, the host controller will
4
RW issue an interrupt. The interrupt is acknowledged by software clearing the Host System
Error bit.
Frame List Rollover Enable (frame_list_rollover_enable): When this bit is a one,

0h and the Frame List Rollover bit in the USBSTS register is a one, the host controller will
3
RW issue an interrupt. The interrupt is acknowledged by software clearing the Frame List
Rollover bit
Port Change Interrupt Enable (port_change_intr_enable): When this bit is a one,

0h and the Port Change Detect bit in the USBSTS register is a one, the host controller will
2
RW issue an interrupt. The interrupt is acknowledged by software clearing the Port Change
Detect bit.
USB Error Interrupt Enable (usberrint_enable): When this bit is a one, and the
0h USBERRINT bit in the USBSTS register is a one, the host controller will issue an interrupt
1
RW at the next interrupt threshold. The interrupt is acknowledged by software clearing the
USBERRINT bit.
0h USB Interrupt Enable (usbint_enable): When this bit is a one, and the USBINT bit in
0 the USBSTS register is a one, the host controller will issue an interrupt at the next
RW interrupt threshold. The interrupt is acknowledged by software clearing the USBINT bit.
16.6.2.7 USB Frame Index (FRINDEX)—Offset 1Ch

This register is used by the host controller to index into the periodic frame list.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_14
frame_index
Bit Default &

Range Access

31:14

Bit Default &

Range Access
Frame Index (frame_index): The value of this register increments at the end of each
time frame (e.g. micro-frame). Bits [N:3] are used for the Frame List current index. This
means that each location of the frame list is accessed 8 times (frames or micro-frames)
before moving to the next index. The following illustrates values of N based on the value
of the Frame List Size field in the USBCMD register. USBCMD[Frame List Size] Number
Elements N
0h USBCMD[Frame List Size] = 00 (1024), N = 12
13:0 USBCMD[Frame List Size] = 01 (512), N = 11
RW USBCMD[Frame List Size] = 10 (256), N = 10
USBCMD[Frame List Size] = 11 (Reserved)
This register must be written as a DWord. Byte writes produce undefined results. This
register cannot be written unless the Host Controller is in the Halted state as indicated
by the HCHalted bit in USBSTS register. A write to this register while the Run/Stop bit of
USBCMD register is set to a one produces undefined results. Writes to this register also
affect the SOF value.
16.6.2.8 4 Gigabyte Memory Segment Selector (CTRLDSSEGMENT)—Offset 20h

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
seg_4g_selector
Bit Default &

Range Access
4 Gigabyte Memory Segment Selector (seg_4g_selector): This 32-bit register

corresponds to the most significant address bits [63:32] for all EHCI data structures.
If the 64-bit Addressing Capability field in HCCPARAMS is a zero, then this register is not
0h used. Software cannot write to it and a read from this register will return zeros.
31:0 If the 64-bit Addressing Capability field in HCCPARAMS is a one, then this register is
RO used with the link pointers to construct 64-bit addresses to EHCI control data
structures. This register allows the host software to locate all control data structures
within the same 4 Gigabyte memory segment.
NOTE: 64-bit Addressing is not supported
16.6.2.9 Periodic Frame List Base Address (PERIODICLISTBASE)—Offset 24h

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
base_address
reserved_11_0
Bit Default &
Range Access
Base Address (base_address): This field contains bits [31:12]of the 32 bit address of
the Periodic Frame List in the system memory. System software loads this register prior
0h to starting the schedule execution by the Host Controller. The memory structure
31:12 referenced by this physical memory pointer is assumed to be 4-Kbyte aligned. The
RW contents of this register are combined with the Frame Index Register (FRINDEX) to
enable the Host Controller to step through the Periodic Frame List in sequence. Writes
must be DWord Writes.

11:0
RO Must be written as 0s. During runtime the value of these bits is undefined.
16.6.2.10 Asynchronous List Address (ASYNCLISTADDR)—Offset 28h

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_4_0
lpl
Bit Default &

Range Access
Link Pointer Low (lpl): This field contains bit [31:5] of the address of the next
asynchronous queue head to be executed. The memory structure referenced by this
0h physical memory pointer is assumed to be 4-Kbyte aligned. The contents of this register
31:5
RW are combined with the Frame Index Register (FRINDEX) to enable the Host Controller to
step through the Periodic Frame List in sequence. This field may only reference a Queue
Head (QH).

4:0
RO These bits are reserved and their value has no effect on operation.
16.6.2.11 Configure Flag (CONFIGFLAG)—Offset 50h

This register is in the auxiliary power well. It is only reset by hardware when the
auxiliary power is initially applied or in response to a host controller reset.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_1
cf
Bit Default &
Range Access

31:1
Configure Flag (cf): Host software sets this bit as the last action in its process of
configuring the Host Controller. This bit controls the default port-routing control logic.
0h Bit values and side-effects are listed below.
0
RW 0: Port routing control logic default-routes each port to an implementation dependent
classic host controller.
1: Port routing control logic default-routes all ports to this host controller
16.6.2.12 Port Status/Control[0-1] (PORTSC[0-1])—Offset 54h, Count 2, Stride

4h
This register is in the auxiliary power well. It is only reset by hardware when the
auxiliary power is initially applied or in response to a host controller reset.
Access Method
(Size: 32 bits)

Default: 00002000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0
device_address
port_reset
over_current_change
connect_status_change
suspend_status
port_test_ctrl
port_indicator_ctrl
port_owner
force_port_resume
port_en_dis_change
current_connect_status
wkoc_e
wkdscnnt_e
wkcnnt_e
pp
line_status
suspend
over_current_active
port_enable_disable
suspend_using_l1
Bit Default &

Range Access
Device Address (device_address): The 7-bit USB device address for the device
0h attached to and immediately downstream of the associated root port. A value of zero
31:25
RO indicates no device is present or support for this feature is not present.
NOTE: This field is not supported.

Bit Default &

Range Access
0h Suspend Status (suspend_status): These two bits are used by software to

24:23 determine whether the most recent L1 suspend request was successful.
RO NOTE: This field is not supported.
0h Wake on Over-current Enable (wkoc_e): Writing this bit to a one enables the port
22 to be sensitive to over-current conditions as wake-up events. This field is zero if Port
RW Power is zero.
0h Wake on Disconnect Enable (wkdscnnt_e): Writing this bit to a one enables the
21 port to be sensitive to device disconnects as wake-up events. This field is zero if Port
RW Power is zero.
0h Wake on Connect Enable (wkcnnt_e): Writing this bit to a one enables the port to
20 be sensitive to device connects as wake-up events. This field is zero if Port Power is
RW zero.
Port Test Control (port_test_ctrl): When this field is zero, the port is NOT operating
in a test mode. A non-zero value indicates that it is operating in test mode and the
specific test mode is indicated by the specific value. The encoding of the test mode bits
are (0110 - 1111 are reserved):
0h 0000: Test mode not enabled
19:16
RW 0001: Test J_STATE
0010: Test K_STATE
0011: Test SE0_NAK
0100: Test Packet
0101: Test FORCE_ENABLE
Port Indicator Control (port_indicator_ctrl): Writing to these bits has no effect if

the Port Indicator (P_INDICATOR) bit in the HCSPARAMS register is a zero.
If P_INDICATOR bit is a one, then the bit encodings are:
00h 00: Port indicators are off
15:14 01: Amber
RW 10: Green
11: Undefined
This field is zero if Port Power is zero.
NOTE: Port Indicator is not supported.
Port Owner (port_owner): This bit unconditionally goes to a 0 when the Configured
bit in the CONFIGFLAG register makes a 0b to 1b transition. This bit unconditionally
1h goes to 1 whenever the Configured bit is zero. System software uses this field to release
13 ownership of the port to a selected host controller (in the event that the attached device
RW is not a high-speed device). Software writes a one to this bit when the attached device
is not a high-speed device. A one in this bit means that a companion host controller
owns and controls the port.
Port Power (pp): The function of this bit depends on the value of the Port Power
Control (PPC) field in the HCSPARAMS register. The behavior is as follows:
If PPC=0, PP=1 and RO: Host controller does not have port power control switches.
Each port is hard-wired to power.
If PPC=0, PP=1 or 0 and R/W: Host controller has port power control switches. This bit
0h represents the current setting of the switch (0 = off, 1 = on). When power is not
12
RW available on port (i.e. PP equals a 0),the port is non-functional and will not report
attaches, detaches, etc.
When an over-current condition is detected on a powered port and PPC is a one, the PP
bit in each affected port may be transitioned by the host controller from a 1 to 0
(removing power from the port).
NOTE: Per Port Power control is supported (PPC=1)
Line Status (line_status): These bits reflect the current logical levels of the D+ (bit
11) and D- (bit 10) signal lines. These bits are used for detection of low-speed USB
devices prior to the port reset and enable sequence. This field is valid only when the port
enable bit is zero and the current connect status bit is set to a one. The encoding of the
0h bits are:
11:10 Bits[11:10] USB State Interpretation
RO 00: USB State is SE0, Not Low-speed device, perform EHCI reset
10: USB State is J-state, Not Low-speed device, perform EHCI reset
01: USB State is K-state, Low-speed device, release ownership of port
11: USB State is Undefined, Not Low-speed device, perform EHCI reset
The value of this field is undefined if Port Power is zero.
0h Suspend using l1 (suspend_using_l1): NOTE: suspend using l1 is not supported as

9
RO part of LPM.

Bit Default &

Range Access
Port Reset (port_reset): 1: Port is in Reset

0: Port is not in Reset
When software writes a one to this bit (from a zero), the bus reset sequence as defined
in the USB Specification Revision 2.0 is started. Software writes a zero to this bit to
terminate the bus reset sequence. Software must keep this bit at a one long enough to
ensure the reset sequence, as specified in the USB Specification Revision 2.0,
completes.
When software writes this bit to a one, it must also write a zero to the Port Enable bit.
0h Note that when software writes a zero to this bit there may be a delay before the bit
8 status changes to a zero. The bit status will not read as a zero until after the reset has
RW completed. If the port is in high-speed mode after reset is complete, the host controller
will automatically enable this port (e.g. set the Port Enable bit to a one). A host
controller must terminate the reset and stabilize the state of the port within 2ms of
software transitioning this bit from a one to a zero. For example: if the port detects that
the attached device is high-speed during reset, then the host controller must have the
port in the enabled state within 2ms of software writing this bit to a zero.
The HCHalted bit in the USBSTS register should be a zero before software attempts to
use this bit. The host controller may hold Port Reset asserted to a one when the
HCHalted bit is a one. This field is zero if Port Power is zero.
Suspend (suspend): 1: Port in suspend state

0: Port not in suspend state
Port Enabled Bit and Suspend bit of this register define the port states as follows:
Bit [Port Enabled, Suspend]: Port State
0X: Disable
10: Enable
11: Suspend
When in suspend state, downstream propagation of data is blocked on this port, except
0h for port reset. The blocking occurs at the end of the current transaction, if a transaction
7 was in progress when this bit was written to 1. In the suspend state, the port is
RW sensitive to resume detection. Note that the bit status does not change until the port is
suspended and that there may be a delay in suspending a port if there is a transaction
currently in progress on the USB.
A write of zero to this bit is ignored by the host controller. The host controller will
unconditionally set this bit to a zero when:
Software sets the Force Port Resume bit to a zero (from a one).
Software sets the Port Reset bit to a one (from a zero).
If host software sets this bit to a one when the port is not enabled (i.e. Port enabled bit
is a zero) the results are undefined. This field is zero if Port Power is zero.
Force Port Resume (force_port_resume): 1: Resume detected/driven on port

0: No resume (K-state) detected/driven on port
This functionality defined for manipulating this bit depends on the value of the Suspend
bit.
For example, if the port is not suspended (Suspend and Enabled bits are a one) and
software transitions this bit to a one, then the effects on the bus are undefined.
Software sets this bit to a 1 to drive resume signaling. The Host Controller sets this bit
to a 1 if a J-to-K transition is detected while the port is in the Suspend state. When this
0h bit transitions to a one because a J-to-K transition is detected, the Port Change Detect
6 bit in the USBSTS register is also set to a one. If software sets this bit to a one, the host
RW controller must not set the Port Change Detect bit. Note that when the EHCI controller
owns the port, the resume sequence follows the defined sequence documented in the
USB Specification Revision 2.0. The resume signaling (Full-speed 'K') is driven on the
port as long as this bit remains a one. Software must appropriately time the Resume
and set this bit to a zero when the appropriate amount of time has elapsed. Writing a
zero (from one) causes the port to return to high-speed mode (forcing the bus below
the port into a high-speed idle). This bit will remain a one until the port has switched to
the high-speed idle. The host controller must complete this transition within 2
milliseconds of software setting this bit to a zero. This field is zero if Port Power is zero.
0h Over Current Change (over_current_change): This bit gets set to a one when there
5 is a change to Over-current Active. Software clears this bit by writing a one to this bit
RW/1C position.
Over Current Active (over_current_active): 1: This port currently has an over-

0h current condition
4 0: This port does not have an over-current condition
RO This bit will automatically transition from a one to a zero when the over current
condition is removed.

Bit Default &

Range Access
Port Enable/Disable Change (port_en_dis_change): 1: Port enabled/disabled

status has changed
0h 0: No change
3
RW/1C For the root hub, this bit gets set to a one only when a port is disabled due to the
appropriate conditions existing at the EOF2 point Software clears this bit by writing a 1
to it. This field is zero if Port Power is zero.
Port Enable/Disable (port_enable_disable): 1: Enable

0: Disable
Ports can only be enabled by the host controller as a part of the reset and enable.
Software cannot enable a port by writing a one to this field. The host controller will only
0h set this bit to a one when the reset sequence determines that the attached device is a
2 high-speed device. Ports can be disabled by either a fault condition (disconnect event or
RW other fault condition) or by host software. Note that the bit status does not change until
the port state actually changes. There may be a delay in disabling or enabling a port due
to other host controller and bus events. When the port is disabled (0) downstream
propagation of data is blocked on this port, except for reset. This field is zero if Port
Power is zero.
Connect Status Change (connect_status_change): 1: Change in Current Connect

Status
0: No change
0h Indicates a change has occurred in the ports Current Connect Status. The host controller
1 sets this bit for all changes to the port device connect status, even if system software
RW/1C has not cleared an existing connect status change. For example, the insertion status
changes twice before system software has cleared the changed condition, hub hardware
will be setting an already set bit. Software sets this bit to 0 by writing a 1 to it.
Current Connect Status (current_connect_status): 1: Device is present on port

0h 0: No device is present
0 This value reflects the current state of the port, and may not correspond directly to the
RO event that caused the Connect Status Change bit (Bit 1) to be set.
16.6.2.13 Programmable Microframe Base Value (INSNREG00)—Offset 90h

This register allows changing the microframe length value (default is microframe SOF =
125ms) to reduce the simulation time.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_14
uframe_count_en
uframe_count
Bit Default &

Range Access

31:14

Bit Default &

Range Access
0h
13:1 uframe_count: 1-microframe counter with byte interface (8-bits).
RW
0b
0 uframe_count_en: 1-microframe counter is enabled when this bit is set to 1.
RW
16.6.2.14 Programmable Packet Buffer OUT/IN Thresholds (INSNREG01)—

Offset 94h
This register allows setting the packet buffer OUT/IN thresholds. The value specified by
the thresholds is in number of DWORDs (32-bit entries). The minimum threshold
amount that can be programmed is the highest possible INCRX burst value (INCR16)
i.e the minimum OUT and IN threshold value should be 64 bytes (16 DWords). With the
implemented packet buffer depth of 512 bytes, OUT and IN threshold values can be
equal to the packet buffer depth only when isochronous/interrupt transactions are not
initiated by the host controller.
Access Method
(Size: 32 bits)

Default: 00200020h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0
reserved_31_24
reserved_15_8
IN_Threshold
OUT_Threshold
Bit Default &

Range Access

31:24
Out Threshold (OUT_Threshold): The OUT threshold is used to start the USB transfer
20h as soon as the OUT threshold amount of data is fetched from system memory. It is also
23:16
RW used to disconnect the data fetch, if the threshold amount of space is not available in
the Packet Buffer.

15:8
IN Threshold (IN_Threshold): The IN threshold is used to start the memory transfer

20h as soon as the IN threshold amount of data is available in the Packet Buffer. It is also
7:0
RW used to disconnect the data write, if the threshold amount of data is not available in the
Packet Buffer.
16.6.2.15 Programmable Packet Buffer Depth (INSNREG02)—Offset 98h

Access Method


(Size: 32 bits)

Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
reserved_31_12
pkt_buffer_depth
Bit Default &
Range Access

31:12
80h Programmable Packet Buffer Depth (pkt_buffer_depth): The value specified here
11:0
RW is the number of DWORDs (32-bit entries).
16.6.2.16 Programmable Controller Settings (INSNREG03)—Offset 9Ch

Access Method

Default: 00002001h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1
reserved_31_14
Time_available_offset
Break_Memory
TestSE0
Tx_Tx_turnaround_dly
Periodic_Frame_List
Bit Default &

Range Access

31:14
TestSE0 NAK (TestSE0): When set to 1 (default), the core ignores the line state
checking when transmitting SOF during the SE0_NAK test mode. When Set to 0, the
1h port state machine disables the port if it does not find the line state to be in SE0 when
13
RW transmitting SOF during the SE0_NAK testing. While doing impedance measurement
during the SE0_NAK testing, the line state could go to non SE0 forcing the core to
disable the port. This bit is used to control the port behavior during this.

Bit Default &

Range Access
Transmit to Transmit turnaround delay (Tx_Tx_turnaround_dly): This field

specifies the extra delays in phy_clks to be added to the Transmit to Transmit
turnaround delay value maintained in the core. The default value of this register field is
0h 0. This default value of 0 is sufficient for most PHYs. But for some PHYs which puts wait
12:10 states during the token packet, it may be required to program a value greater than 0 to
RW meet the transmit to transmit minimum turnaround time. The recommendation to use
the default value of 0 and change it only if there is an issue with minimum transmit-to-
transmit turnaround time. This value should be programmed during core initialization
and should not be changed afterwards.
Periodic Frame List Fetch (Periodic_Frame_List): Setting this bit will force the host
0h controller to fetch the periodic frame list in every microframe of a frame. If not set, then
9 the periodic frame list will be fetched only in microframe 0 of every frame. The default is
RW 0 (not set). This bit can be changed only during core initialization and should not be
changed afterwards.
Time Available Offset (Time_available_offset): This value indicates the additional

number of bytes to be accommodated for the time-available calculation. The USB traffic
0h on the bus can be started only when sufficient time is available to complete the packet
8:1 within the EOF1 point. This time-available calculation is done in the hardware, and can
RW be further offset by programming a value in this location. Note that time-available
calculation is added for future flexibility and the application is not required to program
this field by default.
Break Memory Transfer (Break_Memory): 1: Enables this function

1h 0: Disables this function
0
RO Used in conjunction with INSNREG01 to enable breaking memory transactions into
chunks once the OUT/IN threshold value is reached.
16.6.2.17 Programmable Controller Settings (INSNREG04)—Offset A0h

Bits [2:0] are used only for debug purposes.
Access Method
Type: Memory Mapped I/O Register Offset: [BAR0] + A0h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
automatic_feature
reserved_31_6
reserved_3
port_enum
NAK
sys_res
Bit Default &

Range Access

31:6

Bit Default &

Range Access
Automatic Suspend Feature (automatic_feature): 0: enables the automatic

feature. The Suspend signal is deasserted (logic level 1) when run/stop is reset by
software, but the hchalted bit is not yet set.
1: disables the automatic feature, which takes all ports out of suspend when software
0h clears the run/stop bit. This is for backward compatibility.
5 For systems where the host is halted without waking up all ports out of suspend, the
RW port can become stuck because the PHYCLK is not running when the halt is
programmed. To avoid this, the DWC H20AHB host core automatically pulls ports out of
suspend when the host is halted by software. This bit is used to disable this automatic
function.
This is a functional bit.
0h NAK reload fix enabled (NAK): When 1 NAK reload fix disabled.
4 This is a functional bit for backward compatibility with Synopsys USB 2.0 Host-AHB core
RW Release 2.40c.

3
0h Port Enumeration Time (port_enum): Scales down port enumeration time. This is a
2
RW debug bit.
System Reset (sys_res): When sys_res[1]=1, HCCPARAMS bits 17,15:4 become

0h writable.
1:0
RW When sys_res[0]=1, HCCPARAMS register become writable Upon system reset, this field
is 0. This is a debug bit.
16.6.2.18 UTMI Configuration (INSNREG05)—Offset A4h

This register gives access to implementation specific PHY vendor control and status
registers. NOTE: Vendor control and status registers are not implemented. Read from
Vendor Status will always return 0. Write to Vendor Control will have no effect.
Access Method
(Size: 32 bits)

Default: 00001000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_18
Vbusy
Vport
VControlLoadM
Vcontrol
Vstatus
Bit Default &

Range Access

31:18
0h Vbusy: Hardware indicator that a write to this register has occurred and the hardware is
17 currently processing the operation defined by the data written. When processing is
RO/V finished, this bit is cleared.

Bit Default &

Range Access
Vport: Port number from/to the PHY Vendor Status and Control value is read/written in
Vstatus field.
0h Vport meaningful values depend on the number of ports implemented. Allowed values
16:13
RW are 4'h1 and 4'h2, any other value will return Vstatus=8'h0.
Once software writes to VPort, from that write onwards, any writes to this register is
ignored.
1h Vendor Control Load (VControlLoadM): 0: Load.

12
RW 1: NOP.
0h Vendor control register (Vcontrol): Not implemented - write to Vendor Control will
11:8
RW have no effect.
0h Vendor Status register (Vstatus): Not implemented - read from Vendor Status will
7:0
RO always return 0.
16.6.3 USB OHCI

Default
Value
0h 3h “OHCI Revision (HCREVISION)—Offset 0h” on page 570 00000010h
4h 7h “Host Controller Control (HCCONTROL)—Offset 4h” on page 570 00000000h
8h Bh “Host Controller Command Status (HCCMDSTATUS)—Offset 8h” on page 571 00000000h
Ch Fh “Host Controller Interrupt Status (HCINTRSTATUS)—Offset Ch” on page 573 00000000h
10h 13h “Host Controller Interrupt Enable (HCINTRENABLE)—Offset 10h” on page 574 00000000h
14h 17h “Host Controller Interrupt Disable (HCINTRDISABLE)—Offset 14h” on page 575 00000000h
18h 1Bh “Host Controller Communication Area (HCHCCA)—Offset 18h” on page 576 00000000h
“Host Controller Current Isochronous or Interrupt Endpoint (HCPRDCURED)—

1Ch 1Fh 00000000h
“Host Controller Current First Control Endpoint (HCCTRLHEADED)—Offset 20h” on

20h 23h 00000000h
page 577
“Host Controller Current Control Endpoint (HCCTRLCURED)—Offset 24h” on

24h 27h 00000000h
page 578
28h 2Bh “Host Controller First Bulk Endpoint (HCBULKHEADED)—Offset 28h” on page 578 00000000h
2Ch 2Fh “Host Controller Current Bulk Endpoint (HCBULKCURED)—Offset 2Ch” on page 579 00000000h
“Host Controller Last Completed Descriptor (HCDONEHEAD)—Offset 30h” on

30h 33h 00000000h
page 580
34h 37h “Host Controller Frame Interval (HCFMINTERVAL)—Offset 34h” on page 580 00002EDFh
38h 3Bh “Host Controller Remaining Frame (HCFMREMAINING)—Offset 38h” on page 581 00000000h
3Ch 3Fh “Host Controller Frame Number (HCFMNUMBER)—Offset 3Ch” on page 582 00000000h
40h 43h “Host Controller Periodic List Start (HCPERIODICSTART)—Offset 40h” on page 583 00000000h
44h 47h “Host Controller LS Threshold (HCLSTHRESHOLD)—Offset 44h” on page 583 00000628h
48h 4Bh “Host Controller Root Hub Descriptor A (HCRHDESPA)—Offset 48h” on page 584 02000902h
4Ch 4Fh “Host Controller Root Hub Descriptor B (HCRHDESPB)—Offset 4Ch” on page 585 00000000h
50h 53h “Host Controller Root Hub Status (HCRHSTATUS)—Offset 50h” on page 586 00000000h
54h 57h “Host Controller Root Hub Port Status (HCRHPORTSTS)—Offset 54h” on page 587 00000000h

16.6.3.1 OHCI Revision (HCREVISION)—Offset 0h

Access Method
Type: Memory Mapped I/O Register HCREVISION: [BAR0] + 0h
(Size: 32 bits)

Default: 00000010h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
reserved_31_8
revision
Bit Default &
Description
Range Access
000000h reserved_31_8 (reserved_31_8): Reserved bits.

31: 8
10h Revision (revision): This read-only field contains the BCD representation of the
7: 0
RO version of the HCI specification that is implemented by this HC.
16.6.3.2 Host Controller Control (HCCONTROL)—Offset 4h

The HCCONTROL register defines the operating modes for the Host Controller (HC).
Most of the fields in this register are modified only by the Host Controller Driver (HCD),
except hc_function_state and rmtwkup_connected.
Access Method
(Size: 32 bits) HCCONTROL: [BAR0] + 4h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rmtwkup_connected
intr_routing
hc_function_state
reserved_31_11
ctrlbulk_serviceratio
rmtwkup_enable
ctrllist_enable
isoc_enable
bulklist_enable
periodiclist_enable
Bit Default &

Description
Range Access

31: 11

Bit Default &

Description
Range Access
Remote Wakeup Enable (rmtwkup_enable): This bit is used by HCD to enable or

0h disable the remote wakeup feature upon the detection of upstream resume signaling.
10 When this bit is set and the resume_detected bit in HCINTRSTATUS is set, a remote
RW wakeup is signaled to the host system. Setting this bit has no impact on the generation
of hardware interrupts.
Remote Wakeup Connected (rmtwkup_connected): This bit indicates whether HC

0h supports remote wakeup signaling. If remote wakeup is supported and used by the
9
RW system it is the responsibility of system firmware to set this bit during POST. HC clears
the bit upon a hardware reset but does not alter it upon a software reset.
Interrupt Routing (intr_routing): This bit determines the routing of interrupts

generated by events registered in HCINTRSTATUS. If clear, all interrupts are routed to
0h the normal host bus interrupt mechanism. If set, interrupts are routed to the System
8
RW Management Interrupt. HCD clears this bit upon a hardware reset, but it does not alter
this bit upon a software reset. HCD uses this bit as a tag to indicate the ownership of
HC.
Host Controller Functional State for USB (hc_function_state): 00: USBRESET

01: USBRESUME
10: USBOPERATIONAL
11: USBSUSPEND
A transition to USBOPERATIONAL from another state causes SOF generation to begin
0h 1ms later. HCD may determine whether HC has begun sending SOFs by reading the
7: 6
RW start_of_frame field of HCINTRSTATUS. This field may be changed by HC only when in
the USBSUSPEND state. HC may move from the USBSUSPEND state to the USBRESUME
state after detecting the resume signaling from a downstream port. HC enters
USBSUSPEND after software reset, whereas it enters USBRESET after a hardware reset.
The latter also resets the Root Hub and asserts subsequent reset signaling to
downstream ports.
Bulk List Enable (bulklist_enable): This bit is set to enable the processing of the
Bulk list in the next Frame. If cleared by HCD, processing of the Bulk list does not occur
0h after the next SOF. HC checks this bit whenever it determines to process the list. When
5
RW disabled, HCD may modify the list. If HCBULKCURED is pointing to an ED (Endpoint
Descriptor) to be removed, HCD must advance the pointer by updating HCBULKCURED
before re-enabling processing of the list.
Control List Enable (ctrllist_enable): This bit is set to enable the processing of the
Control list in the next Frame. If cleared by HCD, processing of the Control list does not
0h occur after the next SOF. HC must check this bit whenever it determines to process the
4
RW list. When disabled, HCD may modify the list. If HCCTRLCURED is pointing to an ED to
be removed, HCD must advance the pointer by updating HCCTRLCURED before re-
enabling processing of the list.
Isochronous Enable (isoc_enable): This bit is used by HCD to enable/disable

processing of isochronous ED's. While processing the periodic list in a Frame, HC checks
0h the status of this bit when it finds an Isochronous ED (F=1). If set (enabled), HC
3 continues processing the ED's. If cleared (disabled), HC halts processing of the periodic
RW list (which now contains only isochronous ED's) and begins processing the Bulk/Control
lists. Setting this bit is guaranteed to take effect in the next Frame (not the current
Frame).
0h Periodic List Enable (periodiclist_enable): This bit is set to enable the processing of
2 the periodic list in the next Frame. If cleared by HCD, processing of the periodic list does
RW not occur after the next SOF. HC must check this bit before it starts processing the list.
Control Bulk Service Ratio (ctrlbulk_serviceratio): This specifies the service ratio
between Control and Bulk ED's. Before processing any of the nonperiodic lists, HC must
0h compare the ratio specified with its internal count on how many nonempty Control ED's
1: 0
RW have been processed, in determining whether to continue serving another Control ED or
switching to Bulk ED's. The internal count will be retained when crossing the frame
boundary. In case of reset, HCD is responsible for restoring this value.
16.6.3.3 Host Controller Command Status (HCCMDSTATUS)—Offset 8h

The HCCMDSTATUS register is used by the Host Controller to receive commands issued
by the Host Controller Driver, as well as reflecting the current status of the Host
Controller. To the Host Controller Driver, it appears to be a 'write to set' register. The
Host Controller must ensure that bits written as 1 become set in the register while bits

written as 0 remain unchanged in the register. The Host Controller Driver may issue
multiple distinct commands to the Host Controller without concern for corrupting
previously issued commands. The Host Controller Driver has normal read access to all
bits. The sch_overrun_cnt field indicates the number of frames with which the Host
Controller has detected the scheduling overrun error. This occurs when the Periodic list
does not complete before EOF. When a scheduling overrun error is detected, the Host
Controller increments the counter and sets the sch_overrun field in the HCINTRSTATUS
register.
Access Method
Type: Memory Mapped I/O Register HCCMDSTATUS: [BAR0] + 8h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_18
sch_overrun_cnt
reserved_15_4
hcreset
ownerchange_req
bulklist_filled
ctrllist_filled
Bit Default &
Description
Range Access

31: 18
Scheduling Overrun Count (sch_overrun_cnt): These bits are incremented on each

0h scheduling overrun error. It is initialized to 00b and wraps around at 11b. This will be
17: 16 incremented when a scheduling overrun is detected even if sch_overrun in
RO HCINTRSTATUS has already been set. This is used by HCD to monitor any persistent
scheduling problems

15: 4
Ownership Change Request (ownerchange_req): This bit is set by an OS HCD to

0h request a change of control of the HC. When set HC will set the ownerchange_req field
3
RW/1C in HCINTRSTATUS. After the changeover, this bit is cleared and remains so until the next
request from OS HCD.
BulkListFilled (bulklist_filled): This bit is used to indicate whether there are any TDs
on the Bulk list. It is set by HCD whenever it adds a TD to an ED in the Bulk list. When
HC begins to process the head of the Bulk list, it checks bulklist_filled. As long as
0h bulklist_filled is 0, HC will not start processing the Bulk list. If bulklist_filled is 1, HC will
2
RW start processing the Bulk list and will set bulklist_filled to 0. If HC finds a TD on the list,
then HC will set bulklist_filled to 1 causing the Bulk list processing to continue. If no TD
is found on the Bulk list, and if HCD does not set bulklist_filled, then bulklist_filled will
still be 0 when HC completes processing the Bulk list and Bulk list processing will stop.
Control List Filled (ctrllist_filled): This bit is used to indicate whether there are any
TDs on the Control list. It is set by HCD whenever it adds a TD to an ED in the Control
list. When HC begins to process the head of the Control list, it checks ctrllist_filled. As
0h long as ctrllist_filled is 0, HC will not start processing the Control list. If CF is 1, HC will
1 start processing the Control list and will set ctrllist_filled to 0. If HC finds a TD on the
RW list, then HC will set ctrllist_filled to 1 causing the Control list processing to continue. If
no TD is found on the Control list, and if the HCD does not set ctrllist_filled, then
ctrllist_filled will still be 0 when HC completes processing the Control list and Control list
processing will stop.

Bit Default &

Description
Range Access
Host Controller Reset (hcreset): This bit is set by HCD to initiate a software reset of
HC. Regardless of the functional state of HC, it moves to the USBSUSPEND state in
0h which most of the operational registers are reset except those stated otherwise; e.g.,
0 the intr_routing field of HCCONTROL, and no Host bus accesses are allowed. This bit is
RW cleared by HC upon the completion of the reset operation. The reset operation must be
completed within 10ms. This bit, when set, should not cause a reset to the Root Hub
and no subsequent reset signaling should be asserted to its downstream ports.
16.6.3.4 Host Controller Interrupt Status (HCINTRSTATUS)—Offset Ch

This register provides status on various events that cause hardware interrupts. When
an event occurs, Host Controller sets the corresponding bit in this register. When a bit
becomes set, a hardware interrupt is generated if the interrupt is enabled in the
HCINTRENABLE register and the mstr_intr_enable bit is set. The Host Controller Driver
may clear specific bits in this register by writing 1 to bit positions to be cleared. The
Host Controller Driver may not set any of these bits. The Host Controller will never
clear the bit.
Access Method
Type: Memory Mapped I/O Register HCINTRSTATUS: [BAR0] + Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
owner_change
start_of_frame
rhub_stschange
reserved_31
reserved_29_07
fmnum_overflow
unrecov_err
wrback_donehead
sch_overrun
resume_detected
Bit Default &
Description
Range Access
0h reserved_31 (reserved_31): Reserved bits.

31
Ownership Change (owner_change): This bit is set by HC when HCD sets the
0h ownerchange_req field in HCCMDSTATUS. This event, when unmasked, will always
30 generate a System Management Interrupt (SMI) immediately. This bit is tied to 0 when
RW/1C the SMI pin is not implemented.
NOTE: OHCI SMIs are ignored by the system.

29: 7
0h Root Hub Status Change (rhub_stschange): This bit is set when the content of
6 HCRHSTATUS or the content of any of HCRHPORTSTS[NumberofDownstreamPort] has
RW/1C changed.
0h Frame Number Overflow (fmnum_overflow): This bit is set when the MSb of
5 HCFMNUMBER (bit 15) changes value, from 0 to 1 or from 1 to 0, and after
RW/1C HccaFrameNumber has been updated.
0h Unrecoverable Error (unrecov_err): This bit is set when HC detects a system error
4 not related to USB. HC should not proceed with any processing or signaling before the
RW/1C system error has been corrected. HCD clears this bit after HC has been reset.

Bit Default &

Description
Range Access
Resume Detected (resume_detected): This bit is set when HC detects that a device
0h on the USB is asserting resume signaling. It is the transition from no resume signaling
3
RW/1C to resume signaling causing this bit to be set. This bit is not set when HCD sets the
USBRESUME state.
0h Start of Frame (start_of_frame): This bit is set by HC at each start of a frame and
2 after the update of HccaFrameNumber. HC also generates a SOF token at the same
RW/1C time.
Writeback Done Head (wrback_donehead): This bit is set immediately after HC has
0h written HCDONEHEAD to HccaDoneHead. Further updates of the HccaDoneHead will not
1
RW/1C occur until this bit has been cleared. HCD should only clear this bit after it has saved the
content of HccaDoneHead.
0h Scheduling Overrun (sch_overrun): This bit is set when the USB schedule for the
0 current Frame overruns and after the update of HccaFrameNumber. A scheduling
RW/1C overrun will also cause the sch_overrun_cnt of HCCMDSTATUS to be incremented.
16.6.3.5 Host Controller Interrupt Enable (HCINTRENABLE)—Offset 10h

Each enable bit in the HCINTRENABLE register corresponds to an associated interrupt
bit in the HCINTRSTATUS register. The HCINTRENABLE register is used to control which
events generate a hardware interrupt. When a bit is set in the HCINTRSTATUS register
AND the corresponding bit in the HCINTRENABLE register is set AND the
mstr_intr_enable bit is set, then a hardware interrupt is requested on the host bus.
Writing a '1' to a bit in this register sets the corresponding bit, whereas writing a '0' to
a bit in this register leaves the corresponding bit unchanged. On read, the current value
of this register is returned.
Access Method
Type: Memory Mapped I/O Register HCINTRENABLE: [BAR0] + 10h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
mstr_intr_enable
reserved_29_07
unrecov_err_enb
owner_change_enb
sch_overrun_enb
fmnum_overflow_enb
resume_detected_enb
start_of_frame_enb
rhub_stschange_enb
wrback_donehead_enb
Bit Default &

Description
Range Access
0h Master Interrupt Enable (mstr_intr_enable): A 0 written to this field is ignored by

31 HC. A 1 written to this field enables interrupt generation due to events specified in the
RW/1S other bits of this register. This is used by HCD as a Master Interrupt Enable.
0h Owner Change Enable (owner_change_enb): 0: Ignore

30
RW/1S 1: Enable interrupt generation due to Ownership Change.

29: 7

Bit Default &

Description
Range Access
0h Root Hub Status Change Enable (rhub_stschange_enb): 0: Ignore

6
RW/1S 1: Enable interrupt generation due to Root Hub Status Change
0h Frame Number Overflow Enable (fmnum_overflow_enb): 0: Ignore

5
RW/1S 1: Enable interrupt generation due to Frame Number Overflow
0h Unrecoverable Error Enable (unrecov_err_enb): 0: Ignore

4
RW/1S 1: Enable interrupt generation due to Unrecoverable Error
0h Resume Detected Enable (resume_detected_enb): 0: Ignore

3
RW/1S 1: Enable interrupt generation due to Resume Detect
0h Start of Frame Enable (start_of_frame_enb): 0: Ignore

2
RW/1S 1: Enable interrupt generation due to Start of Frame
0h HCDONEHEAD Writeback Enable (wrback_donehead_enb): 0: Ignore

1
RW/1S 1: Enable interrupt generation due to HCDONEHEAD Writeback
0h Scheduling Overrun Enable (sch_overrun_enb): 0: Ignore

0
RW/1S 1: Enable interrupt generation due to Scheduling Overrun
16.6.3.6 Host Controller Interrupt Disable (HCINTRDISABLE)—Offset 14h

Each disable bit in the HCINTRDISABLE register corresponds to an associated interrupt
bit in the HCINTRSTATUS register. The HCINTRDISABLE register is coupled with the
HCINTRENABLE register. Thus, writing a '1' to a bit in this register clears the
corresponding bit in the HCINTRENABLE register, whereas writing a '0' to a bit in this
register leaves the corresponding bit in the HCINTRENABLE register unchanged. On
read, the current value of the HCINTRENABLE register is returned.
Access Method
Type: Memory Mapped I/O Register HCINTRDISABLE: [BAR0] + 14h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
unrecov_err_dbl
mstr_intr_enable
owner_change_dbl
reserved_29_07
fmnum_overflow_dbl
sch_overrun_dbl
wrback_donehead_dbl
resume_detected_dbl
rhub_stschange_dbl
start_of_frame_dbl
Bit Default &

Description
Range Access
0h Master Interrupt Enable (mstr_intr_enable): A '0' written to this field is ignored by

31 HC. A '1' written to this field disables interrupt generation due to events specified in the
RW/1C other bits of this register. This field is set after a hardware or software reset.
0h Owner Change Disable (owner_change_dbl): 0: Ignore

30
RW/1C 1: Disable interrupt generation due to Ownership Change

29: 7

Bit Default &

Description
Range Access
0h Root Hub Status Change Disable (rhub_stschange_dbl): 0: Ignore

6
RW/1C 1: Disable interrupt generation due to Root Hub Status Change
0h Frame Number Overflow Disable (fmnum_overflow_dbl): 0: Ignore

5
RW/1C 1: Disable interrupt generation due to Frame Number Overflow
0h Unrecoverable Error Disable (unrecov_err_dbl): 0: Ignore

4
RW/1C 1: Disable interrupt generation due to Unrecoverable Error
0h Resume Detected Disable (resume_detected_dbl): 0: Ignore

3
RW/1C 1: Disable interrupt generation due to Resume Detect
0h Start of Frame Disable (start_of_frame_dbl): 0: Ignore

2
RW/1C 1: Disable interrupt generation due to Start of Frame
0h HCDONEHEAD Writeback Disable (wrback_donehead_dbl): 0: Ignore

1
RW/1C 1: Disable interrupt generation due to HCDONEHEAD Writeback
0h Scheduling Overrun Disable (sch_overrun_dbl): 0: Ignore

0
RW/1C 1: Disable interrupt generation due to Scheduling Overrun
16.6.3.7 Host Controller Communication Area (HCHCCA)—Offset 18h

The HCHCCA register contains the physical address of the Host Controller
Communication Area. The Host Controller Driver determines the alignment restrictions
by writing all 1s to HCHCCA and reading the content of HCHCCA. The alignment is
evaluated by examining the number of zeroes in the lower order bits. The minimum
alignment is 256 bytes; therefore, bits 0 through 7 must always return '0' when read.
Detailed description can be found in Chapter 4. This area is used to hold the control
structures and the Interrupt table that are accessed by both the Host Controller and the
Host Controller Driver.
Access Method
Type: Memory Mapped I/O Register HCHCCA: [BAR0] + 18h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_7_0
hccabase
Bit Default &

Description
Range Access
0h Host Controller Communication Area Base (hccabase): This is the base address of
31: 8
RW the Host Controller Communication Area

7: 0

16.6.3.8 Host Controller Current Isochronous or Interrupt Endpoint

(HCPRDCURED)—Offset 1Ch
The HcPeriodCurrentED register contains the physical address of the current
Isochronous or Interrupt Endpoint Descriptor.
Access Method
Type: Memory Mapped I/O Register HCPRDCURED: [BAR0] + 1Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
period_cur_ed
reserved_3_0
Bit Default &
Description
Range Access
Period Current ED (period_cur_ed): This is used by HC to point to the head of one of

0h the Periodic lists which will be processed in the current Frame. The content of this
31: 4
RO register is updated by HC after a periodic ED has been processed. HCD may read the
content in determining which ED is currently being processed at the time of reading.

3: 0
16.6.3.9 Host Controller Current First Control Endpoint (HCCTRLHEADED)—

Offset 20h
The HCCTRLHEADED register contains the physical address of the first Endpoint
Descriptor of the Control list.
Access Method
Type: Memory Mapped I/O Register HCCTRLHEADED: [BAR0] + 20h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_3_0
ctrl_head_ed

Bit Default &

Description
Range Access
0h Control Head ED (ctrl_head_ed): HC traverses the Control list starting with the
31: 4 HCCTRLHEADED pointer. The content is loaded from HCCA during the initialization of
RW HC.

3: 0
16.6.3.10 Host Controller Current Control Endpoint (HCCTRLCURED)—Offset 24h

The HCCTRLCURED register contains the physical address of the current Endpoint
Descriptor of the Control list.
Access Method
Type: Memory Mapped I/O Register HCCTRLCURED: [BAR0] + 24h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_3_0
ctrl_cur_ed
Bit Default &

Description
Range Access
Control Current ED (ctrl_cur_ed): This pointer is advanced to the next ED after

serving the present one. HC will continue processing the list from where it left off in the
last Frame. When it reaches the end of the Control list, HC checks the ctrllist_filled of in
0h HCCMDSTATUS. If set, it copies the content of HCCTRLHEADED to HCCTRLCURED and
31: 4
RW clears the bit. If not set, it does nothing. HCD is allowed to modify this register only
when the ctrllist_enable of HCCONTROL is cleared. When set, HCD only reads the
instantaneous value of this register. Initially, this is set to zero to indicate the end of the
Control list.

3: 0
16.6.3.11 Host Controller First Bulk Endpoint (HCBULKHEADED)—Offset 28h

The HCBULKHEADED register contains the physical address of the first Endpoint
Descriptor of the Bulk list.
Access Method
Type: Memory Mapped I/O Register HCBULKHEADED: [BAR0] + 28h
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
bulk_head_ed
reserved_3_0
Bit Default &
Description
Range Access
0h Bulk Head ED (bulk_head_ed): HC traverses the Bulk list starting with the
31: 4 HCBULKHEADED pointer. The content is loaded from HCCA during the initialization of
RW HC.

3: 0
16.6.3.12 Host Controller Current Bulk Endpoint (HCBULKCURED)—Offset 2Ch

The HCBULKCURED register contains the physical address of the current endpoint of
the Bulk list. As the Bulk list will be served in a round-robin fashion, the endpoints will
be ordered according to their insertion to the list.
Access Method
Type: Memory Mapped I/O Register HCBULKCURED: [BAR0] + 2Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_3_0
bulk_cur_ed
Bit Default &

Description
Range Access
Bulk Current ED (bulk_cur_ed): This is advanced to the next ED after the HC has
served the present one. HC continues processing the list from where it left off in the last
Frame. When it reaches the end of the Bulk list, HC checks the ctrllist_filled of
0h HCCONTROL. If set, it copies the content of HCBULKHEADED to HCBULKCURED and
31: 4
RW clears the bit. If it is not set, it does nothing. HCD is only allowed to modify this register
when the bulklist_enable of HCCONTROL is cleared. When set, the HCD only reads the
instantaneous value of this register. This is initially set to zero to indicate the end of the
Bulk list.

3: 0

16.6.3.13 Host Controller Last Completed Descriptor (HCDONEHEAD)—Offset

30h
The HCDONEHEAD register contains the physical address of the last completed Transfer
Descriptor that was added to the Done queue. In normal operation, the Host Controller
Driver should not need to read this register as its content is periodically written to the
HCCA.
Access Method
Type: Memory Mapped I/O Register HCDONEHEAD: [BAR0] + 30h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_3_0
donehead
Bit Default &

Description
Range Access
Done Head (donehead): When a TD is completed, HC writes the content of

0h HCDONEHEAD to the NextTD field of the TD. HC then overwrites the content of
31: 4
RO HCDONEHEAD with the address of this TD. This is set to zero whenever HC writes the
content of this register to HCCA. It also sets the wrback_donehead of HCINTRSTATUS.

3: 0
16.6.3.14 Host Controller Frame Interval (HCFMINTERVAL)—Offset 34h

The HCFMINTERVAL register contains a 14-bit value which indicates the bit time
interval in a Frame, (i.e., between two consecutive SOFs), and a 15-bit value indicating
the Full Speed maximum packet size that the Host Controller may transmit or receive
without causing scheduling overrun. The Host Controller Driver may carry out minor
adjustment on the fm_interval by writing a new value over the present one at each
SOF. This provides the programmability necessary for the Host Controller to
synchronize with an external clocking resource and to adjust any unknown local clock
offset.
Access Method
Type: Memory Mapped I/O Register HCFMINTERVAL: [BAR0] + 34h
(Size: 32 bits)

Default: 00002EDFh

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 0 1 1 0 1 1 1 1 1
fm_interval_tgl
fm_interval
fsmps
reserved_15_14
Bit Default &
Description
Range Access
0h Frame Interval Toggle (fm_interval_tgl): HCD toggles this bit whenever it loads a
31
RW new value to fm_interval.
FS Largest Data Packet (fsmps): This field specifies a value which is loaded into the
0h Largest Data Packet Counter at the beginning of each frame. The counter value
30: 16 represents the largest amount of data in bits which can be sent or received by the HC in
RW a single transaction at any given time without causing scheduling overrun. The field
value is calculated by the HCD.

15: 14
Frame Interval (fm_interval): This specifies the interval between two consecutive
2EDFh SOFs in bit times. The nominal value is set to be 11,999.HCD should store the current
13: 0 value of this field before resetting HC. By setting the hcreset field of HCCMDSTATUS as
RW this will cause the HC to reset this field to its nominal value. HCD may choose to restore
the stored value upon the completion of the Reset sequence.
16.6.3.15 Host Controller Remaining Frame (HCFMREMAINING)—Offset 38h

The HCFMREMAINING register is a 14-bit down counter showing the bit time remaining
in the current Frame. NOTE: In OpenHCI 1.0a, the Frame Remaining and Frame
Remaining Toggle bits in the HCFMREMAINING register are read-only to the Host
Controller Driver. However, it is allowed to write to these bits for debugging purposes.
Though these bits are writable, the Host Controller Driver must not write into these
bits. Doing so yields undefined results.
Access Method
Type: Memory Mapped I/O Register HCFMREMAINING: [BAR0] + 38h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
fm_remaining_tgl
reserved_30_14
fm_remaining

Bit Default &

Description
Range Access
Frame Remaining Toggle (fm_remaining_tgl): This bit is loaded from the

0h fm_interval_tgl field of fm_interval whenever fm_remaining reaches 0. This bit is used
31 by HCD for the synchronization between fm_interval and fm_remaining.
RW NOTE: Though these bits are writable, the Host Controller Driver must not write into
these bits. Doing so yields undefined results.

30: 14
Frame Remaining (fm_remaining): This counter is decremented at each bit time.

When it reaches zero, it is reset by loading the fm_interval value specified in fm_interval
0h at the next bit time boundary. When entering the USBOPERATIONAL state, HC re-loads
13: 0 the content with the fm_interval of HCFMINTERVAL and uses the updated value from the
RW next SOF.
NOTE: Though these bits are writable, the Host Controller Driver must not write into
16.6.3.16 Host Controller Frame Number (HCFMNUMBER)—Offset 3Ch

The HCFMNUMBER register is a 16-bit counter. It provides a timing reference among
events happening in the Host Controller and the Host Controller Driver. The Host
Controller Driver may use the 16-bit value specified in this register and generate a 32-
bit frame number without requiring frequent access to the register. NOTE: In OpenHCI
1.0a, the Frame Number bits in the HCFMNUMBER register are read-only to the Host
Controller Driver. However, it is allowed to write to these bits for debugging purposes.
Though these bits are writable, the Host Controller Driver must not write into these
bits. Doing so yields undefined results.
Access Method
Type: Memory Mapped I/O Register HCFMNUMBER: [BAR0] + 3Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_16
fmnumber
Bit Default &

Description
Range Access

31: 16
Frame Number (fmnumber): This is incremented when fm_remaining is re-loaded. It

will be rolled over to 0 after FFFF. When entering the USBOPERATIONAL state, this will
be incremented automatically. The content will be written to HCCA after HC has
0h incremented the FrameNumber at each frame boundary and sent a SOF but before HC
15: 0
RW reads the first ED in that Frame. After writing to HCCA, HC will set the start_of_frame in
HCINTRSTATUS.
NOTE: Though these bits are writable, the Host Controller Driver must not write into

16.6.3.17 Host Controller Periodic List Start (HCPERIODICSTART)—Offset 40h

The HcPeriodicStart register has a 14-bit programmable value which determines when
is the earliest time HC should start processing the periodic list.
Access Method
(Size: 32 bits) HCPERIODICSTART: [BAR0] + 40h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
periodic_start
reserved_31_14
Bit Default &

Description
Range Access

31: 14
Periodic Start (periodic_start): After hardware reset, this field is cleared. This is
then set by HCD during the HC initialization. The value is calculated roughly as 10% off
0h from fm_interval. A typical value will be h3E67. When fm_remaining reaches the value
13: 0
RW specified, processing of the periodic lists will have priority over Control/Bulk processing.
HC will therefore start processing the Interrupt list after completing the current Control
or Bulk transaction that is in progress.
16.6.3.18 Host Controller LS Threshold (HCLSTHRESHOLD)—Offset 44h

The HcLSThreshold register contains an 11-bit value used by the Host Controller to
determine whether to commit to the transfer of a maximum of 8-byte LS packet before
EOF. Neither the Host Controller nor the Host Controller Driver is allowed to change this
value.
Access Method
Type: Memory Mapped I/O Register HCLSTHRESHOLD: [BAR0] + 44h
(Size: 32 bits)

Default: 00000628h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 1 0 0 0
reserved_31_12
ls_threshold

Bit Default &

Description
Range Access

31: 12
LS Threshold (ls_threshold): This field contains a value which is compared to the

fm_remaining field prior to initiating a Low Speed transaction. The transaction is started
0628h only if fm_remaining this field. The value is calculated by HCD with the consideration of
11: 0
RW transmission and setup overhead.
NOTE: Neither the Host Controller nor the Host Controller Driver are allowed to change
this value.
16.6.3.19 Host Controller Root Hub Descriptor A (HCRHDESPA)—Offset 48h

The HCRHDESPA register is the first register of two describing the characteristics of the
Root Hub.
Access Method
Type: Memory Mapped I/O Register HCRHDESPA: [BAR0] + 48h
(Size: 32 bits)

Default: 02000902h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 1 0
reserved_23_13
no_overcur_prot
overcur_prot_mode
pwr_switch_mode
pwron_to_pwrgood_time
device_type
no_pwr_switching
Bit Default & ndp

Description
Range Access
Power On To Power Good Time (pwron_to_pwrgood_time): This byte specifies

02h the duration HCD has to wait before accessing a powered-on port of the Root Hub. It is
31: 24
RW implementation-specific. The unit of time is 2ms. The duration is calculated as
pwron_to_pwrgood_time*2ms.

23: 13
No Over Current Protection (no_overcur_prot): This bit describes how the

0h overcurrent status for the Root Hub ports is reported. When this bit is cleared, the
12 overcur_prot_mode field specifies global or per-port reporting.
RW 0: Over-current status is reported collectively for all downstream ports
1: No overcurrent protection supported
Over Current Protection Mode (overcur_prot_mode): This bit describes how the
overcurrent status for the Root Hub ports is reported. At reset, this field should reflect
1h the same mode as pwr_switch_mode. This field is valid only if the no_overcur_prot field
11
RW is cleared.
0: over-current status is reported collectively for all downstream ports
1: over-current status is reported on a per-port basis
0h Device Type (device_type): This bit specifies that the Root Hub is not a compound
10 device. The Root Hub is not permitted to be a compound device. This field should always
RO read/write 0.

Bit Default &

Description
Range Access
No Power Switching (no_pwr_switching): These bits are used to specify whether

0h power switching is supported or port are always powered. It is implementation-specific.
9 When this bit is cleared, the pwr_switch_mode specifies global or per-port switching.
RW 0: Ports are power switched
1: Ports are always powered on when the HC is powered on
Power Switching Mode (pwr_switch_mode): This bit is used to specify how the
power switching of the Root Hub ports is controlled. It is implementation-specific. This
field is only valid if the no_pwr_switching field is cleared.
1h 0: all ports are powered at the same time.
8 1: each port is powered individually. This mode allows port power to be controlled by
RW either the global switch or per-port switching. If the port_pwr_ctrlmask bit is set, the
port responds only to port power commands (Set/ClearPortPower). If the port mask is
cleared, then the port is controlled only by the global power switch (Set/
ClearGlobalPower).
02h Number Downstream Ports (ndp): These bits specify the number of downstream
7: 0 ports (ndp) supported by the Root Hub. It is implementation-specific. The minimum
RO number of ports is 1.
16.6.3.20 Host Controller Root Hub Descriptor B (HCRHDESPB)—Offset 4Ch

The HCRHDESPB register is the second register of two describing the characteristics of
the Root Hub. These fields are written during initialization to correspond with the
system implementation.
Access Method
(Size: 32 bits) HCRHDESPB: [BAR0] + 4Ch

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved_31_18
reserved_16_3
device_removable
port_pwr_ctrlmask
reserved_1
Bit Default &

Description
Range Access

31: 19
Port Power Control Mask (port_pwr_ctrlmask): Each bit indicates if a port is

affected by a global power control command when pwr_switch_mode is set. When set,
the port's power state is only affected by per-port power control (Set/ClearPortPower).
When cleared, the port is controlled by the global power switch (Set/ClearGlobalPower).
0h If the device is configured to global switching mode (pwr_switch_mode=0), this field is
18: 17
RW not valid.
bit 0: Reserved
bit 1: Ganged-power mask on Port #1
bit 2: Ganged-power mask on Port #2
NOTE: Per Port Power Control is supported

16: 3

Bit Default &

Description
Range Access
Device Removable (device_removable): Each bit is dedicated to a port of the Root

Hub. When cleared, the attached device is removable. When set, the attached device is
0h not removable.
2: 1
RW bit 0: Reserved
bit 1: Device attached to Port #1
bit 2: Device attached to Port #2
0h reserved_1 (reserved_1): Reserved bit. These bits are reserved and should be set to
0
RO zero.
16.6.3.21 Host Controller Root Hub Status (HCRHSTATUS)—Offset 50h

The HCRHSTATUS register is divided into two parts. The lower word of a Dword
represents the Hub Status field and the upper word represents the Hub Status Change
field. Reserved bits should always be written '0'.
Access Method
Type: Memory Mapped I/O Register HCRHSTATUS: [BAR0] + 50h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
overcur_ind_change
local_pwrsts_change
reserved_30_18
reserved_14_2
dev_rmtwkup_enb
clr_rmtwkup_enb
overcur_ind
local_pwrsts
Bit Default &
Description
Range Access
0h Clear Remote Wakeup Enable (clr_rmtwkup_enb): Writing a '1' clears

31
WO dev_rmtwkup_enb. Writing a '0' has no effect.

30: 18
0h Over Current Indicator Change (overcur_ind_change): This bit is set by hardware

17 when a change has occurred to the OCI field of this register. The HCD clears this bit by
RW/1S writing a 1.Writing a 0 has no effect.
Local Power Status Change / SetGlobalPower (local_pwrsts_change): (read)

Local Power Status Change.
The Root Hub does not support the local power status feature and this bit is always read
as 0.
0h (write) SetGlobalPower.
16
RW/1S In global power mode (pwr_switch_mode=0), this bit is written to 1 to turn on power to
all ports (clear pps).
In per-port power mode, it sets pps only on ports whose port_pwr_ctrlmask bit is not
set.
Writing a 0 has no effect.

Bit Default &

Description
Range Access
Device Remote Wakeup Enable / SetRemoteWakeupEnable

(dev_rmtwkup_enb): (read) Device Remote Wakeup Enable.
This bit enables a csc bit as a resume event, causing a USBSUSPEND to USBRESUME
0h state transition and setting the resume_detected interrupt.
15
RW/1S 0: csc is not a remote wakeup event.
1: csc is a remote wakeup event.
(write) SetRemoteWakeupEnable.
Writing a '1' sets dev_rmtwkup_enb. Writing a '0' has no effect.

14: 2
Over Current Indicator (overcur_ind): This bit reports overcurrent conditions when
0h the global reporting is implemented. When set, an overcurrent condition exists. When
1
RO cleared, all power operations are normal. If per-port overcurrent protection is
implemented this bit is always 0
Local Power Status / ClearGlobalPower (local_pwrsts): (read)Local Power Status.

The Root Hub does not support the local power status feature; thus, this bit is always
read as 0.
0h (write) ClearGlobalPower.
0 In global power mode (pwr_switch_mode=0), this bit is written to 1 to turn off power to
RW/1S all ports (clear pps).
In per-port power mode, it clears pps only on ports whose port_pwr_ctrlmask bit is not
set.
Writing a 0 has no effect.
16.6.3.22 Host Controller Root Hub Port Status (HCRHPORTSTS)—Offset 54h

The HCRHPORTSTS[1:ndp] register is used to control and report port events on a per-
port basis. ndp represents the number of HCRHPORTSTS registers that are
implemented in hardware. The lower word is used to reflect the port status, whereas
the upper word reflects the status change bits. Some status bits are implemented with
special write behavior (see below). If a transaction (token through handshake) is in
progress when a write to change port status occurs, the resulting port status change
must be postponed until the transaction completes. Reserved bits should always be
written '0'.
Access Method
Type: Memory Mapped I/O Register HCRHPORTSTS: [BAR0] + 54h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
lsda
poci
prsc
pesc
csc
pps
prs
pss
pes
ccs
reserved_31_21
ocic
pssc
reserved_15_10
reserved_7_5
Bit Default &

Description
Range Access

31: 21

Bit Default &

Description
Range Access
Port Reset Status Change (prsc): This bit is set at the end of the 10-ms port reset
0b signal. The HCD writes a 1 to clear this bit. Writing a 0 has no effect.
20
RW/1C 0: port reset is not complete
1: port reset is complete
Port Over Current Indicator Change (ocic): This bit is valid only if overcurrent
0b conditions are reported on a per-port basis. This bit is set when Root Hub changes the
19 poci bit. The HCD writes a 1 to clear this bit. Writing a 0 has no effect.
RW/1C 0: no change in poci
1: poci has changed
Port Suspend Status Change (pssc): This bit is set when the full resume sequence
has been completed. This sequence includes the 20-s resume pulse, LS EOP, and 3-ms
0b resychronization delay. The HCD writes a 1 to clear this bit. Writing a 0 has no effect.
18
RW/1C This bit is also cleared when prsc is set.
0: resume is not completed
1: resume completed
Port Enable Status Change (pesc): This bit is set when hardware events cause the
0b pes bit to be cleared. Changes from HCD writes do not set this bit. The HCD writes a 1 to
17 clear this bit. Writing a 0 has no effect.
RW/1C 0: no change in pes
1: change in pes
Connect Status Change (csc): This bit is set whenever a connect or disconnect event
occurs. The HCD writes a 1 to clear this bit. Writing a 0 has no effect. If ccs is cleared
when a SetPortReset, SetPortEnable, or SetPortSuspend write occurs, this bit is set to
0b force the driver to re-evaluate the connection status since these writes should not occur
16 if the port is disconnected.
RW/1C 0: no change in ccs
1 = change in ccs
Note: If the device_removable[ndp] bit is set, this bit is set only after a Root Hub reset
to inform the system that the device is attached.

15: 10
Speed Device Attached / Clear Port Power (lsda): (read) Low or Full Speed Device
Attached.
This bit indicates the speed of the device attached to this port. When set, a Low Speed
0b device is attached to this port. When clear, a Full Speed device is attached to this port.
9 This field is valid only when the ccs is set.
RW/1C 0: full speed device attached
1: low speed device attached
(write) ClearPortPower.
The HCD clears the pps bit by writing a 1 to this bit. Writing a 0 has no effect.
Port Power Status / Set Port Power (pps): (read) Port Power Status.
This bit reflects the port power status, regardless of the type of power switching
implemented. This bit is cleared if an overcurrent condition is detected. HCD sets this bit
by writing SetPortPower or SetGlobalPower. HCD clears this bit by writing
ClearPortPower or ClearGlobalPower. Which power control switches are enabled is
determined by pwr_switch_mode and PortPortControlMask[ndp]. In global switching
0b mode , only Set/ClearGlobalPower controls this bit. In per-port power switching
8 (pwr_switch_mode=1), if the port_pwr_ctrlmask[ndp] bit for the port is set, only Set/
RW/1C ClearPortPower commands are enabled. If the mask is not set, only Set/
ClearGlobalPower commands are enabled. When port power is disabled, ccs, pes, pss,
and prs should be reset.
0: port power is off
1: port power is on
(write) SetPortPower.
The HCD writes a 1 to set the pps bit. Writing a 0 has no effect.

7: 5

Bit Default &

Description
Range Access
Port Reset Status / Set Port Reset (prs): (read) Port Reset Status.
When this bit is set by a write to SetPortReset, port reset signaling is asserted. When
reset is completed, this bit is cleared when prsc is set. This bit cannot be set if ccs is
cleared.
0b 0: port reset signal is not active
4
RW/1C 1: port reset signal is active
(write) SetPortReset.
The HCD sets the port reset signaling by writing a 1 to this bit. Writing a 0 has no effect.
If ccs is cleared, this write does not set prs, but instead sets csc. This informs the driver
that it attempted to reset a disconnected port.
Port Over Current Indicator / Clear Suspend Status (poci): (read) Port Over
Current Indicator.
This bit is only valid when the Root Hub is configured in such a way that overcurrent
conditions are reported on a per-port basis. If per-port overcurrent reporting is not
supported, this bit is set to 0. If cleared, all power operations are normal for this port. If
0b set an overcurrent condition exists on this port. This bit always reflects the overcurrent
3
RW/1C input signal
0: no overcurrent condition.
1: overcurrent condition detected.
(write) ClearSuspendStatus.
The HCD writes a 1 to initiate a resume. Writing a 0 has no effect. A resume is initiated
only if pss is set.
Port Suspend Status / Set Port Suspend (pss): Port Suspend Status.
This bit indicates the port is suspended or in the resume sequence. It is set by a
SetSuspendState write and cleared when pssc is set at the end of the resume interval.
This bit cannot be set if ccs is cleared. This bit is also cleared when prsc is set at the end
of the port reset or when the HC is placed in the USBRESUME state. If an upstream
0b resume is in progress, it should propagate to the HC.
2
RW/1C 0: port is not suspended
1: port is suspended
(write) SetPortSuspend.
The HCD sets the pss bit by writing a 1 to this bit. Writing a 0 has no effect. If ccs is
cleared, this write does not set pss; instead it sets csc. This informs the driver that it
attempted to suspend a disconnected port.
Port Enable Status / Set Port Enable (pes): (read) Port Enable Status.
This bit indicates whether the port is enabled or disabled. The Root Hub may clear this
bit when an overcurrent condition, disconnect event, switched-off power, or operational
bus error such as babble is detected. This change also causes pesc to be set. HCD sets
this bit by writing SetPortEnable and clears it by writing ClearPortEnable. This bit cannot
0b be set when ccs is cleared. This bit is also set, if not already, at the completion of a port
1 reset when prsc is set or port suspend when pssc is set.
RW/1C 0: port is disabled
1: port is enabled
(write) SetPortEnable.
The HCD sets pes by writing a 1. Writing a 0 has no effect. If ccs is cleared, this write
does not set pes, but instead sets csc. This informs the driver that it attempted to
enable a disconnected port.
Current Connect Status / Clear Port Enable (ccs): (read) Current Connect Status.
This bit reflects the current state of the downstream port.
0: no device connected
0b 1: device connected
0 (write) ClearPortEnable
RW/1C The HCD writes a 1 to this bit to clear the pes bit. Writing a 0 has no effect. The ccs is
not affected by any write.
Note: This bit is always read 1 when the attached device is non-removable
(device_removable[ndp]).
§§


SDIO/SD/eMMC—Intel® Quark™ SoC X1000
17.0 SDIO/SD/eMMC
The Intel® Quark™ SoC X1000 provides an SDIO/SD/eMMC controller that supports a
single port configurable as:
• One SDIO 3.0 interface
• One SD 3.0 interface
• One eMMC 4.41 interface
3


Table 109. SDIO/SD/eMMC Interface Signals

Direction/
Type
O SD Card Clock
SD_CLK
3.3V Clock frequency up to 50 MHz.
SD Card Data
Bidirectional port used to transfer data to and from SD/eMMC
I/O card.
SD_DATA[7:0]
3.3V By default, after power up or reset, only D[0] is used for data
transfer. A wider data bus can be configured for data transfer,
using D[0]-D[7].
SD Card Detect
I
SD_CD_B Active low when a card is present. Floating (pulled high with
3.3V internal PU) when a card is not present.
SD Card Command
I/O This signal is used for card initialization and transfer of
SD_CMD
MG commands. It has two modes—open-drain for initialization, and
push-pull for fast command transfer.
I SD Card Write Protect

SD_WP
MG Active high to protect from write.
O SD Card Access LED Control

SD_LED
MG Controls the LED to indicate that the card is being accessed
O
SD_PWR SD Card Power Supply Control
MG

Intel® Quark™ SoC X1000—SDIO/SD/eMMC
17.2 Features
17.2.1 SDIO/SD/eMMC Features

Table 110 summarizes the SDIO/SD/eMMC supported/non-supported features.
Table 110. SDIO/SD/eMMC Features

Features Supported
Meets SD Memory Card Specification version 3.0 Yes
Meets SD Host Controller Standard Specification Version 3.0 Yes
Meets SDIO card specification version 3.0 Yes
Meets eMMC Specification version 4.41 Yes
Supports both DMA and Non-DMA mode of operation Yes
Supports SDMA Yes
Supports ADMA1 and ADMA2 Yes
eMMC supports 1 bit, 4 bit and 8 bit bus modes Yes
SD/SDIO supports 1 bit and 4 bit bus modes Yes
SDXC Capacity up to 2TB Yes
High Speed (SD Clock up to 50 MHz - 25 MByte/s for SDIO/SD and 50 MByte/s for
Yes
eMMC)
Integrated ADMA Controller Yes
SD SPI mode Yes
Multi SD/SDIO card Slots No
Boot OS from SD/eMMC device (see Note) Yes
SDIO Dual Voltage support 1.8V/3.3V No
SDIO/SD UHS-I Support No
Wakeup On Card Insertion No
Note:
1. Must pair together with SPI chip.
2. Refer to the Intel® X1000 – Board Support Package (BSP) UEFI EMMC Patch on eMMC enabling for software
portion.
17.2.2 SD 3.0/ SDIO 3.0 / eMMC 4.41 Interfaces

This section provides a very high level overview of the SD, SDIO, eMMC 4.41
specification. Refer to the SD and eMMC specification for complete details.
17.2.2.1 SD 3.0 Bus Topology

The SD Memory Card bus has a single master, single slave (card), synchronous
topology (refer to Figure 33). During the initialization process, commands are sent to
the card allowing the application to detect the card and assign logical addresses to the
physical slot. All data communication in the Card Identification Mode uses the
command line (CMD) only.
The SD bus allows dynamic configuration of the number of data lines. After power up,
by default, the SD Memory Card uses only D[0] for data transfer. After initialization the
host can change the bus width (number of active data lines). This feature allows easy
trade off between hardware cost and system performance. Note that while D[1:3] are
not in use, the related host’s data lines should be in tri-state (input mode).

Figure 33. SD Memory Card Bus Topology
CLK
VDD
Vss SD
SD Host
Memory Card
D[3:0]
CMD
17.2.2.2 SDIO 3.0 Interface

The SDIO card interface is very much like the SD Memory Card interface. The SoC
supports one SDIO card slot.
Figure 34. SDIO Card Bus Topology
CLK CLK
CMD CMD
D[0] D[0]
SD Host SD I/O Card
D[1] D[1]/Interrupt
D[2] D[2]/Wait for Read
D[3] CD/D[3]

17.2.2.3 eMMC Interface
Figure 35. eMMC Interface
CLK
CMD
eMMC Host eMMC Device
D[0:7]
RST#
17.2.3 SDIO/SD/eMMC Host Controller

The Secure Digital Host Controller 3.0 specification defines a standardized host
controller for interfacing to eMMC devices, as well as Secure Digital memory and I/O
cards. The specification encompasses the following:
• A register map and register set
• Data buffer model
• Data movement model, including both Programmed I/O (PIO) and Direct Memory
Access (DMA)
• Three different DMA modes:
— Single DMA (SDMA)
— Advanced DMA Mode 1; this is linked list DMA with the restriction that data
buffers must reside at 4k boundaries.
— Advanced DMA Mode 2; this is linked list DMA with the restriction of 4k aligned
data buffers removed; data buffers may reside at any arbitrary alignment.
64-bit addressing is also supported.
• Interrupt model
• Suspend and resume mechanism
• Power state definition
• Method for automatic generation of STOP_TRANSMISSION commands (Auto
CMD12)
• Test registers
In short, the SD Host Controller specification defines a standard software model for
accessing SD/SDIO/eMMC devices, and makes it possible for standards-compliant host
controllers to work with off-the-shelf device drivers.

17.2.3.1 SD DMA
A new DMA transfer algorithm, called ADMA (Advanced DMA), is defined in the SD Host
Controller Standard Specification Version 2.00. The DMA algorithm defined in the SD
Host Controller Standard Specification Version 1.00 is called SDMA (Single Operation
DMA). SDMA had the disadvantage that a DMA Interrupt generated at every page
boundary disturbs the CPU to reprogram the new system address. This SDMA algorithm
forms a performance bottleneck by interruption at every page boundary. Only one SD
command transaction can be executed per a SDMA operation.
ADMA adopts a scatter-gather DMA algorithm, so that higher data transfer speed is
available. The Host Driver can program a list of data transfers between system memory
and SD card to the Descriptor Table before executing ADMA. It enables ADMA to
operate without interrupting the Host Driver.
There are two types of ADMA; ADMA1 and ADMA2. ADMA1 can support data transfer of
only 4 KByte aligned data in system memory. ADMA2 improves the restriction so that
data of any location and any size can be transferred in system memory. The format of
Descriptor Table is different between them. The Host Controller Specification Ver2.00
defines ADMA2 as standard ADMA and recommends supporting ADMA2 rather than
ADMA1.
The SDIO/SD/eMMC controller supports all three flavors of DMA described in the SD
Host Controller 2.0 specification - SDMA, ADMA1, and ADMA2.
Note: Although the SD Host Controller Standard Specification Version 3.00 states that ADMA1
is not supported in Standard Host Controller versions 3.0 and latter, the SoC SDIO/SD/
eMMC controller supports both ADMA1 and ADMA2.
17.3 References
The SDIO/SD/eMMC controller is a Secure Digital I/O (SDIO), Secure Digital (SD),
MultiMediaCard (eMMC) host controller that is configured to comply with:
• SD Specification Part 1 Physical Layer Specification version 3.00, April 16, 2009
https://www.sdcard.org
• SD Specification Part E1 SDIO Specification version 3.00, December 16, 2010
https://www.sdcard.org
• SD Specification Part A2 SD Host Controller Standard Specification version 3.00,
February 18, 2010 https://www.sdcard.org
• Embedded MultiMediaCard (eMMC) Product Standard v4.41, JESD84-A441 http://
www.jedec.org/.
17.4 Register Map


Figure 36. SDIO/SD/eMMC Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI
CAM
(I/O)
Bus 0
PCI SDIO/
ECAM eMMC PCI
Headers Memory
(Mem)
D:20,F:0 Space
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
D:21
SPI1 F:1
BAR
I2C*/GPIO F:2
Legacy Bridge
D:31, F:0 SDIO/eMMC
Mem
SDIO/eMMC F:0
Registers
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7

Default
Value
2h 3h “Device ID (DEVICE_ID)—Offset 2h” on page 598 08A7h


Default
Value

86h 86h 00h
page 608

Access Method
(Size: 16 bits)
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value
Bit Default &

Description
Range Access
8086h
RO


Access Method
(Size: 16 bits)
Default: 08A7h
15 12 8 4 0
0 0 0 0 1 0 0 0 1 0 1 0 0 1 1 1
value
Bit Default &
Description
Range Access
08A7h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
IntrDis
RSVD
SERREn
RSVD
MasEn
MEMen
RSVD
Bit Default &
Description
Range Access
0h
RO

10
0h
RO
8
00h
RO

2
RW requests.
1
0h
RO


Access Method
(Size: 16 bits)
Default: 0010h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
capable_66Mhz
RcdMasAb
RSVD0
DEVSEL
FastB2B
hasCapList
RSVD1
SigSysErr
RSVD
RSVD
RSVD
IntrStatus
Bit Default &
Description
Range Access
0h
RO
14
13
0h
RO
0b
RO
0h
RO
0b
RO
0h
RO
0b
RO

4
RO register sets.
RO used
0h
RO

Access Method
Default: 08050110h

31 28 24 20 16 12 8 4 0
0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0
progIntf
classCode
subClassCode
rev_id
Bit Default &
Description
Range Access
31: 24
RO performs.
23: 16
15: 8
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
RO

Access Method
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
cfgHdrFormat
multiFnDev
Bit Default &

Description
Range Access

7
RO function device

6: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
start_bist
RSVD
comp_code
BIST_capable
Bit Default &

Description
Range Access

7
0h
RO
0h
RO

Bit Default &

Description
Range Access

3: 0

Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0
address 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
prefetchable
memType
isIO
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
RO

Access Method
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

30h
Access Method
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD
AddrDecodeEn
ROM_base_addr
Bit Default &
Description
Range Access

31: 11
000h
RO

RW also set to 1

Access Method
(Size: 32 bits)
Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
value
RSVD0
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
vector information.

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
7: 0

Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 00h

7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
Default: 01h
7 4 0
0 0 0 0 value 0 0 0 1
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 16 bits) PMC: [B:0, D:20, F:0] + 82h

Default: 4803h
15 12 8 4 0
0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1
PME_support
D2_support
D1_support
aux_curr
RSVD
PME_clock
version
DSI
Bit Default &
Description
Range Access
15: 11
0h
RO
0h
RO
0h
RO
0h
RO
0h
RO
2: 0

84h
Access Method
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
PME_status
no_soft_reset
Data_select
PME_en
RSVD
RSVD
Data_scale
power_state
Bit Default &

Description
Range Access
15
0h
RO

Bit Default &

Description
Range Access
0h
RO
8
0h
RO

3
0h
RO
1: 0

Offset 86h
Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access

7: 0
RO to 0.

87h
Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO


Access Method
(Size: 8 bits)
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value
Bit Default &
Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0
RO in the chain

Access Method
(Size: 16 bits)
Default: 0100h
15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
perVecMskCap
bit64Cap
multiMsgCap
MSIEnable
RSVD0
multiMsgEn

Bit Default &

Description
Range Access
0h
RO

8
7

6: 4

3: 1
RW MSI.

Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD0
Bit Default &
Description
Range Access
31: 2
0h
RO

Access Method
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData

Bit Default &

Description
Range Access
15: 0
changed by hardware

Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
MSIMask
Bit Default &
Description
Range Access
0h
RO
0

Access Method
(Size: 32 bits) PER_VEC_PEND: [B:0, D:20, F:0] + B0h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
RSVD0
Bit Default &

Description
Range Access
0h
RO
0h
RO


Default
Value
0h 3h “SDMA System Address Register (SYS_ADR)—Offset 0h” on page 613 00000000h
4h 5h “Block Size Register (BLK_SIZE)—Offset 4h” on page 614 0000h
6h 7h “Block Count Register (BLK_COUNT)—Offset 6h” on page 615 0000h
8h Bh “Argument Register (ARGUMENT)—Offset 8h” on page 616 00000000h
Ch Dh “Transfer Mode Register (TX_MODE)—Offset Ch” on page 616 0000h
Eh Fh “Command Register (CMD)—Offset Eh” on page 618 0000h
10h 13h “Response Register 0 (RESPONSE0)—Offset 10h” on page 619 00000000h
14h 17h “Response Register 2 (RESPONSE2)—Offset 14h” on page 620 00000000h
18h 1Bh “Response Register 4 (RESPONSE4)—Offset 18h” on page 620 00000000h
1Ch 1Fh “Response Register 6 (RESPONSE6)—Offset 1Ch” on page 621 00000000h
20h 23h “Buffer Data Port Register (BUF_DATA_PORT)—Offset 20h” on page 621 00000000h
24h 27h “Present State Register (PRE_STATE)—Offset 24h” on page 622 1FF00000h
28h 28h “Host Control Register (HOST_CTL)—Offset 28h” on page 627 00h
29h 29h “Power Control Register (PWR_CTL)—Offset 29h” on page 628 00h
2Ah 2Ah “Block Gap Control Register (BLK_GAP_CTL)—Offset 2Ah” on page 628 00h
2Ch 2Dh “Clock Control Register (CLK_CTL)—Offset 2Ch” on page 630 0000h
2Eh 2Eh “Timeout Control Register (TIMEOUT_CTL)—Offset 2Eh” on page 632 00h
2Fh 2Fh “Software Reset Register (SW_RST)—Offset 2Fh” on page 633 00h
30h 31h “Normal Interrupt Status Register (NML_INT_STATUS)—Offset 30h” on page 634 0000h
32h 33h “Error Interrupt Status Register (ERR_INT_STATUS)—Offset 32h” on page 636 0000h
“Normal Interrupt Status Enable (NRM_INT_STATUS_EN)—Offset 34h” on

34h 35h 0000h
page 638
“Error Interrupt Status Enable Register (ERR_INT_STAT_EN)—Offset 36h” on

36h 37h 0000h
page 639
“Normal Interrupt Signal Enable Register (NRM_INT_SIG_EN)—Offset 38h” on

38h 39h 0000h
page 640
“Error Interrupt Signal Enable Register (ERR_INT_SIG_EN)—Offset 3Ah” on

3Ah 3Bh 0000h
page 642
3Ch 3Dh “Auto CMD12 Error Status Register (CMD12_ERR_STAT)—Offset 3Ch” on page 643 0000h
3Eh 3Fh “Host Control 2 Register (HOST_CTRL_2)—Offset 3Eh” on page 644 0000h
40h 43h “Capabilities Register (CAPABILITIES)—Offset 40h” on page 645 01EC32B2h
44h 47h “Capabilities Register 2 (CAPABILITIES_2)—Offset 44h” on page 647 03000000h
“Maximum Current Capabilities Register (MAX_CUR_CAP)—Offset 48h” on

48h 4Bh 00000001h
page 648
“Force Event Register for Auto CMD12 Error Status

50h 51h 0000h
(FORCE_EVENT_CMD12_ERR_STAT)—Offset 50h” on page 649
“Force Event Register for Error Interrupt Status (FORCE_EVENT_ERR_INT_STAT)—

52h 53h 0000h
54h 54h “ADMA Error Status Register (ADMA_ERR_STAT)—Offset 54h” on page 651 00h
58h 5Bh “ADMA System Address Register (ADMA_SYS_ADDR)—Offset 58h” on page 652 00000000h


Default
Value
“initialization Preset Values Register (3.3v or 1.8v) (PRESET_VALUE_0)—Offset

60h 61h 0040h
60h” on page 653
“Default Speed Preset Values Register (PRESET_VALUE_1)—Offset 62h” on

62h 63h 0001h
page 653
64h 65h “High Speed Preset Values Register (PRESET_VALUE_2)—Offset 64h” on page 654 0000h
66h 67h “SDR12 Preset Values Register (PRESET_VALUE_3)—Offset 66h” on page 654 0001h
68h 69h “SDR25 Preset Values Register (PRESET_VALUE_4)—Offset 68h” on page 655 0000h
6Ah 6Bh “SDR50 Preset Values Register (PRESET_VALUE_5)—Offset 6Ah” on page 656 0000h
6Ch 6Dh “SDR104 Preset Values Register (PRESET_VALUE_6)—Offset 6Ch” on page 656 0000h
6Eh 6Fh “DDR50 Preset Values Register (PRESET_VALUE_7)—Offset 6Eh” on page 657 0000h
70h 73h “Boot Time-out control register (BOOT_TIMEOUT_CTRL)—Offset 70h” on page 658 00000000h
74h 74h “Debug Selection Register (DEBUG_SEL)—Offset 74h” on page 658 00h
E0h E3h “Shared Bus Control Register (SHARED_BUS)—Offset E0h” on page 659 00000000h
F0h F0h “SPI Interrupt Support Register (SPI_INT_SUP)—Offset F0h” on page 660 00h
FCh FDh “Slot Interrupt Status Register (SLOT_INT_STAT)—Offset FCh” on page 661 0000h
FEh FFh “Host Controller Version Register (HOST_CTRL_VER)—Offset FEh” on page 661 A702h
17.6.1 SDMA System Address Register (SYS_ADR)—Offset 0h

This register contains the physical system memory address used for DMA transfers or
the second argument for the Auto CMD23
Access Method
Type: Memory Mapped I/O Register SYS_ADR: [BAR0] + 0h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
sys_adr

Bit Default &

Description
Range Access
SDMA System Address / Auto CMD23 Argument 2 (sys_adr): This register

contains the physical system memory address used for DMA transfers (1) or the second
argument for the Auto CMD23 (2).
(1) SDMA System Address This register contains the system memory address for a
SDMA transfer. When the Host Controller stops a SDMA transfer, this register shall point
to the system address of the next contiguous data position. It can be accessed only if no
transaction is executing (i.e., after a transaction has stopped). Read operations during
transfers may return an invalid value. The Host Driver shall initialize this register before
starting a SDMA transaction. After SDMA has stopped, the next system address of the
next contiguous data position can be read from this register. The SDMA transfer waits at
0h the every boundary specified by the Host SDMA Buffer Boundary in the Block Size
31: 0 register. The Host Controller generates DMA Interrupt to request the Host Driver to
RW update this register. The Host Driver sets the next system address of the next data
position to this register. When the most upper byte of this register (003h) is written, the
Host Controller restarts the SDMA transfer. When restarting SDMA by the Resume
command or by setting Continue Request in the Block Gap Control register, the Host
Controller shall start at the next contiguous address stored here in the SDMA System
Address register. ADMA does not use this register
(2) Argument 2 This register is used with the Auto CMD23 to set a 32-bit block count
value to the argument of the CMD23 while executing Auto CMD23. If Auto CMD23 is
used with ADMA, the full 32-bit block count value can be used. If Auto CMD23 is used
without AMDA, the available block count value is limited by the Block Count register.
65535 blocks is the maximum value in this case.
17.6.2 Block Size Register (BLK_SIZE)—Offset 4h

This register is used to configure the number of bytes in a data block.
Access Method
Type: Memory Mapped I/O Register BLK_SIZE: [BAR0] + 4h
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
boundary
tr_blk_size
tx_blk_size_12
Bit Default &

Description
Range Access
0h Transfer Block Size [12] (tx_blk_size_12): Transfer Block Size 12th bit. This bit is
15
RW added to support 4Kb Data block transfer.

Bit Default &

Description
Range Access
Host SDMA Buffer Boundary (boundary): The large contiguous memory space may
not be available in the virtual memory system. To perform long SDMA transfer, SDMA
System Address register shall be updated at every system memory boundary during
SDMA transfer. These bits specify the size of contiguous buffer in the system memory.
The SDMA transfer shall wait at the every boundary specified by these fields and the
Host Controller generates the DMA Interrupt to request the Host Driver to update the
SDMA System Address register. At the end of transfer, the Host Controller may issue or
may not issue DMA Interrupt. In particular, DMA Interrupt shall not be issued after
Transfer Complete Interrupt is issued. In case of this register is set to 0 (buffer size =
4K bytes), lower 12-bit of byte address points data in the contiguous buffer and the
000b upper 20-bit points the location of the buffer in the system memory. The SDMA transfer
14: 12 stops when the Host Controller detects carry out of the address from bit 11 to 12. These
RW bits shall be supported when the SDMA Support in the Capabilities register is set to 1
and this function is active when the DMA Enable in the Transfer Mode register is set to 1.
ADMA does not use this register.
000b 4K bytes (Detects A11 carry out)
010b 16K Bytes (Detects A13 carry out)
Transfer Block Size (tr_blk_size): This register specifies the block size of data
transfers for CMD17, CMD18, CMD24, CMD25, and CMD53. Values ranging from 1 up to
the maximum buffer size can be set. In case of memory, it shall be set up to 512 bytes
(Refer to SD Host Controller Simplified Specification Version 3.00 - Implementation Note
in Section 1.7.2). It can be accessed only if no transaction is executing (i.e., after a
transaction has stopped). Read operations during transfers may return an invalid value,
and write operations shall be ignored.
000h 0800h 2048 Bytes
11: 0 ... ...
RW 0200h 512 Bytes
01FFh 511 Bytes
... ...
0004h 4 Bytes
0003h 3 Bytes
0002h 2 Bytes
0001h 1 Byte
0000h No data transfer
17.6.3 Block Count Register (BLK_COUNT)—Offset 6h

This register is enabled when Block Count Enable in the Transfer Mode register is set to
1 and is valid only for multiple block transfers.
Access Method
(Size: 16 bits) BLK_COUNT: [BAR0] + 6h

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
blk_count

Bit Default &

Description
Range Access
Blocks Count For Current Transfer (blk_count): The Host Driver shall set this
register to a value between 1 and the maximum block count. The Host Controller
decrements the block count after each block transfer and stops when the count reaches
zero. Setting the block count to 0 results in no data blocks is transferred. This register
should be accessed only when no transaction is executing (i.e., after transactions are
stopped). During data transfer, read operations on this register may return an invalid
0000h value and write operations are ignored. When a suspend command is completed, the
15: 0 number of blocks yet to be transferred can be determined by reading this register.
RW Before issuing a resume command, the Host Driver shall restore the previously saved
block count.
FFFFh 65535 blocks
... ...
0002h 2 blocks
0001h 1 block
0000h Stop Count
17.6.4 Argument Register (ARGUMENT)—Offset 8h

This register contains the SD Command Argument.
Access Method
Type: Memory Mapped I/O Register ARGUMENT: [BAR0] + 8h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
argument
Bit Default &

Description
Range Access
0h Command Argument 1 (argument): The SD Command Argument is specified as

31: 0
RW bit39-8 of Command-Format in the SD Physical Layer Specification.
17.6.5 Transfer Mode Register (TX_MODE)—Offset Ch

This register is used to control the operation of data transfers. The Host Driver shall set
this register before issuing a command which transfers data (Refer to Data Present
Select in the Command register), or before issuing a Resume command. The Host
Driver shall save the value of this register when the data transfer is suspended (as a
result of a Suspend command) and restore it before issuing a Resume command. To
prevent data loss, the Host Controller shall implement write protection for this register
during data transactions. Writes to this register shall be ignored when the Command
Inhibit (DAT) in the Present State register is 1.
Access Method
Type: Memory Mapped I/O Register TX_MODE: [BAR0] + Ch
(Size: 16 bits)


Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
cmd_comp_ata
blk_sel
dma_en
data_tr_dir
cmd12_en
blk_count_en
rsvd
Bit Default &
Description
Range Access
000h
15: 7 RSVD (rsvd): Reserved
RO
Command Completion Signal Enable for CE-ATA Device (cmd_comp_ata): 1 -

0b Device will send command completion Signal
6 0 - Device will not send command completion Signal
RW
NOTE: This filed is not part of the SD Host Controller Specification v3.00.
Multi / Single Block Select (blk_sel): This bit is set when issuing multiple-block
transfer commands using DAT line. For any other commands, this bit shall be set to 0. If
0h this bit is 0, it is not necessary to set the Block Count register. (Refer to Table 2-8 on SD
5
RW Host Controller Simplified Specification Version 3.00)
1 Multiple Block
0 Single Block
Data Transfer Direction Select (data_tr_dir): This bit defines the direction of DAT
0h line data transfers. The bit is set to 1 by the Host Driver to transfer data from the SD
4 card to the SD Host Controller and it is set to 0 for all other commands.
RW 1 Read (Card to Host)
0 Write (Host to Card)
Auto CMD Enable (cmd12_en): This field determines use of auto command functions.
00b Auto Command Disabled
01b Auto CMD12 Enable
10b Auto CMD23 Enable
11b Reserved
There are two methods to stop Multiple-block read and write operation.
(1) Auto CMD12 Enable The Host Controller issues CMD12 automatically when last block
transfer is completed. Auto CMD12 error is indicated to the Auto CMD Error Status
register. The Host Driver shall not set this bit if the command does not require CMD12.
In particular, secure commands defined in the Part 3 File Security specification do not
require CMD12.
00b (2) Auto CMD23 Enable The Host Controller issues a CMD23 automatically before issuing
3: 2 a command specified in the Command Register. The following conditions are required to
RW use the Auto CMD23:
- Auto CMD23 Supported
- A memory card that supports CMD23 (SCR[33]=1)
- If DMA is used, it shall be ADMA.
- Only when CMD18 or CMD25 is issued (Note, the Host Controller does not check
command index.)
Auto CMD23 can be used with or without ADMA. By writing the Command register, the
Host Controller issues a CMD23 first and then issues a command specified by the
Command Index in Command register. If response errors of CMD23 are detected, the
second command is not issued. A CMD23 error is indicated in the Auto CMD Error Status
register. 32-bit block count value for CMD23 is set to SDMA System Address / Argument
2 register
Block Count Enable (blk_count_en): This bit is used to enable the Block Count
register, which is only relevant for multiple block transfers. When this bit is 0, the Block
Count register is disabled, which is useful in executing an infinite transfer. (Refer to
0b Table 2-8 on SD Host Controller Simplified Specification Version 3.00) If ADMA2 data
1
RW transfer is more than 65535 blocks, this bit shall be set to 0. In this case, data transfer
length is designated by Descriptor Table.
1 Enable
0 Disable

Bit Default &

Description
Range Access
DMA Enable (dma_en): DMA can be enabled only if it is supported as indicated in the
Capabilities register. One of the DMA modes can be selected by DMA Select in the Host
0b Control 1 register. If DMA is not supported, this bit is meaningless and shall always read
0 0. If this bit is set to 1, a DMA operation shall begin when the Host Driver writes to the
RW upper byte of Command register (00Fh).
1 DMA Data transfer
0 No data transfer or Non DMA data transfer
17.6.6 Command Register (CMD)—Offset Eh

The Host Driver shall check the Command Inhibit (DAT) bit and Command Inhibit
(CMD) bit in the Present State register before writing to this register. Writing to the
upper byte of this register triggers SD command generation. The Host Driver has the
responsibility to write this register because the Host Controller does not protect for
writing when Command Inhibit (CMD) is set.
Access Method
Type: Memory Mapped I/O Register CMD: [BAR0] + Eh
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
data_pr_sel
cmd_index_chk_en
resp_type_sel
cmd_crc_chk_en
cmd_index
rsvd
reserved
cmd_type
Bit Default &

Description
Range Access
0h
RO
0h Command Index (cmd_index): These bits shall be set to the command number
13: 8 (CMD0-63, ACMD0-63) that is specified in bits 45-40 of the Command-Format in the
RW Physical Layer Specification and SDIO Card Specification

Bit Default &

Description
Range Access
Command Type (cmd_type): There are three types of special commands: Suspend,
Resume and Abort. These bits shall be set to 00b for all other commands.
(1) Suspend Command If the Suspend command succeeds, the Host Controller shall
assume the SD Bus has been released and that it is possible to issue the next command,
which uses the DAT line. The Host Controller shall de-assert Read Wait for read
transactions and stop checking busy for write transactions. The interrupt cycle shall
start, in 4-bit mode. If the Suspend command fails, the Host Controller shall maintain its
current state, and the Host Driver shall restart the transfer by setting Continue Request
in the Block Gap Control register.
00b (2) Resume Command The Host Driver re-starts the data transfer by restoring the
7: 6
RW registers in the range of 000-00Dh. The Host Controller shall check for busy before
starting write transfers.
(3) Abort Command If this command is set when executing a read transfer, the Host
Controller shall stop reads to the buffer. If this command is set when executing a write
transfer, the Host Controller shall stop driving the DAT line. After issuing the Abort
command, the Host Driver should issue a software reset.
11b Abort CMD12, CMD52 for writing I/O Abort in CCCR
10b Resume CMD52 for writing Function Select in CCCR
01b Suspend CMD52 for writing Bus Suspend in CCCR
00b Normal Other commands
Data Present Select (data_pr_sel): This bit is set to 1 to indicate that data is present
and shall be transferred using the DAT line. It is set to 0 for the following:
(1) Commands using only CMD line (ex. CMD52).
0b (2) Commands with no data transfer but using busy signal on DAT[0] line (R1b or R5b
5 ex. CMD38)
RW (3) Resume command
1 Data Present
0 No Data Present
Command Index Check Enable (cmd_index_chk_en): If this bit is set to 1, the

Host Controller shall check the Index field in the response to see if it has the same value
0b as the command index. If it is not, it is reported as a Command Index Error. If this bit is
4
RW set to 0, the Index field is not checked.
1 Enable
0 Disable
Command CRC Check Enable (cmd_crc_chk_en): If this bit is set to 1, the Host
Controller shall check the CRC field in the response. If an error is detected, it is reported
0b as a Command CRC Error. If this bit is set to 0, the CRC field is not checked. The position
3 of CRC field is determined according to the length of the response. (Refer to definition in
RW D01-00 and Table 2-10 in the SD Host Controller Simplified Specification Version 3.00).
1 Enable
0 Disable
0h
2 Reserved (reserved): Reserved
RO
Response Type Select (resp_type_sel): 00 No Response

0h 01 Response Length 136
1: 0
RW 10 Response Length 48
11 Response Length 48 check Busy after response
17.6.7 Response Register 0 (RESPONSE0)—Offset 10h

This register is used to store responses from SD cards
Access Method
Type: Memory Mapped I/O Register RESPONSE0: [BAR0] + 10h
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
cmd_resp
Bit Default &
Description
Range Access
Command Response (cmd_resp): Section 2.2.7 and Table 2-12 in the SD Host
0h Controller Simplified Specification Version 3.00 describe the mapping of command
31: 0 responses from the SD Bus to this register for each response type. In the table, R[]
RO refers to a bit range within the response data as transmitted on the SD Bus, REP[] refers
to a bit range within the Response register

Access Method
Type: Memory Mapped I/O Register RESPONSE2: [BAR0] + 14h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
cmd_resp
Bit Default &

Description
Range Access

Access Method
(Size: 32 bits) RESPONSE4: [BAR0] + 18h

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
cmd_resp
Bit Default &
Description
Range Access
17.6.10 Response Register 6 (RESPONSE6)—Offset 1Ch

Access Method
Type: Memory Mapped I/O Register RESPONSE6: [BAR0] + 1Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
cmd_resp
Bit Default &

Description
Range Access
17.6.11 Buffer Data Port Register (BUF_DATA_PORT)—Offset 20h

32-bit data port register to access internal buffer
Access Method
(Size: 32 bits) BUF_DATA_PORT: [BAR0] + 20h

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
buf_data
Bit Default &
Description
Range Access
0h Buffer Data (buf_data): The Host Controller buffer can be accessed through this 32-
31: 0 bit Data Port register. Refer to section 1.7 in the SD Host Controller Simplified
RW Specification Version 3.00
17.6.12 Present State Register (PRE_STATE)—Offset 24h

This bit indicates whether one of the DAT line on SD bus is in use. The Host Driver can
get status of the Host Controller from this 32-bit read only register
Access Method
Type: Memory Mapped I/O Register PRE_STATE: [BAR0] + 24h
(Size: 32 bits)

Default: 1FF00000h
31 28 24 20 16 12 8 4 0
0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rd_tx_active
crd_st_stable
wr_tx_active
dat_ln_active
reserved2
cmd_ln_sig_lvl
dat_sig_lvl
wr_prot_sw_pin_lvl
reserved1
buf_rd_en
data_ln_sig_lvl
cmd_inhibit_cmd
crd_ins
buf_wr_en
reserved
crd_det_pin_lvl
re_tune_req
cmd_inhibit_dat
Bit Default &
Description
Range Access
0h
31: 29 Reserved2 (reserved2): Reserved
RO
DAT[7:4] Line Signal Level (dat_sig_lvl): This status is used to check DAT line level
to recover from errors, and for debugging.
Fh D28 - DAT[7]
28: 25 D27 - DAT[6]
RO D26 - DAT[5]
D25 - DAT[4]
NOTE: This filed is not part of the SD Host Controller Specification v3.00.
1b CMD Line Signal Level (cmd_ln_sig_lvl): This status is used to check the CMD line
24
RO level to recover from errors, and for debugging.
DAT[3:0] Line Signal Level (data_ln_sig_lvl): This status is used to check the DAT
line level to recover from errors, and for debugging. This is especially useful in detecting
Fh the busy signal level from DAT[0].
23: 20 D23 - DAT[3]
RO D22 - DAT[2]
D21 - DAT[1]
D20 - DAT[0]

Bit Default &

Description
Range Access
Write Protect Switch Pin Level (wr_prot_sw_pin_lvl): The Write Protect Switch is
0b supported for memory and combo cards. This bit reflects the inverse value of the
19 SD_WP pin.
RO 1 Write enabled (SD_WP=0)
0 Write protected (SD_WP=1)
Card Detect Pin Level (crd_det_pin_lvl): This bit reflects the inverse value of the
SD_CD_B pin. Debouncing is not performed on this bit. This bit may be valid when Card
0b State Stable is set to 1, but it is not guaranteed because of propagation delay. Use of
18
RO this bit is limited to testing since it must be debounced by software.
1 Card present (SD_CD_B=0)
0 No card present (SD_CD_B=1)
Card State Stable (crd_st_stable): This bit is used for testing. If it is 0, the Card
Detect Pin Level is not stable. If this bit is set to 1, it means the Card Detect Pin Level is
0b stable. No Card state can be detected by this bit is set to 1 and Card Inserted is set to 0.
17
RO The Software Reset For All in the Software Reset register shall not affect this bit.
1 No Card or Inserted (stable)
0 Reset or Debouncing (unstable)
Card Inserted (crd_ins): This bit indicates whether a card has been inserted. The
Host Controller shall debounce this signal so that the Host Driver will not need to wait
for it to stabilize. Changing from 0 to 1 generates a Card Insertion interrupt in the
Normal Interrupt Status register and changing from 1 to 0 generates a Card Removal
interrupt in the Normal Interrupt Status register. The Software Reset For All in the
Software Reset register shall not affect this bit. If a card is removed while its power is on
0b and its clock is oscillating, the Host Controller shall clear SD Bus Power in the Power
16
RO Control register and SD Clock Enable in the Clock Control register. When this bit is
changed from 1 to 0, the Host Controller shall immediately stop driving CMD and
DAT[3:0] (tri-state). In addition, the Host Driver should clear the Host Controller by the
Software Reset For All in Software Reset register. The card detect is active regardless of
the SD Bus Power.
1 Card Inserted
0 Reset or Debouncing or No Card
0h
RO
Buffer Read Enable (buf_rd_en): This status is used for non-DMA read transfers. The
Host Controller may implement multiple buffers to transfer data efficiently. This read
only flag indicates that valid data exists in the host side buffer. If this bit is 1, readable
0b data exists in the buffer. A change of this bit from 1 to 0 occurs when all the block data
11
RO is read from the buffer. A change of this bit from 0 to 1 occurs when block data is ready
in the buffer and generates the Buffer Read Ready interrupt.
1 Read enable
0 Read disable
Buffer Write Enable (buf_wr_en): This status is used for non-DMA write transfers.
The Host Controller can implement multiple buffers to transfer data efficiently. This read
only flag indicates if space is available for write data. If this bit is 1, data can be written
to the buffer. A change of this bit from 1 to 0 occurs when all the block data is written to
0b the buffer. A change of this bit from 0 to 1 occurs when top of block data can be written
10
RO to the buffer and generates the Buffer Write Ready interrupt. The Host Controller should
neither set Buffer Write Enable nor generate Buffer Write Ready Interrupt after the last
block data is written to the Buffer Data Port Register.
1 Write enable
0 Write disable
Read Transfer Active (rd_tx_active): This status is used for detecting completion of
a read transfer. (Refer to Section 3.12.3 in the SD Host Controller Simplified
Specification Version 3.00) This bit is set to 1 for either of the following conditions:
(1) After the end bit of the read command.
(2) When read operation is restarted by writing a 1 to Continue Request in the Block
Gap Control register.
0b This bit is cleared to 0 for either of the following conditions:
9 (1) When the last data block as specified by block length is transferred to the System.
RO (2) In case of ADMA2, end of read operation is designated by Descriptor Table.
(3) When all valid data blocks in the Host Controller have been transferred to the
System and no current block transfers are being sent as a result of the Stop At Block
Gap Request being set to 1.
A Transfer Complete interrupt is generated when this bit changes to 0.
1 Transferring data
0 No valid data

Bit Default &

Description
Range Access
Write Transfer Active (wr_tx_active): This status indicates a write transfer is active.
If this bit is 0, it means no valid write data exists in the Host Controller. (Refer to Section
3.12.4 in the SD Host Controller Simplified Specification Version 3.00) This bit is set in
either of the following cases:
(1) After the end bit of the write command.
(2) When write operation is restarted by writing a 1 to Continue Request in the Block
Gap Control register.
This bit is cleared in either of the following cases:
0b (1) After getting the CRC status of the last data block as specified by the transfer count
8 (Single and Multiple) In case of ADMA2, transfer count is designated by Descriptor
RO Table.
(2) After getting the CRC status of any block where data transmission is about to be
stopped by a Stop At Block Gap Request.
During a write transaction, a Block Gap Event interrupt is generated when this bit is
changed to 0, as the result of the Stop At Block Gap Request being set. This status is
useful for the Host Driver in determining non DAT line commands can be issued during
write busy.
1 Transferring data
0 No valid data
0h
7: 4 Reserved (reserved): Reserved
RO
Re-Tuning Request (re_tune_req): Host Controller may request Host Driver to

execute re-tuning sequence by setting this bit when the data window is shifted by
temperature drift and a tuned sampling point does not have a good margin to receive
0b correct data. This bit is cleared when a command is issued with setting Execute Tuning
3
RO in the Host Control 2 register. Changing of this bit from 0 to 1 generates Re-Tuning
Event. Refer to Normal Interrupt Status registers for more detail. This bit isn't set to 1 if
Sampling Clock Select in the Host Control 2 register is set to 0 (using fixed sampling
clock). Refer to Re-Tuning Modes in the Capabilities register for more detail.

Bit Default &

Description
Range Access
DAT Line Active (dat_ln_active): This bit indicates whether one of the DAT line on
SD Bus is in use.
(a) In the case of read transactions
This status indicates whether a read transfer is executing on the SD Bus. Changing this
value from 1 to 0 generates a Block Gap Event interrupt in the Normal Interrupt Status
register, as the result of the Stop At Block Gap Request being set. Refer to Section
3.12.3 for details on timing. This bit shall be set in either of the following cases:
(1) After the end bit of the read command.
(2) When writing a 1 to Continue Request in the Block Gap Control register to restart a
read transfer.
This bit shall be cleared in either of the following cases: (1) When the end bit of the last
data block is sent from the SD Bus to the Host Controller. In case of ADMA2, the last
block is designated by the last transfer of Descriptor Table.
(2) When a read transfer is stopped at the block gap initiated by a Stop At Block Gap
Request.
The Host Controller shall stop read operation at the start of the interrupt cycle of the
next block gap by driving Read Wait or stopping SD clock. If the Read Wait signal is
already driven (due to data buffer cannot receive data), the Host Controller can continue
to stop read operation by driving the Read Wait signal. It is necessary to support Read
Wait in order to use suspend / resume function.
0b (b) In the case of write transactions
2 This status indicates that a write transfer is executing on the SD Bus. Changing this
RO value from 1 to 0 generate a Transfer Complete interrupt in the Normal Interrupt Status
register. Refer to Section 3.12.4 for sequence details. This bit shall be set in either of the
following cases:
(1) After the end bit of the write command.
(2) When writing to 1 to Continue Request in the Block Gap Control register to continue
a write transfer.
This bit shall be cleared in either of the following cases:
(1) When the SD card releases write busy of the last data block. If SD card does not
drive busy signal for 8 SD Clocks, the Host Controller shall consider the card drive -Not
Busy-. In case of ADMA2, the last block is designated by the last transfer of Descriptor
Table.
(2) When the SD card releases write busy prior to waiting for write transfer as a result of
a Stop At Block Gap Request.
(c) Command with busy
This status indicates whether a command indicates busy (ex. erase command for
memory) is executing on the SD Bus. This bit is set after the end bit of the command
with busy and cleared when busy is de-asserted. Changing this bit from 1 to 0 generate
a Transfer Complete interrupt in the Normal Interrupt Status register. Refer Figure 2-11
to Figure 2-13 on SD Host Controller Simplified Specification Version 3.00.
1 DAT Line Active
0 DAT Line Inactive
Command Inhibit (DAT) (cmd_inhibit_dat): This status bit is generated if either the
DAT Line Active or the Read Transfer Active is set to 1. If this bit is 0, it indicates the
Host Controller can issue the next SD Command. Commands with busy signal belong to
0b Command Inhibit (DAT) (ex. R1b, R5b type). Changing from 1 to 0 generates a Transfer
1 Complete interrupt in the Normal Interrupt Status register. Note: The SD Host Driver
RO can save registers in the range of 000-00Dh for a suspend transaction after this bit has
changed from 1 to 0.
1 Cannot issue command which uses the DAT line
0 Can issue command which uses the DAT line
Command Inhibit (CMD) (cmd_inhibit_cmd): If this bit is 0, it indicates the CMD

line is not in use and the Host Controller can issue a SD Command using the CMD line.
This bit is set immediately after the Command register (00Fh) is written. This bit is
cleared when the command response is received. Auto CMD12 and Auto CMD23 consist
of two responses. In this case, this bit is not cleared by the response of CMD12 or
CMD23 but cleared by the response of a read/write command. Status issuing Auto
CMD12 is not read from this bit. So if a command is issued during Auto CMD12
0b operation, Host Controller shall manage to issue two commands: CMD12 and a
0 command set by Command register. Even if the Command Inhibit (DAT) is set to 1,
RO commands using only the CMD line can be issued if this bit is 0. Changing from 1 to 0
generates a Command Complete Interrupt in the Normal Interrupt Status register. If the
Host Controller cannot issue the command because of a command conflict error or
because of Command Not Issued By Auto CMD12 Error (Refer to Section (Refer to
Section 2.2.18 and 2.2.23 SD Host Controller Simplified Specification Version 3.00), this
bit shall remain 1 and the Command Complete is not set.
1 Cannot issue command
0 Can issue command using only CMD line

17.6.13 Host Control Register (HOST_CTL)—Offset 28h

Access Method
Type: Memory Mapped I/O Register HOST_CTL: [BAR0] + 28h
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
crd_det_sig_sel
data_tx_wid
sd8_bit_mode
dma_sel
crd_det_tst_lvl
led_ctl
hi_spd_en
Bit Default &
Description
Range Access
Card Detect Signal Selection (crd_det_sig_sel): This bit selects source for the card
detection. When the source for the card detection is switched, the interrupt should be
0b disabled during the switching period by clearing the Interrupt Status/Signal Enable
7 register in order to mask unexpected interrupt being caused by the glitch. The Interrupt
RW Status/Signal Enable should be disabled during over the period of debouncing.
1 The Card Detect Test Level is selected (for test purpose)
0 SD_CD_B is selected (for normal use)
Card Detect Test Level (crd_det_tst_lvl): This bit is enabled while the Card Detect
0h Signal Selection is set to 1 and it indicates card inserted or not. Generates (card ins or
6 card removal) interrupt when the normal int sts enable bit is set.
RW 1 - Card Inserted
0 - No Card
Extended Data Transfer Width (sd8_bit_mode): This bit controls 8-bit bus width
mode for embedded device. Support of this function is indicated in 8-bit Support for
Embedded Device in the Capabilities register. If a device supports 8-bit bus mode, this
0h bit may be set to 1. If this bit is 0, bus width is controlled by Data Transfer Width in the
5 Host Control 1 register. This bit is not effective when multiple devices are installed on a
RW bus slot (Slot Type is set to 10b in the Capabilities register). In this case, each device
bus width is controlled by Bus Width Preset field in the Shared Bus Control register.
1 8-bit Bus Width
0 Bus Width is Selected by Data Transfer Width
DMA Select (dma_sel): One of supported DMA modes can be selected. The host driver
shall check support of DMA modes by referring the Capabilities register.
00 - SDMA is selected
00b 01 - 32-bit Address ADMA1 is selected
4: 3
RW 10 - 32-bit Address ADMA2 is selected
11 - 64-bit Address ADMA2 is selected
NOTE: Codes 01 and 11 are not part of the SD Host Controller Simplified Specification
Version 3.00
High Speed Enable (hi_spd_en): This bit is optional. Before setting this bit, the HD
shall check the High Speed Support in the capabilities register. If this bit is set to 0
(default), the HC outputs CMD line and DAT lines at the falling edge of the SD clock (up
to 25 MHz/ 20MHz for MMC). If this bit is set to 1, the HC outputs CMD line and DAT
0b lines at the rising edge of the SD clock (up to 50 MHz for SD/52MHz for MMC)/ 208Mhz
2
RW (for SD3.0) If Preset Value Enable in the Host Control 2 register is set to 1, Host Driver
needs to reset SD Clock Enable before changing this field to avoid generating clock
glitches. After setting this field, the Host Driver sets SD Clock Enable again
1 - High Speed Mode
0 - Normal Speed Mode
Data Transfer Width (SD1 or SD4) (data_tx_wid): This bit selects the data width of
0h the Host Controller. The Host Driver shall set it to match the data width of the SD card.
1
RW 1 4-bit mode
0 1-bit mode

Bit Default &

Description
Range Access
LED Control (led_ctl): This bit is used to caution the user not to remove the card while
the SD card is being accessed. If the software is going to issue multiple SD commands,
0h this bit can be set during all these transactions. It is not necessary to change for each
0
RW transaction.
1 LED on
0 LED off
17.6.14 Power Control Register (PWR_CTL)—Offset 29h

Access Method
(Size: 8 bits) PWR_CTL: [BAR0] + 29h

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
sd_bus_volt_sel
sd_bus_pwr
rsvd
hw_rst
Bit Default &

Description
Range Access
0h
RO
0b HW reset (hw_rst): Hardware reset signal is generated for eMMC4.4 card when this
4 bit is set.
RW NOTE: Not part of the SD Host Controller Simplified Specification Version 3.00
SD Bus Voltage Select (sd_bus_volt_sel): By setting these bits, the Host Driver
selects the voltage level for the SD card. Before setting this register, the Host Driver
0h shall check the Voltage Support bits in the Capabilities register. If an unsupported
3: 1 voltage is selected, the Host System shall not supply SD Bus voltage.
RW 111 3.3V (Typ.)
110 Reserved
101 Reserved
SD Bus Power (sd_bus_pwr): Before setting this bit, the SD Host Driver shall set SD
Bus Voltage Select. If the Host Controller detects the No Card state, this bit shall be
0b cleared. If this bit is cleared, the Host Controller shall immediately stop driving CMD and
0 DAT[3:0] (tri-state) and drive SDCLK to low level (Refer to Section 2.2.14 of SD Host
RW Controller Simplified Specification Version 3.00).
1 Power on
0 Power off
17.6.15 Block Gap Control Register (BLK_GAP_CTL)—Offset 2Ah

Access Method
Type: Memory Mapped I/O Register BLK_GAP_CTL: [BAR0] + 2Ah
(Size: 8 bits)


Default: 00h
7 4 0
0 0 0 0 0 0 0 0
rd_wait_ctl
int_blk_gap
rsvd
alt_boot_en
boot_en
cont_req
spi_mode
stp_blk_gap_req
Bit Default &
Description
Range Access
0b
7 RSVD (rsvd): Reserved
RO
Alternate Boot Mode Enable (alt_boot_en): To start boot code access in alternative
0b mode.
6 1 - To start alternate boot mode access
RW 0 - To stop alternate boot mode access
NOTE: Not part of the SD Host Controller Simplified Specification Version 3.00
Boot Enable (boot_en): To start boot code access

0b 1 - To start boot code access
5
RW 0 - To stop boot code access
SPI mode enable (spi_mode): SPI mode enable bit.

0b 1 - SPI mode
4
RW 0 - SD mode
Interrupt At Block Gap (int_blk_gap): This bit is valid only in 4-bit mode of the
SDIO card and selects a sample point in the interrupt cycle. Setting to 1 enables
interrupt detection at the block gap for a multiple block transfer. Setting to 0 disables
0b interrupt detection during a multiple block transfer. If the SD card cannot signal an
3 interrupt during a multiple block transfer, this bit should be set to 0. When the Host
RW Driver detects an SD card insertion, it shall set this bit according to the CCCR of the
SDIO card.
1 Enabled
0 Disabled
Read Wait Control (rd_wait_ctl): The read wait function is optional for SDIO cards.
If the card supports read wait, set this bit to enable use of the read wait protocol to stop
read data using the DAT[2] line. Otherwise, the Host Controller has to stop the SD Clock
0b to hold read data, which restricts commands generation. When the Host Driver detects
2 an SD card insertion, it shall set this bit according to the CCCR of the SDIO card. If the
RW card does not support read wait, this bit shall never be set to 1 otherwise DAT line
conflict may occur. If this bit is set to 0, Suspend/Resume cannot be supported.
1 Enable Read Wait Control
0 Disable Read Wait Control
Continue Request (cont_req): This bit is used to restart a transaction, which was
stopped using the Stop At Block Gap Request. To cancel stop at the block gap, set Stop
At Block Gap Request to 0 and set this bit 1 to restart the transfer. The Host Controller
automatically clears this bit in either of the following cases: (1) In the case of a read
0b transaction, the DAT Line Active changes from 0 to 1 as a read transaction restarts.
1 (2) In the case of a write transaction, the Write Transfer Active changes from 0 to 1 as
RW the write transaction restarts.
Therefore, it is not necessary for Host Driver to set this bit to 0. If Stop At Block Gap
Request is set to 1, any write to this bit is ignored.
1 Restart
0 Not affect

Bit Default &

Description
Range Access
Stop At Block Gap Request (stp_blk_gap_req): This bit is used to stop executing
read and write transaction at the next block gap for non-DMA, SDMA and ADMA
transfers. The Host Driver shall leave this bit set to 1 until the Transfer Complete is set
to 1. Clearing both Stop At Block Gap Request and Continue Request shall not cause the
transaction to restart. When Host Controller version is 1.00, the Host Driver can set this
bit if the card supports Read Wait Control. When Host Controller version is 2.00 or later,
0b the Host Driver can set this bit regardless of the card supports Read Wait Control. The
0 Host Controller shall stop read transfer by using Read Wait or stopping SD clock. In case
RW of write transfers in which the Host Driver writes data to the Buffer Data Port register,
the Host Driver shall set this bit after all block data is written. If this bit is set to 1, the
Host Driver shall not write data to Buffer Data Port register. This bit affects Read
Transfer Active, Write Transfer Active, DAT Line Active and Command Inhibit (DAT) in
the Present State register.
1 Stop
0 Transfer
17.6.16 Clock Control Register (CLK_CTL)—Offset 2Ch

This register is used configure the frequency of the SDIO controller, and enable the
clock.
Access Method
Type: Memory Mapped I/O Register CLK_CTL: [BAR0] + 2Ch
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
clk_gen_sel
sdclk_freq_sel
upr_sdclk_freq_sel
int_clk_stable
rsvd
sd_clk_en
int_clk_en

Bit Default &

Description
Range Access
SDCLK Frequency Select (sdclk_freq_sel): This register is used to select the

frequency of the SDCLK pin. The frequency is not programmed directly; rather this
register holds the divisor of the Base Clock Frequency For SD clock in the capabilities
register. Only the following settings are allowed.
(1) 8-bit Divided Clock Mode
80h - base clock divided by 256
00h - base clock(10MHz-63MHz)
Setting 00h specifies the highest frequency of the SD Clock. When setting multiple bits,
the most significant bit is used as the divisor. But multiple bits should not be set. The
two default divider values can be calculated by the frequency that is defined by the Base
Clock Frequency For SD Clock in the Capabilities register.
- 400KHz divider value
- 25MHz divider value
- 50MHz divider value
The frequency of the SDCLK is set by the following formula: Clock Frequency = (Base
clock) / divisor. Thus choose the smallest possible divisor which results in a clock
frequency that is less than or equal to the target frequency. Maximum Frequency for SD
= 50Mhz (base clock) Maximum Frequency for MMC = 52Mhz (base clock) Minimum
Frequency = 195.3125Khz (50Mhz / 256), same clock for MMC also.
For example, if the Base Clock Frequency For SD Clock in the Capabilities register has
the value 33MHz, and the target frequency is 25MHz, then choosing the divisor value of
01h will yield 16.5MHz, which is the nearest frequency less than or equal to the target.
00h Similarly, to approach a clock value of 400KHz, the divisor value of 40h yields the
15: 8 optimal clock value of 258KHz.
RW (2) 10-bit Divided Clock Mode
Host Controller Version 3.00 supports this mandatory mode instead of the 8-bit Divided
Clock Mode. The length of divider is extended to10 bits and all divider values shall be
supported.
3FFh 1/2046 Divided Clock
...
N 1/2N Divided Clock (Duty 50%)
...
002h 1/4 Divided Clock
001h 1/2 Divided Clock
000h Base Clock (10MHz-255MHz)
(3) Programmable Clock Mode
Host Controller Version 3.00 supports this mode as optional. A non-zero value set to
Clock Multiplier in the Capabilities register indicates support of this clock mode. The
multiplier enables the Host System to select a finer grain SD clock frequency. It is not
necessary to support all frequency generation specified by this field because
programmable clock generator is vendor specific and dependent on the implementation.
Therefore, this mode is used with Preset Value registers. The Host Controller vendor
provides possible settings and the Host System vendor sets appropriate values to the
Preset Value registers.
3FFh Base Clock * M / 1024
...
N - 1 Base Clock * M / N
...
002h Base Clock * M / 3
001h Base Clock * M / 2
000h Base Clock * M
This field depends on setting of Preset Value Enable in the Host Control 2 register. If
Preset Value Enable = 0, this field is set by Host Driver. If the Preset Value Enable = 1,
this field is automatically set to a value specified in one of Preset Value registers.
Upper Bits of SDCLK Frequency Select (upr_sdclk_freq_sel): Host Controller

00b Version 1.00 and 2.00 do not support these bits and they are treated as 00b fixed value
7: 6 (ROC). Host Controller Version 3.00 shall support these bits to expand SDCLK
RW Frequency Select to 10-bit. Bit 07-06 is assigned to bit 09-08 of clock divider in SDCLK
Frequency Select.

Bit Default &

Description
Range Access
Clock Generator Select (clk_gen_sel): This bit is used to select the clock generator
mode in SDCLK Frequency Select. If the Programmable Clock Mode is supported (non-
zero value is set to Clock Multiplier in the Capabilities register), this bit attribute is RW,
0b and if not supported, this bit attribute is RO and zero is read. This bit depends on the
5 setting of Preset Value Enable in the Host Control 2 register. If the Preset Value Enable =
RW 0, this bit is set by Host Driver. If the Preset Value Enable = 1, this bit is automatically
set to a value specified in one of Preset Value registers.
1 Programmable Clock Mode
0 Divided Clock Mode
00b
RO
SD Clock Enable (sd_clk_en): The Host Controller shall stop SDCLK when writing this
bit to 0. SDCLK Frequency Select can be changed when this bit is 0. Then, the Host
0b Controller shall maintain the same clock frequency until SDCLK is stopped (Stop at
2 SDCLK=0). If the Card Inserted in the Present State register is cleared, this bit shall be
RW cleared.
1 Enable
0 Disable
Internal Clock Stable (int_clk_stable): This bit is set to 1 when SD Clock is stable
after writing to Internal Clock Enable in this register to 1. The SD Host Driver shall wait
0b to set SD Clock Enable until this bit is set to 1. Note: This is useful when using PLL for a
1
RO clock oscillator that requires setup time.
1 Ready
0 Not Ready
Internal Clock Enable (int_clk_en): This bit is set to 0 when the Host Driver is not
using the Host Controller or the Host Controller awaits a wakeup interrupt. The Host
Controller should stop its internal clock to go very low power state. Still, registers shall
0b be able to be read and written. Clock starts to oscillate when this bit is set to 1. When
0
RW clock oscillation is stable, the Host Controller shall set Internal Clock Stable in this
register to 1. This bit shall not affect card detection.
1 Oscillate
0 Stop
17.6.17 Timeout Control Register (TIMEOUT_CTL)—Offset 2Eh

At the initialization of the Host Controller, the Host Driver shall set the Data Timeout
Counter Value according to the Capabilities register.
Access Method
Type: Memory Mapped I/O Register TIMEOUT_CTL: [BAR0] + 2Eh
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
data_timeout_cnt_val
reserved

Bit Default &

Description
Range Access
0h
RO
Data Timeout Counter Value (data_timeout_cnt_val): This value determines the

interval by which DAT line time-outs are detected. Refer to the Data Time-out Error in
the Error Interrupt Status register for information on factors that dictate time-out
generation. Time-out clock frequency will be generated by dividing the sdclockTMCLK by
0h this value. When setting this register, prevent inadvertent time-out events by clearing
3: 0 the Data Time-out Error Status Enable (in the Error Interrupt Status Enable register).
RW 1111 - Reserved
1110 - TMCLK * 2^27
---
0001 - TMCLK * 2^14
0000 - TMCLK * 2^13
17.6.18 Software Reset Register (SW_RST)—Offset 2Fh

A reset pulse is generated when writing 1 to each bit of this register. After completing
the reset, the Host Controller shall clear each bit. Because it takes some time to
complete software reset, the SD Host Driver shall confirm that these bits are 0.
Access Method
Type: Memory Mapped I/O Register SW_RST: [BAR0] + 2Fh
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
sw_rst_dat_ln
sw_rst_cmd_ln
sw_rst_all
rsvd
Bit Default &

Description
Range Access
0h
RO

Bit Default &

Description
Range Access
Software Reset For DAT Line (sw_rst_dat_ln): Only part of data circuit is reset.
The following registers and bits are cleared by this bit:
- Buffer Data Port Register: Buffer is cleared and Initialized.
- Present State register
Buffer read Enable
Buffer write Enable
Read Transfer Active
Write Transfer Active
DAT Line Active
0b Command Inhibit (DAT)
2
RW - Block Gap Control register
Continue Request
Stop At Block Gap Request
- Normal Interrupt Status register
Buffer Read Ready
Buffer Write Ready
Block Gap Event
Transfer Complete
1 - Reset
0 - Work
Software Reset For CMD Line (sw_rst_cmd_ln): Only part of command circuit is
reset. The following registers and bits are cleared by this bit:
- Present State register
0h Command Inhibit (CMD)
1
RW - Normal Interrupt Status register
Command Complete
1 Reset
0 Work
Software Reset For All (sw_rst_all): This reset affects the entire Host Controller
except for the card detection circuit. Register bits of type ROC, RW, RW1C, RWAC are
cleared to 0. During its initialization, the Host Driver shall set this bit to 1 to reset the
0b Host Controller. The Host Controller shall reset this bit to 0 when Capabilities registers
0 are valid and the Host Driver can read them. Additional use of Software Reset For All
RW may not affect the value of the Capabilities registers. If this bit is set to 1, the host
driver should issue reset command and reinitialize the SD card.
1 Reset
0 Work
17.6.19 Normal Interrupt Status Register (NML_INT_STATUS)—Offset

30h
The Normal Interrupt Status Enable affects reads of this register, but Normal Interrupt
Signal Enable does not affect these reads. An interrupt is generated when the Normal
Interrupt Signal Enable is enabled and at least one of the status bits is set to 1. Writing
1 to a bit of RW1C attribute clears it; writing 0 keeps the bit unchanged. Writing 1 to a
bit of ROC attribute keeps the bit unchanged. More than one status can be cleared with
a single register write. The Card Interrupt is cleared when the card stops asserting the
interrupt; that is, when the Card Driver services the interrupt condition.
Access Method
Type: Memory Mapped I/O Register NML_INT_STATUS: [BAR0] + 30h
(Size: 16 bits)

Default: 0000h

15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
boot_ter_int
buf_rd_rdy
err_int
boot_ck_rcv
int_a
crd_int
buf_wr_rdy
dma_int
blk_gap_event
int_c
crd_ins
re_tune
int_b
tx_comp
cmd_comp
crd_rm
Bit Default &
Description
Range Access
0b Error Interrupt (err_int): If any of the bits in the Error Interrupt Status Register are
15 set, then this bit is set. Therefore the HD can test for an error by checking this bit first.
RO 0 - No Error. 1 - Error.
0h Boot Terminate Interrupt (boot_ter_int): This status is set if the boot operation get
14 terminated. 0 - Boot operation is not terminated. 1 - Boot operation is terminated
RW/1C NOTE: Not part of the SD Host Controller Simplified Specification Version 3.00
0h Boot Acknowledge Received (boot_ck_rcv): This status is set if the boot

13 acknowledge is received from device. 0 - Boot ack is not received. 1 - Boot ack is
RW/1C received. NOTE: Not part of the SD Host Controller Simplified Specification Version 3.00
Re-Tuning Event (re_tune): This status is set if Re-Tuning Request in the Present
0h State register changes from 0 to 1. Host Controller requests Host Driver to perform re-
12 tuning for next data transfer. Current data transfer (not large block count) can be
RO completed without re-tuning. 1 - Re-Tuning should be performed, 0 - Re-Tuning is not
required
0h INT_C (int_c): This status is set if INT_C is enabled and INT_C# pin is in low level.
11 Writing this bit to 1 does not clear this bit. It is cleared by resetting the INT_C interrupt
RO factor
0h INT_B (int_b): This status is set if INT_B is enabled and INT_B# pin is in low level.
10 Writing this bit to 1 does not clear this bit. It is cleared by resetting the INT_B interrupt
RO factor
0h INT_A (int_a): This status is set if INT_A is enabled and INT_A# pin is in low level.
9 Writing this bit to 1 does not clear this bit. It is cleared by resetting the INT_A interrupt
RO factor
Card Interrupt (crd_int): Writing this bit to 1 does not clear this bit. It is cleared by
resetting the SD card interrupt factor. In 1-bit mode, the HC shall detect the Card
Interrupt without SD Clock to support wakeup. In 4-bit mode, the card interrupt signal
is sampled during the interrupt cycle, so there are some sample delays between the
interrupt signal from the card and the interrupt to the Host system. when this status has
been set and the HD needs to start this interrupt service, Card Interrupt Status Enable
0b in the Normal Interrupt Status register shall be set to 0 in order to clear the card
8 interrupt statuses latched in the HC and stop driving the Host System. After completion
RO of the card interrupt service (the reset factor in the SD card and the interrupt signal
may not be asserted), set Card Interrupt Status Enable to 1 and start sampling the
interrupt signal again. Interrupt detected by DAT[1] is supported when there is a card
per slot. In case of shared bus, interrupt pins are used to detect interrupts. If 000b is
set to Interrupt Pin Select in the Shared Bus Control register, this status is effective.
Non-zero value is set to Interrupt Pin Select, INT_A, INT_B or INT_C is then used to
device interrupts. 0 - No Card Interrupt, 1 - Generate Card Interrupt
Card Removal (crd_rm): This status is set if the Card Inserted in the Present State
0b register changes from 1 to 0. When the HD writes this bit to 1 to clear this status the
7 status of the Card Inserted in the Present State register should be confirmed. Because
RW/1C the card detect may possibly be changed when the HD clear this bit an Interrupt event
may not be generated. 0 - Card State Stable or Debouncing, 1 - Card Removed
Card Insertion (crd_ins): This status is set if the Card Inserted in the Present State
0b register changes from 0 to 1. When the HD writes this bit to 1 to clear this status the
6 status of the Card Inserted in the Present State register should be confirmed. Because
RW/1C the card detect may possibly be changed when the HD clear this bit an Interrupt event
may not be generated. 0 - Card State Stable or Debouncing, 1 - Card Inserted
0b Buffer Read Ready (buf_rd_rdy): This status is set if the Buffer Read Enable changes
5 from 0 to 1. Buffer Read Ready is set to 1 for every CMD19 execution in tuning
RW/1C procedure. 0 - Not Ready to read Buffer. 1 - Ready to read Buffer.

Bit Default &

Description
Range Access
0b Buffer Write Ready (buf_wr_rdy): This status is set if the Buffer Write Enable
4
RW/1C changes from 0 to 1. 0 - Not Ready to Write Buffer. 1 - Ready to Write Buffer.
0b DMA Interrupt (dma_int): This status is set if the HC detects the Host DMA Buffer
3 Boundary in the Block Size register. 0 - No DMA Interrupt, 1 - DMA Interrupt is
RW/1C Generated
Block Gap Event (blk_gap_event): If the Stop At Block Gap Request in the Block Gap
Control Register is set, this bit is set. Read Transaction: This bit is set at the falling edge
0b of the DAT Line Active Status (When the transaction is stopped at SD Bus timing. The
2
RW/1C Read Wait must be supported in order to use this function). Write Transaction: This bit is
set at the falling edge of Write Transfer Active Status (After getting CRC status at SD
Bus timing). 0 - No Block Gap Event, 1 - Transaction stopped at Block Gap
Transfer Complete (tx_comp): This bit is set when a read / write transfer and a
command with busy is completed.
(1) In the case of a Read Transaction
This bit is set at the falling edge of Read Transfer Active Status. This interrupt is
generated in two cases. The first is when a data transfer is completed as specified by
data length (After the last data has been read to the Host System). The second is when
data has stopped at the block gap and completed the data transfer by setting the Stop
At Block Gap Request in the Block Gap Control register (After valid data has been read
to the Host System). Refer to Section 3.12.3 of SD Host Controller Simplified
Specification Version 3.00 for more details on the sequence of events.
(2) In the case of a Write Transaction
This bit is set at the falling edge of the DAT Line Active Status. This interrupt is
generated in two cases. The first is when the last data is written to the SD card as
0b specified by data length and the busy signal released. The second is when data transfers
1
RW/1C are stopped at the block gap by setting Stop At Block Gap Request in the Block Gap
Control register and data transfers completed. (After valid data is written to the SD card
and the busy signal released). Refer to Section 3.12.4 for more details on the sequence
of events.
(3) In the case of a command with busy
This bit is set when busy is de-asserted. Refer to DAT Line Active and Command Inhibit
(DAT) in the Present State register. Table on page 66 of SD Host Controller Simplified
Specification Version 3.00, Relation between Transfer Complete and Data Timeout Error,
shows that Transfer Complete has higher priority than Data Timeout Error. If both bits
are set to 1, execution of a command can be considered to be completed.
1 Command execution is completed
0 Not complete
While performing tuning procedure (Execute Tuning is set to 1), Transfer Complete is
not set to 1.
Command Complete (cmd_comp): This bit is set when get the end bit of the
command response. Auto CMD12 and Auto CMD23 consist of two responses. Command
Complete is not generated by the response of CMD12 or CMD23 but generated by the
response of a read/write command. Refer to Command Inhibit (CMD) in the Present
0b State register for how to control this bit. Table on page 67 of SD Host Controller
0
RW/1C Simplified Specification Version 3.00, Relation between command complete and
command time-out error, shows that Command Timeout Error has higher priority than
Command Complete. If both bits are set to 1, it can be considered that the response
was not received correctly.
1 Command complete 0 No command complete
17.6.20 Error Interrupt Status Register (ERR_INT_STATUS)—Offset 32h

Signals defined in this register can be enabled by the Error Interrupt Status Enable
register, but not by the Error Interrupt Signal Enable register. The interrupt is
generated when the Error Interrupt Signal Enable is enabled and at least one of the
statuses is set to 1. Writing to 1 clears the bit and writing to 0 keeps the bit unchanged.
More than one status can be cleared at the one register write.
Access Method

Type: Memory Mapped I/O Register ERR_INT_STATUS: [BAR0] + 32h

(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
vend_spec_err_status
adma_err
data_end_bit_err
cmd_timeout_err
ceata_err
tgt_rsp_err
tune_err
cmd12_err
data_crc_err
data_timeout_err
cur_limit_err
cmd_index_err
cmd_end_bit_err
cmd_crc_err
rsvd
Bit Default &

Description
Range Access
00b
15: 14 Vendor Specific Error Status (vend_spec_err_status): Reserved
RW
0b CEATA Error Status (ceata_err): Occurs when ATA command termination has
13 occurred due to an error condition the device has encountered. 0 - no error, 1 - error
0b Target Response Error (tgt_rsp_err): Occurs when detecting ERROR in m_hresp

12 (dma transaction) 0 - no error, 1 - error NOTE: Not part of the SD Host Controller
RW Simplified Specification Version 3.00
0b
RO
Tuning Error (tune_err): This bit is set when an unrecoverable error is detected in a
tuning circuit except during tuning procedure (Occurrence of an error during tuning
procedure is indicated by Sampling Select). By detecting Tuning Error, Host Driver needs
0b to abort a command executing and perform tuning. To reset tuning circuit, Sampling
10 Clock shall be set to 0 before executing tuning procedure. The Tuning Error is higher
RW priority than the other error interrupts generated during data transfer. By detecting
Turning Error, the Host Driver should discard data transferred by a current read/write
command and retry data transfer after the Host Controller retrieved from tuning circuit
error. 1 - Error, 0 - No Error
0b ADMA Error (adma_err): This bit is set when the Host Controller detects errors during
9 ADMA based data transfer. The state of the ADMA at an error occurrence is saved in the
RW ADMA Error Status Register. 1 - Error, 0 - No Error
Auto CMD Error (cmd12_err): Auto CMD12 and Auto CMD23 use this error status.
0b This bit is set when detecting that one of the bits D00-D04 in Auto CMD Error Status
8 register has changed from 0 to 1. In case of Auto CMD12, this bit is set to 1, not only
RW when the errors in Auto CMD12 occur but also when Auto CMD12 is not executed due to
the previous command error. 0 - No Error, 1 - Error
Current Limit Error (cur_limit_err): By setting the SD Bus Power bit in the Power
Control Register, the HC is requested to supply power for the SD Bus. If the HC supports
0b the Current Limit Function, it can be protected from an Illegal card by stopping power
7 supply to the card in which case this bit indicates a failure status. Reading 1 means the
RW HC is not supplying power to SD card due to some failure. Reading 0 means that the HC
is supplying power and no error has occurred. This bit shall always set to be 0, if the HC
does not support this function. 0 - No Error, 1 - Power Fail
0b Data End Bit Error (data_end_bit_err): Occurs when detecting 0 at the end bit
6 position of read data which uses the DAT line or the end bit position of the CRC status. 0
RW - No Error, 1 - Error
0b Data CRC Error (data_crc_err): Occurs when detecting CRC error when transferring
5 read data which uses the DAT line or when detecting the Write CRC Status having a
RW value of other than 010. 0 - No Error, 1 - Error

Bit Default &

Description
Range Access
Data Timeout Error (data_timeout_err): Occurs when detecting one of following

0b timeout conditions. 1. Busy Time-out for R1b, R5b type. 2. Busy Time-out after Write
4
RW CRC status 3. Write CRC status Time-out 4. Read Data Time-out 0 - No Error, 1 -
Timeout
0b Command Index Error (cmd_index_err): Occurs if a Command Index error occurs

3
RW in the Command Response. 0 - No Error, 1 - Error
0b Command End Bit Error (cmd_end_bit_err): Occurs when detecting that the end bit
2
RW of a command response is 0. 0 - No Error, 1 - End Bit Error Generated
Command CRC Error (cmd_crc_err): Command CRC Error is generated in two cases.
1. If a response is returned and the Command Time-out Error is set to 0, this bit is set
to 1 when detecting a CRT error in the command response 2. The HC detects a CMD line
0b conflict by monitoring the CMD line when a command is issued. If the HC drives the CMD
1
RW line to 1 level, but detects 0 level on the CMD line at the next SDCLK edge, then the HC
shall abort the command (Stop driving CMD line) and set this bit to 1. The Command
Time-out Error shall also be set to 1 to distinguish CMD line conflict. 0 - No Error, 1 -
CRC Error Generated
Command Timeout Error (cmd_timeout_err): Occurs only if the no response is

0b returned within 64 SDCLK cycles from the end bit of the command. If the HC detects a
0 CMD line conflict, in which case Command CRC Error shall also be set. This bit shall be
RW set without waiting for 64 SDCLK cycles because the command will be aborted by the
HC. 0 - No Error, 1 - Timeout
17.6.21 Normal Interrupt Status Enable (NRM_INT_STATUS_EN)—

Offset 34h
Setting to 1 enables Interrupt Status.
Access Method
Type: Memory Mapped I/O Register NRM_INT_STATUS_EN: [BAR0] + 34h
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
fixed_0
boot_ack_rcv_stat_en
buf_rd_rdy_stat_en
blk_gap_event_stat_en
cmd_comp_stat_en
int_b_stat_en
int_a_stat_en
crd_int_stat_en
boot_ter_int_stat_en
re_tune_stat_en
int_c_stat_en
crd_rm_stat_en
tx_comp_stat_en
crd_ins_stat_en
buf_wr_rdy_stat_en
dma_int_stat_en
Bit Default &

Description
Range Access
0h Fixed to 0 (fixed_0): The HC shall control error Interrupts using the Error Interrupt
15
RO Status Enable register.
0h Boot Terminate Interrupt Status Enable (boot_ter_int_stat_en): 0 - Masked, 1 -

14
RW Enabled NOTE: Not part of the SD Host Controller Simplified Specification Version 3.00
0h Boot ACK Receive Enable (boot_ack_rcv_stat_en): 0 - Masked, 1 - Enabled NOTE:

13
RW Not part of the SD Host Controller Simplified Specification Version 3.00

Bit Default &

Description
Range Access
0h
12 Re-Tuning Event Status Enable (re_tune_stat_en): 0 - Masked, 1 - Enabled
RW
INT_C Status Enable (int_c_stat_en): If this bit is set to 0, the Host Controller shall
0h clear the interrupt request to the System. The Host Driver may clear this bit before
11
RW servicing the INT_C and may set this bit again after all interrupt requests to INT_C pin
are cleared to prevent inadvertent interrupts.
INT_B Status Enable (int_b_stat_en): If this bit is set to 0, the Host Controller shall
10
RW servicing the INT_B and may set this bit again after all interrupt requests to INT_B pin
INT_A Status Enable (int_a_stat_en): If this bit is set to 0, the Host Controller shall
9
RW servicing the INT_A and may set this bit again after all interrupt requests to INT_A pin
Card Interrupt Status Enable (crd_int_stat_en): If this bit is set to 0, the HC shall
clear Interrupt request to the System. The Card Interrupt detection is stopped when this
0b bit is cleared and restarted when this bit is set to 1. The HD may clear the Card
8
RW Interrupt Status Enable before servicing the Card Interrupt and may set this bit again
after all Interrupt requests from the card are cleared to prevent inadvertent Interrupts.
0 - Masked, 1 - Enabled
0h
7 Card Removal Status Enable (crd_rm_stat_en): 0 - Masked, 1 - Enabled
RW
0h
6 Card Insertion Status Enable (crd_ins_stat_en): 0 - Masked, 1 - Enabled
RW
0h
5 Buffer Read Ready Status Enable (buf_rd_rdy_stat_en): 0 - Masked, 1 - Enabled
RW
0h
4 Buffer Write Ready Status Enable (buf_wr_rdy_stat_en): 0 - Masked, 1 - Enabled
RW
0h
3 DMA Interrupt Status Enable (dma_int_stat_en): 0 - Masked, 1 - Enabled
RW
0h
2 Block Gap Event Status Enable (blk_gap_event_stat_en): 0 - Masked, 1 - Enabled
RW
0h
1 Transfer Complete Status Enable (tx_comp_stat_en): 0 - Masked, 1 - Enabled
RW
0h
0 Command Complete Status Enable (cmd_comp_stat_en): 0 - Masked, 1 - Enabled
RW
17.6.22 Error Interrupt Status Enable Register (ERR_INT_STAT_EN)—

Offset 36h
Setting to 1 enables Interrupt Status.
Access Method
(Size: 16 bits) ERR_INT_STAT_EN: [BAR0] + 36h

Default: 0000h

15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ceata_err_en
data_timeout_err_stat_en
tgt_rsp_err_en
tune_err_stat_en
adma_err_stat_en
cmd12_err_stat_en
data_end_bit_err_stat_en
cmd_crc_err_stat_en
cur_limit_err_stat_en
data_crc_err_stat_en
cmd_ind_err_stat_en
cmd_end_bit_err_stat_en
cmd_timeout_err_stat_en
rsvd
rsvd0
Bit Default &

Description
Range Access
0b
15: 14 Vendor Specific Error Status Enable (rsvd0): Reserved
RO
0b CEATA Error Status Enable (ceata_err_en): 0 - Masked, 1 - Enabled NOTE: Not part
13
RW of the SD Host Controller Simplified Specification Version 3.00
0b Target Response Error Status Enable (tgt_rsp_err_en): 0 - Masked, 1 - Enabled

12
0h
RO
0h
10 Tuning Error Status Enable (tune_err_stat_en): 0 - Masked, 1 - Enabled
RW
0h
9 ADMA Error Status Enable (adma_err_stat_en): 0 - Masked, 1 - Enabled
RW
0h
8 Auto CMD12 Error Status Enable (cmd12_err_stat_en): 0 - Masked, 1 - Enabled
RW
0h Current Limit Error Status Enable (cur_limit_err_stat_en): 0 - Masked, 1 -

7
RW Enabled
0h Data End Bit Error Status Enable (data_end_bit_err_stat_en): 0 - Masked, 1 -

6
RW Enabled
0h
5 Data CRC Error Status Enable (data_crc_err_stat_en): 0 - Masked, 1 - Enabled
RW
0h Data Timeout Error Status Enable (data_timeout_err_stat_en): 0 - Masked, 1 -

4
RW Enabled
0h Command Index Error Status Enable (cmd_ind_err_stat_en): 0 - Masked, 1 -

3
RW Enabled
0h Command End Bit Error Status Enable (cmd_end_bit_err_stat_en): 0 - Masked,

2
RW 1 - Enabled
0h Command CRC Error Status Enable (cmd_crc_err_stat_en): 0 - Masked, 1 -

1
RW Enabled
0h Command Timeout Error Status Enable (cmd_timeout_err_stat_en): 0 -

0
RW Masked, 1 - Enabled
17.6.23 Normal Interrupt Signal Enable Register (NRM_INT_SIG_EN)—

Offset 38h
This register is used to select which interrupt status is indicated to the Host System as
the interrupt. These status bits all share the same1 bit interrupt line. Setting any of
these bits to 1 enables interrupt generation.

Access Method
Type: Memory Mapped I/O Register NRM_INT_SIG_EN: [BAR0] + 38h
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
blk_gap_event_sig_en
fixed_0
re_tune_sig_en
boot_ack_rcv_sig_en
int_c_sig_en
buf_wr_rdy_sig_en
int_b_sig_en
crd_ins_sig_en
buf_rd_rdy_sig_en
tx_comp_sig_en
cmd_comp_sig_en
boot_ter_int_sig_en
int_a_sig_en
crd_int_sig_en
crd_rm_sig_en
dma_int_sig_en
Bit Default &
Description
Range Access
0h Fixed to 0 (fixed_0): The HD shall control error Interrupts using the Error Interrupt
15
RO Signal Enable register.
0h Boot Terminate Interrupt Signal Enable (boot_ter_int_sig_en): 0 - Masked, 1 -

14
0h Boot ACK Received Signal Enable (boot_ack_rcv_sig_en): 0 - Masked, 1 -

13
0h
12 Re-Tuning Event Signal Enable (re_tune_sig_en): 0 - Masked, 1 - Enabled
RW
0h
11 INT_C Signal Enable (int_c_sig_en): 0 - Masked, 1 - Enabled
RW
0h
10 INT_B Signal Enable (int_b_sig_en): 0 - Masked, 1 - Enabled
RW
0h
9 INT_A Signal Enable (int_a_sig_en): Reserved.
RW
0h
8 Card Interrupt Signal Enable (crd_int_sig_en): 0 - Masked, 1 - Enabled
RW
0h
7 Card Removal Signal Enable (crd_rm_sig_en): 0 - Masked, 1 - Enabled
RW
0h
6 Card Insertion Signal Enable (crd_ins_sig_en): 0 - Masked, 1 - Enabled
RW
0h
5 Buffer Read Ready Signal Enable (buf_rd_rdy_sig_en): 0 - Masked, 1 - Enabled
RW
0h
4 Buffer Write Ready Signal Enable (buf_wr_rdy_sig_en): 0 - Masked, 1 - Enabled
RW
0h
3 DMA Interrupt Signal Enable (dma_int_sig_en): 0 - Masked, 1 - Enabled
RW
0h
2 Block Gap Event Signal Enable (blk_gap_event_sig_en): 0 - Masked, 1 - Enabled
RW
0h
1 Transfer Complete Signal Enable (tx_comp_sig_en): 0 - Masked, 1 - Enabled
RW

Bit Default &

Description
Range Access
0h
0 Command Complete Signal Enable (cmd_comp_sig_en): 0 - Masked, 1 - Enabled
RW
17.6.24 Error Interrupt Signal Enable Register (ERR_INT_SIG_EN)—

Offset 3Ah
This register is used to select which interrupt status is notified to the Host System as
the interrupt. These status bits all share the same 1 bit interrupt line. Setting any of
these bits to 1 enables interrupt generation.
Access Method
(Size: 16 bits) ERR_INT_SIG_EN: [BAR0] + 3Ah

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ceata_err_sig_en
cur_limit_err_sig_en
tgt_err_rsp_sig_en
adma_err_sig_en
data_end_bit_err_sig_en
data_timeout_err_stat_en
cmd_end_bit_err_stat_en
cmd_crc_err_stat_en
cmd12_err_sig_en
data_crc_err_sig_en
cmd_ind_err_stat_en
cmd_timeout_err_stat_en
rsvd
tune_err_sig
rsvd0
Bit Default &

Description
Range Access
0b
15: 14 Vendor Specific Error Signal Enable (rsvd0): Reserved
RO
0b CEATA Error Signal Enable (ceata_err_sig_en): 0 - Masked, 1 - Enabled NOTE: Not

13
RW part of the SD Host Controller Simplified Specification Version 3.00
0b Target Response Error Signal Enable (tgt_err_rsp_sig_en): 0 - Masked, 1 -

12
0h
RO
0h
10 Tuning Error Signal Enable (tune_err_sig): 0 - Masked, 1 - Enabled
RW
0h
9 ADMA Error Signal Enable (adma_err_sig_en): 0 - Masked, 1 - Enabled
RW
0h
8 Auto CMD12 Error Signal Enable (cmd12_err_sig_en): 0 - Masked, 1 - Enabled
RW
0h
7 Current Limit Error Signal Enable (cur_limit_err_sig_en): 0 - Masked, 1 - Enabled
RW
0h Data End Bit Error Signal Enable (data_end_bit_err_sig_en): 0 - Masked, 1 -

6
RW Enabled

Bit Default &

Description
Range Access
0h
5 Data CRC Error Signal Enable (data_crc_err_sig_en): 0 - Masked, 1 - Enabled
RW
0h Data Timeout Error Signal Enable (data_timeout_err_stat_en): 0 - Masked, 1 -

4
RW Enabled
0h Command Index Error Signal Enable (cmd_ind_err_stat_en): 0 - Masked, 1 -

3
RW Enabled
0h Command End Bit Error Signal Enable (cmd_end_bit_err_stat_en): 0 - Masked,

2
RW 1 - Enabled
0h Command CRC Error Signal Enable (cmd_crc_err_stat_en): 0 - Masked, 1 -

1
RW Enabled
0h Command Timeout Error Signal Enable (cmd_timeout_err_stat_en): 0 - Masked,

0
RW 1 - Enabled
17.6.25 Auto CMD12 Error Status Register (CMD12_ERR_STAT)—Offset

3Ch
This register is used to indicate CMD12 response error of Auto CMD12 and CMD23
response error of Auto CMD23. The Host driver can determine what kind of Auto
CMD12 / CMD23 errors occur by this register. Auto CMD23 errors are indicated in bit
04-01.This register is valid only when the Auto CMD Error is set.
Access Method
Type: Memory Mapped I/O Register CMD12_ERR_STAT: [BAR0] + 3Ch
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rsvd1
cmd_not_iss_cmd12_err
rsvd2
cmd12_ind_err
cmd12_crc_err
cmd12_timeout_err
cmd12_end_bit_err
cmd12_not_exe
Bit Default &

Description
Range Access
0h
15: 8 RSVD1 (rsvd1): Reserved
RO
Command Not Issued By Auto CMD12 Error (cmd_not_iss_cmd12_err):

0b CMD_wo_DAT is not executed due to an Auto CMD12 error (D04 - D01) in this register.
7
RO This bit is set to 0 when Auto CMD Error is generated by Auto CMD23. 0 - No Error, 1 -
Not Issued
0h
RO

Bit Default &

Description
Range Access
0b Auto CMD Index Error (cmd12_ind_err): Occurs if the Command Index error occurs
4
RO in response to a command. 0 - No Error, 1 - Error
0b Auto CMD End Bit Error (cmd12_end_bit_err): Occurs when detecting that the end
3
RO bit of command response is 0. 0 - No Error, 1 - End Bit Error Generated
0b Auto CMD CRC Error (cmd12_crc_err): Occurs when detecting a CRC error in the
2
RO command response. 0 - No Error, 1 - CRC Error Generated
0b Auto CMD Timeout Error (cmd12_timeout_err): Occurs if the no response is

1 returned within 64 SDCLK cycles from the end bit of the command. If this bit is set to 1,
RO the other error status bits (D04 - D02) are meaningless. 0 - No Error, 1 - Timeout
Auto CMD12 Not Executed (cmd12_not_exe): If memory multiple block data

transfer is not started due to command error, this bit is not set because it is not
0b necessary to issue Auto CMD12. Setting this bit to 1 means the HC cannot issue Auto
0
RO CMD12 to stop memory multiple block transfer due to some error. If this bit is set to 1,
other error status bits (D04 - D01) are meaningless. This bit is set to 0 when Auto CMD
Error is generated by Auto CMD23. 0 - Executed, 1 - Not Executed
17.6.26 Host Control 2 Register (HOST_CTRL_2)—Offset 3Eh

Access Method
(Size: 16 bits) HOST_CTRL_2: [BAR0] + 3Eh

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
execute_tuning
driver_strength
vl
preset_value
uhs_mode
async_int
rsvd0
sampling_clock
Bit Default &

Description
Range Access
Preset Value Enable (preset_value): Host Controller Version 3.00 supports this bit.
As the operating SDCLK frequency and I/O driver strength depend on the Host System
implementation, it is difficult to determine these parameters in the Standard Host
Driver. When Preset Value Enable is set to automatic This bit enables the functions
0b defined in the Preset Value registers. 1 Automatic Selection by Preset Value are Enabled
15 0 SDCLK and Driver Strength are controlled by Host Driver If this bit is set to 0, SDCLK
RW Frequency Select, Clock Generator Select in the Clock Control register and Driver
Strength Select in Host Control 2 register are set by Host Driver. If this bit is set to 1,
SDCLK Frequency Select, Clock Generator Select in the Clock Control register and Driver
Strength Select in Host Control 2 register are set by Host Controller as specified in the
Preset Value registers.
Asynchronous Interrupt Enable (async_int): This bit can be set to 1 if a card

support asynchronous interrupt and Asynchronous Interrupt Support is set to 1 in the
0b Capabilities register. Asynchronous interrupt is effective when DAT[1] interrupt is used
14 in 4-bit SD mode (and zero is set to Interrupt Pin Select in the Shared Bus Control
RW register). If this bit is set to 1, the Host Driver can stop the SDCLK during asynchronous
interrupt period to save power. During this period, the Host Controller continues to
deliver Card Interrupt to the host when it is asserted by the Card. 1 Enabled, 0 Disabled
00h
RO

Bit Default &

Description
Range Access
Sampling Clock Select (sampling_clock): This bit is set by tuning procedure when
Execute Tuning is cleared. Writing 1 to this bit is meaningless and ignored. Setting 1
0b means that tuning is completed successfully and setting 0 means that tuning is failed.
7 Host Controller uses this bit to select sampling clock to receive CMD and DAT. This bit is
RW cleared by writing 0. Change of this bit is not allowed while the Host Controller is
receiving response or a read data block. 1 Tuned clock is used to sample data, 0 Fixed
clock is used to sample data
Execute Tuning (execute_tuning): This bit is set to 1 to start tuning procedure and
0b automatically cleared when tuning procedure is completed. The result of tuning is
6
RW/AC indicated to Sampling Clock Select. Tuning procedure is aborted by writing 0 for more
detail about tuning procedure. 1 Execute Tuning, 0 Not Tuned or Tuning Completed
Driver Strength Select (driver_strength): Host Controller output driver in 1.8V

signaling is selected by this bit. In 3.3V signaling, this field is not effective. This field can
be set depends on Driver Type A, C and D support bits in the Capabilities register. This
0b bit depends on setting of Preset Value Enable. If Preset Value Enable = 0, this field is set
5: 4
RW by Host Driver. If Preset Value Enable = 1, this field is automatically set by a value
specified in the one of Preset Value registers. 00b Driver Type B is Selected (Default),
01b Driver Type A is Selected, 10b Driver Type C is Selected, 11b Driver Type D is
Selected
0b
3 Reserved (vl): Reserved
RO
UHS Mode Select (uhs_mode): This field is used to select one of UHS-I modes and
effective when 1.8V Signaling Enable is set to 1. If Preset Value Enable in the Host
Control 2 register is set to 1, Host Controller sets SDCLK Frequency Select, Clock
Generator Select in the Clock Control register and Driver Strength Select according to
0b Preset Value registers. In this case, one of preset value registers is selected by this field.
2: 0 Host Driver needs to reset SD Clock Enable before changing this field to avoid
RW generating clock glitch. After setting this field, Host Driver sets SD Clock Enable again.
000b - SDR12, 001b - SDR25, 010b - SDR50, 011b - SDR104, 100b - DDR50, 101b -
111 Reserved. When SDR50, SDR104 or DDR50 is selected for SDIO card, interrupt
detection at the block gap shall not be used. Read Wait timing is changed for these
modes. Refer to the SDIO Specification Version 3.00 for more detail.
17.6.27 Capabilities Register (CAPABILITIES)—Offset 40h

This register provides the Host Driver with information specific to the Host Controller
implementation. The Host Controller may implement these values as fixed or loaded
from flash memory during power on initialization. Refer to Software Reset For All in the
Software Reset register for loading from flash memory and completion timing control.
Access Method
(Size: 32 bits) CAPABILITIES: [BAR0] + 40h

Default: 01EC32B2h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 1 1 1 1 0 1 1 0 0 0 0 1 1 0 0 1 0 1 0 1 1 0 0 1 0
timeout_clock_frequency
async_int_support
adma2_support
base_clock_frequency_sd_clk
sys_bus_support_64b
volt_support_1p8v
volt_support_3p0v
volt_support_3p3v
suspend_resume_support
sdma_support
high_speed_support
extended_media_bus_support
max_block_length
timeout_clock_unit
slot_type
rsvd4
rsvd5
rsvd6
Bit Default &
Description
Range Access
Slot Type (slot_type): This field indicates usage of a slot by a specific Host System. (A
host controller register set is defined per slot.) Embedded slot for one device (01b)
means that only one non-removable device is connected to a SD bus slot. Shared Bus
0b Slot (10b) can be set if Host Controller supports Shared Bus Control register. The
31: 30 Standard Host Driver controls only a removable card or one embedded device is
RO connected to a SD bus slot. If a slot is configured for shared bus (10b), the Standard
Host Driver does not control embedded devices connected to a shared bus. Shared bus
slot is controlled by a specific host driver developed by a Host System. 00b - Removable
Card Slot, 01b - Embedded Slot for One Device, 10b - Shared Bus Slot, 11b - Reserved
0b Asynchronous Interrupt Support (async_int_support): Refer to SDIO

29 Specification Version 3.00 about asynchronous interrupt. 1 - Asynchronous Interrupt
RO Supported, 0 - Asynchronous Interrupt Not Supported
0b 64-bit System Bus Support (sys_bus_support_64b): 1 - supports 64 bit system

28
RO address, 0 - Does not support 64 bit system address
0b
27 RSVD4 (rsvd4): Reserved
RO
0b Voltage Support 1.8V (volt_support_1p8v): 0 - 1.8V Not Supported, 1 - 1.8V

26
RO Supported

25
RO Supported

24
RO Supported
Suspend/Resume Support (suspend_resume_support): This bit indicates whether

1b the HC supports Suspend / Resume functionality. If this bit is 0, the Suspend and
23
RO Resume mechanism are not supported and the HD shall not issue either Suspend /
Resume commands. 0 - Not Supported, 1 - Supported
1b SDMA Support (sdma_support): This bit indicates whether the HC is capable of using
22 DMA to transfer data between system memory and the HC directly. 0 - SDMA Not
RO Supported, 1 - SDMA Supported.
High Speed Support (high_speed_support): This bit indicates whether the HC and
1b the Host System support High Speed mode and they can supply SD Clock frequency
21
RO from 25MHz to 50 MHz (for SD)/ 20MHz to 52MHz (for MMC). 0 - High Speed Not
Supported, 1 - High Speed Supported
0b
RO
1b
19 ADMA2 Support (adma2_support): 1 - ADMA2 support. 0 - ADMA2 not support
RO
Extended Media Bus Support (extended_media_bus_support): This bit indicates

1b whether the Host Controller is capable of using 8-bit bus width mode. This bit is not
18 effective when Slot Type is set to 10b. In this case, refer to Bus Width Preset in the
RO Shared Bus resister. 1 - Extended Media Bus Supported, 0 - Extended Media Bus not
Supported

Bit Default &

Description
Range Access
Max Block Length (max_block_length): This value indicates the maximum block
00b size that the HD can read and write to the buffer in the HC. The buffer shall transfer this
17: 16
RO block size without wait cycles. Three sizes can be defined as indicated below. 00 - 512
byte, 01 - 1024 byte, 10 - 2048 byte, 11 - 4096 byte
Base Clock Frequency for SD Clock (base_clock_frequency_sd_clk): (1) 6-bit

Base Clock Frequency This mode is supported by the Host Controller Version 1.00 and
2.00. Upper 2-bit is not effective and always 0. Unit values are 1MHz. The supported
clock range is 10MHz to 63MHz. 11xx xxxxb - Not supported, 0011 1111b - 63MHz,
0000 0010b - 2MHz, 0000 0001b - 1MHz, 0000 0000b - Get information via another
32h method. (2) 8-bit Base Clock Frequency This mode is supported by the Host Controller
15: 8 Version 3.00.Unit values are 1MHz. The supported clock range is 10MHz to 255MHz. FFh
RO - 255MHz, 02h - 2MHz, 01h - 1MHz, 00h - Get information via another method. If the
real frequency is 16.5MHz, the lager value shall be set 0001 0001b (17MHz) because
the Host Driver use this value to calculate the clock divider value (Refer to the SDCLK
Frequency Select in the Clock Control register.) and it shall not exceed upper limit of the
SD Clock frequency. If these bits are all 0, the Host System has to get information via
another method.
1b Timeout Clock Unit (timeout_clock_unit): This bit shows the unit of base clock
7
RO frequency used to detect Data Timeout Error. 0 - KHz, 1 - MHz
0b
RO
32h Timeout Clock Frequency (timeout_clock_frequency): This bit shows the base
5: 0 clock frequency used to detect Data Timeout Error. Not 0 - 1KHz to 63KHz or 1MHz to
RO 63MHz, 000000b - Get Information via another method.
17.6.28 Capabilities Register 2 (CAPABILITIES_2)—Offset 44h

This register provides the host driver with information specific to the host controller
implementation.
Access Method
(Size: 32 bits) CAPABILITIES_2: [BAR0] + 44h

Default: 03000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
spi_blk_mode
driver_type_c_sup
retune_modes
spi_mode
tim_cnt_for_retune
driver_type_a_sup
driver_type_d_sup
use_tuning_for_sdr50
rsvd0
clk_mult
rsvd1
rsvd2
ddr50_support
ddr104_support
rsvd3
sdr50_support
Bit Default &

Description
Range Access
0b
RO
1b SPI Block Mode (spi_blk_mode): SPI block mode. 0 - Not Supported, 1 - Supported
25
RO NOTE: Not part of the SD Host Controller Simplified Specification Version 3.00

Bit Default &

Description
Range Access
1b SPI Mode (spi_mode): SPI mode. 0 - Not Supported, 1 - Supported NOTE: Not part of
24
RO the SD Host Controller Simplified Specification Version 3.00
Clock Multiplier (clk_mult): This field indicates clock multiplier value of

0b programmable clock generator. Refer to Clock Control register. Setting 00h means that
23: 16 Host Controller does not support programmable clock generator. FFh Clock Multiplier M
RO = 256, ...., 02h Clock Multiplier M = 3, 01h Clock Multiplier M = 2, 00h Clock Multiplier is
Not Supported
Re-Tuning Modes (retune_modes): This field defines the re-tuning capability of a

Host Controller and how to manage the data transfer length and a Re-Tuning Timer by
0b the Host Driver 00 - Mode1, 01 - Mode2, 10 - Mode3, 11 - Reserved. There are two re-
15: 14
RO tuning timings: Re-Tuning Request and expiration of a Re-Tuning Timer. By receiving
either timing, the Host Driver executes the re-tuning procedure just before a next
command issue
0b Use Tuning for SDR50 (use_tuning_for_sdr50): If this bit is set to 1, this Host
13 Controller requires tuning to operate SDR50. (Tuning is always required to operate
RO SDR104.) 1 SDR50 requires tuning 0, SDR50 does not require tuning
0b
RO
Timer count for Re-Tuning (tim_cnt_for_retune): This field indicates an initial

0h value of the Re-Tuning Timer for Re-Tuning Mode 1 to 3. 0h - Get information via other
11: 8
RO source, 1h = 1 seconds, 2h = 2 seconds, 3h = 4 seconds, 4h = 8 seconds, --, n = 2(n-
1) seconds, --, Bh = 1024 seconds, Fh - Ch = Reserved
0b
RO
0b Driver Type D Support (driver_type_d_sup): This bit indicates support of Driver

6 Type D for 1.8 Signaling. 1 - Driver Type D is Supported, 0 - Driver Type D is Not
RO Supported.
0b Driver Type C Support (driver_type_c_sup): This bit indicates support of Driver

5 Type C for 1.8 Signaling. 1 - Driver Type C is Supported, 0 - Driver Type C is Not
RO Supported.
0b Driver Type A Support (driver_type_a_sup): This bit indicates support of Driver

4 Type A for 1.8 Signaling. 1 - Driver Type A is Supported, 0 - Driver Type A is Not
RO Supported.
0b
RO
0b DDR50 Support (ddr50_support): 1 - DDR50 is Supported, 0 - DDR50 is Not

2
RO Supported
0b DDR104 Support (ddr104_support): 1 - DDR104 is Supported, 0 - DDR104 is Not

1
RO Supported
0b SDR50 Support (sdr50_support): If SDR104 is supported, this bit shall be set to 1.

0 Bit 40 indicates whether SDR50 requires tuning or not. 1 - SDR50 is Supported, 0 -
RO SDR50 is Not Supported
17.6.29 Maximum Current Capabilities Register (MAX_CUR_CAP)—

Offset 48h
These registers indicate maximum current capability for each voltage. The value is
meaningful if Voltage Support is set in the Capabilities register. If this information is
supplied by the Host System via another method, all Maximum Current Capabilities
register shall be 0.
Access Method

Type: Memory Mapped I/O Register MAX_CUR_CAP: [BAR0] + 48h

(Size: 32 bits)

Default: 00000001h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
max_cur_1p8v
max_cur_3p0v
max_cur_3p3v
rsvd
Bit Default &

Description
Range Access
0h
RO
00h
23: 16 Maximum Current for 1.8V (max_cur_1p8v): Maximum current capability for 1.8V
RO
00h
RO
01h
RO
17.6.30 Force Event Register for Auto CMD12 Error Status

(FORCE_EVENT_CMD12_ERR_STAT)—Offset 50h
The Force Event Register is not a physically implemented register. Rather, it is an
address at which the Auto CMD Error Status Register can be written. Writing 1 : set
each bit of the Auto CMD Error Status Register Writing 0 : no effect
Access Method
(Size: 16 bits) FORCE_EVENT_CMD12_ERR_STAT: [BAR0] + 50h

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
non_cmd12_err
cmd_timeout_err
cmd_ind_err
cmd_end_bit_err
cmd_crc_err
cmd_not_exe
reserved
reserved0
Bit Default &

Description
Range Access
00h
RO

Bit Default &

Description
Range Access
0b Force Event for Command Not Issued By Auto CMD12 Error (non_cmd12_err):
7
RW 1 - Interrupt is generated, 0 - No interrupt
00b
RO
0b Force Event for Auto CMD Index Error (cmd_ind_err): 1 - Interrupt is generated, 0
4
RW - No interrupt
0b Force Event for Auto CMD End Bit Error (cmd_end_bit_err): 1 - Interrupt is
3
RW generated, 0 - No interrupt
0b Force Event for Auto CMD CRC Error (cmd_crc_err): 1 - Interrupt is generated, 0 -
2
RW No interrupt
0b Force Event for Auto CMD Timeout Error (cmd_timeout_err): 1 - Interrupt is

1
0b Force Event for Auto CMD Not Executed (cmd_not_exe): 1 - Interrupt is

0
17.6.31 Force Event Register for Error Interrupt Status

(FORCE_EVENT_ERR_INT_STAT)—Offset 52h
The Force Event Register is not a physically implemented register. Rather, it is an
address at which the Error Interrupt Status register can be written. The effect of a write
to this address will be reflected in the Error Interrupt Status Register if the
corresponding bit of the Error Interrupt Status Enable Register is set. Writing 1 : set
each bit of the Error Interrupt Status Register Writing 0 : no effect Note: By setting this
register, the Error Interrupt can be set in the Error Interrupt Status register. In order to
generate interrupt signal, both the Error Interrupt Status Enable and Error Interrupt
Signal Enable shall be set.
Access Method
Type: Memory Mapped I/O Register FORCE_EVENT_ERR_INT_STAT: [BAR0] + 52h
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rsvd0
ceata_err
adma_err
cmd12_err
data_end_bit_err
data_crc_err
data_timeout_err
cmd_crc_err
tgt_rsp_err
cur_limit_err
cmd_end_bit_err
cmd_timeout_err
rsvd
cmd_ind_err
Bit Default &

Description
Range Access
00b
15: 14 Force Event for Vendor Specific Error Status (rsvd0): Reserved
RO
0b
13 Force Event for CEATA error (ceata_err): 1 - Interrupt is generated, 0 - No interrupt
RW

Bit Default &

Description
Range Access
0b Force Event for Target Response Error (tgt_rsp_err): 1 - Interrupt is generated, 0

12
RW - No interrupt
0h
RO
0h
9 Force Event for ADMA Error (adma_err): 1 - Interrupt is generated, 0 - No interrupt
RW
0h Force Event for Auto CMD Error (cmd12_err): 1 - Interrupt is generated, 0 - No

8
RW interrupt
0h Force Event for Current Limit Error (cur_limit_err): 1 - Interrupt is generated, 0 -

7
RW No interrupt
0h Force Event for Data End Bit Error (data_end_bit_err): 1 - Interrupt is generated,
6
RW 0 - No interrupt
0h Force Event for Data CRC Error (data_crc_err): 1 - Interrupt is generated, 0 - No

5
RW interrupt
0h Force Event for Data Timeout Error (data_timeout_err): 1 - Interrupt is

4
0h Force Event for Command Index Error (cmd_ind_err): 1 - Interrupt is generated,

3
RW 0 - No interrupt
0h Force Event for Command End Bit Error (cmd_end_bit_err): 1 - Interrupt is

2
0h Force Event for Command CRC Error (cmd_crc_err): 1 - Interrupt is generated, 0 -

1
RW No interrupt
0h Force Event for Command Timeout Error (cmd_timeout_err): 1 - Interrupt is

0
17.6.32 ADMA Error Status Register (ADMA_ERR_STAT)—Offset 54h

When ADMA Error Interrupt is occurred, the ADMA Error States field in this register
holds the ADMA state and the ADMA System Address Register holds the address around
the error descriptor. For recovering the error, the Host Driver requires the ADMA state
to identify the error descriptor address as follows: ST_STOP: Previous location set in
the ADMA System Address register is the error descriptor address ST_FDS: Current
location set in the ADMA System Address register is the error descriptor address
ST_CADR: This sate is never set because do not generate ADMA error in this state.
ST_TFR: Previous location set in the ADMA System Address register is the error
descriptor address In case of write operation, the Host Driver should use ACMD22 to
get the number of written block rather than using this information, since unwritten data
may exist in the Host Controller. The Host Controller generates the ADMA Error
Interrupt when it detects invalid descriptor data (Valid=0) at the ST_FDS state. In this
case, ADMA Error State indicates that an error occurs at ST_FDS state. The Host Driver
may find that the Valid bit is not set in the error descriptor.
Access Method
Type: Memory Mapped I/O Register ADMA_ERR_STAT: [BAR0] + 54h
(Size: 8 bits)

Default: 00h

7 4 0
0 0 0 0 0 0 0 0
adma_len_mis_err
adma_err_state
rsvd
Bit Default &
Description
Range Access
0h
RO
ADMA Length Mismatch Error (adma_len_mis_err): This error occurs in the

0b following 2 cases. 1. While Block Count Enable being set, the total data length specified
2 by the Descriptor table is different from that specified by the Block Count and Block
RO Length. 2. Total data length can not be divided by the block length. 1 - Error, 0 - No
error
ADMA Error State (adma_err_state): This field indicates the state of ADMA when
error is occurred during ADMA data transfer. This field never indicates 10 because ADMA
00b never stops in this state. D01 D00: ADMA Error State when error is occurred Contents of
1: 0
RO SYS_SDR register, 00 - ST_STOP (Stop DMA) Points next of the error descriptor, 01 -
ST_FDS (Fetch Descriptor) Points the error descriptor, 10 - Never set this state (Not
used), 11 - ST_TFR (Transfer Data) Points the next of the error descriptor
17.6.33 ADMA System Address Register (ADMA_SYS_ADDR)—Offset

58h
This register contains the physical Descriptor address used for ADMA data transfer.
Access Method
(Size: 32 bits) ADMA_SYS_ADDR: [BAR0] + 58h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
adma_sys_addr
Bit Default &

Description
Range Access
ADMA System Address (adma_sys_addr): This register holds byte address of

executing command of the Descriptor table. 32-bit Address Descriptor uses lower 32- bit
of this register. At the start of ADMA, the Host Driver shall set start address of the
0h Descriptor table. The ADMA increments this register address, which points to next line,
31: 0 when every fetching a Descriptor line. When the ADMA Error Interrupt is generated, this
RW register shall hold valid Descriptor address depending on the ADMA state. The Host
Driver shall program Descriptor Table on 32-bit boundary and set 32-bit boundary
address to this register. ADMA2 ignores lower 2-bit of this register and assumes it to be
00b.

17.6.34 initialization Preset Values Register (3.3v or 1.8v)

(PRESET_VALUE_0)—Offset 60h
Access Method
Type: Memory Mapped I/O Register PRESET_VALUE_0: [BAR0] + 60h
(Size: 16 bits)

Default: 0040h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
driver_strength_sel_val
rsvd
clock_gen_sel_val
sdclk_freq_sel_val
Bit Default &
Description
Range Access
00b
15: 14 Reserved (driver_strength_sel_val): Reserved
RO
000b
13: 11 Reserved (rsvd): Reserved
RO
0b Clock Generator Select Value (clock_gen_sel_val): This bit is effective when Host
10 Controller supports programmable clock generator. 1 - Programmable Clock Generator,
RO 0 - Host Controller Ver2.00 Compatible Clock Generator
040h SDCLK Frequency Select Value (sdclk_freq_sel_val): 10-bit preset value to set
9: 0
RO SDCLK Frequency Select in the Clock Control Register is described by a host system.
17.6.35 Default Speed Preset Values Register (PRESET_VALUE_1)—

Offset 62h
Access Method
(Size: 16 bits)

Default: 0001h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
clock_gen_sel_val
sdclk_freq_sel_val
rsvd

Bit Default &

Description
Range Access
00b
RO
000b
RO

9: 0
17.6.36 High Speed Preset Values Register (PRESET_VALUE_2)—Offset

64h
Access Method
(Size: 16 bits) PRESET_VALUE_2: [BAR0] + 64h

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
clock_gen_sel_val
sdclk_freq_sel_val
rsvd
Bit Default &

Description
Range Access
00b
RO
000b
RO

9: 0
17.6.37 SDR12 Preset Values Register (PRESET_VALUE_3)—Offset 66h

Access Method


(Size: 16 bits)

Default: 0001h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
sdclk_freq_sel_val
clock_gen_sel_val
rsvd
Bit Default &

Description
Range Access
00b
RO
000b
RO

9: 0
17.6.38 SDR25 Preset Values Register (PRESET_VALUE_4)—Offset 68h

Access Method
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
clock_gen_sel_val
sdclk_freq_sel_val
rsvd

Bit Default &

Description
Range Access
00b
RO
000b
RO

9: 0
17.6.39 SDR50 Preset Values Register (PRESET_VALUE_5)—Offset 6Ah

Access Method
(Size: 16 bits) PRESET_VALUE_5: [BAR0] + 6Ah

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
clock_gen_sel_val
sdclk_freq_sel_val
rsvd
Bit Default &

Description
Range Access
00b
RO
000b
RO

9: 0
17.6.40 SDR104 Preset Values Register (PRESET_VALUE_6)—Offset 6Ch

Access Method
(Size: 16 bits) PRESET_VALUE_6: [BAR0] + 6Ch


Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
clock_gen_sel_val
sdclk_freq_sel_val
rsvd
Bit Default &

Description
Range Access
00b
RO
000b
RO

9: 0
17.6.41 DDR50 Preset Values Register (PRESET_VALUE_7)—Offset 6Eh

Access Method
Type: Memory Mapped I/O Register PRESET_VALUE_7: [BAR0] + 6Eh
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
sdclk_freq_sel_val
clock_gen_sel_val
rsvd
Bit Default &

Description
Range Access
00b
RO
000b
RO

Bit Default &

Description
Range Access

9: 0
17.6.42 Boot Time-out control register (BOOT_TIMEOUT_CTRL)—Offset

70h
Access Method
(Size: 32 bits) BOOT_TIMEOUT_CTRL: [BAR0] + 70h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
time_cnt_val
Bit Default &

Description
Range Access
0h Boot Data Timeout Counter Value (time_cnt_val): This value determines the
31: 0 interval by which DAT line time-outs are detected during boot operation for eMMC4.4
RW card. The value is in number of sd clock.
17.6.43 Debug Selection Register (DEBUG_SEL)—Offset 74h

Access Method
Type: Memory Mapped I/O Register DEBUG_SEL: [BAR0] + 74h
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
debug_sel
rsvd
Bit Default &

Description
Range Access
00h
RO

Bit Default &

Description
Range Access
0b Debug Select (debug_sel): 1- cmd register, Interrupt status, transmitter module,

0 ahb_iface module and clk sdcard signals are probed out.
WO 0 - receiver module and fifo_ctrl module signals are probed out
17.6.44 Shared Bus Control Register (SHARED_BUS)—Offset E0h

This register is optional. The devices on shared bus are not intended to be controlled by
the Standard Host Driver. This is because shared bus configuration depends on a host
system; the devices on shared bus may be controlled by a specific driver of a host
system.
Access Method
Type: Memory Mapped I/O Register SHARED_BUS: [BAR0] + E0h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
rsvd0
pwr_ctrl
rsvd1
rsvd2
rsvd3
rsvd4
rsvd5
int_pin
clk_pin
bus_width
num_int_pin
num_clk_pin
Bit Default &
Description
Range Access
0b
31 Reserved (rsvd0): Reserved
RO
Back-End Power Control (pwr_ctrl): Each bit of this field controls back-end power
supply for an embedded device. Host interface voltage (VDDH) is not controlled by this
field. The number of devices supported is specified by Number of Clock Pins and a
maximum of 7 devices can be controlled.
D16 Back-end Power Control for Device 1
0h D19 Back-end Power Control for Device 4
30: 24 D20 Back-end Power Control for Device 5
RW D21 Back-end Power Control for Device 6
The function of each bit is defined as follows:
0 Back-end Power is Off
1 Back-end Power is Supplied
Back-End power control is effective for embedded memory devices in the Sleep State
that support the Sleep command (CMD14) to reduce power consumption and embedded
SDIO devices when IOEx is set to 0.
0b
RO
Interrupt Pin Select (int_pin): Interrupt pin inputs are enabled by this field. Enable
of unsupported interrupt pin is meaningless.
0h 000b - Interrupt is detected by Interrupt Cycle,
22: 20
RW xx1b - INT_A is Enabled,
x1xb - INT_B is Enabled,
1xxb - INT_C is Enabled
0b
RO

Bit Default &

Description
Range Access
Clock Pin Select (clk_pin): One of clock pin outputs is selected by this field. Select of
unsupported clock pin is meaningless. Refer to Figure 2-38 (An Example Timing of
Selecting Clock Pin) on SD Host Controller Simplified Specification Version 3.00 for the
0h timing of clock outputs.
18: 16 000b - Clock Pins are Disabled,
RW 001b - CLK[1] is Selected,
010b - CLK[2] is Selected
---
111b - CLK[7] is Selected
0b
RO
Bus Width Preset (bus_width): Shared bus supports mixing of 4-bit and 8- bit bus
width devices. Each bit of this field specifies the bus width for each embedded device.
The number of devices supported is specified by Number of Clock Pins and a maximum
of 7 devices are supported.This field is effective when multiple devices are connected to
a shared bus (Slot Type is set to 10b in the Capabilities register). In the other case,
0h Extended Data Transfer Width in the Host Control 1 register is used to select 8-bit bus
14: 8
RO width. As use of 1-bit mode is not intended for shared bus, Data Transfer Width in the
Host Control 1 register should be set to 1. D08 - Bus width preset for Device 1, D09 -
Bus width preset for Device 2, D10 - Bus width preset for Device 3, D11 - Bus width
preset for Device 4, D12 - Bus width preset for Device 5, D13 - Bus width preset for
Device 6, D14 - Bus width preset for Device 7 The function of each bit is defined as
follows: 0 - 4 bit bus width mode, 1 - 8 bit bus width mode
00b
7: 6 Reserved (rsvd4): Reserved
RO
Number of Interrupt Input Pins (num_int_pin): This field indicates support of

interrupt input pins for shared bus system.Three asynchronous interrupt pins are
0h defined, INT_A#, INT_B# and INT_C#. Which interrupt pin is used is determined by the
5: 4
RO system. Each one is driven by open drain and then wired or connection is possible. 00b
- Interrupt Input Pin is Not Supported, 01b - INTA is Supported, 10b - INTA and INTB
are Supported, 11b - INTA, INTB and INTC are Supported
0b
RO
Number of Clock Pins (num_clk_pin): This field indicates support of clock pins to
select one of devices for shared bus system. Up to 7 clock pins can be supported.Shared
0h bus is supported by specific system. Then Standard Host Driver does not support control
2: 0
RO of these clock pins. 000b - Shared bus is not supported, 001b - 1 SDCLK pin is
supported, 010b - 2 SDCLK pins are supported, ..... , 111b - 7 SDCLK pins are
supported
17.6.45 SPI Interrupt Support Register (SPI_INT_SUP)—Offset F0h

Access Method
Type: Memory Mapped I/O Register SPI_INT_SUP: [BAR0] + F0h
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
spi_int_support

Bit Default &

Description
Range Access
SPI Interrupt Support (spi_int_support): This bit is set to indicate the assertion of
00h interrupts in the SPI mode at any time, irrespective of the status of the card select (CS)
7: 0 line. If this bit is zero, then SDIO card can only assert the interrupt line in the SPI mode
RW when the CS line is asserted.
17.6.46 Slot Interrupt Status Register (SLOT_INT_STAT)—Offset FCh

Access Method
Type: Memory Mapped I/O Register SLOT_INT_STAT: [BAR0] + FCh
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
reserved
int_sig_slot
Bit Default &
Description
Range Access
00h
RO
Interrupt Signal For Each Slot (int_sig_slot): These status bit indicate the logical
OR of Interrupt signal and wake up signal for each slot. A maximum of 8 slots can be
00h defined. If one interrupt signal is associated with multiple slots. the HD can know which
7: 0
RO interrupt is generated by reading these status bits. By a power on reset or by Software
Reset For All, the Interrupt signal shall be de asserted and this status shall read 00h. Bit
00 - Slot 1, Bit 01 - Slot 2, Bit 02 - Slot 3, ----- -----, Bit 07 - Slot 8
17.6.47 Host Controller Version Register (HOST_CTRL_VER)—Offset FEh

Access Method
(Size: 16 bits) HOST_CTRL_VER: [BAR0] + FEh

Default: A702h
15 12 8 4 0
1 0 1 0 0 1 1 1 0 0 0 0 0 0 1 0
vend_ver_num
spec_ver_num

Bit Default &

Description
Range Access
a7h Vendor Version Number (vend_ver_num): This status is reserved for the vendor
15: 8
RO version number. The HD should not use this status.
Specification Version Number (spec_ver_num): This status indicates the Host

02h Controller Spec. Version. The upper and lower 4- bits indicate the version. 00 - SD Host
7: 0
RO Specification version 1.0, 01 - SD Host Specification version 2.00 including only the
feature of the Test Register, 02 - SD Host Specification Version 3.00, others - Reserved
§§


High Speed UART—Intel® Quark™ SoC X1000
18.0 High Speed UART
The Intel® Quark™ SoC X1000 implements two instances of a 16550 compliant UART
controller that supports baud rates between 300 and 2764800. Hardware flow control is
also supported.
Note: Only one UART controller (UART0) provides MODEM pins.


Table 113. UART 0 Interface Signals

Direction/
Type
SIU0_RXD I UART 0 - Serial Input
SIU0_TXD O UART 0 - Serial Output
SIU0_RTS_B O UART 0 - MODEM Request to Send
SIU0_CTS_B I UART 0 - MODEM Clear to Send
SIU0_DCD_B I UART 0 - MODEM Data Carrier Detect
SIU0_DSR_B I UART 0 - MODEM Data Set Ready
SIU0_DTR_B O UART 0 - MODEM Data Terminal Ready
SIU0_RI_B I UART 0 - MODEM Ring Indicator
Table 114. UART 1 Interface Signals

Direction/
Type
SIU1_RXD I UART 1 - Serial Input
SIU1_TXD O UART 1 -Serial Output
SIU1_RTS_B O UART 1 - MODEM Request to Send
SIU1_CTS_B I UART 1 - MODEM Clear to Send

Intel® Quark™ SoC X1000—High Speed UART
18.2 Features
18.2.1 UART Function

The UART transmits and receives data in bit frames as shown in Figure 37.
• Each data frame is between 7 and 12 bits long, depending on the size of data
programmed and if parity and stop bits are enabled.
• The frame begins with a start bit that is represented by a high-to-low transition.
• Next, 5 to 8 bits of data are transmitted, beginning with the Least Significant Bit
(LSB). An optional parity bit follows, which is set if even parity is enabled and an
odd number of ones exist within the data byte; or, if odd parity is enabled and the
data byte contains an even number of ones.
• The data frame ends with one, one and a half, or two stop bits (as programmed by
users), which is represented by one or two successive bit periods of a logic one.
Figure 37. UART Data Transfer Flow
Start Data Data Data Data Data Data Data Data Parity Stop Stop
Bit [0] [1] [2] [3] [4] [5] [6] [7] Bit Bit 1 Bit 2
TXD or RXD pin
LSB MSB
Shaded bits are optional that users can program.
18.2.2 Baud Rate Generator

The baud rates for the UARTs are generated from the base frequency (Fbase) indicated
in Table 115 by programming the DLH and DLL registers as divisor. The hexadecimal
value of the divisor is (IER_DLH[7:0]<<8) | RBR_THR_DLL[7:0].
The output baud rate is equal to the base frequency divided by sixteen times the value
of the divisor, as follows: baud rate = (Fbase) / (16 * divisor).
Table 115. Baud Rates Achievable with Different DLAB Settings
DLH,DLL Divisor
DLH,DLL Divisor Baud Rate
Hexadecimal
Fbase: 44236800 Hz
1 0001 2764800
3 0003 921600
6 0006 460800
9 0009 307200
12 000C 230400
15 000F 184320
18 0012 153600
24 0018 115200
48 0030 57600
72 0048 38400

Table 115. Baud Rates Achievable with Different DLAB Settings

DLH,DLL Divisor
DLH,DLL Divisor Baud Rate
Hexadecimal
144 0090 19200

288 0120 9600
384 0180 7200
576 0240 4800
768 0300 3600
1152 0480 2400
1536 0600 1800
2304 0900 1200
4608 1200 600
9216 2400 300
18.3 Usage
Each UART has a Transmit FIFO and a Receive FIFO and each FIFO holds 16 bytes of
data. There are three separate methods for moving data into and out of the FIFOs:
DMA, Interrupts, and Polling.
18.3.1 DMA Mode Operation

Each UART has an associated DMA Controller (DMAC) that is enabled by setting
DMA_CFG_REG.DMA_EN to 1. Two DMA channels are used — one for transmit data and
one for receive data.
A hardware interface between the UART and the DMA is used to signal when data can
be read from the Receive FIFO and to signal when the Transmit FIFO is either empty or
has reached a programmed threshold level. This interface gives the DMA all
responsibility for the transfer of data and it must be programmed accordingly. Using the
UART to set the DMA mode via IIR_FCR[3] has no effect. An interrupt is generated
upon the completion of a DMA transfer.
NOTE: RX data can overrun or be corrupted when operating the UART DMA with
baud rate > 115200. It is recommended to disable the DMA for high baud rate or high
UART bandwidth operations.
18.3.1.1 Receiver DMA

The DMA Controller uses Channel 0 to transfer data from the UART to Host Memory.
The UART requests a DMA transfer to memory under the following condition:
• When the Receiver FIFO is at or above programmed trigger level in FIFO mode
To transfer data from the UART, the source transfer width (CTL0_L.SRC_TR_WIDTH) is
set to 8-bits to match the size of the FIFO entries and the destination transfer width
(CTL0_L.DST_TR_WIDTH) is set to 32-bits.
The receive watermark level (IIR_FCR[7:6]) should be set such that DMA requests to
transfer data to memory are made often enough for the Receiver FIFO to accept serial
transfers continuously. This will prevent the Receiver FIFO from overflowing.
To prevent Receiver FIFO underflow, the source burst length must be set such that the
FIFO can be emptied, but not underflowed, at the completion of the burst transaction.
For optimal operation, CTL0_L.SRC_MSIZE should be set at the receive watermark
level; that is:

• CTL0_L.SRC_MSIZE = decoded level of IIR_FCR[7:6]
18.3.1.2 Transmitter DMA

The DMA Controller uses Channel 1 to transfer data from Host Memory to the UART.
The UART requests a DMA transfer from memory under the following conditions:
• When the transmitter FIFO is empty in FIFO mode with Programmable THRE
interrupt mode disabled
• When the transmitter FIFO is at, or below the programmed threshold with FIFO and
Programmable THRE interrupt mode enabled
To transfer data to the UART, the destination transfer width (CTL1_L.DST_TR_WIDTH)

is set to 8-bits to match the size of the FIFO entries and the source transfer width
(CTL1_L.SRC_TR_WIDTH) is set to 32-bits.
The transmit watermark level (IIR_FCR[5:4) should be set such that DMA requests to
transfer data from memory are made often enough for the Transmitter FIFO to be able
to perform serial transfers continuously. This will prevent the Transmitter FIFO from
underflowing.
The Transmitter FIFO can overflow if CTL1_L.DEST_MSIZE is programmed to a value

greater than the transmit watermark level as there may not be not enough space in the
Transmitter FIFO to service the destination burst request. To avoid overflow,
CTL1_L.DEST_MSIZE should be programmed as follows:
• CTL1_L.DEST_MSIZE <= UART FIFO DEPTH − decoded level of IIR_FCR[5:4]
For optimal operation, CTL1_L.DEST_MSIZE should be set at the FIFO level that
triggers a transmit DMA request; that is:
• CTL1_L.DEST_MSIZE = UART FIFO DEPTH − decoded level of IIR_FCR[5:4]
18.3.2 FIFO Interrupt-Mode Operation
18.3.2.1 Receiver Interrupt

When the Receive FIFO and receiver interrupts are enabled (IIR_FCR[0]=1 and
IER_DLH[0]=1), receiver interrupts occur as follows:
• The Receive Data Available Interrupt is asserted when the FIFO has reached its
programmed trigger level. The interrupt is cleared when the FIFO drops below the
programmed trigger level.
• The IIR Receive Data Available indication also occurs when the FIFO trigger level is
reached, and like the interrupt, the bits are cleared when the FIFO drops below the
trigger level.
• The Data Ready bit (LSR.DR) is set to 1 as soon as a character is transferred from
the shift register to the Receive FIFO. This bit is reset to 0 when the FIFO is empty.
18.3.2.2 Transmitter Interrupt

When the transmitter FIFO and transmitter interrupt are enabled (IIR_FCR[0]=1,
IER_DLH[1]=1), transmit interrupts occur as follows:
• When Programmable THRE Interrupt Mode is disabled (IER_DLH[7] set to 0), the
transmitter interrupt occurs when the transmitter FIFO is empty. It is cleared when
the transmitter FIFO is no longer empty or the interrupt identification register (IIR)
is read.
• When Programmable THRE Interrupt Mode is enabled (IER_DLH[7] set to 1), the
transmitter interrupt occurs when the number of entries in the transmitter FIFO is

at or below a programmed transmitter FIFO empty threshold level (IIR_FCR[5:4]).

It is cleared when the number of entries in the transmitter FIFO is greater than the
programmed transmitter FIFO empty threshold.
Users could cause the UART Transmit FIFO to overflow when too many characters are
written. FIFO underflow does not cause an error as the UART waits for the Transmit
FIFO to be serviced.
18.3.3 FIFO Polled-Mode Operation

With the FIFOs enabled (IIR_FCR[0] set to 1), clearing IER_DLH[7] and IER_DLH[3:0]
puts the serial port in the FIFO Polled Operation mode. Because the receiver and the
transmitter are controlled separately, either one or both can be in Polled Operation
mode. In this mode, software checks Receiver and Transmitter status using the Line
Status Register (LSR). The processor polls the following bits for Receive and Transmit
Data Service.
18.3.3.1 Receive Data Service

The processor checks the Data Ready bit (LSR.DR) which is set when 1 or more bytes
remains in the Receive FIFO or Receive Buffer Register (RBR_THR_DLL).
18.3.3.2 Transmit Data Service

The processor checks transmit data request LSR.THRE bit, which is set when the
transmitter needs data.
The processor can also check transmitter empty LSR.TEMT, which is set when the
Transmit FIFO or Holding register is empty.
18.3.4 Autoflow Control

Autoflow Control uses Clear-to-Send (nCTS) and Request-to-Send (nRTS) signals to
automatically control the flow of data between the UART and external modem. When
autoflow is enabled, the remote device is not allowed to send data unless the UART
asserts nRTS low. If the UART de-asserts nRTS while the remote device is sending data,
the remote device is allowed to send one additional byte after nRTS is de-asserted. An
overflow could occur if the remote device violates this rule. Likewise, the UART is not
allowed to transmit data unless the remote device asserts nCTS low. This feature
increases system efficiency and eliminates the possibility of a Receive FIFO Overflow
error due to long interrupt latency.
Autoflow mode can be used in two ways: Full autoflow, automating both nCTS and
nRTS, and half autoflow, automating only nCTS. Full Autoflow is enabled by writing a 1
to bits 1 and 5 of the Modem Control Register (MCR). Auto-nCTS-Only mode is enabled
by writing a 1 to bit 5 and a 0 to bit 1 of the MCR register.
NOTE: It is recommended to turn on the autoflow control at all times to prevent data
corruption or overrun issue.
18.3.4.1 RTS (UART Output)

When in Full-autoflow mode, nRTS is asserted when the UART FIFO is ready to receive
data from the remote transmitter. This occurs when the amount of data in the Receive
FIFO is below the programmable threshold value. When the amount of data in the
Receive FIFO reaches the programmable threshold, nRTS is de-asserted. It will be
asserted once again when enough bytes are removed from the FIFO to lower the data
level below the threshold.

18.3.4.2 CTS (UART Input)

When in Full or Half-autoflow mode, nCTS is asserted by the remote receiver when the
receiver is ready to receive data from the UART. The UART checks nCTS before sending
the next byte of data and will not transmit the byte until nCTS is low. If nCTS goes high
while the transfer of a byte is in progress, the transmitter completes this byte.
18.4 Register Map

Figure 38. HSUART Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI
CAM
(I/O)
Bus 0
PCI
ECAM HSUART
(Mem) PCI Memory
Headers Space
D:20,F:1,F5
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
D:21
SPI1 F:1 BAR1

DMAC
I2C*/GPIOF:2 BAR0 Mem
Registers
Legacy Bridge
D:31, F:0 HSUART
SDIO/eMMC F:0 Mem
Registers
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7


Registers listed are for Function 1 (HS-UART0). Function 5 (HS-UART1) contains the
same registers. Differences between the HS-UARTs are noted in individual registers.

Default
Value

86h 86h 00h
page 681

Access Method


(Size: 16 bits)
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value
Bit Default &
Description
Range Access
8086h
RO

Access Method
(Size: 16 bits)
Default: 0936h
15 12 8 4 0
0 0 0 0 1 0 0 1 0 0 1 1 0 1 1 0
value
Bit Default &

Description
Range Access
0936h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IntrDis
SERREn
MasEn
MEMen
RSVD0
RSVD
RSVD
RSVD
Bit Default &

Description
Range Access
0h
RO

10
0h
RO

Bit Default &

Description
Range Access
8
00h
RO

2
RW requests.
1
0h
RO

Access Method
(Size: 16 bits)
Default: 0010h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
capable_66Mhz
RSVD0
SigSysErr
DEVSEL
FastB2B
hasCapList
RSVD1
RSVD
RSVD
RSVD
RcdMasAb
IntrStatus
Bit Default &
Description
Range Access
0h
RO
14
13
0h
RO
0b
RO
0h
RO
0b
RO
0h
RO
0b
RO

4
RO register sets.

Bit Default &

Description
Range Access
RO used
0h
RO

Access Method
(Size: 32 bits)
Default: 07000210h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0
rev_id
classCode
subClassCode
progIntf
Bit Default &
Description
Range Access
31: 24
RO performs.
23: 16
15: 8
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0


Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
cfgHdrFormat
multiFnDev
Bit Default &

Description
Range Access

7
RO function device

6: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
BIST_capable
comp_code
start_bist
RSVD

Bit Default &

Description
Range Access

7
0h
RO
0h
RO

3: 0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
prefetchable
memType
address
RSVD
isIO
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder

Access Method
(Size: 32 bits)
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD
prefetchable
memType
isIO
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

30h
Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ROM_base_addr
AddrDecodeEn
RSVD
Bit Default &

Description
Range Access

31: 11
000h
RO

RW also set to 1

Access Method


(Size: 32 bits)
Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
RSVD0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 8 bits) INTR_LINE: [B:0, D:20, F:1] + 3Ch
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
vector information.

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
Bit Default & value Description

Range Access
0h
RO

Access Method
(Size: 8 bits) MAX_LAT: [B:0, D:20, F:1] + 3Fh
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 01h

7 4 0
0 0 0 0 0 0 0 1
value
Bit Default &
Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 16 bits)
Default: 4803h
15 12 8 4 0
0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1
D2_support
D1_support
PME_support
aux_curr
DSI
RSVD
PME_clock
version
Bit Default &

Description
Range Access
15: 11
0h
RO
0h
RO

Bit Default &

Description
Range Access
0h
RO
0h
RO
0h
RO
2: 0

84h
Access Method
(Size: 16 bits)
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
Data_scale
no_soft_reset
power_state
Data_select
PME_en
RSVD
RSVD
PME_status
Bit Default &

Description
Range Access
15
0h
RO
0h
RO
8
0h
RO

3
0h
RO
1: 0


Offset 86h
Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access

7: 0
RO to 0.

87h
Access Method
(Size: 8 bits) DATA_REGISTER: [B:0, D:20, F:1] + 87h
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits) MSI_CAP_ID: [B:0, D:20, F:1] + A0h
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value
Bit Default &

Description
Range Access
7: 0


Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
7: 0
RO in the chain

Access Method
(Size: 16 bits) MESSAGE_CTRL: [B:0, D:20, F:1] + A2h
Default: 0100h
15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
perVecMskCap
multiMsgEn
bit64Cap
multiMsgCap
MSIEnable
RSVD0
Bit Default &

Description
Range Access
0h
RO

8
7

6: 4

3: 1
RW MSI.

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD0
Bit Default &
Description
Range Access
31: 2
0h
RO

Access Method
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData
Bit Default &

Description
Range Access
15: 0
changed by hardware

Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MSIMask
RSVD0

Bit Default &

Description
Range Access
0h
RO
0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
RSVD0
Bit Default &
Description
Range Access
0h
RO
0h
RO
18.6.1 UART Registers

Offset Default
Start Value
“Receive Buffer / Transmit Holding / Divisor Latch Low (RBR_THR_DLL)—Offset 0h” on

0h 3h 00000000h
page 685
4h 7h “Interrupt Enable / Divisor Latch High (IER_DLH)—Offset 4h” on page 685 00000000h
8h Bh “Interrupt Identification / FIFO Control (IIR_FCR)—Offset 8h” on page 686 00000001h
Ch Fh “Line Control (LCR)—Offset Ch” on page 688 00000000h
10h 13h “MODEM Control (MCR)—Offset 10h” on page 688 00000000h
14h 17h “Line Status (LSR)—Offset 14h” on page 689 00000060h
18h 1Bh “MODEM Status (MSR)—Offset 18h” on page 691 00000000h
1Ch 1Fh “Scratchpad (SCR)—Offset 1Ch” on page 692 00000000h
7Ch 7Fh “UART Status (USR)—Offset 7Ch” on page 693 00000000h
A4h A7h “Halt Transmission (HTX)—Offset A4h” on page 693 00000000h
A8h ABh “DMA Software Acknowledge (DMASA)—Offset A8h” on page 694 00000000h

18.6.1.1 Receive Buffer / Transmit Holding / Divisor Latch Low

(RBR_THR_DLL)—Offset 0h
Receive Buffer Register(RBR), reading this register when the DLAB bit (LCR[7]) is zero;
Transmit Holding Register (THR), writing to this register when the DLAB is zero; Divisor
Latch Low (DLL), when DLAB bit is one
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
FIELD
Bit Default &
Range Access
0b
RO
Receive Buffer / Transmit Holding / Divisor Latch Low (FIELD): Different UART
registers are accessed depending on read/write transfer type and Line control Register
(LCR) DLAB bit value.
RBR, Receive Buffer Register

- Access - Read AND DLAB (LCR[7]) =0
Data byte received on the serial input port. Valid only if the Data Ready (DR) bit in the
LSR is set. If FIFOs are disabled (FCR[0] set to zero), the data must be read before the
next data arrives, otherwise it is overwritten, resulting in an over-run error. If FIFOs are
enabled (FCR[0] set to one), this register accesses the head of the receive FIFO. If the
receive FIFO is full and this register is not read before the next data character arrives,
then the data already in the FIFO is preserved, but any incoming data are lost and an
over-run error occurs.
00h THR, Transmit Holding Register.

7:0 - Access - Write AND DLAB (LCR[7]) =0
RW Data to be transmitted on the serial output port. Data should only be written to the THR
when the THR Empty (THRE) bit (LSR[5]) is set. If FIFOs are disabled (FCR[0] = 0) and
THRE is set, writing a single character to the THR clears the THRE. Any additional writes
to the THR before the THRE is set again causes the THR data to be overwritten. If FIFOs
are enabled (FCR[0] = 1) and THRE is set, a total of 16 characters (FIFO Depth) can be
written to the THR before the FIFO is full. Any attempt to write data when the FIFO is
full results in the write data being lost.
DLL, Lower part of the Divisor Latch.

- Access - Read/Write AND DLAB (LCR[7]) =1
Lower 8 bits of a 16-bit, read/write, Divisor Latch register that contains the baud rate
divisor for the UART. This register may be accessed only when the DLAB bit (LCR[7]) is
set. Note that with the Divisor Latch Registers (DLL and DLH) set to zero, the baud clock
is disabled and no serial communications occur. Also, once the DLL is set, at least 8 clock
cycles of sclk should be allowed to pass before transmitting or receiving data.
18.6.1.2 Interrupt Enable / Divisor Latch High (IER_DLH)—Offset 4h

Interrupt Enable Register (IER), when the DLAB bit is zero; Divisor Latch High (DLH),
when the DLAB bit is one
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
FIELD
Bit Default &
Range Access
0b
RO
Interrupt Enable / Divisor Latch High (FIELD): Different UART registers are
accessed depending on the Line control Register (LCR) DLAB bit value.
IER, Interrupt Enable Register
Access - Read/write AND DLAB (LCR[7]) =0
Interrupt Enable Register: Each of the bits used has a different function ( 0 = disabled 1
= enabled):
0 - ERBFI, Enable Received Data Available Interrupt

1 - ETBEI, Enable Transmit Holding Register Empty Interrupt
2 - ELSI, Enable Receiver Line Status Interrupt
3 - EDSSI, Enable Modem Status Interrupt
4 - RESERVED
00h 5 - RESERVED
7:0 6 - RESERVED
RW 7 - PTIME, Programmable THRE Interrupt Mode Enable
DLH, Divisor Latch (High)

Access - Read/write AND DLAB (LCR[7]) =1
This register makes up the upper 8-bits of a 16-bit, read/write, Divisor Latch register
that contains the baud rate divisor for the UART. This register can be accessed only
when the DLAB bit (LCR[7]) is set.
NOTE: if the Divisor Latch Registers (DLL and DLH) are set to zero, the baud clock is
disabled and no serial communications occur. Also, once the Divisor Latch Registers (DLL
and DLH) are set, a mimimum of 50ns should be awaited before transmitting or
receiving data as this time is required in order for changes to take effect due to internal
synchronization processes.
18.6.1.3 Interrupt Identification / FIFO Control (IIR_FCR)—Offset 8h

Interrupt Identification Register (IIR) if reading; FIFO Control Register (FCR) if writing
Access Method
(Size: 32 bits)

Default: 00000001h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
RSV
FIELD

Bit Default &

Range Access
0b
RO
Interrupt Identification / FIFO Control (FIELD): Different UART registers are

accessed depending on read/write transfer type.
IIR, Interrupt Identification Register

- Access - Read only
- Reset - 0x01
Each of the bits used has a different function:
[3:0] - IID, Interrupt ID. This indicates the highest priority pending interrupt which can
be one of the following types:
0000 = modem status.
0001 = no interrupt pending.
0010 = THR empty.
0100 = received data available.
0110 = receiver line status.
0111 = busy detect. NEVER INDICATED
1100 = character timeout.
[5:3] - RESERVED read as zero
[7:6] - FIFOSE, FIFOs Enabled.

This is used to indicate whether the FIFO's are enabled or disabled:
00 = disabled
11 = enabled
FCR, FIFO Control Register

Access - Write only
Used to control the FIFOs. Different functions:
[0] - FIFOE, FIFO Enable.

Enables/disables the transmit (XMIT) and receive (RCVR) FIFO's. Whenever the value of
this bit is changed both the XMIT and RCVR controller portion of FIFO's will be reset.
01h
7:0
RW [1] - RFIFOR, RCVR FIFO Reset
Resets the control portion of the receive FIFO and treats the FIFO as empty. This will
also de-assert the DMA RX request and single signals. NOTE that this bit is 'self-clearing'
and it is not necessary to clear this bit.
[2] - XFIFOR, XMIT FIFO Reset

Resets the control portion of the transmit FIFO and treats the FIFO as empty. This will
also de-assert the DMA TX request and single signals. NOTE that this bit is 'self-clearing'
and it is not necessary to clear this bit.
[3] - DMAM, DMA Mode

Not used in UART due to the use of extra DMA handshake interface signals
[5:4] - TET, TX Empty Trigger

Used to select the empty threshold level at which the THRE Interrupts will be generated
when the mode is active. It also determines when the DMA support is requested if the
DMA is enabled.
The following trigger levels are supported:
00 = FIFO empty
01 = 2 characters in the FIFO
10 = FIFO 1/4 full
11 = FIFO 1/2 full
[7:6] - RT, RCVR Trigger

Used to select the trigger level in the receiver FIFO at which the Received Data Available
Interrupt will be generated. In auto flow control mode it is used to determine when the
rts_n signal will be de-asserted. It also determines when the DMA support is requested
if the DMA is enabled.
The following trigger levels are supported:
00 = 1 character in the FIFO
01 = FIFO 1/4 full
10 = FIFO 1/2 full
11 = FIFO 2 less than full

18.6.1.4 Line Control (LCR)—Offset Ch

Used to specify the format of the asynchronous data communication exchange.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
BC
STOP
DLAB
PEN
RSV
STICK_PAR
EPS
DLS
Bit Default &
Range Access
0b
RO
0h Divisor Latch Access Bit (DLAB): Used to enable reading and writing of the Divisor
7 Latch register (DLL and DLH) to set the baud rate of the UART. This bit must be cleared
RW after initial baud rate setup in order to access other registers.
Break Control Bit (BC): Used to cause a break condition to be transmitted to the
0h receiving device. If set to one the serial output is forced to the spacing (logic 0) state.
6
RW When not in Loopback Mode, as determined by MCR[4], the serial output line is forced
low until the Break bit is cleared.
0h
5 Reserved for future use (STICK_PAR): Reserved.
RW
0h Even Parity Select (EPS): Used to select between even and odd parity, when parity is
4 enabled (PEN set to one). If set to one, an even number of logic '1's is transmitted or
RW checked. If set to zero, an odd number of logic '1's is transmitted or checked.
Parity Enable (PEN): Used to enable and disable parity generation and detection in
0h transmitted and received serial character respectively.
3
RW 0 = parity disabled
1 = parity enabled
Number of stop bits (STOP): Used to select the number of stop bits per character
that the peripheral will transmit and receive. If set to zero, one stop bit is transmitted in
the serial data. If set to one and the data bits are set to 5 (LCR[1:0] set to zero) one
0h and a half stop bits are transmitted. Otherwise, two stop bits are transmitted.
2 NOTE: regardless of the number of stop bits selected the receiver will only check the
RW first stop bit.
0 = 1 stop bit
1 = 1.5 stop bits when DLS (LCR[1:0]) == 00
1 = 2 stop bits when DLS (LCR[1:0]) different from 00
Data Length Select (DLS): Used to select the number of data bits per character that
the peripheral will transmit and receive. The number of bit that may be selected are as
0h follows:
1:0 00 = 5 bits
RW 01 = 6 bits
10 = 7 bits
11 = 8 bits
18.6.1.5 MODEM Control (MCR)—Offset 10h

Controls the interface with the MODEM or data set (or a peripheral device emulating a
MODEM)

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
AFCE
RSV
LOOPBACK
RSVD
DTR
RTS
Bit Default &
Range Access
0b
RO
Auto Flow Control Enable (AFCE): When FIFOs are enabled and the Auto Flow
0h Control Enable (AFCE) bit is set, Auto Flow Control features are enabled
5
RW 0 = Auto Flow Control Mode disabled
1 = Auto Flow Control Mode enabled
LoopBack Bit (LOOPBACK): Used to put the UART into a diagnostic mode for test
purposes. Data on the serial out line is held high, while serial data output is looped back
0h to the serial in line, internally. In this mode all the interrupts are fully functional. Also, in
4
RW loopback mode, the modem control inputs (dsr_n, cts_n, ri_n, dcd_n) are disconnected
and the modem control outputs (dtr_n, rts_n, out1_n, out2_n) are looped back to the
inputs, internally.
0h
RO
Request to Send (RTS): Used to directly control the Request to Send (rts_n) output.
The Request To Send (rts_n) output is used to inform the modem or data set that the
UART is ready to exchange data. When Auto RTS Flow Control is not enabled (MCR[5]
set to zero), the rts_n signal is set low by programming MCR[1] (RTS) to a high. In Auto
0h Flow Control, MCR[5] set to one and FIFO's enable (FCR[0] set to one, the rts_n output
1
RW is controlled in the same way, but is also gated with the receiver FIFO threshold trigger
(rts_n is inactive high when above the threshold). The rts_n signal will be de-asserted
when MCR[1] is set low. Note that in Loopback mode (MCR[4] set to one), the rts_n
output is held inactive high while the value of this location is internally looped back to an
input.
Data Terminal Ready (DTR): Used to directly control the Data Terminal Ready (dtr_n)
output. The value written to this location is inverted and driven out on dtr_n, that is:
0 = dtr_n de-asserted (logic 1)
0h 1 = dtr_n asserted (logic 0)
0
RW The Data Terminal Ready output is used to inform the modem or data set that the UART
is ready to establish communications. Note that in Loopback mode (MCR[4] set to one),
the dtr_n output is held inactive high while the value of this location is internally looped
back to an input.
18.6.1.6 Line Status (LSR)—Offset 14h

Provides status information concerning the data transfer.
Access Method
(Size: 32 bits)


Default: 00000060h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0
DR
RSV
RFE
TEMT
THRE
FE
PE
OE
BI
Bit Default &
Range Access
0b
RO
Receiver FIFO Error bit (RFE): This bit is only relevant when FIFO are enabled
(FCR[0] set to one). This is used to indicate if there is at least one parity error, framing
0h error, or break indication in the FIFO. That is:
7
RO 0 = no error in RX FIFO 1 = error in RX FIFO
This bit is cleared when the LSR is read and the character with the error is at the top of
the receiver FIFO and there are no subsequent errors in the FIFO.
Transmitter Empty bit (TEMT): If FIFO are enabled (FCR[0] set to one), this bit is set
1h whenever the Transmitter Shift Register and the FIFO are both empty.
6
RO If FIFO are disabled, this bit is set whenever the Transmitter Holding Register and the
Transmitter Shift Register are both empty.
Transmit Holding Register Empty bit (THRE): If THRE mode is disabled (IER[7] set
to zero) and regardless of FIFO's being enabled or not, this bit indicates that the THR or
TX FIFO is empty. This bit is set whenever data is transferred from the THR or TX FIFO
1h to the transmitter shift register and no new data has been written to the THR or TX
5
RO FIFO. This also causes a THRE Interrupt to occur, if the THRE Interrupt is enabled.
If the THRE mode and FIFO are enabled (IER[7] and FCR[0] set to one), the
functionality is switched to indicate the transmitter FIFO is full, and no longer controls
THRE interrupts, which are then controlled by the FCR[5:4] threshold setting.
Break Interrupt bit (BI): Used to indicate the detection of a break sequence on the
serial input data. If in UART mode it is set whenever the serial input is held in a logic '0'
state for longer than the sum of start time + data bits + parity + stop bits.
0h A break condition on serial input causes one and only one character, consisting of all
4
RO zeros, to be received by the UART. In the FIFO mode, the character associated with the
break condition is carried through the FIFO and is revealed when the character is at the
top of the FIFO. Reading the LSR clears the BI bit. In the non-FIFO mode, the BI
indication occurs immediately and persists until the LSR is read.
Framing Error bit (FE): Used to indicate the occurrence of a framing error in the
receiver. A framing error occurs when the receiver does not detect a valid STOP bit in
the received data. In the FIFO mode, since the framing error is associated with a
character received, it is revealed when the character with the framing error is at the top
0h of the FIFO. When a framing error occurs the UART will try resynchronize. It does this by
3 assuming that the error was due to the start bit of the next character and then
RO continues receiving the other bit i.e. data, and/or parity and stop. It should be noted
that the Framing Error (FE) bit (LSR[3]) will be set if a break interrupt has occurred, as
indicated by Break Interrupt (BI) bit (LSR[4]).
0 = no framing error, 1 = framing error
Reading the LSR clears the FE bit.
Parity Error bit (PE): Used to indicate the occurrence of a parity error in the receiver if
the Parity Enable (PEN) bit (LCR[3]) is set. In the FIFO mode, since the parity error is
associated with a character received, it is revealed when the character with the parity
0h error arrives at the top of the FIFO. It should be noted that the Parity Error (PE) bit
2 (LSR[2]) will be set if a break interrupt has occurred, as indicated by Break Interrupt
RO (BI) bit (LSR[4]).
0 = no parity error, 1 = parity error
Reading the LSR clears the PE bit.

Bit Default &

Range Access
Overrun error bit (OE): Used to indicate the occurrence of an overrun error. This
occurs if a new data character was received before the previous data was read. In the
non-FIFO mode, the OE bit is set when a new character arrives in the receiver before
the previous character was read from the RBR. When this happens, the data in the RBR
0h is overwritten. In the FIFO mode, an overrun error occurs when the FIFO is full and a
1
RO new character arrives at the receiver. The data in the FIFO is retained and the data in
the receive shift register is lost.
0 = no overrun error, 1 = overrun error
Reading the LSR clears the OE bit.
Data Ready bit (DR): Used to indicate that the receiver contains at least one character
in the RBR or the receiver FIFO.
0h 0 = no data ready, 1 = data ready
0
RO
This bit is cleared when the RBR is read in the non-FIFO mode, or when the receiver
FIFO is empty, in the FIFO mode.
18.6.1.7 MODEM Status (MSR)—Offset 18h

Provides the current state of the control lines from the MODEM (or peripheral device)
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
DCD
DDCD
CTS
DCTS
RI
DSR
TERI
DDSR
Bit Default &
Range Access
0b
RO
Data Carrier Detect (DCD): Used to indicate the current state of the modem control
line dcd_n. That is this bit is the complement dcd_n. When the Data Carrier Detect input
0h (dcd_n) is asserted it is an indication that the carrier has been detected by the modem
7 or data set.
RO 0 = dcd_n input is de-asserted (logic 1)
1 = dcd_n input is asserted (logic 0)
In Loopback Mode (MCR[4] set to one), DCD is the same as MCR[3] (Out2).
Ring Indicator (RI): Used to indicate the current state of the modem control line ri_n.
That is this bit is the complement ri_n. When the Ring Indicator input (ri_n) is asserted
0h it is an indication that a telephone ringing signal has been received by the modem or
6
RO data set.
0 = ri_n input is de-asserted (logic 1) 1 = ri_n input is asserted (logic 0)
In Loopback Mode (MCR[4] set to one), RI is the same as MCR[2] (Out1).
Data Set Ready (DSR): Used to indicate the current state of the modem control line
dsr_n. That is this bit is the complement dsr_n. When the Data Set Ready input (dsr_n)
0h is asserted it is an indication that the modem or data set is ready to establish
5
RO communications with the UART.
0 = dsr_n input is de-asserted (logic 1) 1 = dsr_n input is asserted (logic 0)
In Loopback Mode (MCR[4] set to one), DSR is the same as MCR[0] (DTR).

Bit Default &

Range Access
Clear to Send (CTS): Used to indicate the current state of the modem control line
cts_n. That is, this bit is the complement cts_n. When the Clear to Send input (cts_n) is
0h asserted it is an indication that the modem or data set is ready to exchange data with
4
RO the UART
0 = cts_n input is de-asserted (logic 1) 1 = cts_n input is asserted (logic 0)
In Loopback Mode (MCR[4] set to one), CTS is the same as MCR[1] (RTS).
Delta Data Carrier Detect (DDCD): Used to indicate that the modem control line
dcd_n has changed since the last time the MSR was read. That is:
0 = no change on dcd_n since last read of MSR 1 = change on dcd_n since last read of
0h MSR
3
RO Reading the MSR clears the DDCD bit. In Loopback Mode (MCR[4] set to one), DDCD
reflects changes on MCR[3] (Out2). Note, if the DDCD bit is not set and the dcd_n signal
is asserted (low) and a reset occurs (software or otherwise), then the DDCD bit will get
set when the reset is removed if the dcd_n signal remains asserted.
Trailing Edge of Ring Indicator (TERI): Used to indicate that a change on the input
ri_n (from an active low, to an inactive high state) has occurred since the last time the
0h MSR was read. That is:
2
RO 0 = no change on ri_n since last read of MSR 1 = change on ri_n since last read of MSR
Reading the MSR clears the TERI bit. In Loopback Mode (MCR[4] set to one), TERI
reflects when MCR[2] (Out1) has changed state from a high to a low.
Delta Data Set Ready (DDSR): Used to indicate that the modem control line dsr_n
has changed since the last time the MSR was read. That is:
0 = no change on dsr_n since last read of MSR 1 = change on dsr_n since last read of
0h MSR
1
RO Reading the MSR clears the DDSR bit. In Loopback Mode (MCR[4] set to one), DDSR
reflects changes on MCR[0] (DTR). Note, if the DDSR bit is not set and the dsr_n signal
is asserted (low) and a reset occurs (software or otherwise), then the DDSR bit will get
set when the reset is removed if the dsr_n signal remains asserted.
Delta Clear to Send (DCTS): used to indicate that the modem control line cts_n has
changed since the last time the MSR was read. That is:
0 = no change on cts_n since last read of MSR 1 = change on cts_n since last read of
0h MSR
0
RO Reading the MSR clears the DCTS bit. In Loopback Mode (MCR[4] set to one), DCTS
reflects changes on MCR[1] (RTS). Note, if the DCTS bit is not set and the cts_n signal is
asserted (low) and a reset occurs (software or otherwise), then the DCTS bit will get set
when the reset is removed if the cts_n signal remains asserted.
18.6.1.8 Scratchpad (SCR)—Offset 1Ch

Used by the programmer to hold data temporarily
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SCR
RSV
Bit Default &

Range Access
0b
RO
0h Scratchpad Register (SCR): This register is for programmers to use as a temporary

7:0
RW storage space. It has no defined purpose in the UART.

18.6.1.9 UART Status (USR)—Offset 7Ch

Provides internal status information
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RFF
RSV0
TFNF
RFNE
TFE
Bit Default &
Range Access
0h
RO
Receive FIFO Full (RFF): Used to indicate that the receive FIFO is completely full.
0h That is:
4
RO 0 = Receive FIFO not full 1 = Receive FIFO Full
This bit is cleared when the RX FIFO is no longer full.
Receive FIFO Not Empty (RFNE): Used to indicate that the receive FIFO contains one
0h or more entries.
3
RO 0 = Receive FIFO is empty 1 = Receive FIFO is not empty
This bit is cleared when the RX FIFO is empty.
Transmit FIFO Empty (TFE): Used to indicate that the transmit FIFO is completely
0h empty.
2
RO 0 = Transmit FIFO is not empty 1 = Transmit FIFO is empty
This bit is cleared when the TX FIFO is no longer empty.
0h Transmit FIFO Not Full (TFNF): Used to indicate that the transmit FIFO in not full.
1 0 = Transmit FIFO is full 1 = Transmit FIFO is not full
RO This bit is cleared when the TX FIFO is full.
0h
RO
18.6.1.10 Halt Transmission (HTX)—Offset A4h

Halt Transmission
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
HTX

Bit Default &

Range Access
0b
RO
Halt Transmission (HTX): Used to halt transmissions for testing, so that the transmit
0h FIFO can be filled by the master when FIFO's are enabled. Note, if FIFO's are not
0
RW enabled the setting of the halt TX register will have no effect on operation.
0 = Halt TX disabled 1 = Halt TX enabled
18.6.1.11 DMA Software Acknowledge (DMASA)—Offset A8h

DMA software acknowledge if a transfer needs to be terminated due to an error
condition.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DMASA
RSV
Bit Default &

Range Access
0b
RO
DMA Software Acknowledge (DMASA): Used to perform DMA software acknowledge

0h if a transfer needs to be terminated due to an error condition. For example, if the DMA
0 disables the channel, then the UART should clear its request. This will cause the DMA
RW operation to be concluded.
NOTE: this bit is 'self-clearing'. It is not necessary to clear it.
18.6.2 DMA Controller Registers

Offset Default
Start Value
0h 3h “Channel 0 Source Address (SAR0)—Offset 0h” on page 696 00000000h
8h Bh “Channel 0 Destination Address (DAR0)—Offset 8h” on page 696 00000000h
10h 13h “Channel 0 Linked List Pointer (LLP0)—Offset 10h” on page 697 00000000h
18h 1Bh “Channel 0 Control LOWER (CTL0_L)—Offset 18h” on page 697 00304801h
1Ch 1Fh “Channel 0 Control UPPER (CTL0_U)—Offset 1Ch” on page 699 00000002h
20h 23h “Channel 0 Source Status (SSTAT0)—Offset 20h” on page 700 00000000h
28h 2Bh “Channel 0 Destination Status (DSTAT0)—Offset 28h” on page 700 00000000h
30h 33h “Channel 0 Source Status Address (SSTATAR0)—Offset 30h” on page 701 00000000h


Offset Default
Start Value
38h 3Bh “Channel 0 Destination Status Address (DSTATAR0)—Offset 38h” on page 701 00000000h
40h 43h “Channel 0 Configuration LOWER (CFG0_L)—Offset 40h” on page 702 00000E00h
44h 47h “Channel 0 configuration UPPER (CFG0_U)—Offset 44h” on page 703 00000004h
48h 4Bh “Channel 0 Source Gather (SGR0)—Offset 48h” on page 704 00000000h
50h 53h “Channel 0 Destination Scatter (DSR0)—Offset 50h” on page 705 00000000h
58h 5Bh “Channel 1 Source Address (SAR1)—Offset 58h” on page 705 00000000h
60h 63h “Channel 1 Destination Address (DAR1)—Offset 60h” on page 706 00000000h
68h 6Bh “Channel 1 Linked List Pointer (LLP1)—Offset 68h” on page 706 00000000h
70h 73h “Channel 1 Control LOWER (CTL1_L)—Offset 70h” on page 707 00304801h
74h 77h “Channel 1 Control UPPER (CTL1_U)—Offset 74h” on page 709 00000002h
78h 7Bh “Channel 1 Source Status (SSTAT1)—Offset 78h” on page 710 00000000h
80h 83h “Channel 1 Destination Status (DSTAT1)—Offset 80h” on page 710 00000000h
88h 8Bh “Channel 1 Source Status Address (SSTATAR1)—Offset 88h” on page 711 00000000h
90h 93h “Channel 1 Destination Status Address (DSTATAR1)—Offset 90h” on page 711 00000000h
98h 9Bh “Channel 1 Configuration LOWER (CFG1_L)—Offset 98h” on page 712 00000E20h
9Ch 9Fh “Channel 1 configuration UPPER (CFG1_U)—Offset 9Ch” on page 713 00000004h
A0h A3h “Channel 1 Source Gather (SGR1)—Offset A0h” on page 714 00000000h
A8h ABh “Channel 1 Destination Scatter (DSR1)—Offset A8h” on page 714 00000000h
2C0h 2C3h “Raw Status for IntTfr Interrupt (RAW_TFR)—Offset 2C0h” on page 715 00000000h
2C8h 2CBh “Raw Status for IntBlock Interrupt (RAW_BLOCK)—Offset 2C8h” on page 715 00000000h
2D0h 2D3h “Raw Status for IntSrcTran Interrupt (RAW_SRC_TRAN)—Offset 2D0h” on page 716 00000000h
2D8h 2DBh “Raw Status for IntDstTran Interrupt (RAW_DST_TRAN)—Offset 2D8h” on page 716 00000000h
2E0h 2E3h “Raw Status for IntErr Interrupt (RAW_ERR)—Offset 2E0h” on page 717 00000000h
2E8h 2EBh “Status for IntTfr Interrupt (STATUS_TFR)—Offset 2E8h” on page 717 00000000h
2F0h 2F3h “Status for IntBlock Interrupt (STATUS_BLOCK)—Offset 2F0h” on page 718 00000000h
2F8h 2FBh “Status for IntSrcTran Interrupt (STATUS_SRC_TRAN)—Offset 2F8h” on page 718 00000000h
300h 303h “Status for IntDstTran Interrupt (STATUS_DST_TRAN)—Offset 300h” on page 719 00000000h
308h 30Bh “Status for IntErr Interrupt (STATUS_ERR)—Offset 308h” on page 719 00000000h
310h 313h “Mask for IntTfr Interrupt (MASK_TFR)—Offset 310h” on page 720 00000000h
318h 31Bh “Mask for IntBlock Interrupt (MASK_BLOCK)—Offset 318h” on page 720 00000000h
320h 323h “Mask for IntSrcTran Interrupt (MASK_SRC_TRAN)—Offset 320h” on page 721 00000000h
328h 32Bh “Mask for IntDstTran Interrupt (MASK_DST_TRAN)—Offset 328h” on page 722 00000000h
330h 333h “Mask for IntErr Interrupt (MASK_ERR)—Offset 330h” on page 722 00000000h
338h 33Bh “Clear for IntTfr Interrupt (CLEAR_TFR)—Offset 338h” on page 723 00000000h
340h 343h “Clear for IntBlock Interrupt (CLEAR_BLOCK)—Offset 340h” on page 723 00000000h
348h 34Bh “Clear for IntSrcTran Interrupt (CLEAR_SRC_TRAN)—Offset 348h” on page 724 00000000h
350h 353h “Clear for IntDstTran Interrupt (CLEAR_DST_TRAN)—Offset 350h” on page 724 00000000h
358h 35Bh “Clear for IntErr Interrupt (CLEAR_ERR)—Offset 358h” on page 725 00000000h
360h 363h “Combined Interrupt Status (STATUS_INT)—Offset 360h” on page 725 00000000h
368h 36Bh “Source Software Transaction Request (REQ_SRC_REG)—Offset 368h” on page 726 00000000h


Offset Default
Start Value
“Destination Software Transaction Request register (REQ_DST_REG)—Offset 370h” on

370h 373h 00000000h
page 726
378h 37Bh “Source Single Transaction Request (SGL_REQ_SRC_REG)—Offset 378h” on page 727 00000000h
“Destination Single Software Transaction Request (SGL_REQ_DST_REG)—Offset 380h”

380h 383h 00000000h
on page 728
388h 38Bh “Source Last Transaction Request (LST_SRC_REG)—Offset 388h” on page 728 00000000h
390h 393h “Destination Single Transaction Request (LST_DST_REG)—Offset 390h” on page 729 00000000h
398h 39Bh “DMA Configuration (DMA_CFG_REG)—Offset 398h” on page 729 00000000h
3A0h 3A3h “Channel Enable (CH_EN_REG)—Offset 3A0h” on page 730 00000000h
18.6.2.1 Channel 0 Source Address (SAR0)—Offset 0h

Source Address of DMA transfer The starting source address is programmed by
software before the DMA channel is enabled, or by an LLI update before the start of the
DMA transfer. While the DMA transfer is in progress, this register is updated to reflect
the source address of the current transfer.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SAR
Bit Default &

Range Access
Current Source Address of DMA transfer (SAR): Updated after each source transfer.
The SINC field in the CTL0_L register determines whether the address increments,
0b decrements, or is left unchanged on every source transfer throughout the block transfer.
31:0
RW
NOTE: Channel 0 is dedicated to the UART controller Received data. Source Address is
the UART controller RBR_THR_DLL register address: 0xFFFF_F000
18.6.2.2 Channel 0 Destination Address (DAR0)—Offset 8h

Destination address of DMA transfer The starting destination address is programmed by
the destination address of the current transfer.
Access Method

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DAR
Bit Default &
Range Access
Current Destination address of DMA transfer (DAR): Updated after each

destination transfer. The DINC field in the CTL0_L register determines whether the
0b address increments, decrements, or is left unchanged on every destination transfer
31:0 throughout the block transfer.
RW
NOTE: Channel 0 is dedicated to the UART controller Received data. Destination Address
is a memory address
18.6.2.3 Channel 0 Linked List Pointer (LLP0)—Offset 10h

Program this register to point to the first Linked List Item (LLI) in memory prior to
enabling the channel if block chaining is enabled
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
LOC
Bit Default &

Range Access
Starting Address In Memory (LOC): Starting Address In Memory of next LLI if block
chaining is enabled.
0b Note that the two LSBs of the starting address are not stored because the address is
31:2
RW assumed to be aligned to a 32-bit boundary.
LLI accesses are always 32-bit accesses (Hsize = 2) aligned to 32-bit boundaries and
cannot be changed or programmed to anything other than 32-bit.
0b
RO
18.6.2.4 Channel 0 Control LOWER (CTL0_L)—Offset 18h

Contains fields that control the DMA transfer. Part of the block descriptor (linked list
item - LLI) when block chaining is enabled. It can be varied on a block-by-block basis
within a DMA transfer when block chaining is enabled.
Access Method

Default: 00304801h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 1
SMS
DMS
SRC_MSIZE
DEST_MSIZE
DST_TR_WIDTH
SRC_TR_WIDTH
RSV1
TT_FC
RSV0
SINC
DINC
DST_GATHER_EN
SRC_GATHER_EN
INT_EN
Bit Default &
Range Access
0b
RO
0b Source AMBA Layer (SMS): Hardcoded the Master interface attached to the source of
26:25
RO channel 0.
0b Destination AMBA Layer (DMS): Hardcoded the Master interface attached to the
24:23
RO destination of channel 0
Transfer Type and Flow Control (TT_FC): The following transfer types are
supported:
Code - Type - Flow Controller
--------------------------------------------------
000 - Memory to Memory - DMAC
011b 001 - Memory to Peripheral - DMAC
22:20
RW 010 - Peripheral to Memory - DMAC
011 - Peripheral to Peripheral - DMAC
100 - Peripheral to Memory - Peripheral
101 - Peripheral to Peripheral - Source Peripheral
110 - Memory to Peripheral - Peripheral
111 - Peripheral to Peripheral - Destination Peripheral
0b
RO
Destination scatter enable (DST_GATHER_EN): 0 = Scatter disabled

0b 1 = Scatter enabled
18
RW Scatter on the destination side is applicable only when the CTL0_L.DINC bit indicates an
incrementing or decrementing address control.
Source gather enable (SRC_GATHER_EN): 0 = Gather disabled

0b 1 = Gather enabled
17
RW Gather on the source side is applicable only when the CTL0_L.SINC bit indicates an
Source Burst Transaction Length (SRC_MSIZE): Number of data items, each of

width CTL0_L.SRC_TR_WIDTH, to be read from the source every time a source burst
transaction request is made from either the corresponding hardware or software
handshaking interface. Value - Size(#TR_WITDH)
--------------------
001b 000 - 1
16:14 001 - 4
RW 010 - 8
011 - 16
100 - 32
101 - 64
110 - 128
111 - 256

Bit Default &

Range Access
Destination Burst Transaction Length (DEST_MSIZE): Destination Burst

Transaction Length. Number of data items, each of width CTL0_L.DST_TR_WIDTH, to be
written to the destination every time a destination burst transaction request is made
from either the corresponding hardware or software handshaking interface. Value -
Size(#TR_WITDH)
--------------------
001b 000 - 1
13:11
RW 001 - 4
010 - 8
011 - 16
100 - 32
101 - 64
110 - 128
111 - 256
Source Address Increment (SINC): Indicates whether to increment or decrement

the source address on every source transfer. If the device is fetching data from a source
peripheral FIFO with a fixed address,
0b then set this field to No change.
10:9 00 = Increment
RW 01 = Decrement
1x = No change
NOTE: Incrementing or decrementing is done for alignment to the next SRC_TR_WIDTH
boundary.
Destination Address Increment (DINC): Indicates whether to increment or

decrement the destination address on every destination transfer. If your device is
writing data to a destination peripheral FIFO with a
0b fixed address, then set this field to No change
8:7 00 = Increment
RW 01 = Decrement
1x = No change
NOTE: Incrementing or decrementing is done for alignment to the next DST_TR_WIDTH
boundary.
Source transfer width (SRC_TR_WIDTH): Decoding for this field:

Value - Size(bits)
--------------------
000 - 8
0b 001 - 16
6:4
RW 010 - 32
011 - 64
100 - 128
101 - 256
11x - 256
Destination transfer width (DST_TR_WIDTH): Decoding for this field:

Value - Size(bits)
--------------------
000 - 8
0b 001 - 16
3:1
RW 010 - 32
011 - 64
100 - 128
101 - 256
11x - 256
1b Interrupt enable (INT_EN): If set, then all interrupt-generating sources are enabled.
0 Functions as a global mask bit for all interrupts for the channel. RAW interrupt registers
RW still assert if INT_EN = 0.
18.6.2.5 Channel 0 Control UPPER (CTL0_U)—Offset 1Ch

Contains fields that control the DMA transfer. It can be varied on a block-by-block basis
within a DMA transfer when block chaining is enabled. If status write-back is enabled,
the content is written to the control register location of the LLI in system memory at
the end of the block transfer.
Access Method


(Size: 32 bits)

Default: 00000002h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0
RSV
BLOCK_TS
Bit Default &
Range Access
0b
RO
Block length (BLOCK_TS): When the DMAC is the flow controller, the user writes this
002h field before the channel is enabled in order to indicate the block size. The number
11:0 programmed into BLOCK_TS indicates the total number of single transactions to
RW perform for every block transfer. A single transaction is mapped to a single AMBA beat.
Width: The width of the single transaction is determined by CTL0_L.SRC_TR_WIDTH.
18.6.2.6 Channel 0 Source Status (SSTAT0)—Offset 20h

This register is a temporary placeholder for the source status information on its way to
the SSTAT0 register location of the LLI. The source status information should be
retrieved by software from the SSTAT0 register location of the LLI, and not by a read of
this register over the DMAC slave interface.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SSTAT
Bit Default &

Range Access
0b Channel Source Status (SSTAT): Source status information retrieved by hardware

31:0
RW from the address pointed to by the contents of the SSTATAR0 register.
18.6.2.7 Channel 0 Destination Status (DSTAT0)—Offset 28h

This register is a temporary placeholder for the destination status information on its
way to the DSTAT0 register location of the LLI. The destination status information
should be retrieved by software from the DSTAT0 register location of the LLI and not by
a read of this register over the DMAC slave interface
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DSTAT
Bit Default &
Range Access
0b Channel Destination Status (DSTAT): Destination status information retrieved by

31:0
RW hardware from the address pointed to by the contents of the DSTATAR0 register
18.6.2.8 Channel 0 Source Status Address (SSTATAR0)—Offset 30h

After the completion of each block transfer, hardware can retrieve the source status
information from the address pointed to by the contents of this register
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SSTATAR
Bit Default &

Range Access
0b Channel Source Status Address (SSTATAR): Pointer from where hardware can fetch
31:0 the source status information, which is registered in the SSTAT0 register and written out
RW to the SSTAT0 register location of the LLI before the start of the next block.
18.6.2.9 Channel 0 Destination Status Address (DSTATAR0)—Offset 38h

After the completion of each block transfer, hardware can retrieve the destination
status information from the address pointed to by the contents of this register
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DSTATAR
Bit Default &
Range Access
0b Channel Destination Status Address (DSTATAR): Pointer from where hardware can
31:0 fetch the destination status information, which is registered in the DSTAT0 register and
RW written out to the DSTAT0 register location of the LLI before the start of the next block.
18.6.2.10 Channel 0 Configuration LOWER (CFG0_L)—Offset 40h

Contains fields that configure the DMA transfer. The channel configuration register
remains fixed for all blocks of a multi-block transfer. This register need to be
programmed before enable the channel.
Access Method

Default: 00000E00h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0
HS_SEL_DST
RELOAD_DST
FIFO_EMPTY
RELOAD_SRC
CH_SUSP
SRC_HS_POL
DST_HS_POL
HS_SEL_SRC
RSV2
RSV1
RSV0
CH_PRIOR
Bit Default &

Range Access
0b Reload destination enable (RELOAD_DST): The DARx register can be automatically

31 reloaded from its initial value at the end of every block for multi-block transfers. A new
RW block transfer is then initiated.
0b Reload source enable (RELOAD_SRC): The SAR register can be automatically

0b
RO
0b Source handshake polarity (SRC_HS_POL): 0 = Active high

19
RW 1 = Active low
0b Destination handshake polarity (DST_HS_POL): 0 = Active high

18
RW 1 = Active low
0b
RW
Source Handshake select (HS_SEL_SRC): Used to select which handshake interface

1b is active for source requests on this channel
11 0 = HW handshake. SW ones are ignored
RW 1 = SW handshake. HW ones are ignored
If source peripheral is memory this bit is ignored

Bit Default &

Range Access
Destination Handshake select (HS_SEL_DST): Used to select which handshake

1b interface is active for destination requests on this channel
If destination peripheral is memory this bit is ignored
Channel FIFO empty status (FIFO_EMPTY): Indicates if there is data left in the
1b channel FIFO. Can be used in conjunction with CH_SUSP to cleanly disable a channel.
9
RO 1 = Channel FIFO empty
0 = Channel FIFO not empty
Channel Suspend control (CH_SUSP): Suspends all DMA data transfers from the
source until this bit is cleared. There is no guarantee that the current transaction will
0b complete. Can also be used in conjunction
8 with FIFO_EMPTY to cleanly disable a channel without
RW losing any data.
0 = Not suspended.
1 = Suspend DMA transfer from the source.
0b Channel Priority (CH_PRIOR): Priority value equal to 7 is the highest priority, and 0
7:5 is the lowest. This field must be programmed within the following range: 0: 1
RW A programmed value outside this range will cause erroneous behavior.
0b
RO
18.6.2.11 Channel 0 configuration UPPER (CFG0_U)—Offset 44h

Access Method
(Size: 32 bits)

Default: 00000004h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
RSV5
DEST_PER
RSV4
SS_UPD_EN
DS_UPD_EN
FIFO_MODE
SRC_PER
PROTCTL
FCMODE
Bit Default &

Range Access
0b
RO
Destination hardware interface (DEST_PER): Assigns a hardware handshaking

interface (0-1) to the channel destination if the CFG0_L.HS_SEL_DST field is 0;
0b otherwise, this field is ignored. The channel can then communicate with the destination
11
RW peripheral connected to that interface through the assigned hardware handshaking
interface. NOTE: For correct DMA operation, only one peripheral (source or destination)
should be assigned to the same handshaking interface.
0b
RO

Bit Default &

Range Access
Source hardware interface (SRC_PER): Assigns a hardware handshaking interface

(0-1) to the channel source if the CFG0_L.HS_SEL_SRC field is 0; otherwise, this field is
0b ignored. The channel can then communicate with the source peripheral connected to
7
RW that interface through the assigned hardware handshaking interface. NOTE: For correct
DMA operation, only one peripheral (source or destination) should be assigned to the
same handshaking interface.
0b Source Status Update Enable (SS_UPD_EN): Source status information is fetched

6 only from the location pointed to by the SSTATAR0 register, stored in the SSTAT0
RW register and written out to the SSTAT0 location of the LLI if this field is high
0b Destination Status Update Enable (DS_UPD_EN): Destination status information is

5 fetched only from the location pointed to by the DSTATAR0 register, stored in the
RW DSTAT0 register and written out to the DSTAT0 location of the LLI if this field is high
AHB bus protocol bus control (PROTCTL): Protection Control bits used to drive the
AHB HPROT[3:1] bus.
001b The AMBA Specification recommends that the default value of HPROT indicates a non-
4:2 cached, non-buffered, privileged data access. The reset value is used to indicate such an
RW access. HPROT[0] is tied high because all transfers are data accesses, as there are no
opcode fetches. There is a one-to-one mapping of these register bits to the HPROT[3:1]
master interface signals.
Channel FIFO mode control (FIFO_MODE): Determines how much space or data
needs to be available in the FIFO before a burst transaction request is serviced.
0b 0 = Space/data available for single AHB transfer of the specified transfer width.
1
RW 1 = Data available is greater than or equal to half the FIFO depth for destination
transfers and space available is greater than half the fifo depth for source transfers. The
exceptions are at the end of a burst transaction request or at the end of a block transfer.
Channel flow control mode (FCMODE): Determines when source transaction

requests are serviced when the Destination Peripheral is the flow controller.
0 = Source transaction requests are serviced when they occur. Data pre-fetching is
0b enabled.
0 1 = Source transaction requests are not serviced until a
RW destination transaction request occurs.
In this mode, the amount of data transferred from the source is limited so that it is
guaranteed to be transferred to the destination prior to block termination by the
destination. Data pre-fetching is disabled.
18.6.2.12 Channel 0 Source Gather (SGR0)—Offset 48h

The CTL0_L.SINC field controls whether the address increments or decrements. For a
fixed-address control, then the address remains constant throughout the transfer and
this register is ignored.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SGC
SGI
RSV
Bit Default &

Range Access
0b
RO

Bit Default &

Range Access
Source Gather Count (SGC): Source contiguous transfer count between successive
0b gather boundaries. Specifies the number of contiguous source transfers of
24:20
RW CTL0_L.SRC_TR_WIDTH between successive gather intervals. This is defined as a
gather boundary
0b Source Gather Interval (SGI): Specifies the source address increment/decrement in

19:0 multiples of CTL0_L.SRC_TR_WIDTH on a gather boundary when gather mode is
RW enabled for the source transfer.
18.6.2.13 Channel 0 Destination Scatter (DSR0)—Offset 50h

The CTL0_L.DINC field controls whether the address increments or decrements. For a
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DSC
DSI
RSV
Bit Default &

Range Access
0b
RO
Destination Scatter Count (DSC): Source contiguous transfer count between

0b successive scatter boundaries. Specifies the number of contiguous destination transfers
24:20
RW of CTL0_L.DST_TR_WIDTH between successive scatter intervals. This is defined as a
scatter boundary
0b Destination Scatter Interval (DSI): Specifies the destination address increment/

19:0 decrement in multiples of CTL0_L.DST_TR_WIDTH on a scatter boundary when scatter
RW mode is enabled for the destination transfer.
18.6.2.14 Channel 1 Source Address (SAR1)—Offset 58h

Source Address of DMA transfer The starting source address is programmed by
the source address of the current transfer.
Access Method

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SAR
Bit Default &
Range Access
Current Source Address of DMA transfer (SAR): Updated after each source transfer.
The SINC field in the CTL1_L register determines whether the address increments,
0b decrements, or is left unchanged on every source transfer throughout the block transfer.
31:0
RW
NOTE: Channel 1 is dedicated to the UART controller Transmitted data. Source Address
is a memory address
18.6.2.15 Channel 1 Destination Address (DAR1)—Offset 60h

Destination address of DMA transfer The starting destination address is programmed by
the destination address of the current transfer.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DAR
Bit Default &

Range Access
Current Destination address of DMA transfer (DAR): Updated after each

destination transfer. The DINC field in the CTL1_L register determines whether the
0b address increments, decrements, or is left unchanged on every destination transfer
31:0 throughout the block transfer.
RW
NOTE: Channel 0 is dedicated to the UART controller Transmitted data. Destination
Address is the UART controller RBR_THR_DLL register address: 0xFFFF_F000
18.6.2.16 Channel 1 Linked List Pointer (LLP1)—Offset 68h

Program this register to point to the first Linked List Item (LLI) in memory prior to
enabling the channel if block chaining is enabled
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
LOC
Bit Default &
Range Access
Starting Address In Memory (LOC): Starting Address In Memory of next LLI if block
chaining is enabled.
0b Note that the two LSBs of the starting address are not stored because the address is
31:2
RW assumed to be aligned to a 32-bit boundary.
LLI accesses are always 32-bit accesses (Hsize = 2) aligned to 32-bit boundaries and
cannot be changed or programmed to anything other than 32-bit.
0b
RO
18.6.2.17 Channel 1 Control LOWER (CTL1_L)—Offset 70h

Contains fields that control the DMA transfer. Part of the block descriptor (linked list
item - LLI) when block chaining is enabled. It can be varied on a block-by-block basis
within a DMA transfer when block chaining is enabled.
Access Method
(Size: 32 bits)

Default: 00304801h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 1
RSV1
RSV0
INT_EN
SMS
DMS
DST_GATHER_EN
SRC_GATHER_EN
SRC_MSIZE
DEST_MSIZE
SINC
SRC_TR_WIDTH
DST_TR_WIDTH
TT_FC
DINC
Bit Default &

Range Access
0b
RO
0b Source AMBA Layer (SMS): Hardcoded the Master interface attached to the source of
26:25
RO channel 1
0b Destination AMBA Layer (DMS): Hardcoded the Master interface attached to the
24:23
RO destination of channel 1

Bit Default &

Range Access
Transfer Type and Flow Control (TT_FC): The following transfer types are
supported:
Code - Type - Flow Controller
--------------------------------------------------
000 - Memory to Memory - DMAC
011b 001 - Memory to Peripheral - DMAC
22:20
RW 010 - Peripheral to Memory - DMAC
011 - Peripheral to Peripheral - DMAC
100 - Peripheral to Memory - Peripheral
101 - Peripheral to Peripheral - Source Peripheral
110 - Memory to Peripheral - Peripheral
111 - Peripheral to Peripheral - Destination Peripheral
0b
RO
Destination scatter enable (DST_GATHER_EN): 0 = Scatter disabled

0b 1 = Scatter enabled
18
RW Scatter on the destination side is applicable only when the CTL1_L.DINC bit indicates an
Source gather enable (SRC_GATHER_EN): 0 = Gather disabled

0b 1 = Gather enabled
17
RW Gather on the source side is applicable only when the CTL1_L.SINC bit indicates an
Source Burst Transaction Length (SRC_MSIZE): Number of data items, each of

width CTL1_L.SRC_TR_WIDTH, to be read from the source every time a source burst
transaction request is made from either the corresponding hardware or software
handshaking interface. Value - Size(#TR_WITDH)
--------------------
001b 000 - 1
16:14 001 - 4
RW 010 - 8
011 - 16
100 - 32
101 - 64
110 - 128
111 - 256
Destination Burst Transaction Length (DEST_MSIZE): Destination Burst

Transaction Length. Number of data items, each of width CTL1_L.DST_TR_WIDTH, to be
written to the destination every time a destination burst transaction request is made
from either the corresponding hardware or software handshaking interface. Value -
Size(#TR_WITDH)
--------------------
001b 000 - 1
13:11
RW 001 - 4
010 - 8
011 - 16
100 - 32
101 - 64
110 - 128
111 - 256
Source Address Increment (SINC): Indicates whether to increment or decrement

the source address on every source transfer. If the device is fetching data from a source
peripheral FIFO with a fixed address,
0b then set this field to No change.
10:9 00 = Increment
RW 01 = Decrement
1x = No change
NOTE: Incrementing or decrementing is done for alignment to the next SRC_TR_WIDTH
boundary.

Bit Default &

Range Access
Destination Address Increment (DINC): Indicates whether to increment or

decrement the destination address on every destination transfer. If your device is
writing data to a destination peripheral FIFO with a
0b fixed address, then set this field to No change
8:7 00 = Increment
RW 01 = Decrement
1x = No change
NOTE: Incrementing or decrementing is done for alignment to the next DST_TR_WIDTH
boundary.
Source transfer width (SRC_TR_WIDTH): Decoding for this field:

Value - Size(bits)
--------------------
000 - 8
0b 001 - 16
6:4
RW 010 - 32
011 - 64
100 - 128
101 - 256
11x - 256
Destination transfer width (DST_TR_WIDTH): Decoding for this field:

Value - Size(bits)
--------------------
000 - 8
0b 001 - 16
3:1
RW 010 - 32
011 - 64
100 - 128
101 - 256
11x - 256
1b Interrupt enable (INT_EN): If set, then all interrupt-generating sources are enabled.
0 Functions as a global mask bit for all interrupts for the channel. RAW interrupt registers
RW still assert if INT_EN = 0.
18.6.2.18 Channel 1 Control UPPER (CTL1_U)—Offset 74h

Contains fields that control the DMA transfer. It can be varied on a block-by-block basis
within a DMA transfer when block chaining is enabled. If status write-back is enabled,
the content is written to the control register location of the LLI in system memory at
the end of the block transfer.
Access Method
(Size: 32 bits)

Default: 00000002h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0
RSV
BLOCK_TS
Bit Default &

Range Access
0b
RO

Bit Default &

Range Access
Block length (BLOCK_TS): When the DMAC is the flow controller, the user writes this
002h field before the channel is enabled in order to indicate the block size. The number
11:0 programmed into BLOCK_TS indicates the total number of single transactions to
RW perform for every block transfer. A single transaction is mapped to a single AMBA beat.
Width: The width of the single transaction is determined by CTL1_L.SRC_TR_WIDTH.
18.6.2.19 Channel 1 Source Status (SSTAT1)—Offset 78h

This register is a temporary placeholder for the source status information on its way to
the SSTAT1 register location of the LLI. The source status information should be
retrieved by software from the SSTAT1 register location of the LLI, and not by a read of
this register over the DMAC slave interface.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SSTAT
Bit Default &

Range Access
0b Channel Source Status (SSTAT): Source status information retrieved by hardware

31:0
RW from the address pointed to by the contents of the SSTATAR1 register.
18.6.2.20 Channel 1 Destination Status (DSTAT1)—Offset 80h

This register is a temporary placeholder for the destination status information on its
way to the DSTAT1 register location of the LLI. The destination status information
should be retrieved by software from the DSTAT1 register location of the LLI and not by
a read of this register over the DMAC slave interface
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DSTAT

Bit Default &

Range Access
0b Channel Destination Status (DSTAT): Destination status information retrieved by

31:0
RW hardware from the address pointed to by the contents of the DSTATAR1 register
18.6.2.21 Channel 1 Source Status Address (SSTATAR1)—Offset 88h

After the completion of each block transfer, hardware can retrieve the source status
information from the address pointed to by the contents of this register
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SSTATAR
Bit Default &

Range Access
0b Channel Source Status Address (SSTATAR): Pointer from where hardware can fetch
31:0 the source status information, which is registered in the SSTAT1 register and written out
RW to the SSTAT1 register location of the LLI before the start of the next block.
18.6.2.22 Channel 1 Destination Status Address (DSTATAR1)—Offset 90h

After the completion of each block transfer, hardware can retrieve the destination
status information from the address pointed to by the contents of this register
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DSTATAR
Bit Default &

Range Access
0b Channel Destination Status Address (DSTATAR): Pointer from where hardware can
31:0 fetch the destination status information, which is registered in the DSTAT1 register and
RW written out to the DSTAT1 register location of the LLI before the start of the next block.

18.6.2.23 Channel 1 Configuration LOWER (CFG1_L)—Offset 98h

Access Method
(Size: 32 bits)

Default: 00000E20h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 1 0 0 0 0 0
RELOAD_DST
RSV2
DST_HS_POL
RSV1
HS_SEL_DST
CH_SUSP
RSV0
FIFO_EMPTY
SRC_HS_POL
CH_PRIOR
RELOAD_SRC
HS_SEL_SRC
Bit Default &
Range Access
0b Reload destination enable (RELOAD_DST): The DARx register can be automatically

0b Reload source enable (RELOAD_SRC): The SAR register can be automatically

0b
RO
0b Source handshake polarity (SRC_HS_POL): 0 = Active high

19
RW 1 = Active low
0b Destination handshake polarity (DST_HS_POL): 0 = Active high

18
RW 1 = Active low
0b
RW
Source Handshake select (HS_SEL_SRC): Used to select which handshake interface

1b is active for source requests on this channel
If source peripheral is memory this bit is ignored
Destination Handshake select (HS_SEL_DST): Used to select which handshake

1b interface is active for destination requests on this channel
If destination peripheral is memory this bit is ignored
Channel FIFO empty status (FIFO_EMPTY): Indicates if there is data left in the
1b channel FIFO. Can be used in conjunction with CH_SUSP to cleanly disable a channel.
9
RO 1 = Channel FIFO empty
0 = Channel FIFO not empty
Channel Suspend control (CH_SUSP): Suspends all DMA data transfers from the
source until this bit is cleared. There is no guarantee that the current transaction will
0b complete. Can also be used in conjunction
8 with FIFO_EMPTY to cleanly disable a channel without
RW losing any data.
0 = Not suspended.
1 = Suspend DMA transfer from the source.

Bit Default &

Range Access
001b Channel Priority (CH_PRIOR): Priority value equal to 7 is the highest priority, and 0
7:5 is the lowest. This field must be programmed within the following range: 0: 1
RW A programmed value outside this range will cause erroneous behavior.
0b
RO
18.6.2.24 Channel 1 configuration UPPER (CFG1_U)—Offset 9Ch

Access Method

Default: 00000004h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
FIFO_MODE
FCMODE
RSV5
RSV4
SS_UPD_EN
DS_UPD_EN
DEST_PER
SRC_PER
PROTCTL
Bit Default &
Range Access
0b
RO
Destination hardware interface (DEST_PER): Assigns a hardware handshaking

interface (0-1) to the channel destination if the CFG1_L.HS_SEL_DST field is 0;
0b otherwise, this field is ignored. The channel can then communicate with the destination
11
RW peripheral connected to that interface through the assigned hardware handshaking
interface. NOTE: For correct DMA operation, only one peripheral (source or destination)
should be assigned to the same handshaking interface.
0b
RO
Source hardware interface (SRC_PER): Assigns a hardware handshaking interface

(0-1) to the channel source if the CFG1_L.HS_SEL_SRC field is 0; otherwise, this field is
0b ignored. The channel can then communicate with the source peripheral connected to
7
RW that interface through the assigned hardware handshaking interface. NOTE: For correct
DMA operation, only one peripheral (source or destination) should be assigned to the
same handshaking interface.
0b Source Status Update Enable (SS_UPD_EN): Source status information is fetched

6 only from the location pointed to by the SSTATAR1 register, stored in the SSTAT1
RW register and written out to the SSTAT1 location of the LLI if this field is high
0b Destination Status Update Enable (DS_UPD_EN): Destination status information is

5 fetched only from the location pointed to by the DSTATAR1 register, stored in the
RW DSTAT1 register and written out to the DSTAT0 location of the LLI if this field is high

Bit Default &

Range Access
AHB bus protocol bus control (PROTCTL): Protection Control bits used to drive the
AHB HPROT[3:1] bus.
001b The AMBA Specification recommends that the default value of HPROT indicates a non-
4:2 cached, non-buffered, privileged data access. The reset value is used to indicate such an
RW access. HPROT[0] is tied high because all transfers are data accesses, as there are no
opcode fetches. There is a one-to-one mapping of these register bits to the HPROT[3:1]
master interface signals.
Channel FIFO mode control (FIFO_MODE): Determines how much space or data
needs to be available in the FIFO before a burst transaction request is serviced.
0b 0 = Space/data available for single AHB transfer of the specified transfer width.
1
RW 1 = Data available is greater than or equal to half the FIFO depth for destination
transfers and space available is greater than half the fifo depth for source transfers. The
exceptions are at the end of a burst transaction request or at the end of a block transfer.
Channel flow control mode (FCMODE): Determines when source transaction

requests are serviced when the Destination Peripheral is the flow controller.
0 = Source transaction requests are serviced when they occur. Data pre-fetching is
0b enabled.
0 1 = Source transaction requests are not serviced until a
RW destination transaction request occurs.
In this mode, the amount of data transferred from the source is limited so that it is
guaranteed to be transferred to the destination prior to block termination by the
destination. Data pre-fetching is disabled.
18.6.2.25 Channel 1 Source Gather (SGR1)—Offset A0h

The CTL1_L.SINC field controls whether the address increments or decrements. For a
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SGC
SGI
RSV
Bit Default &

Range Access
0b
RO
Source Gather Count (SGC): Source contiguous transfer count between successive
0b gather boundaries. Specifies the number of contiguous source transfers of
24:20
RW CTL1_L.SRC_TR_WIDTH between successive gather intervals. This is defined as a
gather boundary
0b Source Gather Interval (SGI): Specifies the source address increment/decrement in

19:0 multiples of CTL1_L.SRC_TR_WIDTH on a gather boundary when gather mode is
RW enabled for the source transfer.
18.6.2.26 Channel 1 Destination Scatter (DSR1)—Offset A8h

The CTL1_L.DINC field controls whether the address increments or decrements. For a

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DSC
DSI
RSV
Bit Default &

Range Access
0b
RO
Destination Scatter Count (DSC): Source contiguous transfer count between

0b successive scatter boundaries. Specifies the number of contiguous destination transfers
24:20
RW of CTL1_L.DST_TR_WIDTH between successive scatter intervals. This is defined as a
scatter boundary
0b Destination Scatter Interval (DSI): Specifies the destination address increment/

19:0 decrement in multiples of CTL1_L.DST_TR_WIDTH on a scatter boundary when scatter
RW mode is enabled for the destination transfer.
18.6.2.27 Raw Status for IntTfr Interrupt (RAW_TFR)—Offset 2C0h

DMA Transfer Complete Interrupt. This interrupt is generated on DMA transfer
completion to the destination peripheral
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RAW
RSV
Bit Default &

Range Access
0b
RO
0b Raw Status for IntTfr Interrupt (RAW): Interrupt events are stored in this Raw
1:0 Interrupt Status register before masking. Each bit in this register is cleared by writing a
RW 1 to the corresponding location in the correspondent Clear register
18.6.2.28 Raw Status for IntBlock Interrupt (RAW_BLOCK)—Offset 2C8h

Block Transfer Complete Interrupt. This interrupt is generated on DMA block transfer
completion to the destination peripheral.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RAW
RSV
Bit Default &
Range Access
0b
RO
0b Raw Status for IntBlock Interrupt (RAW): Interrupt events are stored in this Raw
18.6.2.29 Raw Status for IntSrcTran Interrupt (RAW_SRC_TRAN)—Offset 2D0h

Source Transaction Complete Interrupt. Generated after completion of the last AHB
transfer of the requested single/burst transaction from the handshaking interface
(either the hardware or software handshaking interface) on the source side. NOTE: If
the source is memory, then IntSrcTran interrupt should be ignored, as there is no
concept of a DMA transaction level for memory
Access Method
(Size: 32 bits) Offset: [BAR1] + 2D0h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
RAW
Bit Default &
Range Access
0b
RO
0b Raw Status for IntSrcTran Interrupt (RAW): Interrupt events are stored in this Raw
18.6.2.30 Raw Status for IntDstTran Interrupt (RAW_DST_TRAN)—Offset 2D8h

Destination Transaction Complete Interrupt. Generated after completion of the last AHB
transfer of the requested single/burst transaction from the handshaking interface
(either the hardware or software handshaking interface) on the destination side. NOTE:
If the destination for a channel is memory, then that channel will never generate the
IntDstTran interrupt. Because of this, the corresponding bit in this field will not be set.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RAW
RSV
Bit Default &
Range Access
0b
RO
0b Raw Status for IntDstTran Interrupt (RAW): Interrupt events are stored in this
1:0 Raw Interrupt Status register before masking. Each bit in this register is cleared by
RW writing a 1 to the corresponding location in the correspondent Clear register
18.6.2.31 Raw Status for IntErr Interrupt (RAW_ERR)—Offset 2E0h

Error Interrupt. This interrupt is generated when an ERROR response is received from
an AHB slave on the HRESP bus during a DMA transfer. In addition, the DMA transfer is
cancelled and the channel is disabled.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
RAW
Bit Default &
Range Access
0b
RO
0b Raw Status for IntErr Interrupt (RAW): Interrupt events are stored in this Raw
18.6.2.32 Status for IntTfr Interrupt (STATUS_TFR)—Offset 2E8h

DMA Transfer Complete Interrupt status
Access Method
(Size: 32 bits)


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
STATUS
Bit Default &
Range Access
0b
RO
0b Status for IntTfr Interrupt (STATUS): Stores all interrupt events from channels after
1:0
RO masking. One bit allocated per channel. Used to generate the DMAC interrupt signals
18.6.2.33 Status for IntBlock Interrupt (STATUS_BLOCK)—Offset 2F0h

Block Transfer Complete Interrupt status
Access Method
(Size: 32 bits) Offset: [BAR1] + 2F0h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
STATUS
Bit Default &
Range Access
0b
RO
0b Status for IntBlock Interrupt (STATUS): Stores all interrupt events from channels
1:0 after masking. One bit allocated per channel. Used to generate the DMAC interrupt
RO signals
18.6.2.34 Status for IntSrcTran Interrupt (STATUS_SRC_TRAN)—Offset 2F8h

Source Transaction Complete Interrupt status
Access Method
Type: Memory Mapped I/O Register Offset: [BAR1] + 2F8h
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
STATUS
Bit Default &
Range Access
0b
RO
0b Status for IntSrcTran Interrupt (STATUS): Stores all interrupt events from channels
RO signals
18.6.2.35 Status for IntDstTran Interrupt (STATUS_DST_TRAN)—Offset 300h

Destination Transaction Complete Interrupt status
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
STATUS
Bit Default &
Range Access
0b
RO
0b Status for IntDstTran Interrupt (STATUS): Stores all interrupt events from
1:0 channels after masking. One bit allocated per channel. Used to generate the DMAC
RO interrupt signals
18.6.2.36 Status for IntErr Interrupt (STATUS_ERR)—Offset 308h

Error Interrupt status
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
STATUS
Bit Default &
Range Access
0b
RO
0b Status for IntErr Interrupt (STATUS): Stores all interrupt events from channels
RO signals
18.6.2.37 Mask for IntTfr Interrupt (MASK_TFR)—Offset 310h

DMA Transfer Complete Interrupt mask. The contents of the Raw Status register is
masked with the contents of the Mask register.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
INT_MASK
INT_MASK_WE
Bit Default &

Range Access
0b
RO
0b Interrupt Mask Write Enable (INT_MASK_WE): 0 = write disabled

9:8
RW 1 = write enabled
0b
RO
Mask for the interrupt (INT_MASK): Written only if the corresponding mask write
0b enable bit in the INT_MASK_WE field is asserted on the same AHB write transfer. This
1:0 allows software to set a mask bit without performing a read-modified write operation.
RW 0 = masked
1 = unmasked
18.6.2.38 Mask for IntBlock Interrupt (MASK_BLOCK)—Offset 318h

Block Transfer Complete Interrupt mask. The contents of the Raw Status register is
masked with the contents of the Mask register.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
INT_MASK
INT_MASK_WE
Bit Default &
Range Access
0b
RO

9:8
0b
RO
RW 0 = masked
1 = unmasked
18.6.2.39 Mask for IntSrcTran Interrupt (MASK_SRC_TRAN)—Offset 320h

Source Transaction Complete Interrupt mask. The contents of the Raw Status register
is masked with the contents of the Mask register.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
INT_MASK
INT_MASK_WE
RSV1
RSV0
Bit Default &

Range Access
0b
RO

Bit Default &

Range Access

9:8
0b
RO
RW 0 = masked
1 = unmasked
18.6.2.40 Mask for IntDstTran Interrupt (MASK_DST_TRAN)—Offset 328h

Destination Transaction Complete Interrupt mask. The contents of the Raw Status
register is masked with the contents of the Mask register.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
INT_MASK
INT_MASK_WE
RSV1
RSV0
Bit Default &
Range Access
0b
RO

9:8
0b
RO
RW 0 = masked
1 = unmasked
18.6.2.41 Mask for IntErr Interrupt (MASK_ERR)—Offset 330h

Error Interrupt mask. The contents of the Raw Status register is masked with the
contents of the Mask register.
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
INT_MASK
INT_MASK_WE
Bit Default &
Range Access
0b
RO

9:8
0b
RO
RW 0 = masked
1 = unmasked
18.6.2.42 Clear for IntTfr Interrupt (CLEAR_TFR)—Offset 338h

DMA Transfer Complete Interrupt clear
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
CLEAR
Bit Default &

Range Access
0b
RO
0b Clear for Interrupt (CLEAR): 0 = no effect

1:0
WO 1 = clear interrupt
18.6.2.43 Clear for IntBlock Interrupt (CLEAR_BLOCK)—Offset 340h

Block Transfer Complete Interrupt clear

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLEAR
RSV
Bit Default &
Range Access
0b
RO

1:0
18.6.2.44 Clear for IntSrcTran Interrupt (CLEAR_SRC_TRAN)—Offset 348h

Source Transaction Complete Interrupt clear
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
CLEAR
Bit Default &
Range Access
0b
RO

1:0
18.6.2.45 Clear for IntDstTran Interrupt (CLEAR_DST_TRAN)—Offset 350h

Destination Transaction Complete Interrupt clear
Access Method

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLEAR
RSV
Bit Default &
Range Access
0b
RO

1:0
18.6.2.46 Clear for IntErr Interrupt (CLEAR_ERR)—Offset 358h

Error Interrupt clear
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
STATUS
Bit Default &
Range Access
0b
RO
0b Clear for Interrupt (STATUS): 0 = no effect

1:0
18.6.2.47 Combined Interrupt Status (STATUS_INT)—Offset 360h

The contents of each of the Status registers is ORed to produce a single bit for each
interrupt type in this Combined Interrupt Status register.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ERR
DSTT
TFR
RSV
SRCT
BLOCK

Bit Default &

Range Access
0b
RO
0b
4 OR of the contents of STATUS_ERR (ERR): Reserved.
RO
0b
3 OR of the contents of STATUS_DSTT (DSTT): Reserved.
RO
0b
2 OR of the contents of STATUS_SRCT (SRCT): Reserved.
RO
0b
1 OR of the contents of STATUS_BLOCK (BLOCK): Reserved.
RO
0b
0 OR of the contents of STATUS_TFR (TFR): Reserved.
RO
18.6.2.48 Source Software Transaction Request (REQ_SRC_REG)—Offset 368h

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SRC_REQ
SRC_REQ_WE
RSV1
RSV0
Bit Default &
Range Access
0b
RO
0b Source Software Transaction Request write enable (SRC_REQ_WE): 0 = write

9:8 disabled
0b
RO
0b Source Software Transaction Request register (SRC_REQ): This bit is written only
1:0 if the corresponding channel write enable bit in the Write Enable field is asserted on the
RW same AHB write transfer, and if the channel is enabled in the CH_EN_REG register
18.6.2.49 Destination Software Transaction Request register (REQ_DST_REG)—

Offset 370h
Access Method
(Size: 32 bits)


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DST_REQ
DST_REQ_WE
RSV1
RSV0
Bit Default &
Range Access
0b
RO
0b Destination Software Transaction Request write enable (DST_REQ_WE): 0 =

9:8 write disabled
0b
RO
0b Destination Transaction Request register (DST_REQ): This bit is written only if the
1:0 corresponding channel write enable bit in the Write Enable field is asserted on the same
RW AHB write transfer, and if the channel is enabled in the CH_EN_REG register
18.6.2.50 Source Single Transaction Request (SGL_REQ_SRC_REG)—Offset 378h
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
SRC_SGLREQ_WE
SRC_SGLREQ
Bit Default &

Range Access
0b
RO
0b Source Single Transaction Request write enable (SRC_SGLREQ_WE): 0 = write

9:8 disabled
0b
RO
0b Source Single Transaction Request register (SRC_SGLREQ): This bit is written

1:0 only if the corresponding channel write enable bit in the Write Enable field is asserted on
RW the same AHB write transfer, and if the channel is enabled in the CH_EN_REG register

18.6.2.51 Destination Single Software Transaction Request

(SGL_REQ_DST_REG)—Offset 380h
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DST_SGLREQ_WE
DST_SGLREQ
RSV1
RSV0
Bit Default &
Range Access
0b
RO
0b Destination Single Transaction Request write enable (DST_SGLREQ_WE): 0 =

9:8 write disabled
0b
RO
Destination Single Transaction Request register (DST_SGLREQ): This bit is

0b written only if the corresponding channel write enable bit in the Write Enable field is
1:0
RW asserted on the same AHB write transfer, and if the channel is enabled in the
CH_EN_REG register
18.6.2.52 Source Last Transaction Request (LST_SRC_REG)—Offset 388h

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LSTSRC_WE
RSV1
RSV0
LSTSRC
Bit Default &

Range Access
0b
RO

Bit Default &

Range Access
0b Source Last Transaction Request write enable (LSTSRC_WE): 0 = write disabled

9:8
0b
RO
0b Source Last Transaction Request register (LSTSRC): This bit is written only if the
1:0 corresponding channel write enable bit in the Write Enable field is asserted on the same
RW AHB write transfer, and if the channel is enabled in the CH_EN_REG register
18.6.2.53 Destination Single Transaction Request (LST_DST_REG)—Offset 390h

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LSTDST_WE
LSTDST
RSV1
RSV0
Bit Default &
Range Access
0b
RO
0b Destination Last Transaction Request write enable (LSTDST_WE): 0 = write

9:8 disabled
0b
RO
0b Destination Last Transaction Request register (LSTDST): This bit is written only if
1:0 the corresponding channel write enable bit in the Write Enable field is asserted on the
RW same AHB write transfer, and if the channel is enabled in the CH_EN_REG register
18.6.2.54 DMA Configuration (DMA_CFG_REG)—Offset 398h

Used to enable the DMA controller (DMAC), which must be done before any channel
activity can begin. If the global channel enable bit is cleared while any channel is still
active, then DMA_EN still returns 1 to indicate that there are channels still active until
hardware has terminated all activity on all channels, at which point the DMA_EN bit
returns 0
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
DMA_EN
Bit Default &
Range Access
0b
RO
0b DMA global enable (DMA_EN): 0 = disabled

0
RW 1 = enabled
18.6.2.55 Channel Enable (CH_EN_REG)—Offset 3A0h

Software can read this register in order to find out which channels are currently
inactive if needs to set up a new channel. It can then enable an inactive channel with
the required priority.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
CH_EN
CH_EN_WE
Bit Default &

Range Access
0b
RO
0b
9:8 Channel enable register (CH_EN_WE): Reserved.
RW
0b
RO
Channel enable register (CH_EN): Setting this bit enables a channel. Clearing this
bit disables the channel.
0 = Disable the Channel
0b 1 = Enable the Channel
1:0
RW The CH_EN_REG.CH_EN bit is automatically cleared by hardware to disable the channel
after the last AMBA transfer of the DMA transfer to the destination has completed.
Software can therefore poll this bit to determine when this channel is free for a new
DMA transfer.
§§

I2C* Controller/GPIO Controller—Intel® Quark™ SoC X1000
19.0 I2C* Controller/GPIO Controller
The Intel® Quark™ SoC X1000 has an I2C* controller connected to the I/O Fabric. Both
7-bit and 10-bit addressing modes are supported. This controller operates in master
mode only.
In addition, this PCI function also provides a GPIO controller. The SoC provides a total
of 16 GPIOs that are split between the Legacy Bridge (D:31 F:0) and the GPIO
controller (D:21 F:2). This chapter details the non-Legacy GPIOs provided by the GPIO
controller.
19.1 I2C Controller
19.1.1 Signal Descriptions

I2C is a two-wire bus for inter-IC communication. Data and clock signals carry
information between the connected devices.

Table 119. I2C* Signals

Direction/
Type
I2C_DATA I/O I2C Serial Data
I2C_CLK I/O I2C Serial Clock
19.1.2 Features
19.1.2.1 I2C* Protocol

The I2C bus is a two-wire serial interface, consisting of a serial data line and a serial
clock. These wires carry information between the devices connected to the bus. Each
device is recognized by a unique address and can operate as either a “transmitter” or
“receiver,” depending on the function of the device. Devices are considered slaves when
performing data transfers, as the SoC is always a Master. A master is a device that
initiates a data transfer on the bus and generates the clock signals to permit that
transfer. At that time, any device addressed is considered a slave.
• The SoC is always the I2C master; and it supports multi-master mode.

Intel® Quark™ SoC X1000—I2C* Controller/GPIO Controller
• The SoC can support clock stretching by slave devices.

• The I2C_DATA line is a bidirectional signal and changes only while the I2C_CLK line
is low, except for STOP, START, and RESTART conditions.
• The output drivers are open-drain or open-collector to perform wire-AND functions
on the bus.
• The maximum number of devices on the bus is limited by the maximum
capacitance specification:
— 130 pF for standard and fast mode
• Data is transmitted in byte packages.
19.1.2.2 I2C* Modes of Operation

The I2C module can operate in the following modes:
• Standard mode (with a bit rate up to 100 Kb/s)
• Fast mode (with a bit rate up to 400 Kb/s)
The I2C module can communicate with devices using only these modes as long as they
are attached to the bus. Additionally, fast mode devices are downward compatible.
• Fast mode devices can communicate with standard mode devices in a 0–100 Kb/s
I2C bus system.
However, according to the I2C specification, standard mode devices are not upward
compatible and should not be incorporated in a fast-mode I2C bus system since they
cannot follow the higher transfer rate and unpredictable states would occur.
19.1.2.3 Functional Description

• The I2C master is responsible for generating the clock and controlling the transfer
of data.
• The slave is responsible for either transmitting or receiving data to/from the
master.
• The acknowledgement of data is sent by the device that is receiving data, which
can be either a master or a slave.
• Each slave has a unique address that is determined by the system designer:
— When a master wants to communicate with a slave, the master transmits a
START/RESTART condition that is then followed by the slave's address and a
control bit (R/W), to determine if the master wants to transmit data or receive
data from the slave.
— The slave then sends an acknowledge (ACK) pulse after the address.
• If the master (master-transmitter) is writing to the slave (slave-receiver):
— The receiver gets one byte of data.
— This transaction continues until the master terminates the transmission with a
STOP condition.
• If the master is reading from a slave (master-receiver):
— The slave transmits (slave-transmitter) a byte of data to the master, and the
master then acknowledges the transaction with the ACK pulse.
— This transaction continues until the master terminates the transmission by not
acknowledging (NACK) the transaction after the last byte is received, and then
the master issues a STOP condition or addresses another slave after issuing a
RESTART condition. This behavior is illustrated in Figure 39.

Figure 39. Data Transfer on the I2C* Bus
P or R
D a ta
MSB LSB ACK ACK
fr o m s la v e fro m re c e iv e r
C lo c k
S 1 2 7 8 9 1 2 3 -8 9 R or P
or
R
STOP AND
START or B y t e C o m p le t e S C L h e ld lo w RESTART
RESTART I n t e r r u p t w it h in w h ile s e r v ic in g C o n d it io n s
C o n d it io n s S la v e in t e r r u p t s
19.1.2.3.1 START and STOP Conditions
When the bus is idle, both the clock and data signals are pulled high through external
pull-up resistors on the bus.
When the master wants to start a transmission on the bus, the master issues a START
condition.
• This is defined to be a high-to-low transition of the data signal while clock is a ‘1’.
• When the master wants to terminate the transmission, the master issues a STOP
condition. This is defined to be a low-to-high transition of the data line while clock
is a 1. Figure 40 shows the timing of the START and STOP conditions.
• When data is being transmitted on the bus, the data line must be stable when clock
is a 1.
Figure 40. START and STOP Conditions
Data
Clock
S Change of Data Change of Data P

Allowed Data Line Stable Allowed
Start Conditions Data Line Valid Stop Condition
The signal transitions for the START/STOP conditions, as depicted above, reflect those
observed at the output of the Master driving the I2C bus. Care should be taken when
observing the data/clock signals at the input of the Slave(s), because unequal line
delays may result in an incorrect data/clock timing relationship.
19.1.2.3.2 Addressing Slave Protocol
There are two address formats—7-bit address format and 10-bit address format.
7-bit Address Format

• During the seven-bit address format, the first seven bits (bits 7:1) of the first byte
set the slave address and the LSB bit (bit 0) is the R/W bit as shown in Figure 41.

• When bit 0 (R/W) is set to 0, the master writes to the slave. When bit 0 (R/W) is
set to 1, the master reads from the slave.
Figure 41. 7-Bit Address Format
MSB LS B
S A6 A5 A4 A3 A2 A1 A0 R /W ACK
sen t by
S lave A d dress slave
S = S TA R T con dition s A C K = A ckn ow ledge R /W = R ead / W rite Pu lse
10-bit Address Format

• During 10-bit addressing, 2 bytes are transferred to set the 10-bit address. The
transfer of the first byte contains the following bit definition.
— The first five bits (bits 7:3) notify the slaves that this is a 10-bit transfer. The
next two bits (bits 2:1) set the slave’s address bits 9:8. The LSB bit (bit 0) is
the RW bit.
— The second byte transferred sets bits 7:0 of the slave address.
— Figure 42 shows the 10-bit address format, and Table 120 defines the special
purpose and reserved first byte addresses.
Figure 42. 10-bit Address Format
S ‘1’ ‘1’ ‘1’ ‘1’ ‘0’ A9 A8 R/W ACK A7 A6 A5 A4 A3 A2 A1 A0 ACK
Sent by slave Sent by slave

Reserved for
10-bit Address
S = START condition R/W = Read/Write Pulse ACK = Acknowledge
Table 120. I2C* Definition of Bits in First Byte

Slave Address RW Bit Description
General Call Address—The I2C controller places the data in the receive
0000 000 0
buffer and issues a General Call interrupt.
0000 000 1 START byte—For more details, refer to I2C bus specification section 3.15.
0000 001 X CBUS address—I2C controller ignores these accesses.
0000 010 X Reserved
0000 011 X Reserved
0000 1XX X High-speed master code
1111 1XX X Reserved
1111 0XX X Ten (10)-bit slave addressing

19.1.2.3.3 Transmit and Receive Protocol
The master can initiate data transmission and reception to/from the bus, acting as
either a master-transmitter or master-receiver. A slave responds to requests from the
master by either transmitting data or receiving data to/from the bus, acting as either a
slave-transmitter or slave-receiver, respectively.
Master-Transmitter and Slave-Receiver
All data is transmitted in byte format, with no limit on the number of bytes transferred
per data transfer. After the master sends the address and RW bit or the master
transmits a byte of data to the slave, the slave-receiver must respond with the
acknowledge signal (ACK). When a slave-receiver does not respond with an ACK pulse,
the master aborts the transfer by issuing a STOP condition. The slave must leave the
SDA line high so that the master can abort the transfer.
If the master-transmitter is transmitting data as shown in Figure 43, then the slave-
receiver responds to the master-transmitter with an acknowledge pulse after every
byte of data is received.
Figure 43. Master Transmitter Protocol
F o r 7 - b it A d d r e s s
S S la v e A d d r e s s R /W A DATA A DATA A /A P
0 ' ( w r it e )
F o r 1 0 - b it A d d r e s s
S la v e A d d r e s s S la v e A d d r e s s
S R /W A A DATA A /A P
F ir s t 7 b it s S e c o n d B y te
0 ' ( w r it e )
‘1 1 1 1 0 x x x ’
A = A c k n o w le d g e ( S D A lo w )
F r o m M a s te r to S la v e A = N o A c k n o w le d g e ( S D A h ig h )
S = S T A R T C o n d it io n
F r o m S la v e to M a s te r
P = S T O P C o n d it io n
Master-Receiver and Slave-Transmitter
If the master is receiving data as shown in Figure 44, the master responds to the
Slave-Transmitter with an acknowledge pulse after a byte of data has been received,
except for the last byte. This is the way the Master-Receiver notifies the Slave-
Transmitter that this is the last byte. The Slave-Transmitter relinquishes the SDA line
after detecting the No Acknowledge (NACK) so that the master can issue a STOP
condition.
When a master does not want to relinquish the bus with a STOP condition, the master
can issue a RESTART condition. This is identical to a START condition except it occurs
after the ACK pulse. The master can then communicate with the same slave or a
different slave.

Figure 44. Master Receiver Protocol
S Slave Address R/W A DATA A DATA A P
‘1’ (read)
For 10-bit Address
Slave Address Slave Address Slave Address
S R/W A A Sr R/W A DATA A P
First 7 bits Second Byte First 7 bits
‘11110xxx’ 0' (write) ‘11110xxx’ ‘1’ (read)
A = Acknowledge (SDA low)

From Master to Slave
A = No Acknowledge (SDA high)
From Slave to
Master S = START Condition
P = STOP Condition
R = RESTART Condition
19.1.2.3.4 START BYTE Transfer Protocol
The START BYTE Transfer protocol is set up for systems that do not have an on-board
dedicated I2C hardware module. When the I2C controller is a master, it supports the
generation of START BYTE transfers at the beginning of every transfer in case a slave
device requires it. This protocol consists of 7 ‘0’s being transmitted followed by a 1, as
illustrated in Figure 45. This allows the processor that is polling the bus to under-
sample the address phase until 0s are detected. Once the microcontroller detects a 0, it
switches from the under sampling rate to the correct rate of the master.
Figure 45. START Byte Transfer
Data dummy acknowledge
(HIGH)
Clock 1 2 7 8 9
S Sr
ACK
Start byte 00000001
The START BYTE procedure is as follows:

1. Master generates a START condition.
2. Master transmits the START byte (0000 0001).
3. Master transmits the ACK clock pulse. (Present only to conform with the byte
handling format used on the bus.)
4. No slave sets the ACK signal to 0.
5. Master generates a RESTART (R) condition.
A hardware receiver does not respond to the START BYTE because it is a reserved
address and resets after the RESTART condition is generated.

19.1.3 Use
19.1.3.1 Master Mode Operation

To use the I2C controller as a master, perform the following steps:
1. Disable the I2C controller by writing 0 (zero) to IC_ENABLE.ENABLE.
2. Write to the IC_CON register to set the maximum speed mode supported
(IC_CON.SPEED) and to specify whether the I2C controller starts its transfers in 7/
10 bit addressing mode.
3. Write to the IC_TAR register the address of the I2C device to be addressed. The
desired speed of the I2C controller master-initiated transfers, either 7-bit or 10-bit
addressing, is controlled by the IC_TAR.IC_10BITADDR_MASTER bit field.
4. Enable the I2C controller by writing a 1 in IC_ENABLE.
5. Write the transfer direction and data to be sent to the IC_DATA_CMD register. If the
IC_DATA_CMD register is written before the I2C controller is enabled, the data and
commands are lost as the buffers are kept cleared when I2C controller is not
enabled.
RESTART and STOP conditions are generated only under software control through
specific IC_DATA_CMD fields. When the Transmit FIFO is empty the I2C controller does
not automatically generate a STOP condition and pauses the I2C bus by holding SCL
low. As part of the command definition (IC_DATA_CMD) software must specify if a
particular command is going to be followed by a STOP or RESTART condition by setting
either IC_DATA_CMD.STOP or IC_DATA_CMD.RESTART.
The I2C controller supports switching back and forth between reading and writing
dynamically. To transmit data, write the data to be written to the lower byte of the I2C
Rx/Tx Data Buffer and Command Register (IC_DATA_CMD). The IC_DATA_CMD.CMD
should be written to 0 for I2C write operations. Subsequently, a read command may be
issued by writing “don't cares” to IC_DATA_CMD.DAT register bits, and a 1 should be
written to the IC_DATA_CMD.CMD bit.
19.1.3.2 Disabling I2C* Controller

The register IC_ENABLE is added to allow software to unambiguously determine when
the hardware has completely shutdown in response to the IC_ENABLE.ENABLE register
being set from 1 to 0.
Procedure
1. Define a timer interval (ti2c_poll) equal to 10 times the signaling period for the
highest I2C transfer speed used in the system and supported by the I2C controller.
For example, if the highest I2C transfer mode is 400Kb/s, then this ti2c_poll is 25 µs.
2. Define a maximum time-out parameter, MAX_T_POLL_COUNT, such that if any
repeated polling operation exceeds this maximum value, an error is reported.
3. Execute a blocking thread/process/function that prevents any further I2C master
transactions to be started by software, but allows any pending transfers to be
completed.
4. The variable POLL_COUNT is initialized to zero (0).
5. Set IC_ENABLE.ENABLE to zero (0).
6. Read the IC_ENABLE_STATUS.IC_EN bit. Increment POLL_COUNT by one. If
POLL_COUNT >= MAX_T_POLL_COUNT, exit with the relevant error code.
7. If IC_ENABLE_STATUS.IC_EN is 1, then sleep for ti2c_poll and proceed to the
previous step. Otherwise, exit with a relevant success code.

19.1.4 References
UM10204 I2C-Bus Specification and User Manual, Revision 03
19.2 GPIO Controller


Table 121. GPIO Signals

Direction/
Type
I/O
GPIO[7:0] General Purpose IO available in the S0 state
CMOS3.3
19.2.2 Features
The GPIO Controller provides 8 GPIO pins via Port A of the controller.
• 8 Independently configurable GPIOs
• Separate data register and data direction for each GPIO
• Interrupt source mode supported for each GPIO
• Debounce logic for interrupt sources
• Metastability registers for GPIO read data
19.3 Register Map


Figure 46. I2C*/GPIO Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI I2C*/GPIO
CAM Memory
PCI Header
(I/O) Space
D:21,F:2
Bus 0
PCI
ECAM
(Mem)
BAR1
PCIe* RP0 F:0
D:23
BAR0 GPIO Mem

SPI0 F:0 RP0 F:1 Registers
IO Fabric
D:21
SPI1 F:1
2
I C*/GPIO F:2
I2C* Mem
D:31, F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7

Default
Value


Default
Value

86h 86h 00h
page 752

Access Method
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value
Bit Default &

Description
Range Access
8086h
RO


Access Method
(Size: 16 bits)
Default: 0934h
15 12 8 4 0
0 0 0 0 1 0 0 1 0 0 1 1 0 1 0 0
value
Bit Default &
Description
Range Access
0934h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
IntrDis
RSVD
SERREn
RSVD
MasEn
MEMen
RSVD
Bit Default &
Description
Range Access
0h
RO

10
0h
RO
8
00h
RO

2
RW requests.
1
0h
RO


Access Method
(Size: 16 bits)
Default: 0010h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
capable_66Mhz
RcdMasAb
RSVD0
DEVSEL
FastB2B
hasCapList
RSVD1
SigSysErr
RSVD
RSVD
RSVD
IntrStatus
Bit Default &
Description
Range Access
0h
RO
14
13
0h
RO
0b
RO
0h
RO
0b
RO
0h
RO
0b
RO

4
RO register sets.
RO used
0h
RO

Access Method
Default: 0C800010h

31 28 24 20 16 12 8 4 0
0 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
progIntf
classCode
subClassCode
rev_id
Bit Default &
Description
Range Access
31: 24
RO performs.
23: 16
15: 8
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
RO

Access Method
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
cfgHdrFormat
multiFnDev
Bit Default &

Description
Range Access

7
RO function device

6: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
start_bist
RSVD
comp_code
BIST_capable
Bit Default &

Description
Range Access

7
0h
RO
0h
RO

Bit Default &

Description
Range Access

3: 0

Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD
prefetchable
memType
isIO
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder

Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD
memType
prefetchable
isIO

Bit Default &

Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO


Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

30h
Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ROM_base_addr
RSVD
AddrDecodeEn
Bit Default &
Description
Range Access

31: 11
000h
RO

RW also set to 1

Access Method
(Size: 32 bits)
Default: 00000080h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
value
RSVD0
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 8 bits) INTR_LINE: [B:0, D:21, F:2] + 3Ch
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
vector information.

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
7: 0

Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 01h

7 4 0
0 0 0 0 0 0 0 1
value
Bit Default &
Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 16 bits)
Default: 4803h
15 12 8 4 0
0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1
D2_support
DSI
PME_support
D1_support
aux_curr
PME_clock
RSVD
version
Bit Default &

Description
Range Access
15: 11
0h
RO
0h
RO

Bit Default &

Description
Range Access
0h
RO
0h
RO
0h
RO
2: 0

84h
Access Method
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
PME_status
PME_en
RSVD
RSVD
Data_scale
Data_select
power_state
Bit Default &
Description no_soft_reset
Range Access
15
0h
RO
0h
RO
8
0h
RO

3
0h
RO
1: 0


Offset 86h
Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access

7: 0
RO to 0.

87h
Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value

Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits) MSI_NXT_CAP_PTR: [B:0, D:21, F:2] + A1h
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
7: 0
RO in the chain

Access Method
(Size: 16 bits)
Default: 0100h
15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
MSIEnable
RSVD0
perVecMskCap
bit64Cap
multiMsgCap
multiMsgEn
Bit Default &

Description
Range Access
0h
RO

8
7

6: 4

3: 1

Bit Default &

Description
Range Access
RW MSI.

Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
address
RSVD0
Bit Default &
Description
Range Access
31: 2
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData
Bit Default &

Description
Range Access
15: 0
changed by hardware

Access Method


(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
MSIMask
Bit Default &
Description
Range Access
0h
RO
0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
value
Bit Default &
Description
Range Access
0h
RO
0h
RO
19.5.1 I2C* Controller Memory Mapped Registers

Offset Default
Start Value
0h 3h “Control Register (IC_CON)—Offset 0h” on page 756 00000037h
4h 7h “Master Target Address (IC_TAR)—Offset 4h” on page 757 00000055h
10h 13h “Data Buffer and Command (IC_DATA_CMD)—Offset 10h” on page 758 00000000h


Offset Default
Start Value
14h 17h “Standard Speed Clock SCL High Count (IC_SS_SCL_HCNT)—Offset 14h” on page 759 00000190h
18h 1Bh “Standard Speed Clock SCL Low Count (IC_SS_SCL_LCNT)—Offset 18h” on page 760 000001D6h
1Ch 1Fh “Fast Speed Clock SCL High Count (IC_FS_SCL_HCNT)—Offset 1Ch” on page 760 0000003Ch
20h 23h “Fast Speed Clock SCL Low Count (IC_FS_SCL_LCNT)—Offset 20h” on page 761 00000082h
2Ch 2Fh “Interrupt Status (IC_INTR_STAT)—Offset 2Ch” on page 761 00000000h
30h 33h “Interrupt Mask (IC_INTR_MASK)—Offset 30h” on page 763 0000005Fh
34h 37h “Raw Interrupt Status (IC_RAW_INTR_STAT)—Offset 34h” on page 764 00000000h
38h 3Bh “Receive FIFO Threshold Level (IC_RX_TL)—Offset 38h” on page 766 0000000Fh
3Ch 3Fh “Transmit FIFO Threshold Level (IC_TX_TL)—Offset 3Ch” on page 766 00000000h
40h 43h “Clear Combined and Individual Interrupt (IC_CLR_INTR)—Offset 40h” on page 767 00000000h
44h 47h “Clear RX_UNDER Interrupt (IC_CLR_RX_UNDER)—Offset 44h” on page 767 00000000h
48h 4Bh “Clear RX_OVER Interrupt (IC_CLR_RX_OVER)—Offset 48h” on page 768 00000000h
4Ch 4Fh “Clear TX_OVER Interrupt (IC_CLR_TX_OVER)—Offset 4Ch” on page 768 00000000h
50h 53h “Clear RD_REQ Interrupt (IC_CLR_RD_REQ)—Offset 50h” on page 769 00000000h
54h 57h “Clear TX_ABRT Interrupt (IC_CLR_TX_ABRT)—Offset 54h” on page 769 00000000h
5Ch 5Fh “Clear ACTIVITY Interrupt (IC_CLR_ACTIVITY)—Offset 5Ch” on page 770 00000000h
60h 63h “Clear STOP_DET Interrupt (IC_CLR_STOP_DET)—Offset 60h” on page 770 00000000h
64h 67h “Clear START_DET Interrupt (IC_CLR_START_DET)—Offset 64h” on page 771 00000000h
6Ch 6Fh “Enable (IC_ENABLE)—Offset 6Ch” on page 771 00000000h
70h 73h “Status (IC_STATUS)—Offset 70h” on page 772 00000006h
74h 77h “Transmit FIFO Level (IC_TXFLR)—Offset 74h” on page 773 00000000h
78h 7Bh “Receive FIFO Level (IC_RXFLR)—Offset 78h” on page 774 00000000h
7Ch 7Fh “SDA Hold (IC_SDA_HOLD)—Offset 7Ch” on page 774 00000001h
80h 83h “Transmit Abort Source (IC_TX_ABRT_SOURCE)—Offset 80h” on page 775 00000000h
9Ch 9Fh “Enable Status (IC_ENABLE_STATUS)—Offset 9Ch” on page 776 00000000h
A0h A3h “SS and FS Spike Suppression Limit (IC_FS_SPKLEN)—Offset A0h” on page 777 00000007h
19.5.1.1 Control Register (IC_CON)—Offset 0h

Can be written only when the I2C is disabled (IC_ENABLE==0). Writes while the I2C
controller is enabled have no effect.
Access Method

Default: 00000037h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 1 1
IC_10BITADDR_MASTER
SPEED
MASTER_MODE
RSVD0
RSVD1
IC_RESTART_EN
Bit Default &
Range Access
0b
RO
Restart Support (IC_RESTART_EN): Determines whether RESTART conditions may

be sent when acting as a master. Some older slaves do not support handling RESTART
conditions; however, RESTART conditions are used in several I2C controller operations.
0: disable
1: enable
When RESTART is disabled, the master is prohibited from performing the following
1b functions:
5 - Change direction within a transfer (split)
RW - Send a START BYTE
- Combined format transfers in 7-bit addressing modes
- Read operation with a 10-bit address
- Send multiple bytes per transfer By replacing RESTART condition followed by a STOP
and a subsequent START condition, split operations are broken down into multiple I2C
transfers. If the above operations are performed, it will result in setting TX_ABRT of the
IC_RAW_INTR_STAT register
Master Addressing Mode (IC_10BITADDR_MASTER): Controls whether the I2C

1b controller starts its transfers in 7- or 10-bit addressing mode when acting as a master.
4
RW 0: 7-bit addressing
1: 10-bit addressing
0b
3 RSVD1: Reserved
RO
Speed Mode (SPEED): I2C Master operational speed. Relevant only if Master is
11b enabled (MASTER_MODE==1).
2:1
RW 01: standard mode (100 kbit/s)
10: fast mode (400 kbit/s)
1b Master Mode Enable (MASTER_MODE): controls whether the I2C master is enabled.
0 0: master disabled
RW 1: master enabled
19.5.1.2 Master Target Address (IC_TAR)—Offset 4h

Can be written only when the I2C is disabled (IC_ENABLE==0). Writes while the I2C
controller is enabled have no effect.
Access Method
(Size: 32 bits)

Default: 00000055h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 1 0 1
IC_TAR
RSV
RSVD0
Bit Default &
Range Access
00000h
RO
0b
RO
Master Target Address (IC_TAR): This is the target address for any master
transaction. To generate a START BYTE, the CPU needs to write only once into these
055h bits.
9:0 If the IC_TAR and IC_SAR are the same, loopback exists but the FIFOs are shared
RW between master and slave, so full loopback is not feasible. Only one direction loopback
mode is supported (simplex), not duplex. A master cannot transmit to itself; it can
transmit to only a slave.
19.5.1.3 Data Buffer and Command (IC_DATA_CMD)—Offset 10h

CPU writes to it when filling the TX FIFO and the reads from when retrieving bytes from
RX FIFO.
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RESTART
DAT
RSV
STOP
CMD
Bit Default &

Range Access
0b
RO
Restart Bit Control (RESTART): This bit controls whether a RESTART is issued before
the byte is sent or received.
- 1 if IC_RESTART_EN is 1, a RESTART is issued before the data is sent/received
0b (according to the value of CMD), regardless of whether or not the transfer direction is
10 changing from the previous command; if IC_RESTART_EN is 0, a STOP followed by a
WO START is issued instead.
- 0 If IC_RESTART_EN is 1, a RESTART is issued only if the transfer direction is changing
from the previous command; if IC_RESTART_EN is 0, a STOP followed by a START is
issued instead

Bit Default &

Range Access
Stop Bit Control (STOP): This bit controls whether a STOP is issued after the byte is
sent or received:
- 1 STOP is issued after this byte, regardless of whether or not the Tx FIFO is empty. If
the Tx FIFO is not empty, the master immediately tries to start a new transfer by issuing
0b a START and arbitrating for the bus.
9
WO - 0 STOP is not issued after this byte, regardless of whether or not the Tx FIFO is empty.
If the Tx FIFO is not empty, the master continues the current transfer by sending/
receiving data bytes according to the value of the CMD bit. If the Tx FIFO is empty, the
master holds the SCL line low and stalls the bus until a new command is available in the
Tx FIFO.
Command (CMD): This bit controls whether a read or a write is performed. This bit
controls the direction only in I2C master mode.
0 = Write
1 = Read
0b When a command is entered in the TX FIFO, this bit distinguishes the write and read
8 commands. If a '1' is written to this bit after receiving a RD_REQ interrupt, then a
WO TX_ABRT interrupt occurs.
NOTE: It is possible that while attempting a master I2C read transfer, a RD_REQ
interrupt may have occurred simultaneously due to a remote I2C master addressingI2C
controller. In this type of scenario, the I2C controller ignores the IC_DATA_CMD write,
generates a TX_ABRT interrupt, and waits to service the RD_REQ interrupt.
Data Buffer (DAT): Contains the data to be transmitted or received on the I2C bus.
0b When writing to this register and want to perform a read, DAT field is ignored by the I2C
7:0 controller.
RW When reading this register, DAT return the value of data received on the I2C controller
interface.
19.5.1.4 Standard Speed Clock SCL High Count (IC_SS_SCL_HCNT)—Offset 14h

Sets the SCL clock high-period count for standard speed (SS). Can be written only
when the I2C is disabled (IC_ENABLE==0). Writes at other times have no effect.
Access Method
(Size: 32 bits)

Default: 00000190h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 0 0
IC_SS_SCL_HCNT
RSV
Bit Default &

Range Access
0b
RO

Bit Default &

Range Access
SS SCL clock high-period count (IC_SS_SCL_HCNT): Must be set before any I2C
bus transaction can take place to ensure proper I/O timing.
0190h The minimum valid value is 6; hardware prevents values less than this being written,
15:0 and if attempted results in 6 being set.
RW This register must not be programmed to a value higher than 65525, because I2C
controller uses a 16-bit counter to flag an I2C bus idle condition when this counter
reaches a value of IC_SS_SCL_HCNT + 10.
19.5.1.5 Standard Speed Clock SCL Low Count (IC_SS_SCL_LCNT)—Offset 18h

Sets the SCL clock low-period count for standard speed (SS). Can be written only when
the I2C is disabled (IC_ENABLE==0). Writes at other times have no effect.
Access Method

Default: 000001D6h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 0 1 1 0
RSV
IC_SS_SCL_LCNT
Bit Default &
Range Access
0b
RO
SS SCL clock low-period count (IC_SS_SCL_LCNT): Must be set before any I2C bus
01d6h transaction can take place to ensure proper I/O timing.
15:0
RW The minimum valid value is 8; hardware prevents values less than this being written,
and if attempted results in 8 being set.
19.5.1.6 Fast Speed Clock SCL High Count (IC_FS_SCL_HCNT)—Offset 1Ch

Sets the SCL clock high-period count for fast speed (FS). Can be written only when the
I2C is disabled (IC_ENABLE==0). Writes at other times have no effect.
Access Method

Default: 0000003Ch

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0
IC_FS_SCL_HCNT
RSV
Bit Default &
Range Access
0b
RO
FS SCL clock high-period count (IC_FS_SCL_HCNT): Must be set before any I2C
003ch bus transaction can take place to ensure proper I/O timing.
15:0
19.5.1.7 Fast Speed Clock SCL Low Count (IC_FS_SCL_LCNT)—Offset 20h

Sets the SCL clock low-period count for fast speed (FS). Can be written only when the
I2C is disabled (IC_ENABLE==0). Writes at other times have no effect.
Access Method
(Size: 32 bits)

Default: 00000082h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0
IC_FS_SCL_LCNT
RSV
Bit Default &

Range Access
0b
RO
FS SCL clock low-period count (IC_FS_SCL_LCNT): Must be set before any I2C bus
0082h transaction can take place to ensure proper I/O timing.
15:0
19.5.1.8 Interrupt Status (IC_INTR_STAT)—Offset 2Ch

Each bit in this register has a corresponding mask bit in the IC_INTR_MASK register.
These bits are cleared by reading the matching interrupt clear register. The unmasked
raw versions of these bits are available in the IC_RAW_INTR_STAT register

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
R_START_DET
R_STOP_DET
RSVD1
R_TX_OVER
R_RX_OVER
R_ACTIVITY
R_TX_ABRT
R_RX_FULL
R_RX_UNDER
RSV
R_RD_REQ
R_TX_EMPTY
Bit Default &
Range Access
0b
RO
0b
11 RSVD0: Reserved
RO
0b Start Detected (R_START_DET): Indicates whether a START or RESTART condition

10
RO has occurred on the I2C interface
0b Stop Detected (R_STOP_DET): Indicates whether a STOP condition has occurred on

9
RO the I2C interface.
Activity (R_ACTIVITY): This bit captures I2C controller activity and stays set until it is
cleared. There are four ways to clear it:
- Disabling the controller
0b - Reading the IC_CLR_ACTIVITY register
8 - Reading the IC_CLR_INTR register
RO - System reset
Once this bit is set, it stays set unless one of the four methods is used to clear it. Even if
the controller is idle, this bit remains set until cleared, indicating that there was activity
on the bus
0b
7 RSVD1: Reserved
RO
TX Abort (R_TX_ABRT): This bit indicates if the I2C controller, in transmitter mode, is
unable to complete the intended actions on the contents of the transmit FIFO. This
situation is referred to as a 'transmit abort'. When this bit is set to 1, the
IC_TX_ABRT_SOURCE register indicates the reason why the transmit abort takes
0b places.
6
RO
NOTE: The controller flushes/resets/empties the TX FIFO whenever this bit is set. The
TX FIFO remains in this flushed state until the register IC_CLR_TX_ABRT is read. Once
this read is performed, the TX FIFO is then ready to accept more data bytes for
transmission
Read Requested (R_RD_REQ): This bit is set to 1 when I2C controller is acting as a
slave and another I2C master is attempting to read data from it. The controller holds the
0b I2C bus in a wait state (SCL=0) until this interrupt is serviced, which means that the
5 slave has been addressed by a remote master that is asking for data to be transferred.
RO The processor must respond to this interrupt and then write the requested data to the
IC_DATA_CMD register. This bit is set to 0 just after the processor reads the
IC_CLR_RD_REQ register
TX Empty (R_TX_EMPTY): This bit is set to 1 when the transmit buffer is at or below
the threshold value set in the IC_TX_TL register. It is automatically cleared by hardware
0b when the buffer level goes above the threshold. When the IC_ENABLE bit 0 is 0, the TX
4
RO FIFO is flushed and held in reset. There the TX FIFO looks like it has no data within it, so
this bit is set to 1, provided there is activity in the master state machine. When there is
no longer activity, then with ic_en=0, this bit is set to 0.

Bit Default &

Range Access
TX Overflow (R_TX_OVER): Set during transmit if the transmit buffer is filled to 16

0b items and the processor attempts to issue another I2C command by writing to the
3
RO IC_DATA_CMD register. When the module is disabled, this bit keeps its level until the
master state machines goes into idle, and when ic_en goes to 0, this interrupt is cleared
RX Full (R_RX_FULL): Set when the receive buffer reaches or goes above the RX_TL
threshold in the IC_RX_TL register. It is automatically cleared by hardware when buffer
0b level goes below the threshold. If the module is disabled (IC_ENABLE[0]=0), the RX
2
RO FIFO is flushed and held in reset; therefore the RX FIFO is not full. So this bit is cleared
once the IC_ENABLE bit 0 is programmed with a 0, regardless of the activity that
continues
RX Overflow (R_RX_OVER): Set if the receive buffer is completely filled to 16 items

0b and an additional byte is received from an external I2C device. The I2C Controller
1 acknowledges this, but any data bytes received after the FIFO is full are lost. If the
RO module is disabled (IC_ENABLE[0]=0), this bit keeps its level until the master state
machines goes into idle, and when ic_en goes to 0, this interrupt is cleared.
RX Underflow (R_RX_UNDER): Set if the processor attempts to read the receive

0b buffer when it is empty by reading from the IC_DATA_CMD register. If the module is
0
RO disabled (IC_ENABLE[0]=0), this bit keeps its level until the master state machines
goes into idle, and when ic_en goes to 0, this interrupt is cleared.
19.5.1.9 Interrupt Mask (IC_INTR_MASK)—Offset 30h

These bits mask their corresponding interrupt status bits. They are active high; a value
of 0 prevents a bit from generating an interrupt.
Access Method

Default: 0000005Fh
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 1 1
M_START_DET
M_STOP_DET
M_ACTIVITY
M_TX_ABRT
RSV
M_TX_EMPTY
RSVD0
RSVD1
RSVD2
M_TX_OVER
M_RX_FULL
M_RX_OVER
M_RX_UNDER
Bit Default &
Range Access
0b
RO
0b
11 RSVD0: Reserved
RO
0b Start Detected Mask (M_START_DET): Indicates whether a START or RESTART

10
RW condition has occurred on the I2C interface
0b Stop Detected Mask (M_STOP_DET): Indicates whether a STOP condition has

9
RW occurred on the I2C interface.

Bit Default &

Range Access
Activity Mask (M_ACTIVITY): This bit captures I2C controller activity and stays set
until it is cleared. There are four ways to clear it:
RW - System reset
on the bus
0b
7 RSVD1: Reserved
RO
TX Abort Mask (M_TX_ABRT): This bit indicates if the I2C controller, in transmitter
mode, is unable to complete the intended actions on the contents of the transmit FIFO.
This situation is referred to as a 'transmit abort'. When this bit is set to 1, the
1b places.
6
RW
transmission
0b
5 RSVD2: Reserved
RO
TX Empty Mask (M_TX_EMPTY): This bit is set to 1 when the transmit buffer is at or
below the threshold value set in the IC_TX_TL register. It is automatically cleared by
1b hardware when the buffer level goes above the threshold. When the IC_ENABLE bit 0 is
4
RW 0, the TX FIFO is flushed and held in reset. There the TX FIFO looks like it has no data
within it, so this bit is set to 1, provided there is activity in the master state machine.
When there is no longer activity, then with ic_en=0, this bit is set to 0. Reset value
TX Overflow Mask (M_TX_OVER): Set during transmit if the transmit buffer is filled
1b to 16 items and the processor attempts to issue another I2C command by writing to the
3
RW IC_DATA_CMD register. When the module is disabled, this bit keeps its level until the
master state machine goes into idle, and when ic_en goes to 0, this interrupt is cleared
RX Full Mask (M_RX_FULL): Set when the receive buffer reaches or goes above the
RX_TL threshold in the IC_RX_TL register. It is automatically cleared by hardware when
1b buffer level goes below the threshold. If the module is disabled (IC_ENABLE[0]=0), the
2
RW RX FIFO is flushed and held in reset; therefore the RX FIFO is not full. So this bit is
cleared once the IC_ENABLE bit 0 is programmed with a 0, regardless of the activity
that continues
RX Overflow Mask (M_RX_OVER): Set if the receive buffer is completely filled to 16

1b items and an additional byte is received from an external I2C device. The I2C Controller
RW module is disabled (IC_ENABLE[0]=0), this bit keeps its level until the master state
machine goes into idle, and when ic_en goes to 0, this interrupt is cleared.
RX Underflow Mask (M_RX_UNDER): Set if the processor attempts to read the

1b receive buffer when it is empty by reading from the IC_DATA_CMD register. If the
0
RW module is disabled (IC_ENABLE[0]=0), this bit keeps its level until the master state
19.5.1.10 Raw Interrupt Status (IC_RAW_INTR_STAT)—Offset 34h

Unlike the IC_INTR_STAT register, these bits are not masked so they always show the
true status of the I2C controller
Access Method
(Size: 32 bits)


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RX_UNDER
RX_FULL
RSV
START_DET
STOP_DET
ACTIVITY
TX_ABRT
TX_EMPTY
TX_OVER
RX_OVER
RSVD0
RSVD1
RSVD2
Bit Default &
Range Access
0b
RO
0b
11 RSVD0: Reserved
RO
0b Start Detected (START_DET): Indicates whether a START or RESTART condition has

10
RO occurred on the I2C interface
0b Stop Detected (STOP_DET): Indicates whether a STOP condition has occurred on the
9
RO I2C interface.
Activity (ACTIVITY): This bit captures I2C controller activity and stays set until it is
cleared. There are four ways to clear it:
RO - System reset
on the bus
0b
7 RSVD1: Reserved
RO
TX Abort (TX_ABRT): This bit indicates if the I2C controller, in transmitter mode, is
unable to complete the intended actions on the contents of the transmit FIFO. This
situation is referred to as a 'transmit abort'. When this bit is set to 1, the
0b places.
6
RO
transmission
0b
5 RSVD2: Reserved
RO
TX Empty (TX_EMPTY): This bit is set to 1 when the transmit buffer is at or below the
threshold value set in the IC_TX_TL register. It is automatically cleared by hardware
0b when the buffer level goes above the threshold. When the IC_ENABLE bit 0 is 0, the TX
4
RO FIFO is flushed and held in reset. There the TX FIFO looks like it has no data within it, so
this bit is set to 1, provided there is activity in the master state machine. When there is
no longer activity, then with ic_en=0, this bit is set to 0.
TX Overflow (TX_OVER): Set during transmit if the transmit buffer is filled to 16

0b items and the processor attempts to issue another I2C command by writing to the
3
RO IC_DATA_CMD register. When the module is disabled, this bit keeps its level until the
master state machine goes into idle, and when ic_en goes to 0, this interrupt is cleared
RX Full (RX_FULL): Set when the receive buffer reaches or goes above the RX_TL
threshold in the IC_RX_TL register. It is automatically cleared by hardware when buffer
0b level goes below the threshold. If the module is disabled (IC_ENABLE[0]=0), the RX
2
RO FIFO is flushed and held in reset; therefore the RX FIFO is not full. So this bit is cleared
once the IC_ENABLE bit 0 is programmed with a 0, regardless of the activity that
continues

Bit Default &

Range Access
RX Overflow (RX_OVER): Set if the receive buffer is completely filled to 16 items and
0b an additional byte is received from an external I2C device. The I2C Controller
RO module is disabled (IC_ENABLE[0]=0), this bit keeps its level until the master state
RX Underflow (RX_UNDER): Set if the processor attempts to read the receive buffer
0b when it is empty by reading from the IC_DATA_CMD register. If the module is disabled
0
RO (IC_ENABLE[0]=0), this bit keeps its level until the master state machine goes into idle,
and when ic_en goes to 0, this interrupt is cleared.
19.5.1.11 Receive FIFO Threshold Level (IC_RX_TL)—Offset 38h

Controls the level of entries (or above) that triggers the RX_FULL interrupt (bit 2 in
IC_RAW_INTR_STAT register)
Access Method
(Size: 32 bits)

Default: 0000000Fh
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1
RX_TL
RSV
Bit Default &

Range Access
0b
RO
Receive FIFO Threshold Level (RX_TL): The valid range is 0-255, with the additional
0Fh restriction that hardware does not allow this value to be set to a value larger than the
7:0 depth of the buffer. If an attempt is made to do that, the actual value set will be the
RW maximum depth of the buffer. A value of 0 sets the threshold for 1 entry, and a value of
255 sets the threshold for 256 entries
19.5.1.12 Transmit FIFO Threshold Level (IC_TX_TL)—Offset 3Ch

Controls the level of entries (or below) that trigger the TX_EMPTY interrupt (bit 4 in
IC_RAW_INTR_STAT register).
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TX_TL
RSV
Bit Default &
Range Access
0b
RO
Transmit FIFO Threshold Level (TX_TL): The valid range is 0-255, with the
00h additional restriction that it may not be set to value larger than the depth of the buffer.
7:0 If an attempt is made to do that, the actual value set will be the maximum depth of the
RW buffer. A value of 0 sets the threshold for 0 entries, and a value of 255 sets the
threshold for 255 entries
19.5.1.13 Clear Combined and Individual Interrupt (IC_CLR_INTR)—Offset 40h

Read this register to clear the combined interrupt, all individual interrupts, and the
IC_TX_ABRT_SOURCE register.
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLR_INTR
RSV
Bit Default &

Range Access
0b
RO
0b Clear Combined and Individual Interrupt (CLR_INTR): This bit does not clear
0 hardware clearable interrupts but clears the software clearable interrupts. Refer to Bit 9
RO of the IC_TX_ABRT_SOURCE register for an exception to clearing IC_TX_ABRT_SOURCE
19.5.1.14 Clear RX_UNDER Interrupt (IC_CLR_RX_UNDER)—Offset 44h

Clear a single interrupt type
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLR_RX_UNDER
RSV
Bit Default &
Range Access
0b
RO
0b Clear RX_UNDER (CLR_RX_UNDER): Read this register to clear the RX_UNDER

0
RO interrupt (bit 0) of the IC_RAW_INTR_STAT
19.5.1.15 Clear RX_OVER Interrupt (IC_CLR_RX_OVER)—Offset 48h

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLR_RX_OVER
RSV
Bit Default &

Range Access
0b
RO
0b Clear RX_OVER (CLR_RX_OVER): Read this register to clear the RX_OVER interrupt
0
RO (bit 1) of the IC_RAW_INTR_STAT
19.5.1.16 Clear TX_OVER Interrupt (IC_CLR_TX_OVER)—Offset 4Ch

Access Method

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLR_TX_OVER
RSV
Bit Default &
Range Access
0b
RO
0b Clear TX_OVER (CLR_TX_OVER): Read this register to clear the TX_OVER interrupt
0
RO (bit 3) of the IC_RAW_INTR_STAT
19.5.1.17 Clear RD_REQ Interrupt (IC_CLR_RD_REQ)—Offset 50h

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
CLR_RD_REQ
Bit Default &
Range Access
0b
RO
0b Clear RD_REQ (CLR_RD_REQ): Read this register to clear the RD_REQ interrupt (bit
0
RO 5) of the IC_RAW_INTR_STAT
19.5.1.18 Clear TX_ABRT Interrupt (IC_CLR_TX_ABRT)—Offset 54h

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLR_TX_ABRT
RSV
Bit Default &
Range Access
0b
RO
Clear TX_ABRT (CLR_TX_ABRT): Read this register to clear the TX_ABRT interrupt
0b (bit 6) of the IC_RAW_INTR_STAT register, and the IC_TX_ABRT_SOURCE register. This
0 also releases the TX FIFO from the flushed/reset state, allowing more writes to the TX
RO FIFO. Refer to Bit 9 of the IC_TX_ABRT_SOURCE register for an exception to clearing
IC_TX_ABRT_SOURCE
19.5.1.19 Clear ACTIVITY Interrupt (IC_CLR_ACTIVITY)—Offset 5Ch

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLR_TX_ABRT
RSV
Bit Default &

Range Access
0b
RO
Clear ACTIVITY (CLR_TX_ABRT): Reading this register clears the ACTIVITY interrupt
0b if the I2C is not active anymore. If the I2C module is still active on the bus, the
0 ACTIVITY interrupt bit continues to be set. It is automatically cleared by hardware if the
RO module is disabled and if there is no further activity on the bus. The value read from this
register to get status of the ACTIVITY interrupt (bit 8) of the IC_RAW_INTR_STAT
19.5.1.20 Clear STOP_DET Interrupt (IC_CLR_STOP_DET)—Offset 60h

Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
CLR_STOP_DET
Bit Default &
Range Access
0b
RO
0b Clear STOP_DET (CLR_STOP_DET): Read this register to clear the STOP_DET

0
19.5.1.21 Clear START_DET Interrupt (IC_CLR_START_DET)—Offset 64h

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
CLR_START_DET
Bit Default &

Range Access
0b
RO
0b Clear START_DET (CLR_START_DET): Read this register to clear the START_DET

0
19.5.1.22 Enable (IC_ENABLE)—Offset 6Ch

Controls whether the I2C controller is enabled. Software can disable I2C controller
while it is active.

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ENABLE
RSV
Bit Default &
Range Access
0b
RO
Enable I2C Controller (ENABLE): 0: Disabled (TX/RX FIFOs are held in an erased
state)
1: Enabled
NOTE: ensure that the controller is disabled properly. When disabled, the following
occurs:
0b - The TX FIFO and RX FIFO get flushed.
0
RW - Status bits in the IC_INTR_STAT register are still active until the I2C Controller goes
into IDLE state.
If the module is transmitting, it stops as well as deletes the contents of the transmit
buffer after the current transfer is complete.
If the module is receiving, the controller stops the current transfer at the end of the
current byte and does not acknowledge the transfer.
There is a two I2C clocks delay when enabling or disabling the controller
19.5.1.23 Status (IC_STATUS)—Offset 70h

Read-only register used to indicate the current transfer status and FIFO status. The
status register may be read at any time. None of the bits in this register request an
interrupt. - When the I2C is disabled by writing 0 in bit 0 of the IC_ENABLE register:
bits 1 and 2 are set to 1, bits 3 and 4 are set to 0 - When the master or slave state
machines goes to idle and ic_en=0: bits 5 and 6 are set to 0
Access Method
(Size: 32 bits)

Default: 00000006h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0
MST_ACTIVITY
ACTIVITY
RSV
RFNE
TFE
RSVD0
RFF
TFNF

Bit Default &

Range Access
0b
RO
0b
6 RSVD0: Reserved
RO
Master FSM Activity Status (MST_ACTIVITY): When the Master Finite State
Machine (FSM) is not in the IDLE state, this bit is set.
0b 0: Master FSM is in IDLE state so the Master part is not Active
5
RO 1: Master FSM is not in IDLE state so the Master part is Active
NOTE: IC_STATUS[0]-that is, ACTIVITY bit-is the OR of SLV_ACTIVITY and
MST_ACTIVITY bits
Receive FIFO Completely Full (RFF): When the receive FIFO is completely full, this
0b bit is set. When the receive FIFO contains one or more empty location, this bit is
4 cleared.
RO 0: Receive FIFO is not full
1: Receive FIFO is full
Receive FIFO Not Empty (RFNE): This bit is set when the receive FIFO contains one
0b or more entries; it is cleared when the receive FIFO is empty.
3
RO 0: Receive FIFO is empty
1: Receive FIFO is not empty
Transmit FIFO Completely Empty (TFE): When the transmit FIFO is completely
1b empty, this bit is set. When it contains one or more valid entries, this bit is cleared. This
2 bit field does not request an interrupt.
RO 0: Transmit FIFO is not empty
1: Transmit FIFO is empty
Transmit FIFO Not Full (TFNF): Set when the transmit FIFO contains one or more
1b empty locations, and is cleared when the FIFO is full.
1
RO 0: Transmit FIFO is full
1: Transmit FIFO is not full
0b Activity (ACTIVITY): I2C Activity Status.

0 0: Not Active
RO 1: Active
19.5.1.24 Transmit FIFO Level (IC_TXFLR)—Offset 74h

Contains the number of valid data entries in the transmit FIFO buffer. It is cleared
whenever: - The I2C is disabled - There is a transmit abort that is, TX_ABRT bit is set in
the IC_RAW_INTR_STAT register - The slave bulk transmit mode is aborted. The
register increments whenever data is placed into the transmit FIFO and decrements
when data is taken from the transmit FIFO
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TXFLR
RSV

Bit Default &

Range Access
0b
RO
0b Transmit FIFO Level (TXFLR): Contains the number of valid data entries in the
4:0
RO transmit FIFO
19.5.1.25 Receive FIFO Level (IC_RXFLR)—Offset 78h

Read This register contains the number of valid data entries in the receive FIFO buffer.
It is cleared whenever: - The I2C is disabled - Whenever there is a transmit abort
caused by any of the events tracked in IC_TX_ABRT_SOURCE. The register increments
whenever data is placed into the receive FIFO and decrements when data is taken from
the receive FIFO
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RXFLR
RSV
Bit Default &

Range Access
0b
RO
0b Receive FIFO Level (RXFLR): Contains the number of valid data entries in the receive
4:0
RO FIFO
19.5.1.26 SDA Hold (IC_SDA_HOLD)—Offset 7Ch

This register controls the amount of hold time on the SDA signal after a negative edge
of SCL line in units of I2C clock period. The value programmed must be greater than
the minimum hold time in each mode for the value to be implemented: 1 cycle. Writes
to this register succeed only when I2C controller is disabled (IC_ENABLE=0)
Access Method

Default: 00000001h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
IC_SDA_HOLD
RSV
Bit Default &
Range Access
0b
RO
0001h SDA Hold (IC_SDA_HOLD): Sets the required SDA hold time in units of the I2C clock
15:0
RW period.
19.5.1.27 Transmit Abort Source (IC_TX_ABRT_SOURCE)—Offset 80h

Used to indicate the source of the TX_ABRT interrupt. Except for Bit 9, this register is
cleared whenever the IC_CLR_TX_ABRT register or the IC_CLR_INTR register is read.
To clear Bit 9, the source of the ABRT_SBYTE_NORSTRT must be fixed first; RESTART
must be enabled (IC_CON[5]=1), the SPECIAL bit must be cleared (IC_TAR[11]), or
the GC_OR_START bit must be cleared (IC_TAR[10]). Once the source of the
ABRT_SBYTE_NORSTRT is fixed, then this bit can be cleared in the same manner as
other bits in this register. If the source of the ABRT_SBYTE_NORSTRT is not fixed
before attempting to clear this bit, Bit 9 clears for one cycle and is then re-asserted
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ARB_LOST
ABRT_10B_RD_NORSTRT
ABRT_SBYTE_ACKDET
ABRT_7B_ADDR_NOACK
ABRT_SBYTE_NORSTRT
ABRT_TXDATA_NOACK
ABRT_10ADDR2_NOACK
ABRT_10ADDR1_NOACK
RSV
ABRT_MASTER_DIS
RSVD0
RSVD1
RSVD2
Bit Default &

Range Access
0b
RO
0b
RO
0b
12 Master Lost Arbitration (ARB_LOST): Set if master has lost arbitration
RO

Bit Default &

Range Access
0b Master Disabled (ABRT_MASTER_DIS): Set if user tries to initiate a Master

11
RO operation with the Master mode disabled
0b 10 Bit Address READ and RESTART Disabled (ABRT_10B_RD_NORSTRT): Set if

10 the restart is disabled (IC_RESTART_EN bit (IC_CON[5]) =0) and the master sends a
RO read command in 10-bit addressing mode
START With RESTART Disabled (ABRT_SBYTE_NORSTRT): To clear Bit 9, the

source of the ABRT_SBYTE_NORSTRT must be fixed first; restart must be enabled
(IC_CON[5]=1), the SPECIAL bit must be cleared (IC_TAR[11]), or the GC_OR_START
0b bit must be cleared (IC_TAR[10]). Once the source of the ABRT_SBYTE_NORSTRT is
9
RO fixed, then this bit can be cleared in the same manner as other bits in this register. If the
source of the ABRT_SBYTE_NORSTRT is not fixed before attempting to clear this bit, bit
9 clears for one cycle and then gets reasserted. Set if the restart is disabled
(IC_RESTART_EN bit (IC_CON[5]) =0) and the user is trying to send a START Byte
0b
8 RSVD1: Reserved
RO
0b START Acknowledged (ABRT_SBYTE_ACKDET): Set if master has sent a START

7
RO Byte and the START Byte was acknowledged (wrong behavior).
0b
6:4 RSVD2: Reserved
RO
0b TX Data Not Acknowledged (ABRT_TXDATA_NOACK): Set if master has received

3 an acknowledgement for the address, but when it sent data byte(s) following the
RO address, it did not receive an acknowledge from the remote slave(s).
0b 10 Bit Address Second Not Acknowledged (ABRT_10ADDR2_NOACK): Set if

2 master is in 10-bit address mode and the second address byte of the 10-bit address was
RO not acknowledged by any slave
0b 10 Bit Address First Not Acknowledged (ABRT_10ADDR1_NOACK): Set if master

1 is in 10-bit address mode and the first 10-bit address byte was not acknowledged by
RO any slave
0b 7 Bit Address Not Acknowledged (ABRT_7B_ADDR_NOACK): Set if master is in 7-

0
RO bit addressing mode and the address sent was not acknowledged by any slave
19.5.1.28 Enable Status (IC_ENABLE_STATUS)—Offset 9Ch

Report the I2C hardware status
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
RSVD0
IC_EN
Bit Default &

Range Access
0b
RO

Bit Default &

Range Access
0b
2:1 RSVD0: Reserved
RO
I2C Enable Status (IC_EN): When read as 1, the controller is deemed to be in an

0b enabled state. When read as 0, the controller is deemed completely inactive.
0
RO
NOTE: The CPU can safely read this bit anytime.
19.5.1.29 SS and FS Spike Suppression Limit (IC_FS_SPKLEN)—Offset A0h

Used to store the duration, measured in I2C clock cycles, of the longest spike that is
filtered out by the spike suppression logic when the component is operating in
Standard/Fast Speed modes. The relevant I2C requirement is detailed in the I2C Bus
Specification. This register must be programmed with a minimum value of 2.
Access Method
(Size: 32 bits)

Default: 00000007h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1
RSV
IC_FS_SPKLENRX_TL
Bit Default &
Range Access
0b
RO
07h I2C SS and FS Spike Length (IC_FS_SPKLENRX_TL): Must be set before any I2C
7:0
RW bus transaction can take place to ensure stable operation.
19.5.2 GPIO Controller Memory Mapped Registers

Offset Default
Start Value
0h 3h “Port A Data (GPIO_SWPORTA_DR)—Offset 0h” on page 778 00000000h
4h 7h “Port A Data Direction (GPIO_SWPORTA_DDR)—Offset 4h” on page 778 00000000h
30h 33h “Interrupt Enable (GPIO_INTEN)—Offset 30h” on page 779 00000000h
34h 37h “Interrupt Mask (GPIO_INTMASK)—Offset 34h” on page 779 00000000h


Offset Default
Start Value
38h 3Bh “Interrupt Type (GPIO_INTTYPE_LEVEL)—Offset 38h” on page 780 00000000h
3Ch 3Fh “Interrupt Polarity (GPIO_INT_POLARITY)—Offset 3Ch” on page 781 00000000h
40h 43h “Interrupt Status (GPIO_INTSTATUS)—Offset 40h” on page 781 00000000h
44h 47h “Raw Interrupt Status (GPIO_RAW_INTSTATUS)—Offset 44h” on page 782 00000000h
48h 4Bh “Debounce Enable (GPIO_DEBOUNCE)—Offset 48h” on page 782 00000000h
4Ch 4Fh “Clear Interrupt (GPIO_PORTA_EOI)—Offset 4Ch” on page 783 00000000h
50h 53h “Port A External Port (GPIO_EXT_PORTA)—Offset 50h” on page 784 00000000h
60h 63h “Synchronization Level (GPIO_LS_SYNC)—Offset 60h” on page 784 00000000h
19.5.2.1 Port A Data (GPIO_SWPORTA_DR)—Offset 0h

Contains the GPIO Port data bits
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_SWPORTA_DR
Bit Default &

Range Access
0b
RO
Port Data (GPIO_SWPORTA_DR): Values written to this register are output on the I/
0b O signals for if the corresponding data direction bits are set to Output mode and the
7:0
RW corresponding control bit for the Port is set to Software mode. The value read back is
equal to the last value written to this register
19.5.2.2 Port A Data Direction (GPIO_SWPORTA_DDR)—Offset 4h

Used to control the GPIO Port bits data direction
Access Method
(Size: 32 bits)


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
GPIO_SWPORTA_DDR
RSV
Bit Default &
Range Access
0b
RO
Port Data Direction (GPIO_SWPORTA_DDR): Values written to this register

0b independently control the direction of the corresponding data bit in the Port
7:0
RW - 0 Input (default)
- 1 Output
19.5.2.3 Interrupt Enable (GPIO_INTEN)—Offset 30h

Used to configured Port A bits as interrupt sources
Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_INTEN
Bit Default &

Range Access
0b
RO
Interrupt Enable (GPIO_INTEN): Allows each bit of Port A to be configured for

interrupts. By default the generation of interrupts is disabled. Whenever a 1 is written to
a bit of this register, it configures the corresponding bit on Port A to become an
0b interrupt; otherwise, Port A operates as a normal GPIO signal. Interrupts are disabled
7:0
RW on the corresponding bits of Port A if the corresponding data direction register is set to
Output.
0 Configure Port A bit as normal GPIO signal (default)
1 Configure Port A bit as interrupt
19.5.2.4 Interrupt Mask (GPIO_INTMASK)—Offset 34h

Controls masking for Port A bits configured as interrupt sources

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_INTMASK
Bit Default &
Range Access
0b
RO
Interrupt Mask (GPIO_INTMASK): Controls whether an interrupt on Port A can

create an interrupt for the interrupt controller by not masking it. By default, all
0b interrupts bits are unmasked. Whenever a 1 is written to a bit in this register, it masks
7:0 the interrupt generation capability for this signal; otherwise interrupts are allowed
RW through. The unmasked status can be read as well as the resultant status after masking.
0 Interrupt bits are unmasked (default)
1 Mask interrupt
19.5.2.5 Interrupt Type (GPIO_INTTYPE_LEVEL)—Offset 38h

Controls the type of interrupt associated with Port A bits configured as interrupt source
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_INTYPE_LEVEL
Bit Default &

Range Access
0b
31:8 RSV: Reserved.
RO

Bit Default &

Range Access
Interrupt Type (GPIO_INTYPE_LEVEL): Controls the type of interrupt that can

0b occur on Port A. Whenever a 0 is written to a bit of this register, it configures the
7:0 interrupt type to be level-sensitive; otherwise, it is edge-sensitive.
RW 0 Level-sensitive (default)
1 Edge-sensitive
19.5.2.6 Interrupt Polarity (GPIO_INT_POLARITY)—Offset 3Ch

Controls the interrupt polarity associated with Port A bits configured as interrupt
sources
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_INT_POLARITY
Bit Default &
Range Access
0b
RO
Interrupt Polarity (GPIO_INT_POLARITY): Controls the polarity of edge or level

sensitivity that can occur on input of Port A. Whenever a 0 is written to a bit of this
0b register, it configures the interrupt type to falling-edge or active-low sensitive;
7:0
RW otherwise, it is rising-edge or active-high sensitive.
0 Active-low (default)
1 Active-high
19.5.2.7 Interrupt Status (GPIO_INTSTATUS)—Offset 40h

Stores the interrupt status after masking for Port A bits configured as interrupt sources
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_INTSTATUS
Bit Default &
Range Access
0b
RO
0b Interrupt Status (GPIO_INTSTATUS): After mask. See GPIO_RAW_INTSTATUS for

7:0
RW raw interrupt values and GPIO_INTMASK for interrupt mask configuration
19.5.2.8 Raw Interrupt Status (GPIO_RAW_INTSTATUS)—Offset 44h

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
GPIO_RAW_INTSTATUS
RSV
Bit Default &

Range Access
0b
RO
0b Raw Interrupt Status (GPIO_RAW_INTSTATUS): Raw interrupt of status of Port A

7:0
RW (premasking bits)
19.5.2.9 Debounce Enable (GPIO_DEBOUNCE)—Offset 48h

Controls the debounce logic associated to a Port A bit configured as interrupt source
Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_DEBOUNCE
Bit Default &
Range Access
0b
RO
Debounce Enable (GPIO_DEBOUNCE): Controls whether an external signal that is

the source of an interrupt needs to be debounced to remove any spurious glitches.
0b Writing a 1 to a bit in this register enables the debouncing circuitry. A signal must be
7:0
RW valid for two periods of an external clock before it is internally processed.
0 No debounce (default)
1 Enable debounce
19.5.2.10 Clear Interrupt (GPIO_PORTA_EOI)—Offset 4Ch

Controls edge-type interrupt clearing
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
GPIO_PORTA_EOI
RSV
Bit Default &

Range Access
0b
RO

Bit Default &

Range Access
Clear Interrupt (GPIO_PORTA_EOI): Controls the clearing of edge type interrupts

0b from Port A. When a 1 is written into a corresponding bit of this register, the interrupt is
7:0 cleared. All interrupts are cleared when Port A is not configured for interrupts.
RW 0 No interrupt clear (default)
1 Clear interrupt
19.5.2.11 Port A External Port (GPIO_EXT_PORTA)—Offset 50h

Used by the software to read values from the GPIO Port bits
Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_EXT_PORTA
Bit Default &
Range Access
0b
RO
0b External Port (GPIO_EXT_PORTA): When the Port is configured as Input, then

7:0 reading this location reads the values on the external signal. When the data direction is
RW set as Output, reading this location reads the Port data register contents
19.5.2.12 Synchronization Level (GPIO_LS_SYNC)—Offset 60h

Controls if a level-sensitive interrupt type need to be synchronized to the system clock
Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
GPIO_LS_SYNC
Bit Default &
Range Access
0b
RO
Synchronization Level (GPIO_LS_SYNC): Writing a 1 to this register results in all

0b level-sensitive interrupts being synchronized to the system clock.
0
RW 0 Not Synchronized (default)
1 Synchronized
§§


SPI Interface—Intel® Quark™ SoC X1000
20.0 SPI Interface
The Intel® Quark™ SoC X1000 implements two SPI controllers that support master
mode.


Table 125. SPI Interface Signals

Direction/
Type
O SPI Serial Clock

SPI[0/1]_SCK
CMOS3.3
O SPI Slave Select

SPI[[0/1]_SS_B
CMOS3.3 SPI Slave Select is active low.
O SPI Master Output Slave Input

SPI[0/1]_MOSI
CMOS3.3
I SPI Slave Output Master Input

SPI[0/1]_MISO
CMOS3.3
20.2 Features
Compliant with the Motorola* Serial Peripheral Interface:
• supports master mode only
• supports one slave select output
• supports MSB first transfer only
• supports SCLK frequencies up to 25 MHz
• does not support slave mode operation
• does not support SPI boot
• does not support wait mode
20.2.1 SPI Controller

The SPI unit provides one-channel, 4-wire serial input/output interface to directly
connect SPI-compatible devices.

Intel® Quark™ SoC X1000—SPI Interface
Four pins are used to transfer data between the CPU and external device:
• SCLK defines the bit rate at which serial data is driven onto, and sampled from, the
bus;
• SS_B or CS defines the boundaries of a basic data “unit”, comprised of multiple
serial bits.
• MOSI or SDO is the serial data path for outbound data, from system to peripheral
• MISO or SDI is the serial data path for inbound data, from peripheral to system
Serial data is transferred between the system and an external peripheral through FIFOs
in the SPI. Transfers are initiated by the host processor and data is transferred over a
single transfer. Operation is full duplex. Separate FIFOs and serial data paths allow
simultaneous transfers in both directions to/from the external peripheral. Transfer is
started when either new data is available in the transmit FIFO or memory is available in
the receive FIFO. Transfer is terminated when either new data is not available in the
transmit FIFO or memory is not available in the receive FIFO.
Figure 47. Generic SPI Waveform
20.2.1.1 Processor-Initiated Data Transfer

Transmit data (from system to peripheral) is written by the host processor to the SPI
Transmit FIFO. The FIFO is seen as one 32-bit location by the processor and a write to
the FIFO takes the form of a single 32-bit write transaction to the data register (SSDR).
The SPI moves the data from the Transmit FIFO, serializes it, and sends it over the
serial wire (SDO) to the peripheral. Receive data from the peripheral (on SDI) is
converted to parallel words and stored in the Receive FIFO register file. When passed, a
programmable “fullness” threshold triggers an interrupt to the Interrupt Controller (and
subsequently, if enabled, to an interrupt input to the CPU). The interrupt service
routine responds by identifying the source of the interrupt and then doing a read from
the Receive FIFO. The SPI differentiates between the two FIFOs according to whether
the transfer is a READ or a WRITE transfer. Read bursts automatically target the
Receive FIFO, while write bursts write data to the Transmit FIFO. From a memory map
point of view, they are at the same address. FIFOs are 32DWORDS deep or 128 Bytes
in aggregate.
20.2.1.2 Data Format

Data in the FIFOs is always stored with one sample per 32-bit word regardless of the
format’s data word length. Within each 32-bit field, the stored data sample is right
justified, with the word’s LSB in bit 0, and unused bits packed as zeroes on the left-
hand (MSB) side. Logic in the SPI guarantees that data is properly transmitted on SDO
according to the selected frame format and MSB first.

20.2.1.3 FIFO Operation
20.2.1.3.1 Processor-Initiated Data Transfers
There are two separate and independent FIFOs for “incoming” (from peripheral) and
“outgoing” (to peripheral) serial data. FIFOs are filled or emptied by single transfers or
SRAM-like bursts initiated by the system processor. Both FIFO’s are the same size and
arranged as 32-word positions that are a maximum of 32-bits wide. FIFO word width is
programmable from 4 to 32 bits. When FIFO width is programmed to less than a 32-bit
width, only the programmed numbers of bits are available in all word positions.
Each FIFO consists of a dual-port register file with control circuitry to make it work as a
FIFO, with independent read and write ports. Single FIFO write bursts may be between
1 and 32 words in length, and between 4 and 32 bits per word thus transferring from 4
bits up to 1024 data samples per burst. Also continuous operation is possible by
keeping the transmit FIFO loaded with data for much larger data transfers.
FIFO filling and emptying may be performed by the system processor in response to an
interrupt from the FIFO logic. Each FIFO has a programmable threshold at which an
interrupt is triggered. When the threshold value is exceeded, an interrupt is generated
which, if enabled, signals the host processor to empty an “inbound” FIFO or to refill an
“outbound” FIFO.
The system can also poll status bits to learn how full a FIFO is.
20.2.1.4 Baud Rate Generation

The SCLK is generated in the SPI unit from the 200 MHz input system clock (sys_clk)
according to register configuration.
Two clock synthesis stages are implemented to achieve different SPI baud-rates:
• an internal clock (clk_ssp) is generated from the input reference clock based on the
value of SPI_DDS_RATE.DDS_CLK_RATE[23:0].clk_ssp is used in the SPI unit to
clock the logic interfacing the SPI link and to generate SCLK. The relationship
between sys_clk and ssp_clk frequencies is expressed by
fclk_ssp = fsys_clk (DDS_CLK_RATE/224)
• clk_ssp is used to generate SCLK. The relationship between ssp_clk and SCLK
frequencies is expressed by
fSCLK = fclk_ssp /(2*(SCR+1))
Table 126 lists a subset of the possible register configurations and relative SCLK
frequencies. While the table is not exhaustive of all possible register settings, the
following guidelines should be considered:
• To guarantee negligible jitter and duty cycle deviation from 50% it is recommended
to not set DDS_CLK_RATE to values greater than h33333 and not listed in the
table.
• A duty cycle of 50% and virtually no jitter is guaranteed for any setting where
DDS_CLK_RATE is one hot.
• For a given SCLK frequency not listed in Table 126, best jitter and duty cycle values
are obtained by setting the lower possible value of DDS_CLK_RATE.
• There is no limitation for SCR values.

Table 126. SPI Clock Frequency Settings (Sheet 1 of 2)

SCLK SCLK Duty
DDS_CLK_RATE[23:0] SCLK_INT SCR[7:0]
Frequency Frequency Cycle
(hex) Frequency (dec)
(MHz) (KHz) (%)
800000 100.000 0 25.000 50
800000 100.000 2 16.667 50
800000 100.000 4 10.000 50
666666 80.000 0 40.000 60-40
666666 80.000 1 20.000 50
666666 80.000 2 13.333 54-46
666666 80.000 4 8.000 52-48
666666 80.000 9 4.000 50
666666 80.000 19 2.000 50
666666 80.000 49 800.000 50
666666 80.000 99 400.000 50
666666 80.000 199 200.000 50
400000 50.000 0 25.000 50
400000 50.000 3 6.250 50
400000 50.000 4 5.000 50
400000 50.000 9 2.500 50
400000 50.000 24 1.000 50
400000 50.000 49 500.000 50
400000 50.000 99 250.000 50
200000 25.000 0 12.500 50
200000 25.000 4 2.500 50
200000 25.000 9 1.250 50
200000 25.000 19 625.000 50
200000 25.000 24 500.000 50
200000 25.000 49 250.000 50
200000 25.000 124 100.000 50
100000 12.500 0 6.250 50
100000 12.500 49 125.000 50
100000 12.500 124 50.000 50
80000 6.250 0 3.125 50
80000 6.250 124 25.000 50
40000 3.125 0 1.563 50
20000 1.563 0 781.250 50
20000 1.563 77 10.016 50
20000 1.563 154 5.040 50

Table 126. SPI Clock Frequency Settings (Sheet 2 of 2)

SCLK SCLK Duty
DDS_CLK_RATE[23:0] SCLK_INT SCR[7:0]
Frequency Frequency Cycle
(hex) Frequency (dec)
(MHz) (KHz) (%)
10000 0.781 0 390.625 50
8000 0.391 0 195.313 50
8000 0.391 194 1.002 50
20.3 Register Map

Figure 48. SPI Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
PCI
CAM
(I/O)
Bus 0 SPI PCI Memory
PCI Header Space
ECAM D:21,F:0-1
(Mem)
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
BAR
D:21
SPI1 F:1
I2C*/GPIOF:2
SPI Mem
D:31, F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7


Registers listed are for Function 0 (SPI Controller 0). Function 1 (SPI Controller 1)
contains the same registers. Differences between SPI Controllers Ports will be noted in
individual registers.

Default
Value

86h 86h 00h
page 803

Access Method


(Size: 16 bits)
Default: 8086h
15 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0
value
Bit Default &
Description
Range Access
8086h
RO

Access Method
Default: 0935h
15 12 8 4 0
0 0 0 0 1 0 0 1 0 0 1 1 0 1 0 1
value
Bit Default &

Description
Range Access
0935h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
IntrDis
RSVD
RSVD
MasEn
MEMen
RSVD
SERREn
Bit Default &

Description
Range Access
0h
RO

10

Bit Default &

Description
Range Access
0h
RO
8
00h
RO

2
RW requests.
1
0h
RO

Access Method
(Size: 16 bits) STATUS: [B:0, D:21, F:0] + 6h
Default: 0010h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
SigSysErr
RcdMasAb
RSVD
RSVD
RSVD
IntrStatus
capable_66Mhz
RSVD0
DEVSEL
FastB2B
hasCapList
RSVD1
Bit Default &
Description
Range Access
0h
RO
14
13
0h
RO
0b
RO
0h
RO
0b
RO
0h
RO
0b
RO

Bit Default &

Description
Range Access

4
RO register sets.
RO used
0h
RO

Access Method
Default: 0C800010h
31 28 24 20 16 12 8 4 0
0 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0
subClassCode
rev_id
classCode
Bit Default & progIntf

Description
Range Access
31: 24
RO performs.
23: 16
15: 8
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
7: 0

Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
cfgHdrFormat
multiFnDev
Bit Default &

Description
Range Access

7
RO function device

6: 0

Access Method
(Size: 8 bits)
Default: 00h

7 4 0
0 0 0 0 0 0 0 0
start_bist
RSVD
BIST_capable
comp_code
Bit Default &
Description
Range Access

7
0h
RO
0h
RO

3: 0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
prefetchable
memType
address
RSVD
isIO
Bit Default &
Description
Range Access
31: 12
00h
RO

Hardwired to 0
00b
RO

0
RO address decoder


Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 16 bits)
Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO


30h
Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ROM_base_addr
RSVD
AddrDecodeEn
Bit Default &
Description
Range Access

31: 11
000h
RO

RW also set to 1

Access Method
(Size: 32 bits)
Default: 00000080h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
RSVD0
value
Bit Default &

Description
Range Access
0h
RO

Access Method


(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
vector information.

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
0h
RO


Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &
Description
Range Access
0h
RO

Access Method
(Size: 8 bits)
Default: 01h
7 4 0
0 0 0 0 0 0 0 1
value
Bit Default &

Description
Range Access
7: 0

Access Method
(Size: 8 bits) PM_NXT_CAP_PTR: [B:0, D:21, F:0] + 81h
Default: A0h
7 4 0
1 0 1 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0


Access Method
(Size: 16 bits)
Default: 4803h
15 12 8 4 0
0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 1
D2_support
D1_support
DSI
aux_curr
PME_support
RSVD
PME_clock
version
Bit Default &
Description
Range Access
15: 11
0h
RO
0h
RO
0h
RO
0h
RO
0h
RO
2: 0

84h
Access Method
(Size: 16 bits)
Default: 0008h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
PME_status
Data_scale
power_state
Data_select
no_soft_reset
PME_en
RSVD
RSVD

Bit Default &

Description
Range Access
15
0h
RO
0h
RO
8
0h
RO

3
0h
RO
1: 0

Offset 86h
Access Method
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access

7: 0
RO to 0.

87h
Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value

Bit Default &

Description
Range Access
0h
RO

Access Method
(Size: 8 bits) MSI_CAP_ID: [B:0, D:21, F:0] + A0h
Default: 05h
7 4 0
0 0 0 0 0 1 0 1
value
Bit Default &
Description
Range Access
7: 0

Access Method
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
value
Bit Default &

Description
Range Access
7: 0
RO in the chain

Access Method
(Size: 16 bits) MESSAGE_CTRL: [B:0, D:21, F:0] + A2h
Default: 0100h

15 12 8 4 0
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
perVecMskCap
multiMsgEn
bit64Cap
multiMsgCap
MSIEnable
RSVD0
Bit Default &
Description
Range Access
0h
RO

8
7

6: 4

3: 1
RW MSI.

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
address
Bit Default &

Description
Range Access
31: 2
0h
RO

Access Method
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MsgData
Bit Default &
Description
Range Access
15: 0
changed by hardware

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
MSIMask
Bit Default &
Description
Range Access
0h
RO
0

Access Method
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
value

Bit Default &

Description
Range Access
0h
RO
0h
RO

Offset Default
Start Value
0h 3h “SPI Control Register 0 (SSCR0)—Offset 0h” on page 807 00000000h
4h 7h “SPI Control Register 1 (SSCR1)—Offset 4h” on page 808 00000000h
8h Bh “SPI Status Register (SSSR)—Offset 8h” on page 810 0003E004h
10h 13h “SPI Data Register (SSDR)—Offset 10h” on page 811 00000000h
28h 2Bh “DDS Clock Rate Register (DDS_RATE)—Offset 28h” on page 812 00028F5Ch
20.5.1 SPI Control Register 0 (SSCR0)—Offset 0h

The SPI control register 0 controls various SPI functions.
Access Method
Type: Memory Mapped I/O Register SSCR0: [BAR0] + 0h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SCR
FRF
RSV
SSE
DSS
Bit Default &

Description
Range Access
0b
31:16 RSV: Reserved.
RO
Serial Clock Rate (SCR): This bitfield is used to select the baud, or bit rate, of the SPI.
A total of 256 different bit rates can be selected to further divide the frequency of the
clock obtained by dividing the input system clock according to the DDS_RATE register,
0b see DDS_RATE Register description. The resultant clock is driven on the SCLK pin and is
15:8 used by the SPIs transmit logic to drive data on the MOSI pin, and to latch data on the
RW MISO pin. The SCLK frequency is given by
DDS_FREQ /(2 x (SCR + 1))
where SCR is a decimal integer in the range 0 to 255 and DDS_FREQ is determined by
the value programmed in the DDS_RATE Register.

Bit Default &

Description
Range Access
Synchronous Serial Port Enable (SSE): The SPI enable bit is used to enable and
disable all SPI operations. When the SPI is disabled, all of its clocks are powered down
to minimize power consumption. Note that the SSE is the only control bit within the SPI
which is reset to a known state. It is cleared to zero to ensure the SPI is disabled
following a reset. When the SSE bit is cleared during active operation, the SPI is
0b disabled immediately, causing the current frame being transmitted to be terminated.
7 Clearing SSE resets the SPIs FIFOs and the SPI status bits. However, the SPIs control
RW registers and the Receive FIFO Overrun (ROR) status bit are not reset.
Note: After reset or after clearing the SSE, the user must ensure the SSCR1 and SSSR
registers are properly reconfigured or reset before re-enabling the SPI with the SSE;
other control bits in SSCR0 may be written at the same time as the SSE.
0 : SPI operation disabled
1 : SPI operation enabled
Frame Format (FRF): 00 : Motorola* Serial Peripheral Interface (SPI)

0b 01 : Reserved, undefined operation
6:5
RW 10 : Reserved, undefined operation
11 : Reserved, undefined operation
Data Size Select (DSS): The 5-bit data size select field is used to select the size of the
data transmitted and received by the SPI. Data can be 4 to 32 bits in length. When data
is programmed to be less than 16 bits, received data is automatically right justified and
the upper bits in the receive FIFO are zero-filled by receive logic. Transmit data should
not be left justified by the user before being placed in the transmit FIFO; transmit logic
in the SPI will automatically left justify the data sample according to the value of DSS
before the sample is transmitted on MOSI. Although it is possible to program data sizes
of 1, 2, and 3 bits, these sizes are reserved and produce unpredictable results in the
SPI.
00011 : 4-bit data
00100 : 5-bit data
00101 : 6-bit data
00110 : 7-bit data
00111 : 8-bit data
01000 : 9-bit data
01001 : 10-bit data
0b 01010 : 11-bit data
4:0 01011 : 12-bit data
RW 01100 : 13-bit data
01101 : 14-bit data
01110 : 15-bit data
01111 : 16-bit data
10000 : 17-bit data
10001 : 18-bit data
10010 : 19-bit data
10011 : 20-bit data
10100 : 21-bit data
10101 : 22-bit data
10110 : 23-bit data
10111 : 24-bit data
11000 : 25-bit data
11001 : 26-bit data
11010 : 27-bit data
11011 : 28-bit data
11100 : 29-bit data
11101 : 30-bit data
11110 : 31-bit data
11111 : 32-bit data
20.5.2 SPI Control Register 1 (SSCR1)—Offset 4h

The SPI Control Register 1 controls various SPI functions.
Access Method

Type: Memory Mapped I/O Register SSCR1: [BAR0] + 4h

(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV1
RSV0
LBM
STRF
EFWR
RFT
TFT
SPO
SPH
TIE
RIE
Bit Default &
Description
Range Access
00h
31:18 RSV1: Reserved.
RO
Select FIFO for Enable FIFO Write/Read (STRF): This bit selects whether the
0b Transmit or Receive FIFO is enabled for writes and reads whenever the EFWR is
17 programmed to one, which puts the SPI in a special functional mode.
RW 0 : Transmit FIFO is selected for both writes and reads through the SSDR.
1 : Receive FIFO is selected for both writes and reads through the SSDR
Enable FIFO Write/Read Function (EFWR): This bit enables a special functional
mode for the SPI.
0 : the SPI operates in normal mode.
1 : the SPI enters a mode in which whenever the CPU reads or writes to the SPI Data
0b register it actually reads and writes exclusively to either the Transmit FIFO or the
16
RW Receive FIFO depending on the programmed state of the Select FIFO for EFWR (STRF)
bit. In this special mode, data will not be transmitted on the MOSI pin and data input on
the MISO pin will not be stored. This mode can be used to test, through software,
whether or not the Transmit FIFO or the Receive FIFO operates properly as a first-in-
first-out memory stack.
0b Receive FIFO Interrupt Threshold (RFT): The receive FIFO interrupt threshold sets
15:11 the threshold at or above which the FIFO controller triggers, if enabled, a CPU interrupt
RW request. This level should be set to the desired threshold value minus 1.
0b Transmit FIFO Interrupt Threshold (TFT): The transmit FIFO interrupt threshold
10:6 sets the threshold at or below which the FIFO controller triggers, if enabled, a CPU
RW interrupt request. This level should be set to the desired threshold value minus 1.
0b
5 RSV0: Reserved.
RO
Serial Clock Phase (SPH): The serial clock (SCLK) phase bit determines the phase
relationship between the SCLK and the Chip Select (CS) pins.
0 : SCLK remains in its inactive/idle state (as determined by the SPO setting) for one full
cycle after CS is asserted low at the beginning of a frame. SCLK continues to transition
for the rest of the frame and is then held in its inactive state for one-half of an SCLK
period before CS is deasserted high at the end of the frame.
1 : SCLK remains in its inactive/ idle state (as determined by the SPO setting) for half a
cycle after CS is asserted low at the beginning of a frame. SCLK continues to transition
0b for the rest of the frame and is then held in its inactive state for one full SCLK period
4 before CS is de-asserted high at the end of the frame.
RW The combination of the SPO and SPH settings determines when SCLK is active during
the assertion of CS and which SCLK edge is used to transmit and receive data on the
MOSI and MISO pins.
When SPO and SPH are programmed to the same value (both 0 or 1), transmit data is
driven on the falling edge of SCLK and receive data is latched on the rising edge of
SCLK.
When SPO and SPH are programmed to opposite values (one 0 and the other 1),
transmit date is driven on the rising edge of SCLK and receive data is latched on the
falling edge of SCLK.

Bit Default &

Description
Range Access
Serial Clock Polarity (SPO): The serial clock (SCLK) polarity bit selects the polarity of
the inactive state of the SCLK pin.
0b 0 : the SCLK is held low in the inactive or idle state when the SPI is not transmitting/
3 receiving data.
RW 1 : the SCLK is held high during the inactive/idle state.
The programmed setting of the SPO alone does not determine which SCLK edge is used
to transmit or receive data. The SPO setting in combination with SPH determines this.
Loop Back Mode (LBM): The loop back mode bit is used to enable and disable the
ability of the SPI transmit and receive logic to communicate.
0b 0 : the SPI operates normally. The transmit and receive data paths are independent and
2
RW communicate via their respective pins.
1 : the output of the transmit serial shifter is directly connected to the input of the
receive serial shifter internally.
Transmit FIFO Interrupt Enable (TIE): The Transmit FIFO Interrupt Enable bit is
used to mask or enable the transmit FIFO service request interrupt.
0 : the interrupt is masked and the state of TFS within the SPI Status Register is ignored
0b by the interrupt controller.
1 1 : the interrupt is enabled, and whenever TFS is set to one, the interrupt request is
RW made to the interrupt controller.
Note that programming TIE=0 does not affect the current state of TFS or the transmit
FIFO logics ability to set and clear TFS, it only blocks the generation of the interrupt
request.
Receive FIFO Interrupt Enable (RIE): The Receive FIFO Interrupt Enable bit is used
to mask or enable the Receive FIFO service request interrupt.
0 : the interrupt is masked, and the state of RFS within the SPI Status Register is
0b ignored by the interrupt controller.
0 1 : the interrupt is enabled, and whenever RFS is set to one, the interrupt request is
RW made to the interrupt controller.
Note that programming RIE=0 does not affect the current state of RFS or the receive
FIFO logics ability to set and clear RFS, it only blocks the generation of the interrupt
request.
20.5.3 SPI Status Register (SSSR)—Offset 8h

The SPI status register contains bits which signal overrun errors as well as transmit and
receive FIFO service requests. Each of these hardware-detected events signals an
interrupt request to the interrupt controller. The status register also contains flags that
indicate when the SPI is actively transmitting characters, when the transmit FIFO is not
full, and when the receive FIFO is not empty (no interrupt generated). Bits which cause
an interrupt will signal the request as long as the bit is set. Once the bit is cleared, the
interrupt is cleared. Read/write bits are called status bits; read-only bits are called
flags. Status bits are referred to as sticky (once set by hardware, must be cleared by
software). Writing a one to a sticky status bit clears it, writing a zero has no effect.
Read-only flags are set and cleared by hardware; writes have no effect. Additionally,
some bits which cause interrupts have corresponding mask bits in the control registers.
Access Method
(Size: 32 bits) SSSR: [BAR0] + 8h

Default: 0003E004h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1 0 0
RFL
TFL
ROR
TNF
RSV
RFS
TFS
BSY
RNE
ALT_FRM

Bit Default &

Description
Range Access
00h
RO
1Fh Receive FIFO Level (RFL): This 5-bit value shows how many valid entries are
17:13
RO currently in the Receive FIFO (up to 32).
0b Transmit FIFO Level (TFL): This 5-bit value shows how many valid entries are
12:8
RO currently in the Transmit FIFO (up to 32).
Receiver Overrun Status (ROR): The receiver overrun status bit is a read/write bit
which is set when the receive logic attempts to place data into the receive FIFO after it
has been completely filled. Each time a new piece of data is received, the set signal to
0b the ROR bit is asserted, and the newly received data is discarded. This process is
7 repeated for each new piece of data received until at least one empty FIFO entry exists.
RW/C When the ROR bit is set, an interrupt request is made to the CPU which cannot be locally
masked by any SPI register bit. Writing a 1 to this bit resets ROR status and its interrupt
request; writing a 0 to this bit does not affect ROR status. Receiver Overrun Status is a
non-maskable interrupt
Receive FIFO Service Request Flag (RFS): The receive FIFO service request flag is a
read-only bit which is set when the receive FIFO is nearly filled and requires service to
prevent an overrun. RFS is set any time the receive FIFO has the same or more entries
0b of valid data than indicated by the Receive FIFO Threshold, and it is cleared when it has
6
RO fewer entries than the threshold value. When the RFS bit is set, an interrupt request is
made unless RIE is cleared. After the CPU reads the FIFO such that it has fewer entries
than the RFT value, the RFS flag (and the service request and/or interrupt) is
automatically cleared.
Transmit FIFO Service Request Flag (TFS): The Transmit FIFO service request flag
is a read-only bit which is set when the transmit FIFO is nearly empty and requires
service to prevent an underrun. TFS is set any time the transmit FIFO has the same or
0b fewer entries of valid data than indicated by the Transmit FIFO Threshold, and it is
5
RO cleared when it has more entries of valid data than the threshold value. When the TFS
bit is set, an interrupt request is made unless TIE is cleared. After the CPU fills the FIFO
such that it exceeds the threshold, the TFS flag (and the service request and/or
interrupt) is automatically cleared.
SPI Busy Flag (BSY): The receive FIFO not empty flag is a read-only bit which is set
whenever the receive FIFO contains one or more entries of valid data and is cleared
0b when it no longer contains any valid data. This bit can be polled when using
4
RO programmed I/O to remove remaining bytes of data from the receive FIFO since CPU
interrupt requests are only made when the Receive FIFO Threshold has been met or
exceeded. This bit does not request an interrupt.
Receive FIFO Not Empty Flag (RNE): The receive FIFO not empty flag is a read-only
bit which is set whenever the receive FIFO contains one or more entries of valid data
0b and is cleared when it no longer contains any valid data. This bit can be polled when
3
RO using programmed I/O to remove remaining bytes of data from the receive FIFO since
interrupt requests are only made when the Receive FIFO Threshold has been met or
exceeded. This bit does not request an interrupt.
Transmit FIFO Not Full Flag (TNF): The transmit FIFO not full flag is a read-only bit
1b which is set whenever the transmit FIFO contains one or more entries which do not
2 contain valid data. TNF is cleared when the FIFO is completely full. This bit can be polled
RO when using programmed I/O to fill the transmit FIFO over its threshold level. This bit
does not request an interrupt.
0b Alternative Frame (ALT_FRM): This field is not supported and should be treated as
1:0
RO reserved.
20.5.4 SPI Data Register (SSDR)—Offset 10h

The SPI Data Register is one 32-bit location that can be accessed by 32-bit data
transfers. Transfers can be single word transfer or 2 to 32 -word bursts. The SSDR
represents two physical registers: the first is temporary storage for data on its way out
through the Transmit FIFO; the other is temporary storage for data coming in through
the Receive FIFO. As the register is accessed by the system, FIFO control logic transfers

data automatically between register and FIFO as fast as the system moves it. Data in
the FIFO shifts up or down to accommodate the new word (unless it is an attempted
write to a full Transmit FIFO). For outbound data transfers (write from system to SPI
peripheral), the register may be loaded (written) by the system processor anytime it is
below its threshold level. When a data size of less than 32-bits is selected, the user
should not left-justify data written to the transmit FIFO. Transmit logic left-justifies the
data and ignores any unused bits. Received data less than 32-bits is automatically right
justified in the receive buffer.
Access Method
Type: Memory Mapped I/O Register SSDR: [BAR0] + 10h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Bit Default & SSDR Description

Range Access
00h
31:0 SPI Data (SSDR): Data to be written to/read from Transmit/Receive FIFO
RW
20.5.5 DDS Clock Rate Register (DDS_RATE)—Offset 28h

The DDS Clock Rate Register determines the SCLK frequency, along with SCR field of
SSCR0 register. This register determines the frequency DDS_FREQ of an internal clock
used to clock part of the SPI core logic and to drive the SPI output clock (SCLK). Note
that SCLK frequency is not the same as this clock, but is a divided version according to
SCR bit in SSCR0 register.
Access Method
(Size: 32 bits) DDS_RATE: [BAR0] + 28h

Default: 00028F5Ch
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 1 1 1 1 0 1 0 1 1 1 0 0
DDS_CLK_RATE
RSV

Bit Default &

Description
Range Access
0b
RO
DDS Clock Rate (DDS_CLK_RATE): DDS_CLK_RATE is a 24 bit value that is

incrementally added to a 24 bit wrapping internal counter to cause the clock output to
028F5Ch toggle. The frequency of the internal generated clock is given by:
23:0 DDS_FREQ = SYS_FREQ*(DDS_CLK_RATE/2exp24)
RW where SYS_FREQ is 200MHz.
NOTE: in order to minimize jitter and duty cycle on the generated clock, this register
should be set as described in the SPI section of the design specification document.
§§


Legacy Bridge—Intel® Quark™ SoC X1000
21.0 Legacy Bridge
The Legacy Bridge is a collection of hardware blocks that are critical to implement a PC/
AT compatible platform. Certain legacy hardware functions are required to support
commercially available, shrink-wrap operating systems. In addition, the Legacy Bridge
provides interrupt decoding and routing functionality, power management features, and
a SPI interface for system firmware.
21.1 Features
The key features of the various blocks are as follows:
• General Purpose Input Output
— Legacy control interface for SoC GPIOs
— I/O mapped registers
• 8259 Programmable Interrupt Controller
— Legacy interrupt support
— 15 total interrupts through two cascaded controllers
• I/O Advanced Programmable Interrupt Controller
— Legacy-free interrupt support
— 24 total interrupts
— Memory mapped registers
• 8254 Programmable Interval Timer
— Legacy timer support
— Three timers with fixed uses: System Timer, Refresh Request Signal, and
Speaker Tone
• HPET - High Performance Event Timers
— Legacy-free timer support
— Three timers and one counter
• Real-Time Clock (RTC)
— 242-byte RAM backed by battery (aka CMOS RAM)
— Can generate wake/interrupt when time matches programmed value
— I/O and indexed registers
• Watchdog Timer (WDT)
— Provides ability to trigger a reset in the event of an unresponsive system
— Resolution from 1sec to 17 minutes

Intel® Quark™ SoC X1000—Legacy Bridge

• Serial Peripheral Interface (SPI)
— Support for one SPI Flash, of up to 16 Mbyte, only. No other SPI peripherals are
supported.
— Stores boot FW and system configuration data
— Supports SPI clock frequency of 20 MHz.
• Power Management Controller (PMC)
— Provides a software interface to control many of the power management
features present in the SoC. See the Intel® Quark SoC X1000 UEFI Firmware
Writer’s Guide (Order #330236) for more information.
The following features are not supported by the Legacy Bridge:
• LPC Interface
• Serial IRQ (SERIRQ)
• SMBus
Table 129. Miscellaneous Legacy Signals

Direction/
Type
External System Management Interrupt:

I
SMI_B This signal is typically generated by the external system management
CMOS3.3
controller.
I Thermal Alarm:
THRM_B
CMOS3.3 Generated by external hardware to cause an SMI_B/SCI (if enabled).
21.2 Register Map


Figure 49. Legacy Bridge Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
Memory
Space
PCI
CAM
RCRB Mem
(I/O)
Registers
Bus 0
PCI
ECAM
(Mem) Legacy PCI
Header IO Space
D:31,F:0
PM1 IO
RP0 F:0
PCIe*
Registers
D:23
SPI0 F:0 RP0 F:1

IO Fabric
RCBA GPE0 IO
D:21
SPI1 F:1 Registers

PM1BLK
I2C*/GPIO F:2 GPIO IO
GPE0BLK Registers
Legacy Bridge GBA

WDT IO
D:31,F:0 WDTBA Registers
SDIO/eMMC F:0
Fixed IO
HSUART0 F:1 Registers
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7

Default
Offset Size Register ID—Description
Value
“PCI Device ID and Vendor ID Fields (PCI_DEVICE_VENDOR)—Offset 0h” on

0h 4 095E8086h
page 818
“PCI Status and Command Fields (PCI_STATUS_COMMAND)—Offset 4h” on

4h 4 00000003h
page 818
“PCI Class Code and Revision ID Fields (PCI_CLASS_REVISION)—Offset 8h” on

8h 4 06010000h
page 819
Ch 4 “PCI Miscellaneous Fields (PCI_MISC)—Offset Ch” on page 819 00000000h


Default
Offset Size Register ID—Description
Value
“PCI Subsystem ID and Subsystem Vendor ID Fields (PCI_SUBSYSTEM)—Offset

2Ch 4 00000000h
2Ch” on page 820
44h 4 “GPIO Base Address (GBA)—Offset 44h” on page 821 00000000h
48h 4 “PM1_BLK Base Address (PM1BLK)—Offset 48h” on page 821 00000000h
4Ch 4 “GPE0_BLK Base Address (GPE0BLK)—Offset 4Ch” on page 821 00000000h
58h 4 “ACPI Control (ACTL)—Offset 58h” on page 822 00000003h
“PIRQA, PIRQB, PIRQC and PIRQD Routing Control (PABCDRC)—Offset 60h” on

60h 4 80808080h
page 822
“PIRQE, PIRQF, PIRQG and PIRQH Routing Control (PEFGHRC)—Offset 64h” on

64h 4 80808080h
page 824
84h 4 “Watch Dog Timer Base Address (WDTBA)—Offset 84h” on page 824 00000000h
D4h 4 “BIOS Decode Enable (BCE)—Offset D4h” on page 825 FF000000h
D8h 4 “BIOS Control (BC)—Offset D8h” on page 826 00000100h
F0h 4 “Root Complex Base Address (RCBA)—Offset F0h” on page 827 00000000h
21.3.1 PCI Device ID and Vendor ID Fields (PCI_DEVICE_VENDOR)—

Offset 0h
Access Method
(Size: 32 bits)
Default: 095E8086h
31 28 24 20 16 12 8 4 0
0 0 0 0 1 0 0 1 0 1 0 1 1 1 1 0 1 0 0 0 0 0 0 0
VENDOR_ID 1 0 0 0 0 1 1 0
DEVICE_ID
Bit Default &

Description
Range Access
095Eh
31:16 Device ID (DEVICE_ID): PCI Device ID
RO
8086h
15:0 Vendor ID (VENDOR_ID): PCI Vendor ID for Intel
RO
21.3.2 PCI Status and Command Fields (PCI_STATUS_COMMAND)—

Offset 4h
Access Method
Type: PCI Configuration Register PCI_STATUS_COMMAND: [B:0, D:31, F:0] + 4h
(Size: 32 bits)
Default: 00000003h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1
STATUS
COMMAND
Bit Default &
Description
Range Access
0000h
31:16 Status (STATUS): Hardwired to 0.
RO
0003h
15:0 Command (COMMAND): Hardwired to 0.
RO
21.3.3 PCI Class Code and Revision ID Fields

(PCI_CLASS_REVISION)—Offset 8h
Access Method
(Size: 32 bits) PCI_CLASS_REVISION: [B:0, D:31, F:0] + 8h
Default: 06010000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CLASS_CODE
REVISION_ID
Bit Default &
Description
Range Access
060100h
31:8 Class Code (CLASS_CODE): PCI Class Code for Bridge
RO
00h
7:0 Revision ID (REVISION_ID): PCI Revision ID
RO
21.3.4 PCI Miscellaneous Fields (PCI_MISC)—Offset Ch

Access Method
Type: PCI Configuration Register PCI_MISC: [B:0, D:31, F:0] + Ch
(Size: 32 bits)
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
BIST
HEADER
LATENCY
CACHE_LINE_SIZE
Bit Default &
Description
Range Access
00h
31:24 BIST: PCI BIST Field
RO
00h
23:16 Header Type (HEADER): PCI Header Type Field
RO
00h
15:8 Latency Timer (LATENCY): PCI Latency Timer Field
RO
00h
7:0 Cache Line Size (CACHE_LINE_SIZE): PCI Cache Line Size Field
RO

(PCI_SUBSYSTEM)—Offset 2Ch
This register is initialized to logic 0 by the assertion of RESET#. This register can be
written only once after RESET# de-assertion.
Access Method
Type: PCI Configuration Register PCI_SUBSYSTEM: [B:0, D:31, F:0] + 2Ch
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
SUBSYSTEM_VENDOR_ID
SUBSYSTEM_ID
Bit Default &

Description
Range Access
0000h Subsystem Vendor ID (SUBSYSTEM_VENDOR_ID): This is written by BIOS. No

31:16
RW/O hardware action taken.
0000h
15:0 Subsystem ID (SUBSYSTEM_ID): This is written by BIOS. No hardware action taken.
RW/O

21.3.6 GPIO Base Address (GBA)—Offset 44h

Access Method
Type: PCI Configuration Register GBA: [B:0, D:31, F:0] + 44h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EN
RSV2
RSV1
BA
Bit Default &
Description
Range Access
0b
31 Enable (EN): When set, decode of the I/O range pointed to by the BA is enabled.
RW
0b
RO
0b
15:7 Base Address (BA): Provides the 128 bytes of I/O space for GPIO.
RW
0b
RO
21.3.7 PM1_BLK Base Address (PM1BLK)—Offset 48h

Access Method
Type: PCI Configuration Register PM1BLK: [B:0, D:31, F:0] + 48h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RSV1
BA
EN
Bit Default &

Description
Range Access
0b
RW
0b
RO
0b
15:4 Base Address (BA): Provides the 16 bytes of I/O space for PM1_BLK.
RW
0b
RO
21.3.8 GPE0_BLK Base Address (GPE0BLK)—Offset 4Ch

Access Method
Type: PCI Configuration Register GPE0BLK: [B:0, D:31, F:0] + 4Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
BA
EN
RSV2
RSV1
Bit Default &
Description
Range Access
0b
RW
0b
RO
0b
15:6 Base Address (BA): Provides the 64 bytes of I/O space for GPE0_BLK.
RW
0b
RO
21.3.9 ACPI Control (ACTL)—Offset 58h

Access Method
Type: PCI Configuration Register ACTL: [B:0, D:31, F:0] + 58h
(Size: 32 bits)
Default: 00000003h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1
RSV
SCIS
Bit Default &
Description
Range Access
0b
RO
SCI IRQ Select (SCIS): Specifies to which IRQ the SCI is routed. If not using APIC,
SCI must be routed to IRQ9-11, and that interrupt is not shared with SERIRQ, but is
shared with other interrupts. If using APIC, SCI can be mapped to IRQ20-23, and can be
shared with other interrupts.
000b : IRQ9
001b : IRQ10
011b 010b : IRQ11
2:0 011b : SCI Disabled
RW 100b : IRQ20
101b : IRQ21
110b : IRQ22
111b : IRQ23
When the interrupt is mapped to APIC interrupts 9, 10 or 11, APIC must be programmed
for active-high reception. When the interrupt is mapped to APIC interrupt 20 through
23, PIC must be programmed for active-low reception.
21.3.10 PIRQA, PIRQB, PIRQC and PIRQD Routing Control (PABCDRC)—

Offset 60h
Access Method

Type: PCI Configuration Register PABCDRC: [B:0, D:31, F:0] + 60h

(Size: 32 bits)
Default: 80808080h
31 28 24 20 16 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
RSVC
RENC
IRC
RSVD
RSVB
IRD
IRB
RENB
RSVA
REND
IRA
RENA
Bit Default &
Description
Range Access
1b Interrupt Routing Enable for PIRQD (REND): When cleared, PIRQD is routed to one
31 of the legacy interrupts specified in bits[27:24]. When set, PIRQD is not routed to the
RW 8259.
0b
RO
IRQ for PIRQD (IRD): Indicates how to route PIRQD.

0h - Reserved
1h - Reserved
2h - IRQ2
3h - IRQ3
4h - IRQ4
5h - IRQ5
0b 6h - IRQ6
27:24 7h - IRQ7
RW 8h - IRQ8
9h - IRQ9
Ah - IRQ10
Bh - IRQ11
Ch - IRQ12
Dh - Reserved
Eh - IRQ14
Fh - IRQ15
1b Interrupt Routing Enable for PIRQC (RENC): When cleared, PIRQC is routed to one
23 of the legacy interrupts specified in bits[19:16]. When set, PIRQC is not routed to the
RW 8259.
0b
22:20 Reserved (RSVC): Reserved.
RO
0b
19:16 IRQ for PIRQC (IRC): Indicates how to route PIRQC.
RW
1b Interrupt Routing Enable for PIRQB (RENB): When cleared, PIRQB is routed to one
15 of the legacy interrupts specified in bits[11:8]. When set, PIRQB is not routed to the
RW 8259.
0b
14:12 Reserved (RSVB): Reserved.
RO
0b
11:8 IRQ for PIRQB (IRB): Indicates how to route PIRQB.
RW
1b Interrupt Routing Enable for PIRQA (RENA): When cleared, PIRQA is routed to one
7 of the legacy interrupts specified in bits[3:0]. When set, PIRQA is not routed to the
RW 8259.
0b
6:4 Reserved (RSVA): Reserved.
RO
0b
3:0 IRQ for PIRQA (IRA): Indicates how to route PIRQA.
RW

21.3.11 PIRQE, PIRQF, PIRQG and PIRQH Routing Control (PEFGHRC)—

Offset 64h
Access Method
Type: PCI Configuration Register PEFGHRC: [B:0, D:31, F:0] + 64h
(Size: 32 bits)
Default: 80808080h
31 28 24 20 16 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
RSVE
RSVH
IRE
IRH
RENE
RENH
RSVG
IRG
RENG
RSVF
IRF
RENF
Bit Default &
Description
Range Access
1b Interrupt Routing Enable for PIRQH (RENH): When cleared, PIRQH is routed to one
31 of the legacy interrupts specified in bits[27:24]. When set, PIRQH is not routed to the
RW 8259.
0b
30:28 Reserved (RSVH): Reserved.
RO
0b
27:24 IRQ for PIRQH (IRH): Indicates how to route PIRQH.
RW
1b Interrupt Routing Enable for PIRQG (RENG): When cleared, PIRQG is routed to one
23 of the legacy interrupts specified in bits[19:16]. When set, PIRQG is not routed to the
RW 8259.
0b
22:20 Reserved (RSVG): Reserved.
RO
0b
19:16 IRQ for PIRQG (IRG): Indicates how to route PIRQG.
RW
1b Interrupt Routing Enable for PIRQF (RENF): When cleared, PIRQF is routed to one
15 of the legacy interrupts specified in bits[11:8]. When set, PIRQF is not routed to the
RW 8259.
0b
14:12 Reserved (RSVF): Reserved.
RO
0b
11:8 IRQ for PIRQF (IRF): Indicates how to route PIRQF.
RW
1b Interrupt Routing Enable for PIRQE (RENE): When cleared, PIRQE is routed to one
7 of the legacy interrupts specified in bits[3:0]. When set, PIRQE is not routed to the
RW 8259.
0b
6:4 Reserved (RSVE): Reserved.
RO
0b
3:0 IRQ for PIRQE (IRE): Indicates how to route PIRQE.
RW
21.3.12 Watch Dog Timer Base Address (WDTBA)—Offset 84h

Access Method
Type: PCI Configuration Register WDTBA: [B:0, D:31, F:0] + 84h
(Size: 32 bits)
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
BA
EN
RSV2
RSV1
Bit Default &
Description
Range Access
0b
RW
0b
RO
0b
15:6 Base Address (BA): Provides the 64 bytes of I/O space for WDT.
RW
0b
RO
21.3.13 BIOS Decode Enable (BCE)—Offset D4h

Access Method
(Size: 32 bits) BCE: [B:0, D:31, F:0] + D4h
Default: FF000000h
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
EF8
EF0
EE8
EE0
ED8
ED0
EC8
EC0
Bit Default &

Description
Range Access
F8-FF Enable (EF8): Enables decoding of BIOS range FFF80000h - FFFFFFFFh and
1b FFB80000h - FFBFFFFFh.
31
RO 0 = Disable
1 = Enable
F0-F0 Enable (EF0): Enables decoding of BIOS range FFF00000h - FFF7FFFFh and
1b FFB00000h - FFB7FFFFh.
30
RW 0 = Disable
1 = Enable
E8-EF Enable (EE8): Enables decoding of BIOS range FFE80000h - FFEFFFFFh and
1b FFA80000h - FFAFFFFFh.
29
RW 0 = Disable
1 = Enable
E0-E8 Enable (EE0): Enables decoding of BIOS range FFE00000h - FFE7FFFFh and
1b FFA00000h - FFA7FFFFh.
28
RW 0 = Disable
1 = Enable
D8-DF Enable (ED8): Enables decoding of BIOS range FFD80000h - FFDFFFFFh and
1b FF980000h - FF9FFFFFh.
27
RW 0 = Disable
1 = Enable
D0-D8 Enable (ED0): Enables decoding of BIOS range FFD00000h - FFD7FFFFh and
1b FF900000h - FF97FFFFh.
26
RW 0 = Disable
1 = Enable

Bit Default &

Description
Range Access
C8-CF Enable (EC8): Enables decoding of BIOS range FFC80000h - FFCFFFFFh and
1b FF880000h - FF8FFFFFh.
25
RW 0 = Disable
1 = Enable
C0-C8 Enable (EC0): Enables decoding of BIOS range FFC00000h - FFC7FFFFh and
1b FF800000h - FF87FFFFh.
24
RW 0 = Disable
1 = Enable
0b
RO
21.3.14 BIOS Control (BC)—Offset D8h

Access Method
Type: PCI Configuration Register BC: [B:0, D:31, F:0] + D8h
(Size: 32 bits)
Default: 00000100h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
RSV3
RSV2
RSV1
SMM_WPD
PFE
CD
LE
WPD
Bit Default &
Description
Range Access
0b
RO
Prefetch Enable (PFE): When set, BIOS prefetching is enabled. An access to BIOS
1b causes a 64-byte fetch of the line starting at that region. Subsequent accesses within
8 that region result in data being returned from the prefetch buffer. The prefetch buffer is
RW invalidated when this bit is cleared, or a BIOS access occurs to a different line than what
is currently in the buffer
0b
RO
SMM Write Protect Disable (SMM_WPD): When LE is clear: Setting this bit has no
effect.
0b When LE is set: Setting this bit enables both read and write cycles to the SPI Flash,
5 clearing this bit blocks write cycles to the SPI Flash.
RW
This bit is not writeable unless the processor is in SMM mode.
This bit must be cleared before the processor exits SMM mode to prevent write cycles to
SPI flash when the processor is in a non-SMM mode.
0b
RO
0b
2 Cache Disable (CD): Enable caching in read buffer for direct memory read.
RW
Lock Enable (LE): When cleared, setting the WPD bit will not generate SMIs and the
WPD bit is used to enable write cycles to the SPI Flash.
0b When set, enables setting the WPD bit to generate SMIs and the SMM_WPD bit is used
1
RW/P to enable write cycles to the SPI Flash.
Once set, this bit can only be cleared by a reset.

Bit Default &

Description
Range Access
Write Protect Disable (WPD): When LE is clear: Setting this bit enables both read
and write cycles into the SPI Flash, clearing this bit blocks write cycles to the SPI Flash.
0b
0 When LE is set: Setting this bit will generate an SMI, the SMM_WPD bit must then be
RW used to enable write cycles to the SPI Flash.
This bit must be cleared before the processor exits SMM mode in order to clear the SMI
source.
21.3.15 Root Complex Base Address (RCBA)—Offset F0h

Access Method
Type: PCI Configuration Register RCBA: [B:0, D:31, F:0] + F0h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EN
BA
RSV
Bit Default &
Description
Range Access
0b Base Address (BA): Base Address for the root complex register block decode range.
31:14
RW This address is aligned on a 16 KB boundary.
0b
RO
0b
0 Enable (EN): When set, enables the range specified in BA to be claimed as the RCRB.
RW
21.4.1 Root Complex Register Block

This block describes all registers and base functionality that are related to SoC
configuration but not a specific interface. It contains the root complex register block.
Accesses in this space are limited to 32-bit quantities. Burst accesses are not allowed.
Table 131. Summary of Memory Mapped I/O Registers—RCBA

Offset Default
Start Value
0h 3h “Root Complex Topology Capabilities List (RCTCL)—Offset 0h” on page 828 00010005h
4h 7h “Element Self Description (ESD)—Offset 4h” on page 828 00000102h
3140h 3141h “Interrupt Queue Agent 0 (IRQAGENT0)—Offset 3140h” on page 829 0000h
3400h 3403h “Root Complex Topology Capabilities List (RCTCL)—Offset 0h” on page 828 00000000h

21.4.1.1 Root Complex Topology Capabilities List (RCTCL)—Offset 0h

Access Method
Type: Memory Mapped I/O Register RCTCL: [RCBA] + 0h
(Size: 32 bits)

Default: 00010005h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1
NEXTCAP
CAPVER
CAPID
Bit Default &
Description
Range Access
0b
31:20 Next Capability (NEXTCAP): Indicates next item in the list
RO
1h
19:16 Capability Version (CAPVER): Indicates the version of the capability structure.
RO
0005h Capability ID (CAPID): Indicates this is a PCI Express link capability section of an
15:0
RO RCRB
21.4.1.2 Element Self Description (ESD)—Offset 4h

Provides information about the root complex element containing the Link Declaration
Capability
Access Method
(Size: 32 bits) ESD: [RCBA] + 4h

Default: 00000102h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0
NUMLE
RSV
ETYPE
PORTNUM
COMPID
Bit Default &

Description
Range Access
0b
31:24 Port Number (PORTNUM): A value of 0 to indicate the egress port
RO
0b Component ID (COMPID): Indicates the component ID assigned to this element by

23:16
RW/O software. This is written once by BIOS and is locked until a reset.
01h Number of Link Entries (NUMLE): Indicates that one link entry is described by this
15:8
RO RCRB.

Bit Default &

Description
Range Access
0b
RO
2h
3:0 Element Type (ETYPE): Indicates that the element type is a root complex internal link
RO
21.4.1.3 Interrupt Queue Agent 0 (IRQAGENT0)—Offset 3140h

Access Method
Type: Memory Mapped I/O Register IRQAGENT0: [RCBA] + 3140h
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV_RW
INTAPR
RSV
Bit Default &

Description
Range Access
0b
RO
0b
7:4 Reserved (RSV_RW): Reserved.
RW
0b Interrupt A Pin Route (INTAPR): Indicates which PIRQ routing used for INTA#. Legal
3:0 values are [0x0-0x7] corresponding to PIRQ[A-H]. For example, INTA# will be routed to
RW PIRQG if this field is set to 0x6.

Access Method
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
INTDPR
INTCPR
INTBPR
INTAPR

Bit Default &

Description
Range Access
0b Interrupt D Pin Route (INTDPR): Indicates which PIRQ routing used for INTD#.
15:12 Legal values are [0x0-0x7] corresponding to PIRQ[A-H]. For example, INTD# will be
RW routed to PIRQG if this field is set to 0x6.
0b Interrupt C Pin Route (INTCPR): Indicates which PIRQ routing used for INTC#. Legal
11:8 values are [0x0-0x7] corresponding to PIRQ[A-H]. For example, INTC# will be routed to
0b Interrupt B Pin Route (INTBPR): Indicates which PIRQ routing used for INTB#. Legal
7:4 values are [0x0-0x7] corresponding to PIRQ[A-H]. For example, INTB# will be routed to

Access Method
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 RSV_RW 0 0 0 0 0 0
INTAPR
RSV
Bit Default &

Description
Range Access
0b
RO
0b
7:4 Reserved (RSV_RW): Reserved.
RW

Access Method
(Size: 16 bits)

Default: 0000h

15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
INTDPR
INTCPR
INTBPR
INTAPR
Bit Default &
Description
Range Access
0b Interrupt D Pin Route (INTDPR): Indicates which PIRQ routing used for INTD#.
15:12 Legal values are [0x0-0x7] corresponding to PIRQ[A-H]. For example, INTD# will be
RW routed to PIRQG if this field is set to 0x6.
0b Interrupt C Pin Route (INTCPR): Indicates which PIRQ routing used for INTC#. Legal
11:8 values are [0x0-0x7] corresponding to PIRQ[A-H]. For example, INTC# will be routed to
0b Interrupt B Pin Route (INTBPR): Indicates which PIRQ routing used for INTB#. Legal
7:4 values are [0x0-0x7] corresponding to PIRQ[A-H]. For example, INTB# will be routed to
21.4.1.7 RTC Configuration (RC)—Offset 3400h

Access Method
(Size: 32 bits) RC: [RCBA] + 3400h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
UPR_LOCK
LWR_LOCK
RSVRW
Bit Default &
Description
Range Access
0b
RO
0b
2 Reserved (RSVRW): Reserved.
RW
0b Upper 128 Byte Lock (UPR_LOCK): When set, bytes 38h-3Fh in the upper 128 byte
1 bank of RTC RAM are locked. Writes will be dropped and reads will not return any
RW/O guaranteed data.
0b Lower 128 Byte Lock (LWR_LOCK): When set, bytes 38h-3Fh in the lower 128 byte
0 bank of RTC RAM are locked. Writes will be dropped and reads will not return any
RW/O guaranteed data.

21.5 IO Registers
The Legacy Bridge contains a mix of fixed address I/O Registers and I/O Registers that
are mapped by BARs in the Legacy Bridge configuration space. This sections describes
the Fixed I/O registers and the Legacy ACPI I/O Register.s All other I/O Registers are
described in the relevant sections later in this chapter.
21.5.1 Fixed IO Registers
Table 132. Summary of I/O Registers

Offset Default
Start Value
61h 61h “NMI Status and Control Register (NSC)—Offset 61h” on page 832 00h
70h 70h “NMI Enable and RTC Index Register (NMIE)—Offset 70h” on page 833 80h
B2h B2h “Software SMI Control Port (SWSMICTL)—Offset B2h” on page 833 00h
B3h B3h “Software SMI Status Port (SWSMISTS)—Offset B3h” on page 834 00h
CF9h CF9h “Reset Control Register (RSTC)—Offset CF9h” on page 834 00h
21.5.1.1 NMI Status and Control Register (NSC)—Offset 61h

Access Method
Type: I/O Register NSC: 61h
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
SPKR_ENABLE
SERR_NMI_STATUS
RSVD
RSVD
CNTR2_STATUS
SERR_NMI_ENABLE
CNTR2_ENABLE
CNTR1_TOGGLE_STATUS
Bit Default &

Description
Range Access
0b SERR# NMI Status (SERR_NMI_STATUS): Set on errors from a PCIe port or internal
7 functions that generate SERR#. SERR# NMI Enable in this register must be cleared in
RO order for this bit to be set. To reset the interrupt, set bit 2 to 1 and then set it to 0.
0b
RO
0b Timer Counter 2 Status (CNTR2_STATUS): Reflects the current state of the 8254
5 counter 2 output. Counter 2 must be programmed for this bit to have a determinate
RO value.
0b Refresh Cycle Toggle Status (CNTR1_TOGGLE_STATUS): Reflects the current state

4
RO of 8254 counter 1.
0b
RO

Bit Default &

Description
Range Access
0b SERR# NMI Enable (SERR_NMI_ENABLE): When set, SERR# NMIs are disabled.
2
RW When cleared, SERR# NMIs are enabled.
0b
1 Reserved (SPKR_ENABLE): Reserved
RW
0b Timer Counter 2 Enable (CNTR2_ENABLE): When cleared, counter 2 counting is

0
RW disabled. When set, counting is enabled.
21.5.1.2 NMI Enable and RTC Index Register (NMIE)—Offset 70h

Access Method
Type: I/O Register NMIE: 70h
(Size: 8 bits)
Default: 80h
7 4 0
1 0 0 0 0 0 0 0
NMI_ENABLE
RTC_INDEX
Bit Default &

Description
Range Access
1b NMI Enable (NMI_ENABLE): When set, NMI sources disabled. When cleared, NMI
7
WO sources enabled.
0b Real Time Clock Index (RTC_INDEX): Selects RTC register or CMOS RAM address to
6:0
WO access.
21.5.1.3 Software SMI Control Port (SWSMICTL)—Offset B2h

Access Method
Type: I/O Register SWSMICTL: B2h
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
CONTROL
Bit Default &

Description
Range Access
0b Software SMI Control Port (CONTROL): This port is used to pass a command
7:0 between the OS and the SMI handler. Writes to this port store data, set APM bit of SMI
RW Status register of GPE0 Block, and generate SMI_B when APM is set.

21.5.1.4 Software SMI Status Port (SWSMISTS)—Offset B3h

Access Method
Type: I/O Register SWSMISTS: B3h
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
STATUS
Bit Default &
Description
Range Access
0b Software SMI Status Port (STATUS): This port is used to pass data between the OS
7:0
RW and the SMI handler. This is a scratchpad register.
21.5.1.5 Reset Control Register (RSTC)—Offset CF9h

Access Method
Type: I/O Register RSTC: CF9h
(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
RSV2
RSV1
RSVD
WARM_RST
COLD_RST
RSVD
Bit Default &
Description
Range Access
0b
RO
0b
RO
0b Cold Reset (COLD_RST): This bit causes SLPMODE, and RSTRDY# to be driven low,
3 while SLPRDY# remains high. In response to this, the platform will perform a full power
RW cycle
0b
RO
0b Warm Reset (WARM_RST): This bit causes RSTRDY# to be driven low, with SLPMODE
1 high, while SLPRDY# remains high. In response to this, the platform will pulse
RW RESET_BTN_B low to reset the CPU and all peripherals
0b
RO

21.5.2 ACPI GPE0 Block
Table 133. Summary of I/O Registers—GPE0BLK

Offset Default
Start Value
0h 3h “GPE0 Status Register (GPE0STS)—Offset 0h” on page 835 00000000h
4h 7h “GPE0 Enable Register (GPE0EN)—Offset 4h” on page 836 00000000h
10h 13h “SMI Enable Register (SMIEN)—Offset 10h” on page 837 00000000h
14h 17h “SMI Status Register (SMISTS)—Offset 14h” on page 838 00000000h
18h 1Bh “General Purpose Event Control Register (GPEC)—Offset 18h” on page 839 00000000h
“Power Management Configuration Core Well Register (PMCW)—Offset 28h” on

28h 2Bh 00000000h
page 839
“Power Management Configuration Suspend Well Register (PMSW)—Offset 2Ch” on

2Ch 2Fh 00000000h
page 840
30h 33h “Power Management Configuration RTC Well Register (PMRW)—Offset 30h” on page 841 0000000Bh
21.5.2.1 GPE0 Status Register (GPE0STS)—Offset 0h

Access Method
Type: I/O Register GPE0STS: [GPE0BLK] + 0h
(Size: 32 bits)
GPE0BLK Type: PCI Configuration Register (Size: 32 bits)

GPE0BLK Reference: [B:0, D:31, F:0] + 4Ch
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RSV1
SCLT
RSVD
PCIE
GPIO
EGPE
SWGPE
RMU
THRM
Bit Default &

Description
Range Access
0b
RO
0b PCIE Status (PCIE): Set when an Assert SCI message from PCIe Controller is
17
RW/1C received.
0b Remote Management Unit Status (RMU): Set when an Assert SCI message from the
16
RW/1C Remote Management Unit is received.
0b Device Status (SCLT): Set when the SCI signal from Device:20 or Device:21 goes
15
RW/1C active.
0b
14 GPIO Status (GPIO): Set when a GPIO configured for GPE goes active.
RW/1C
0b
13 External GPE Status (EGPE): Set when the GPE_B signal goes active.
RW/1C
0b Thermal Status (THRM): Set anytime THRM_B is received at the state defined by
12
RW/1C GPEC.TPOL.
0b
11 Software GPE Status (SWGPE): Set when GPEC.SWGPE is set.
RW/1C

Bit Default &

Description
Range Access
0b
RO
0b
RO
21.5.2.2 GPE0 Enable Register (GPE0EN)—Offset 4h

Access Method
Type: I/O Register GPE0EN: [GPE0BLK] + 4h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RMU
THRM
RSV1
PCIE
SCLT
RSVD
GPIO
EGPE
SWGPE
Bit Default &
Description
Range Access
0b
RO
0b
17 PCIe Enable (PCIE): When set enables GPE0STS.PCIE to generate SCI/SMI.
RW
0b Remote Management Unit Enable (RMU): When set enables GPE0STS.RMU to

16
RW generate SCI/SMI.
0b
15 Device Enable (SCLT): When set enables GPE0STS.SCLT to generate SCI/SMI.
RW
0b
14 GPIO Enable (GPIO): When set enables GPE0STS.GPIO to generate SCI/SMI.
RW
0b
13 External GPE Enable (EGPE): When set enables GPE0STS.EGPE to generate SCI/SMI.
RW
0b
12 Thermal Enable (THRM): When set enables GPE0STS.THRM to generate SCI/SMI.
RW
0b Software GPE Enable (SWGPE): When set enables GPE0STS.SWGPE to generate SCI/
11
RW SMI.
0b
RO
0b
RO

21.5.2.3 SMI Enable Register (SMIEN)—Offset 10h

Access Method
Type: I/O Register SMIEN: [GPE0BLK] + 10h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RMU
RSV1
SPI
SLP
APM
SERR
ESMI
PCIE
SCLT
RSVD
RSVD
RSVD
RSVD
SWT
GPIO
BIOS
Bit Default &
Description
Range Access
0b
RO
0b
17 PCIe Enable (PCIE): When set enables SMISTS.PCIE to generate SMI.
RW
0b Remote Management Unit Enable (RMU): When set enables SMISTS.RMU to

16
RW generate SMI
0b
15 Device Enable (SCLT): When set enables SMISTS.SCLT to generate SMI.
RW
0b
RO
0b
RO
0b
10 SERR Enable (SERR): When set enables SMISTS.SERR to generate SMI_B
RW
0b
9 GPIO Enable (GPIO): When set enables SMISTS.GPIO to generate SMI_B
RW
0b
8 External SMI Enable (ESMI): When set enables SMISTS.ESMI to generate SMI_B
RW
0b
RO
0b
RO
0b
RO
0b
4 APM Enable (APM): When set enables SMISTS.APM to generate SMI_B
RW
0b
3 SPI Enable (SPI): When set enables SMISTS.SPI to generate SMI_B
RW
0b
2 Sleep (SLP): When set enables SMISTS.SLP to generate SMI_B
RW
0b
1 Software Timer (SWT): When set enables SMISTS.SWT to generate SMI_B
RW
0b
0 BIOS: When set enables SMISTS.BIOS to generate SMI_B
RW

21.5.2.4 SMI Status Register (SMISTS)—Offset 14h

Access Method
Type: I/O Register SMISTS: [GPE0BLK] + 14h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
RMU
RSV1
SPI
SLP
APM
BRLS
PCIE
SCLT
RSVD
SERR
ESMI
RSVD
RSVD
RSVD
SWT
EOS
GPIO
BIOS
Bit Default &
Description
Range Access
0b End of SMI (EOS): This bit is present only in the SMI Status register and not in SMI
31 Enable register. When set, the Legacy Bridge de-asserts SMI#. Cleared when the Legacy
RW Bridge asserts SMI_B.
0b BIOS Release (BRLS): This bit is present only in the SMI Status register and not in
30
WO SMI Enable register. Causes SCI to be generated by the Legacy Bridge. Always reads 0.
0b
RO
0b PCIe Status (PCIE): Set when an Assert SMI message from PCIe Controller is
17
RW/1C received.
0b Remote Management Unit Status (RMU): Set when an Assert SMI message from the
16
RW/1C Remote Management Unit is received.
0b Device Status (SCLT): Set when the SMI_B signal from Device:20 or Device:21 goes
15
RW/1C active.
0b
RO
0b
RO
0b
10 SERR Status (SERR): Set when DO_SERR message is received by the Legacy Bridge.
RW/1C
0b
9 GPIO Status (GPIO): Set when a GPIO configured for SMI goes active.
RW/1C
0b
8 External GPE Status (ESMI): Set when the SMI_B input signal goes active.
RW/1C
0b
RO
0b
RO
0b
RO
0b
4 APM Status (APM): Set when a write to SWSMICTL is performed.
RW/1C
0b
3 SPI Status (SPI): Set when SPI logic is requesting an SMI
RW/1C
0b
2 Sleep (SLP): Set when a write occurs to PM1C.SLPEN
RW/1C

Bit Default &

Description
Range Access
0b
1 Software Timer (SWT): Set when the software SMI has expired.
RW/1C
0b
0 BIOS Status (BIOS): Set when software sets PM1C.GRLS.
RW/1C
21.5.2.5 General Purpose Event Control Register (GPEC)—Offset 18h

Access Method
Type: I/O Register GPEC: [GPE0BLK] + 18h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TPOL
RSV
SWGPE
Bit Default &
Description
Range Access
0b
RO
0b Software General Purpose Event (SWGPE): Sets GPE0S.SWGPE when written with
1
WO 1. This bit always reads back as 0.
0b Thermal Polarity (TPOL): This bit controls the polarity of THRM_B needed to set
0 GPE0S.THRM. When set, a HIGH value on THRM_B will set GPE0S.THRM. When cleared,
RW a LOW value on THRM_B will set GPE0S.THRM.
21.5.2.6 Power Management Configuration Core Well Register (PMCW)—Offset

28h
Access Method
Type: I/O Register PMCW: [GPE0BLK] + 28h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RES
RSV
PSRS
SMIL

Bit Default &

Description
Range Access
0b Periodic SMI Enable (RES): When set, an SMIS.SWT will be set by the rate specified
31
RW by PSRS.
0b
RO
0b SMI Lock (SMIL): When set, writes to SMIE have no effect. This bit is only cleared by
3
RW/O Core Well reset
Periodic SMI Rate Selection (PSRS): Indicates when the timer will time out and
cause an SMI_B. All values are +/- 30us (RTC Clock). Valid values are:
000b - 1.5ms
001b - 16ms
0b 010b - 32ms
2:0
RW 011b - 64ms
100b - 8 sec
101b - 16 sec
110b - 32 sec
111b - 64 sec
21.5.2.7 Power Management Configuration Suspend Well Register (PMSW)—

Offset 2Ch
Access Method
Type: I/O Register PMSW: [GPE0BLK] + 2Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
CBE
DRAMI
Bit Default &
Description
Range Access
0b
RO
0b
1 CPU BIST Enable (CBE): CPU BIST enable INIT functionality not supported.
RW
0b DRAM Initialization Scratch pad (DRAMI): This bit does not affect hardware
0 functionality. It is provided as a BIOS scratchpad bit that is maintained through warm
RW resets.

21.5.2.8 Power Management Configuration RTC Well Register (PMRW)—Offset

30h
Access Method
Type: I/O Register PMRW: [GPE0BLK] + 30h
(Size: 32 bits)

Default: 0000000Bh
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1
WDTS
RSV2
RSV1
RTCB4
RTCB3
RTCB2
RTCB1
RTCB0
Bit Default &
Description
Range Access
0b
RO
Remote Management Unit Watchdog Trip Status (WDTS): This bit is set when the
0b Remote Management Unit watchdog timer expires, causing a system shutdown. It is
9
RW/1C reset by warm and cold resets. It is maintained through the shutdown sequence that is
initiated via this trip
0b
RO
0b
4 RTC Bias Resistor 4 (RTCB4): Adds 192K when de-asserted
RW
1b
RW
0b
RW
1b
RW
1b
RW
21.5.3 ACPI PM1 Block
Table 134. Summary of I/O Registers—PM1BLK

Offset Default
Start Value
0h 1h “PM1 Status Register (PM1S)—Offset 0h” on page 841 0000h
2h 3h “PM1 Enable Register (PM1E)—Offset 2h” on page 842 0000h
4h 7h “PM1 Control Register (PM1C)—Offset 4h” on page 843 00000000h
8h Bh “Power Management 1 Timer Register (PM1T)—Offset 8h” on page 844 00000000h
21.5.3.1 PM1 Status Register (PM1S)—Offset 0h

Access Method

Type: I/O Register PM1S: [PM1BLK] + 0h

(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV3
RTC
RSV2
RSV1
GLOB
WAKE
PCIEWSTS
TO
Bit Default &
Description
Range Access
0b Wake Status (WAKE): Resume Well. This bit is set when the system is in an Sx state
15 and an enable wake event occurs. Upon setting this bit, the Legacy Bridge will transition
RW/1C the system to the S0 state. This bit is not affected by warm resets
0b PCIe Wake Status (PCIEWSTS): This bit is set by hardware to indicate that the
14
RW/1C system woke due to a PCI Express wakeup event.
0b
RO
0b RTC Status (RTC): Resume Well. This bit is set when the RTC asserts IRQ8#, and is
10
RW/1C not affected by any other enable bit. This bit is not affected by warm resets
0b
RO
0b Global Status (GLOB): Set when SMIS.BRLS is written to '1'. It always cause an SCI
5
RW/1C (regardless of PM1C.SCIEN)
0b
RO
0b Timer Overflow Status (TO): Set anytime bit 22 of PM1T goes low. See PM1E.TO for
0
RW/1C the effect of this bit being set
21.5.3.2 PM1 Enable Register (PM1E)—Offset 2h

Access Method
Type: I/O Register PM1E: [PM1BLK] + 2h
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
TO
RSV4
RSV3
RTC
RSV2
RSV1
PWAKED
GLOB
Bit Default &

Description
Range Access
0b
RO

Bit Default &

Description
Range Access
0b PCIe Wake Disable (PWAKED): This bit disables the inputs to the PCIEWSTS bit for
14 waking the system. Modification of this bit has no impact on the value of the PCIEWSTS
RW bit
0b
RO
0b RTC Enable (RTC): Resume Well. When set, and PM1S.RTC is set, an SMI_B/SCI is
10 generated. This bit is not cleared by any reset other than RTCRST_B, CPU/internal
RW thermal Trip, or internal watchdog trip
0b
RO
0b
5 Global Enable (GLOB): When this bit and PM1S.GLOB are set, SMI_B/SCI is generated
RW
0b
RO
0b Timer Overflow Enable (TO): When set, and PM1S.TO is set, an SMI_B/SCI is
0
RW generated
21.5.3.3 PM1 Control Register (PM1C)—Offset 4h

Access Method
Type: I/O Register PM1C: [PM1BLK] + 4h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
SLPEN
RSV1
SCIEN
SLPTYPE
GRLS
BMRLD
Bit Default &
Description
Range Access
0b
RO
0b Sleep Enable (SLPEN): Reads to this bit always return 0. Setting this bit causes the
13
WO system to sequence into the Sleep state defined by SLPTYP
Sleep Type (SLPTYPE): Resume Well. This field defines the type of sleep the system
should enter when SLPEN is set. These bits are reset by RTCRST_B.
0b 000b - S0 - On
12:10 101b - S3 - Suspend to RAM
RW 110b - S4 - Suspend to Disk
111b - S5 - Soft Off
All other values are reserved
0b
RO
0b
2 Global Release (GRLS): Sets SMIS.BIOS when written to 1. This bit always reads as 0
WO
0b Bus Master Reload (BMRLD): This is treated as a scratchpad bit and has no
1
RW functionality.

Bit Default &

Description
Range Access
0b SCI Enable (SCIEN): When set, events in GPE0_BLK generate SCI. When cleared,
0
RW events generate SMI_B.
21.5.3.4 Power Management 1 Timer Register (PM1T)—Offset 8h

Access Method
Type: I/O Register PM1T: [PM1BLK] + 8h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
VAL
RSV
Bit Default &

Description
Range Access
0b
RO
0b Timer Value (VAL): Returns the running count of the PM timer. This counter runs off a
23:0 3.579545 MHz clock (derived from 14.31818 MHz divided by 4). It is reset on a platform
RO reset, and runs continuously in S0. Any time bit 22 goes from 1 to 0, PM1S.TO is set

21.6 Legacy GPIO

The SoC provides a total of 16 GPIOs. Ten of these GPIOs are available for use during
the S0 ACPI state. The remaining 6 GPIOs are available for use during both the S0 and
S3 ACPI state. The GPIOs are split between the Legacy Bridge (0/31/0) and the GPIO
Controller (0/21/2). The GPIOs within the Legacy Bridge are referred to as Legacy
GPIOs and are described in this section.


Table 135. Legacy GPIO Signals

Direction/
Type
I/O
GPIO[9:8] These Legacy GPIO pins are powered and active in S0 only.
Varies
I/O
GPIO_SUS[5:0] These Legacy GPIO pins are powered and active in S3 and S0.
Varies
21.6.2 Features
GPIOs can generate general purpose events (GPEs) on rising and/or falling edges.
The suspend well GPIOs, GPIO_SUS[5:0], can be used to generate wake events when
the system is in the ACPI S3 state.
21.6.3 Use
Each GPIO has six registers that control how it is used, or report its status:
• Use Select
• I/O Select
• GPIO Level
• Trigger Positive Edge
• Trigger Negative Edge
• Trigger Status
The Use Select register selects a GPIO pin as a GPIO, or leaves it as its programmed
function. This register must be set for all other registers to affect the GPIO.
The I/O Select register determines the direction of the GPIO.
The Trigger Positive Edge and Trigger Negative Edge registers enable general purpose
events on a rising and falling edge respectively. This only applies to GPIOs set as input.

The Trigger Status register is used by software to determine if the GPIO triggered a
GPE. This only applies to GPIOs set as input and with one or both of the Trigger modes
enabled.
21.6.4 Register Map
Figure 50. Legacy GPIO Register Map
Legacy PCI
Header
D:31,F:0
IO Space
Legacy GPIO
GPIO_BASE_ Registers
ADDRESS
21.6.5 IO Mapped Registers
Table 136. Summary of I/O Registers—GBA

Offset Default
Start Value
0h 3h “Core Well GPIO Enable (CGEN)—Offset 0h” on page 847 00000003h
4h 7h “Core Well GPIO Input/Output Select (CGIO)—Offset 4h” on page 847 00000003h
8h Bh “Core Well GPIO Level for Input or Output (CGLVL)—Offset 8h” on page 848 00000000h
Ch Fh “Core Well GPIO Trigger Positive Edge Enable (CGTPE)—Offset Ch” on page 848 00000000h
10h 13h “Core Well GPIO Trigger Negative Edge Enable (CGTNE)—Offset 10h” on page 849 00000000h
14h 17h “Core Well GPIO GPE Enable (CGGPE)—Offset 14h” on page 849 00000000h
18h 1Bh “Core Well GPIO SMI Enable (CGSMI)—Offset 18h” on page 850 00000000h
1Ch 1Fh “Core Well GPIO Trigger Status (CGTS)—Offset 1Ch” on page 850 00000000h
20h 23h “Resume Well GPIO Enable (RGEN)—Offset 20h” on page 851 0000003Fh
24h 27h “Resume Well GPIO Input/Output Select (RGIO)—Offset 24h” on page 851 0000003Fh
28h 2Bh “Resume Well GPIO Level for Input or Output (RGLVL)—Offset 28h” on page 852 00000000h
2Ch 2Fh “Resume Well GPIO Trigger Positive Edge Enable (RGTPE)—Offset 2Ch” on page 852 00000000h

Table 136. Summary of I/O Registers—GBA (Continued)

Offset Default
Start Value
30h 33h “Resume Well GPIO Trigger Negative Edge Enable (RGTNE)—Offset 30h” on page 852 00000000h
34h 37h “Resume Well GPIO GPE Enable (RGGPE)—Offset 34h” on page 853 00000000h
38h 3Bh “Resume Well GPIO SMI Enable (RGSMI)—Offset 38h” on page 854 00000000h
3Ch 3Fh “Resume Well GPIO Trigger Status (RGTS)—Offset 3Ch” on page 854 00000000h
40h 43h “Core Well GPIO NMI Enable (CGNMIEN)—Offset 40h” on page 855 00000000h
44h 47h “Resume Well GPIO NMI Enable (RGNMIEN)—Offset 44h” on page 855 00000000h
21.6.5.1 Core Well GPIO Enable (CGEN)—Offset 0h

Access Method
Type: I/O Register CGEN: [GBA] + 0h
(Size: 32 bits)
GBA Type: PCI Configuration Register (Size: 32 bits)

GBA Reference: [B:0, D:31, F:0] + 44h
Default: 00000003h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1
EN
RSV
Bit Default &

Description
Range Access
0b
RO
11b
1:0 Enable (EN): When set, enables the pin as a GPIO.
RW
21.6.5.2 Core Well GPIO Input/Output Select (CGIO)—Offset 4h

Access Method
Type: I/O Register CGIO: [GBA] + 4h
(Size: 32 bits)

Default: 00000003h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1
RSV
IO
Bit Default &

Description
Range Access
0b
RO
11b Input/output (IO): When set, the GPIO signal (if enabled) is programmed as an
1:0
RW input. When cleared, the GPIO signal is programmed as an output.

21.6.5.3 Core Well GPIO Level for Input or Output (CGLVL)—Offset 8h

Access Method
Type: I/O Register CGLVL: [GBA] + 8h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LVL
RSV
Bit Default &
Description
Range Access
0b
RO
Level (LVL): If the GPIO is programmed to be an output (CGIO.IO[n] cleared), then

0b this bit is used by software to drive a value on the pin. 1 = high, 0 = low. If the GPIO is
1:0 programmed as an input, then this bit reflects the state of the input signal (1 = high, 0
RW = low.) and writes will have no effect. The value of this bit has no meaning if the GPIO is
disabled (CGEN.EN[n] = 0).
21.6.5.4 Core Well GPIO Trigger Positive Edge Enable (CGTPE)—Offset Ch

Access Method
Type: I/O Register CGTPE: [GBA] + Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
Bit Default & TE

Description
Range Access
0b
RO
Trigger Enable (TE): When set, the corresponding GPIO, if enabled as input via
0b CGIO.IO[n], will cause an NMI/SMI/SCI when a 0 to 1 transition occurs. When cleared,
1:0
RW the GPIO is not enabled to trigger an NMI/SMI/SCI on a 0 to 1 transition. This bit has no
meaning if CGIO.IO[n] is cleared (i.e. programmed for output)

21.6.5.5 Core Well GPIO Trigger Negative Edge Enable (CGTNE)—Offset 10h
Access Method
Type: I/O Register CGTNE: [GBA] + 10h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
TE
Bit Default &
Description
Range Access
0b
RO
0b CGIO.IO[n], will cause an NMI/SMI/SCI when a 1 to 0 transition occurs. When cleared,
1:0
meaning if CGIO.IO[n] is cleared (i.e. programmed for output)
21.6.5.6 Core Well GPIO GPE Enable (CGGPE)—Offset 14h

Access Method
Type: I/O Register CGGPE: [GBA] + 14h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
EN
Bit Default &
Description
Range Access
0b
RO
0b Enable (EN): When set, the corresponding GPIO, is enabled to generate an SCI and bit
1:0
RW 14 of GPE0 Status register of GPE0 Block will be set.

21.6.5.7 Core Well GPIO SMI Enable (CGSMI)—Offset 18h

Access Method
Type: I/O Register CGSMI: [GBA] + 18h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EN
RSV
Bit Default &
Description
Range Access
0b
RO
0b Enable (EN): When set, the corresponding GPIO, is enabled to generate an SMI and bit
1:0
RW 9 of SMI Status register of GPE0 Block will be set.
21.6.5.8 Core Well GPIO Trigger Status (CGTS)—Offset 1Ch

Access Method
Type: I/O Register CGTS: [GBA] + 1Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
TS
Bit Default &
Description
Range Access
0b
RO
Trigger Status (TS): When set, the corresponding GPIO, if enabled as input via
0b CGIO.IO[n], triggered an SMI/SCI/NMI. This will be set if a 0 to 1 transition occurred
1:0 and CGTPE.TE[n] was set, or a 1 to 0 transition occurred and CGTNE.TE[n] was set. If
RW/1C both CGTPE.TE[n] and CGTNE.TE[n] are set, then this bit will be set on both a 0 to 1
and a 1 to=0 transition. This bit will not be set if the GPIO is configured as an output.

21.6.5.9 Resume Well GPIO Enable (RGEN)—Offset 20h

Access Method
Type: I/O Register RGEN: [GBA] + 20h
(Size: 32 bits)

Default: 0000003Fh
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1
EN
RSV
Bit Default &
Description
Range Access
0b
RO
3Fh
5:0 Enable (EN): When set, enables the pin as a GPIO.
RW
21.6.5.10 Resume Well GPIO Input/Output Select (RGIO)—Offset 24h

Access Method
Type: I/O Register RGIO: [GBA] + 24h
(Size: 32 bits)

Default: 0000003Fh
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1
RSV
Bit Default &

Description IO
Range Access
0b
RO
3Fh Input/Output (IO): When set, the GPIO signal (if enabled) is programmed as an
5:0
RW input. When cleared, the GPIO signal is programmed as an output.

21.6.5.11 Resume Well GPIO Level for Input or Output (RGLVL)—Offset 28h
Access Method
Type: I/O Register RGLVL: [GBA] + 28h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
LVL
RSV
Bit Default &
Description
Range Access
0b
RO
Level (LVL): If the GPIO is programmed to be an output (RGIO.IO[n] cleared), then

0h this bit is used by software to drive a value on the pin. 1 = high, 0 = low. If the GPIO is
5:0 programmed as an input, then this bit reflects the state of the input signal (1 = high, 0
RW = low.) and writes will have no effect. The value of this bit has no meaning if the GPIO is
disabled (RGEN.EN[n] = 0).
21.6.5.12 Resume Well GPIO Trigger Positive Edge Enable (RGTPE)—Offset 2Ch
Access Method
Type: I/O Register RGTPE: [GBA] + 2Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
TE
Bit Default &

Description
Range Access
0b
RO
0h RGIO.IO[n], will cause an NMI/SMI/SCI when a 0 to 1 transition occurs. When cleared,
5:0
meaning if RGIO.IO[n] is cleared (i.e. programmed for output)
21.6.5.13 Resume Well GPIO Trigger Negative Edge Enable (RGTNE)—Offset 30h
Access Method
Type: I/O Register RGTNE: [GBA] + 30h
(Size: 32 bits)


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
TE
Bit Default &
Description
Range Access
0b
RO
0h RGIO.IO[n], will cause an NMI/SMI/SCI when a 1 to 0 transition occurs. When cleared,
5:0
meaning if RGIO.IO[n] is cleared (i.e. programmed for output)
21.6.5.14 Resume Well GPIO GPE Enable (RGGPE)—Offset 34h

Access Method
Type: I/O Register RGGPE: [GBA] + 34h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EN
RSV
Bit Default &

Description
Range Access
0b
RO
0h Enable (EN): When set, the corresponding GPIO, is enabled to generate an SCI and bit
5:0
RW 14 of GPE0 Status register of GPE0 Block will be set.

21.6.5.15 Resume Well GPIO SMI Enable (RGSMI)—Offset 38h

Access Method
Type: I/O Register RGSMI: [GBA] + 38h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EN
RSV
Bit Default &
Description
Range Access
0b
RO
0h Enable (EN): When set, the corresponding GPIO, is enabled to generate an SMI and bit
5:0
RW 9 of SMI Status register of GPE0 Block will be set.
21.6.5.16 Resume Well GPIO Trigger Status (RGTS)—Offset 3Ch

Access Method
Type: I/O Register RGTS: [GBA] + 3Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
TS
Bit Default &
Description
Range Access
0b
RO
Trigger Status (TS): When set, the corresponding GPIO, if enabled as input via
0h RGIO.IO[n], triggered an SMI/SCI/NMI. This will be set if a 0 to 1 transition occurred
5:0 and RGTPE.TE[n] was set, or a 1 to 0 transition occurred and RGTNE.TE[n] was set. If
RW/1C both RGTPE.TE[n] and RGTNE.TE[n] are set, then this bit will be set on both a 0 to 1 and
a 1 to=0 transition. This bit will not be set if the GPIO is configured as an output.

21.6.5.17 Core Well GPIO NMI Enable (CGNMIEN)—Offset 40h

Access Method
Type: I/O Register CGNMIEN: [GBA] + 40h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EN
RSV
Bit Default &
Description
Range Access
0b
RO
0b
1:0 Enable (EN): When set, the corresponding GPIO, is enabled to generate an NMI.
RW
21.6.5.18 Resume Well GPIO NMI Enable (RGNMIEN)—Offset 44h

Access Method
Type: I/O Register RGNMIEN: [GBA] + 44h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
Bit Default &

Description EN
Range Access
0b
RO
0h
5:0 Enable (EN): When set, the corresponding GPIO, is enabled to generate an NMI.
RW

21.7 Legacy SPI Controller

The Legacy SPI Controller provides an interface to a SPI Flash device that contains the
SoC firmware.


Table 137. Legacy SPI Signals

Direction/
Type
LSPI_MOSI O Legacy SPI Data Output
LSPI_MISO I Legacy SPI Data Input
LSPI_SS_B O Legacy Chip Select Signal
LPSI_SCK O Legacy SPI Clock Output
21.7.2 Features
The Legacy SPI Controller provides access to system firmware that resides on a SPI
Flash device connected to the 4-pin Legacy SPI interface. SPI Flash devices up to
16 MByte in size are supported. A SPI clock frequency of 20 MHz is supported.
The Legacy SPI Controller supports direct memory reads from the processor. All other
operations are controlled via the SPI Host Interface registers that reside in the RCRB
Memory Space in the range 3020h to 308Fh.
To protect the integrity of system firmware, the Legacy SPI Controller provides two
write protection mechanisms, one scheme based on address ranges and one SMI_B-
based scheme. If either mechanism indicates an access should not be allowed, then
that write access is blocked.
21.7.3 Register Map


Figure 51. Legacy SPI Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
Memory
Space
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem) Legacy PCI
Header
D:31,F:0 SPI Host
RP0 F:0 Interface
PCIe*
D:23
Registers
SPI0 F:0 RP0 F:1 RCBA_BAR
IO Fabric
D:21
SPI1 F:1
I/O Space
I2C*/GPIOF:2
Legacy Bridge
D:31,F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7
§§
21.7.4 Legacy SPI Host Interface Registers

The SPI Host Interface registers are memory-mapped in the RCRB Memory Space in
offset range 3020h to 308Fh.
Warning: Address locations that are not listed are considered reserved register locations. Reads
to reserved registers may return non-zero values. Writes to reserved locations may
cause system failure.

Table 138. Summary of Memory Mapped I/O Registers—RCBA

Offset Default
Start Value
3020h 3021h “SPI Status (SPISTS)—Offset 3020h” on page 859 0001h
3022h 3023h “SPI Control (SPICTL)—Offset 3022h” on page 859 4001h
3024h 3027h “SPI Address (SPIADDR)—Offset 3024h” on page 860 00000000h
3028h 302Bh “SPI Data 0 - Lower 32 Bits (SPID0_1)—Offset 3028h” on page 861 00000000h
302Ch 302Fh “SPI Data 0 - Upper 32 Bits (SPID0_2)—Offset 302Ch” on page 861 00000000h
3030h 3033h “SPI Data 1 - Lower 32 Bits (SPID1_1)—Offset 3030h” on page 862 00000000h
3034h 3037h “SPI Data 1 - Upper 32 Bits (SPID1_2)—Offset 3034h” on page 862 00000000h
3070h 3073h “BIOS Base Address (BBAR)—Offset 3070h” on page 867 00000000h
3074h 3075h “Prefix Opcode Configuration (PREOP)—Offset 3074h” on page 867 0004h
3076h 3077h “Opcode Type Configuration (OPTYPE)—Offset 3076h” on page 868 0000h
3078h 307Bh “Opcode Menu Configuration - Lower 32 Bits (OPMENU_1)—Offset 3078h” on page 869 00000005h
307Ch 307Fh “Opcode Menu Configuration - Upper 32 Bits (OPMENU_2)—Offset 307Ch” on page 869 00000000h
3080h 3083h “Protected BIOS Range 0 (PBR0)—Offset 3080h” on page 870 00000000h
3084h 3087h “Protected BIOS Range 1 (PBR1)—Offset 3084h” on page 871 00000000h
3088h 308Bh “Protected BIOS Range 2 (PBR2)—Offset 3088h” on page 871 00000000h

21.7.4.1 SPI Status (SPISTS)—Offset 3020h

Access Method
Type: Memory Mapped I/O Register SPISTS: [RCBA] + 3020h
(Size: 16 bits)

Default: 0001h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
RSV2
RSV1
CIP
CLD
BA
CD
Bit Default &
Description
Range Access
0b SPI Configuration Lock-Down (CLD): When set to 1, the SPI Static Configuration
15 information cannot be overwritten. Once set to 1, this bit can only be cleared by a
RW/O hardware reset
0b
RO
Blocked Access Status (BA): Hardware sets this bit to 1 when an access is blocked
from running on the SPI interface due to one of the protection policies or when any of
0b the programmed cycle registers are written while a programmed access is already in
3
RW/1C progress. This bit is set for both programmed accesses and direct memory reads that
get blocked. This bit remains asserted until cleared by software writing a 1 or hardware
reset.
Cycle Done Status (CD): Hardware sets this bit to 1 when the SPI Cycle completes
(i.e., SCIP bit is 0) after software sets the GO bit. This bit remains asserted until cleared
by software writing a 1 or hardware reset. When this bit is set and the SMI Enable bit in
0b the SPI Control register is set, an internal signal is asserted to the SMI_B generation
2
RW/1C block. Software must make sure this bit is cleared prior to enabling the SPI SMI_B
assertion for a new programmed access. This bit gets set after the Status Register
Polling sequence completes after reset de-asserts. It is cleared before and during that
sequence.
0b
RO
Cycle In Progress (CIP): Hardware sets this bit when software sets the SPI Cycle Go
bit in the SPI Control register. This bit remains set until the cycle completes on the SPI
01h interface. Hardware automatically sets and clears this bit so that software can
0 determine when read data is valid and/or when it is safe to begin programming the next
RO command. Software must only program the next command when this bit is 0. This bit
reports 1b during the Status Register Polling sequence after reset de-asserts; it is
cleared when that sequence completes.
21.7.4.2 SPI Control (SPICTL)—Offset 3022h

Access Method
Type: Memory Mapped I/O Register SPICTL: [RCBA] + 3022h
(Size: 16 bits)

Default: 4001h

15 12 8 4 0
0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1
DBCNT
SOPTR
AR
RSV
COPTR
ACS
DC
CG
SMIEN
Bit Default &

Description
Range Access
0b SMI_B Enable (SMIEN): When set to 1, the SPI asserts an SMI_B request whenever
15
RW the Cycle Done Status bit is 1.
1b Data Cycle (DC): When set to 1, there is data that corresponds to this transaction.
14 When 0, no data is delivered for this cycle, and the DBC and data fields themselves are
RW don't care.
Data Byte Count (DBCNT): Data Byte Count: This field specifies the number of bytes
0b to shift in or out during the data portion of the SPI cycle. The valid settings (in decimal)
13:8 are any value from 0 to 63. The number of bytes transferred is the value of this field
RW plus 1. Note that when this field is 00_0000b, then there is 1 byte to transfer and that
11_1111b means there are 64 bytes to transfer.
0b
7 Reserved (RSV): Reserved.
RO
0b Cycle Opcode Pointer (COPTR): This field selects one of the programmed opcodes in
6:4 the Opcode Menu Configuration register to be used as the SPI Command/Opcode. In the
RW case of an Atomic Cycle Sequence, this determines the second command.
Sequence Prefix Opcode Pointer (SOPTR): This field selects one of the two
programmed prefix opcodes for use when performing an Atomic Cycle Sequence. A
0b value of 0 points to the opcode in the least significant byte of the Prefix Opcode
3
RW Configuration register. By making this programmable, the processor supports flash
devices that have different opcodes for enabling writes to the data space vs. status
register.
Atomic Cycle Sequence (ACS): When set to 1 along with the SCGO assertion, the
processor will execute a sequence of commands on the SPI interface without allowing
the other SPI master component to arbitrate and interleave cycles. The sequence is
0b composed of: Atomic Sequence Prefix Command (8-bit opcode only) Primary Command
2
RW specified by software (can include address and data) Polling the Flash Status Register
(opcode 05h) until bit 0 becomes 0b. The SPI Cycle in Progress bit remains set and the
Cycle Done Status bit in the SPI Status register remains unset until the Busy bit in the
Flash Status Register returns 0.
Cycle Go (CG): This bit always returns 0 on reads. However, a write to this register
with a 1 in this bit starts the SPI cycle defined by the other bits of this register. The SPI
0b Cycle in Progress (SCIP) bit in the SPI Status register gets set by this action. Hardware
1
RW/S must ignore writes to this bit while the SPI Cycle In Progress bit is set. Hardware allows
other bits in this register to be programmed for the same transaction when writing this
bit to 1. This saves an additional memory write.
1b Access Request (AR): This bit is used by the software to request that the other SPI
0 master stop initiating long transactions on the SPI bus. This bit defaults to a 1 and must
RW be cleared by BIOS after completing the accesses for the boot process.
21.7.4.3 SPI Address (SPIADDR)—Offset 3024h

Access Method
Type: Memory Mapped I/O Register SPIADDR: [RCBA] + 3024h
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
CA
CSC
Bit Default &
Description
Range Access
CSC: Chip Select Control: These two bits control which SPI Chip Select is used. Default
00 must always select SS0. Direct read mode always uses SS0.
0b 00 : SS0
31:30
RW 01 : Reserved
10 : Reserved
11 : Reserved
0b
RO
0b
23:0 Cycle Address (CA): This field is shifted out as the SPI Address (MSB first).
RW
21.7.4.4 SPI Data 0 - Lower 32 Bits (SPID0_1)—Offset 3028h

Access Method
(Size: 32 bits) SPID0_1: [RCBA] + 3028h

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
0b Cycle Data (CD): This field is shifted out during the data portion of the SPI cycle. This
31:0
RW register also shifts in the data during the data portion of the SPI cycle
21.7.4.5 SPI Data 0 - Upper 32 Bits (SPID0_2)—Offset 302Ch

Access Method
Type: Memory Mapped I/O Register SPID0_2: [RCBA] + 302Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD

Bit Default &

Description
Range Access
31:0

Access Method
Type: Memory Mapped I/O Register SPID1_1: [RCBA] + 3030h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &
Description
Range Access
31:0
21.7.4.7 SPI Data 1 - Upper 32 Bits (SPID1_2)—Offset 3034h

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &
Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
31:0

Access Method


(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &
Description
Range Access
31:0

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
31:0


Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &
Description
Range Access
31:0

Access Method

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD

Bit Default &

Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &
Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &

Description
Range Access
31:0

Access Method
(Size: 32 bits)

Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CD
Bit Default &
Description
Range Access
31:0
21.7.4.20 BIOS Base Address (BBAR)—Offset 3070h

This register is not writable when the SPI Configuration Lock-Down bit in the SPI Status
register is set.
Access Method
Type: Memory Mapped I/O Register BBAR: [RCBA] + 3070h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV2
BOSF
RSV1
Bit Default &
Description
Range Access
0b
RO
Bottom of System Flash (BOSF): This field determines the bottom of the System
BIOS. The processor will not run Programmed commands nor memory reads whose
address field is less than this value. This field corresponds to bits 23:8 of the 3-byte
address; bits 7:0 are assumed to be 00h for this vector when comparing to a potential
SPI address. Software must always program 1s into the upper, Don't Care bits of this
field based on the flash size. Hardware does not know the size of the flash array and
0b relies upon the correct programming by software. The default value of 0000h results in
23:8
RW/L all cycles allowed.
Note: The SPI Host Controller prevents any Programmed cycle using the Address
Register with an address less than the value in this register. Some flash devices specify
that the Read ID command must have an address of 0000h or 0001h. If this command
must be supported with these devices, it must be performed with the BBAR - BIOS Base
Address programmed to 0h. Some of these devices have actually been observed to
ignore the upper address bits of the Read ID command.
0b
RO
21.7.4.21 Prefix Opcode Configuration (PREOP)—Offset 3074h

register is set.
Access Method

Type: Memory Mapped I/O Register PREOP: [RCBA] + 3074h

(Size: 16 bits)

Default: 0004h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
PO1
PO2
Bit Default &
Description
Range Access
0h Prefix Opcode 1 (PO1): Software programs an SPI opcode into this field that is
15:8
RW/L permitted to run as the first command in an atomic cycle sequence.
04h Prefix Opcode 2 (PO2): Software programs an SPI opcode into this field that is
7:0
RW/L permitted to run as the first command in an atomic cycle sequence.
21.7.4.22 Opcode Type Configuration (OPTYPE)—Offset 3076h

register is set. Entries in this register correspond to the entries in the Opcode Menu
Configuration register. Note that the definition below only provides write protection for
opcodes that have addresses associated with them. Therefore, any erase or write
opcodes that do not use an address should be avoided (for example, Chip Erase. and
Auto-Address Increment Byte Program.).
Access Method
Type: Memory Mapped I/O Register OPTYPE: [RCBA] + 3076h
(Size: 16 bits)

Default: 0000h
15 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
OT7
OT6
OT5
OT4
OT3
OT2
OT1
OT0
Bit Default &

Description
Range Access
0b
15:14 Opcode Type 7 (OT7): See the description for bits 1:0
RW/L
0b
RW/L
0b
RW/L
0b
RW/L
0b
RW/L
0b
RW/L

Bit Default &

Description
Range Access
0b
RW/L
Opcode Type 0 (OT0): This field specifies information about the corresponding Opcode
0. This information allows the hardware to 1) know whether to use the address field and
2) provide BIOS protection capabilities. The hardware implementation also uses the
0b read vs. write information for modifying the behavior of the SPI interface logic. The
1:0 encoding of the two bits is:
RW/L 00 = No Address associated with this Opcode and Read Cycle type
01 = No Address associated with this Opcode and Write Cycle type
10 = Address required; Read cycle type
11 = Address required; Write cycle type
21.7.4.23 Opcode Menu Configuration - Lower 32 Bits (OPMENU_1)—Offset

3078h
register is set. Eight entries are available in this register to give BIOS a sufficient set of
commands for communicating with the flash device, while also restricting what
malicious software can do. This keeps the hardware flexible enough to operate with a
wide variety of SPI devices. It is recommended that BIOS avoid programming Write
Enable opcodes in this menu. Malicious software could then perform writes and erases
to the SPI flash without using the atomic cycle mechanism. Write Enable opcodes
should only be programmed in the Prefix Opcode Configuration register.
Access Method
(Size: 32 bits) OPMENU_1: [RCBA] + 3078h

Default: 00000005h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1
AO3
AO2
AO1
AO0
Bit Default &

Description
Range Access
0b
31:24 Allowable Opcode 3 (AO3): See the description for bits 7:0
RW/L
0b
RW/L
0b
RW/L
05h Allowable Opcode 0 (AO0): Software programs an SPI opcode into this field for use
7:0
RW/L when initiating SPI commands through the Control Register.
21.7.4.24 Opcode Menu Configuration - Upper 32 Bits (OPMENU_2)—Offset

307Ch
register is set. Eight entries are available in this register to give BIOS a sufficient set of
commands for communicating with the flash device, while also restricting what

malicious software can do. This keeps the hardware flexible enough to operate with a
wide variety of SPI devices. It is recommended that BIOS avoid programming Write
Enable opcodes in this menu. Malicious software could then perform writes and erases
to the SPI flash without using the atomic cycle mechanism. Write Enable opcodes
should only be programmed in the Prefix Opcode Configuration register.
Access Method
Type: Memory Mapped I/O Register OPMENU_2: [RCBA] + 307Ch
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
AO7
AO6
AO5
AO4
Bit Default &
Description
Range Access
0b
RW/L
0b
RW/L
0b
RW/L
0h Allowable Opcode 4 (AO4): Software programs an SPI opcode into this field for use
7:0
RW/L when initiating SPI commands through the Control Register.
21.7.4.25 Protected BIOS Range 0 (PBR0)—Offset 3080h

This register cannot be written when the SPI Configuration Lock-Down bit in the SPI
Status register is set to 1.
Access Method
Type: Memory Mapped I/O Register PBR0: [RCBA] + 3080h
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
WPE
RSV
PRL
PRB
Bit Default &

Description
Range Access
Write Protection Enable (WPE): When set, this bit indicates that the Base and Limit
0b fields in this register are valid and that writes directed to addresses between them
31
RW/L (inclusive) must be blocked by hardware. The base and limit fields are ignored when this
bit is cleared.
0b
RO

Bit Default &

Description
Range Access
Protected Range Limit (PRL): This field corresponds to SPI address bits 23:12 and
0b specifies the upper limit of the protected range. Address bits 11:0 are assumed to be
23:12
RW/L FFFh for the limit comparison. Any address greater than the value programmed in this
field is unaffected by this protected range.
Protected Range Base (PRB): This field corresponds to SPI address bits 23:12 and
0b specifies the lower base of the protected range. Address bits 11:0 are assumed to be
11:0
RW/L 000h for the base comparison. Any address less than the value programmed in this field
is unaffected by this protected range.

Access Method
(Size: 32 bits)

Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PRL
PRB
WPE
RSV
Bit Default &

Description
Range Access
31
bit is cleared.
0b
RO
23:12
11:0

Access Method
(Size: 32 bits)


Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PRL
RSV
PRB
WPE
Bit Default &

Description
Range Access
31
bit is cleared.
0b
RO
23:12
11:0

21.8 8254 Programmable Interval Timer

The 8254 contains three counters which have fixed uses. All registers are in the core
well and clocked by a 14.31818 MHz clock.
21.8.1 Features
21.8.1.1 Counter 0, System Timer

This counter functions as the system timer by controlling the state of IRQ0 and is
programmed for Mode 3 operation. The counter produces a square wave with a period
equal to the product of the counter period (838 ns) and the initial count value. The
counter loads the initial count value one counter period after software writes the count
value to the counter I/O address. The counter initially asserts IRQ0 and decrements the
count value by two, each counter period. The counter negates IRQ0 when the count
value reaches 0. It then reloads the initial count value and again decrements the initial
count value by two, each counter period. The counter then asserts IRQ0 when the
count value reaches 0, reloads the initial count value, and repeats the cycle, alternately
asserting and negating IRQ0.
21.8.1.2 Counter 1, Refresh Request Signal

This counter is programmed for Mode 2 operation and impacts the period of the
NSC.RTS (NMI Status and Control Register, bit4, Refresh Cycle Toggle Status).
Programming the counter to anything other than Mode 2 results in undefined behavior.
21.8.1.3 Counter 2, Speaker Tone

This counter is typically programmed for Mode 3 operation.
21.8.2 Use
21.8.2.1 Timer Programming

The counter/timers are programmed in the following fashion:
• Write a control word to select a counter
• Write an initial count for that counter.
• Load the least and/or most significant bytes (as required by Control Word bits 5, 4)
of the 16-bit counter.
• Repeat with other counters
Only two conventions must be observed when programming the counters. First, for
each counter, the control word must be written before the initial count is written.
Second, the initial count must follow the count format specified in the control word
(least significant byte only, most significant byte only, or least significant byte and then
most significant byte).
A new initial count may be written to a counter at any time without affecting the
counter's programmed mode. Counting will be affected as described in the mode
definitions. The new count must follow the programmed count format.
If a counter is programmed to read/write two-byte counts, the following precaution

applies: A program must not transfer control between writing the first and second byte
to another routine which also writes into that same counter. Otherwise, the counter will
be loaded with an incorrect count.

The Control Word Register at port 43h controls the operation of all three counters.
Several commands are available:
• Control Word Command: Specifies which counter to read or write, the operating
mode, and the count format (binary or BCD).
• Counter Latch Command: Latches the current count so that it can be read by the
system. The countdown process continues.
• Read Back Command: Reads the count value, programmed mode, the current
state of the OUT pins, and the state of the Null Count Flag of the selected counter.
Table 139 lists the six operating modes for the interval counters.
Table 139. Counter Operating Modes

Mode Function Description
Output is 0. When count goes to 0, output goes to 1 and

0 Out signal on end of count (=0)
stays at 1 until counter is reprogrammed.
Output is 0. When count goes to 0, output goes to 1 for one

1 Hardware retriggerable one-shot
clock time.
Output is 1. Output goes to 0 for one clock time, then back to

2 Rate generator (divide by n counter)
1 and counter is reloaded.
Output is 1. Output goes to 0 when counter rolls over, and

3 Square wave output counter is reloaded. Output goes to 1 when counter rolls
over, and counter is reloaded, etc.
Output is 1. Output goes to 0 when count expires for one

4 Software triggered strobe
clock time.
Output is 1. Output goes to 0 when count expires for one

5 Hardware triggered strobe
clock time.
21.8.2.2 Reading from the Interval Timer

It is often desirable to read the value of a counter without disturbing the count in
progress. There are three methods for reading the counters: a simple read operation,
counter Latch Command, and the Read-Back Command. Each is explained below.
With the simple read and counter latch command methods, the count must be read
according to the programmed format; specifically, if the counter is programmed for two
byte counts, two bytes must be read. The two bytes do not have to be read one right
after the other. Read, write, or programming operations for other counters may be
inserted between them.
21.8.2.2.1 Simple Read
The first method is to perform a simple read operation. The counter is selected through
port 40h (counter 0), 41h (counter 1), or 42h (counter 2).
Note: Performing a direct read from the counter does not return a determinate value,
because the counting process is asynchronous to read operations. However, in the case
of counter 2, the count can be stopped by writing to the NSC.CNTR2_ENABLE register
field.
21.8.2.2.2 Counter Latch Command
The Counter Latch Command, written to port 43h, latches the count of a specific
counter at the time the command is received. This command is used to ensure that the
count read from the counter is accurate, particularly when reading a two-byte count.
The count value is then read from each counter's Count Register as was programmed
by the Control Register.

The count is held in the latch until it is read or the counter is reprogrammed. The count
is then unlatched. This allows reading the contents of the counters on the fly without
affecting counting in progress. Multiple Counter Latch Commands may be used to latch
more than one counter. Counter Latch Commands do not affect the programmed mode
of the counter in any way.
If a Counter is latched and then, some time later, latched again before the count is
read, the second Counter Latch Command is ignored. The count read will be the count
at the time the first Counter Latch Command was issued.
21.8.2.2.3 Read Back Command
The Read Back Command, written to port 43h, latches the count value, programmed
mode, and current states of the OUT pin and Null Count flag of the selected counter or
counters. The value of the counter and its status may then be read by I/O access to the
counter address.
The Read Back Command may be used to latch multiple counter outputs at one time.
This single command is functionally equivalent to several counter latch commands, one
for each counter latched. Each counter's latched count is held until it is read or
reprogrammed. Once read, a counter is unlatched. The other counters remain latched
until they are read. If multiple count Read Back Commands are issued to the same
counter without reading the count, all but the first are ignored.
The Read Back Command may additionally be used to latch status information of
selected counters. The status of a counter is accessed by a read from that counter's I/
O port address. If multiple counter status latch operations are performed without
reading the status, all but the first are ignored.
Both count and status of the selected counters may be latched simultaneously. This is
functionally the same as issuing two consecutive separate Read Back Commands. If
multiple count and/or status Read Back Commands are issued to the same counters
without any intervening reads, all but the first are ignored.
If both count and status of a counter are latched, the first read operation from that
counter returns the latched status, regardless of which was latched first. The next one
or two reads, depending on whether the counter is programmed for one or two type
counts, return the latched count. Subsequent reads return unlatched count.
21.8.3 Register Map


Figure 52. 8254 Timers Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
Memory
Space
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem) Legacy PCI
Header
D:31,F:0
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
D:21
SPI1 F:1
I2C*/GPIO F:2 IO Space

Fixed IO
D:31,F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7
21.8.4 Timer I/O Registers

The I/O ports listed in Table 140 have multiple register functions depending on the
current programmed state of the 8254. The port numbers referenced in the register
descriptions following Table 140 is one possible combination but not the only one.
Table 140. Register Aliases

Port Alias Register Name Default Value Access
Counter 0 Interval Time Status Byte Format (C0TS) 0xxxxxxxb RO

40h 50h
Counter 0 Counter Access Port Register (C0AP) Undefined RW

41h 51h

Table 140. Register Aliases

Port Alias Register Name Default Value Access

42h 52h
Timer Control Word Register (TCW) Undefined WO
43h - Read Back Command (RBC) xxxxxxx0b WO
Counter Latch Command (CLC) xxxx0000b WO
21.8.4.1 Counter 0 Interval Time Status Byte Format (C0TS)—Offset 40h

Access Method
Type: I/O Register C0TS: 40h
(Size: 8 bits)
7 4 0
0 X X X X X X X
CR
CS
RWS
MD
CT
Bit Default &
Description
Range Access
0b Counter State (CS): When set, OUT of the counter is set. When cleared, OUT of the
7
RO counter is 0.
X Count Register (CR): When cleared, indicates when the last count written to the Count
6 Register (CR) has been loaded into the counting element (CE) and is available for
RO reading. The time this happens depends on the counter mode.
Read/Write Selection (RWS): These reflect the read/write selection made through
bits[5:4] of the control register. The binary codes returned during the status read match
X the codes used to program the counter read/write selection.
5: 4 00 Counter Latch Command
RO 01 Read/Write Least Significant Byte (LSB)
10 Read/Write Most Significant Byte (MSB)
11 Read/Write LSB then MSB
Mode (MD): Returns the counter mode programming. The binary code returned
matches the code used to program the counter mode, as listed under the bit function
above.
Bits Mode Description
X 000 0 Out signal on end of count (=0)
3: 1
RO 001 1 Hardware retriggerable one-shot
x10 2 Rate generator (divide by n counter)
x11 3 Square wave output
100 4 Software triggered strobe
101 5 Hardware triggered strobe
X Countdown Type (CT): Type: 0 for binary countdown or a 1 for binary coded decimal
0
RO (BCD) countdown.

Access Method
(Size: 8 bits)

7 4 0
0 X X X X X X X
CR
CS
RWS
MD
CT
Bit Default &
Description
Range Access
7
RO counter is 0.
Read/Write Selection (RWS): These reflect the read/write selection made through
above.
3: 1
X Countdown Type (CT): 0 for binary countdown or a 1 for binary coded decimal (BCD)
0
RO countdown.

Access Method
(Size: 8 bits)
7 4 0
0 X X X X X X X
CS
RWS
CT
CR
MD
Bit Default &

Description
Range Access
7
RO counter is 0.
Read/Write Select ion (RWS): These reflect the read/write selection made through

Bit Default &

Description
Range Access
above.
3: 1
X Countdown Type (CT): 0 for binary countdown or a 1 for binary coded decimal (BCD)
0
RO countdown.
21.8.4.4 Timer Control Word Register (TCW)—Offset 43h

Access Method
Type: I/O Register TCW: 43h
(Size: 8 bits)
7 4 0
X X X X X X X X
CS
RWS
CMS
BCS
Bit Default &
Description
Range Access
Counter Select (CS): The Counter Selection bits select the counter the control word
acts upon as shown below. The Read Back Command is selected when bits[7:6] are both
X 1.
7: 6 00 Counter 0 select
WO 01 Counter 1 select
10 Counter 2 select
11 Read Back Command
Read/Write Select (RWS): The counter programming is done through the counter
port (40h for counter 0, 41h for counter 1, and 42h for counter 2)
X 00 Counter Latch Command
5: 4
WO 01 Read/Write Least Significant Byte (LSB)
Counter Mode Selection (CMS): Selects one of six modes of operation for the
selected counter.
000 = Out signal on end of count (=0)
X 001 = Hardware retriggerable one-shot
3: 1
WO x10 = Rate generator (divide by n counter)
x11 = Square wave output
100 = Software triggered strobe
101 = Hardware triggered strobe
Binary/BCD Countdown Select (BCS):

X
0 0 = Binary countdown is used. The largest possible binary count is 216
WO
1 = Binary coded decimal (BCD) count is used. The largest possible BCD count is 104
21.8.4.5 Counter 0 Counter Access Port Register (C0AP)—Offset 50h

Access Method
Type: I/O Register C0AP: 50h
(Size: 8 bits)

7 4 0
X X X X X X X X
CP
Bit Default &
Description
Range Access
Counter Port (CP): Each counter port address is used to program the 16-bit Count
X Register. The order of programming, either LSB only, MSB only, or LSB then MSB, is
7: 0 defined with the Interval Counter Control Register at port 43h. The counter port is also
RW used to read the current count from the Count Register, and return the status of the
counter programming following a Read Back Command.

Access Method
(Size: 8 bits)
7 4 0
X X X X X X X X
CP
Bit Default &

Description
Range Access

Access Method
(Size: 8 bits)
7 4 0
X X X X X X X X
CP
Bit Default &

Description
Range Access
21.9 High Precision Event Timer (HPET)

This function provides a set of timers to be used by the operating system for timing
events. One timer block is implemented, containing one counter and three timers.

21.9.1 Features
21.9.1.1 Non-Periodic Mode - All Timers

This mode can be thought of as creating a one-shot. When a timer is set up for non-
periodic mode, it generates an interrupt when the value in the main counter matches
the value in the timer's comparator register. As timers 1 and 2 are 32-bit, they
generate another interrupt when the main counter wraps.
T0CV cannot be programmed reliably by a single 64-bit write in a 32-bit environment

unless only the periodic rate is being changed. If T0CV must be re-initialized, the
following algorithm is performed:
1. Set T0C.TVS
2. Set T0CV[31:0]
3. Set T0C.TVS
4. Set T0CV[63:32]
Every timer is required to support the non-periodic mode of operation.
21.9.1.2 Periodic Mode - Timer 0 Only

In periodic mode, when the main counter value matches the value in T0CV, an interrupt
is generated (if enabled). Hardware then increases T0CV by the last value written to
T0CV. During run-time, T0CV can be read to find out when the next periodic interrupt
will be generated. Software is expected to remember the last value written to T0CV.
Example: if the value written to T0CV is 00000123h, then:

• An interrupt will be generated when the main counter reaches 00000123h.
• T0CV will then be adjusted to 00000246h.
• Another interrupt will be generated when the main counter reaches 00000246h.
• T0CV will then be adjusted to 00000369h.
When the incremented value is greater than the maximum value possible for T0CV, the
value wraps around through 0. For example, if the current value in a 32-bit timer is
FFFF0000h and the last value written to this register is 20000, then after the next
interrupt the value changes to 00010000h.
If software wants to change the periodic rate, it writes a new value to T0CV. When the
timer's comparator matches, the new value is added to derive the next matching point.
If software resets the main counter, the value in the comparator's value register must
also be reset by setting T0C.TVS. To avoid race conditions, this should be done with the
main counter halted. The following usage model is expected:
1. Software clears GCFG.EN to prevent any interrupts.
2. Software clears the main counter by writing a value of 00h to it.
3. Software sets T0C.TVS.
4. Software writes the new value in T0CV.
5. Software sets GCFG.EN to enable interrupts.

21.9.1.3 Interrupts
If each timer has a unique interrupt and the timer has been configured for edge-
triggered mode, then there are no specific steps required. If configured to level-
triggered mode, then its interrupt must be cleared by software by writing a '1' back to
the bit position for the interrupt to be cleared.
Interrupts associated with the various timers have several interrupt mapping options.
Software should mask GCFG.LRE when reprogramming HPET interrupt routing to avoid
spurious interrupts.
21.9.1.3.1 Mapping Option #1: Legacy Option (GCFG.LRE set)
This forces the following mapping:
Table 141. 8254 Interrupt Mapping

Timer 8259 Mapping APIC Mapping Comment
0 IRQ0 IRQ2 The 8254 timer does not cause any interrupts
1 IRQ8 IRQ8 RTC does not cause any interrupts.
2 T2C.IR T2C.IRC
21.9.1.3.2 Mapping Option #2: Standard Option (GCFG.LRE cleared)
Each timer has its own routing control. The interrupts can be routed to various
interrupts in the I/O APIC. T[2:0]C.IRC indicates which interrupts are valid options for
routing. If a timer is set for edge-triggered mode, the timers should not be shared with
any other interrupts.
21.9.2 Register Map


Figure 53. HPET Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
Memory
Space
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem)
Legacy PCI
Header
RP0 F:0
PCIe*
D:31,F:0
D:23
SPI0 F:0 Fixed Mem

RP0 F:1
IO Fabric
Registers
D:21
SPI1 F:1
2
I C*/GPIO F:2
IO Space
Legacy Bridge
D:31,F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7
21.9.3 Memory Mapped Registers

The HPET register space is memory mapped to a 1 KB block starting at address
FED00000h. All registers are in the core well and reset by RESET_BTN_B. Accesses that
cross register boundaries result in undefined behavior.
Table 142. Summary of Memory Mapped I/O Registers—0xFED00000

Offset Default
Start Value
0h 3h “General Capabilities and ID Register - Lower 32 Bits (GCID_1)—Offset 0h” on page 884 8086A201h
4h 7h “General Capabilities and ID Register - Upper 32 Bits (GCID_2)—Offset 4h” on page 885 0429B17Fh
10h 13h “General Capabilities and ID Register - Lower 32 Bits (GCID_1)—Offset 0h” on page 884 00000000h

Table 142. Summary of Memory Mapped I/O Registers—0xFED00000 (Continued)

Offset Default
Start Value
20h 23h “General Interrupt Status Register (GIS)—Offset 20h” on page 885 00000000h
F0h F3h “Main Counter Value Register - Lower 32 Bits (MCV_1)—Offset F0h” on page 886 00000000h
F4h F7h “Main Counter Value Register - Upper 32 Bits (MCV_2)—Offset F4h” on page 886 00000000h
“Timer 0 Config and Capabilities Register - Lower 32 Bits (T0C_1)—Offset 100h” on

100h 103h 00000030h
page 887
“Timer 0 Config and Capabilities Register - Upper 32 Bits (T0C_2)—Offset 104h” on

104h 107h 00F00000h
page 888
“Timer 0 Comparator Value Register - Lower 32 Bits (T0CV_1)—Offset 108h” on

108h 10Bh FFFFFFFFh
page 888
“Timer 0 Comparator Value Register - Upper 32 Bits (T0CV_2)—Offset 10Ch” on

10Ch 10Fh FFFFFFFFh
page 888

120h 123h 00000000h
page 889

124h 127h 00F00000h
page 890
128h 12Bh “Timer 1 Comparator Value Register (T1CV_1)—Offset 128h” on page 890 FFFFFFFFh

140h 143h 00000000h
page 890

144h 147h 00F00800h
page 891
148h 14Bh “Timer 2 Comparator Value Register (T2CV_1)—Offset 148h” on page 892 FFFFFFFFh
21.9.3.1 General Capabilities and ID Register - Lower 32 Bits (GCID_1)—Offset

0h
Access Method
(Size: 32 bits) GCID_1: [0xFED00000] + 0h
Default: 8086A201h
31 28 24 20 16 12 8 4 0
1 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 1 0 1 0 0 0 1 0 0 0 0 0 0 0 0 1
RSV
CS
NT
LRC
VID
RID
Bit Default &

Description
Range Access
8086h
31:16 Vendor ID (VID): Value of 8086h indicates Intel.
RO
1b
15 Legacy Route Capable (LRC): Indicates support for Legacy Interrupt Route.
RO
0b
14 Reserved (RSV): Reserved.
RO
1b
13 Counter Size (CS): This bit is set to indicate that the main counter is 64 bits wide.
RO
00010b
12:8 Number of Timers (NT): Indicates that 3 timers are supported.
RO

Bit Default &

Description
Range Access
01h
7:0 Revision ID (RID): Indicates that revision 1.0 of the specification is implemented.
RO
21.9.3.2 General Capabilities and ID Register - Upper 32 Bits (GCID_2)—Offset

4h
Access Method
Type: Memory Mapped I/O Register GCID_2: [0xFED00000] + 4h
(Size: 32 bits)
Default: 0429B17Fh
31 28 24 20 16 12 8 4 0
0 0 0 0 0 1 0 0 0 0 1 0 1 0 0 1 1 0 1 1 0 0 0 1 0 1 1 1 1 1 1 1
CTP
Bit Default &
Description
Range Access
0429B17Fh
31:0 Counter Tick Period (CTP): Indicates a period of 69.841279ns, 14.1318 MHz clock.
RO
21.9.3.3 General Configuration (GC)—Offset 10h

Access Method
(Size: 32 bits) GC: [0xFED00000] + 10h
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSV
LRE
EN
Bit Default &
Description
Range Access
0b
RO
Legacy Route Enable (LRE): When set, interrupts will be routed as follows: Timer 0
0b will be routed to IRQ0 in 8259 and IRQ2 in the I/O APIC Timer 1 will be routed to IRQ8
1 in 8259 and I/O APIC Timer 2 is routed to IRQ11 in 8259 and Timer 2 will be routed to
RW IOxAPIC as per the routing in T2C.IR When set, the TnC.IR will have no impact for
Timers 0 and 1.
Overall Enable (EN): When set, the timers can generate interrupts. When cleared, the
0b main counter will halt and no interrupts will be caused by any timer. For level-triggered
0
RW interrupts, if an interrupt is pending when this bit is cleared, the GIS.Tx will not be
cleared.
21.9.3.4 General Interrupt Status Register (GIS)—Offset 20h

Access Method

Type: Memory Mapped I/O Register GIS: [0xFED00000] + 20h

(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
T2
T1
T0
RSV
Bit Default &
Description
Range Access
0b
RO
0b Timer 2 Status (T2): In edge triggered mode, this bit always reads as 0. In level
2
RW/1C triggered mode, this bit is set when an interrupt is active.
1
0
21.9.3.5 Main Counter Value Register - Lower 32 Bits (MCV_1)—Offset F0h

Access Method
Type: Memory Mapped I/O Register MCV_1: [0xFED00000] + F0h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CV
Bit Default &

Description
Range Access
0b Counter Value (CV): Reads return the current value of the lower 32 bits of the counter.
31:0
RW Writes load the new value to the lower 32 bits of the counter.
21.9.3.6 Main Counter Value Register - Upper 32 Bits (MCV_2)—Offset F4h

Access Method
Type: Memory Mapped I/O Register MCV_2: [0xFED00000] + F4h
(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CV

Bit Default &

Description
Range Access
0b Counter Value (CV): Reads return the current value of the upper 32 bits of the
31:0 counter. Writes load the new value to the upper 32 bits of the counter. Timers 1 and
RW Timer 2 return 0.

Offset 100h
Access Method
Type: Memory Mapped I/O Register T0C_1: [0xFED00000] + 100h
(Size: 32 bits)
Default: 00000030h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0
RSV3
T32M
RSV2
TYP
RSV1
FID
IR
FE
TVS
TS
IE
IT
PIC
Bit Default &
Description
Range Access
0b
RO
0b
15 FSB Interrupt Delivery (FID): Not Supported
RO
0b
14 FSB Enable (FE): Not supported, since FID is not supported.
RO
Interrupt Route (IR): Indicates the routing for the interrupt to the IOxAPIC. If the
0b value is not supported by this particular timer, the value read back will not match what
13:9
RW is written. If GC.LRE is set, then Timers 0 and 1 have a fixed routing, and this field has
no effect.
0b Timer 32-bit Mode (T32M): When set, this bit forces a 64-bit timer to behave as a
8
RW 32-bit timer.
0b
RO
0b Timer Value Set (TVS): This bit will return 0 when read. Writes will only have an effect
6
WO for Timer 0 if it is set to periodic mode. Writes will have no effect for Timers 1 and 2
1b
5 Timer Size (TS): 1 = 64-bits, 0 = 32-bits. Set for timer 9. Cleared for timers 1 and 2
RO
1b Periodic Interrupt Capable (PIC): When set, hardware supports a periodic mode for
4
RO this timer's interrupt.
0b Timer Type (TYP): If PIC is set, this bit is read/write, and can be used to enable the
3
RW timer to generate a periodic interrupt.
0b Interrupt Enable (IE): When set, enables the timer to cause an interrupt when it
2 times out. When cleared, the timer count and generates status bits, but will not cause
RW an interrupt.
Timer Interrupt Type (IT): When cleared, interrupt is edge triggered. When set,
0b interrupt is level triggered and will be held active until it is cleared by writing 1 to
1
RW GIS.Tn. If another interrupt occurs before the interrupt is cleared, the interrupt remains
active.
0b
RO


Offset 104h
Reads to this register return the current value of the comparator. The default value for
each timer is all 1's for the bits that are implemented. Timer 0 is 64-bits wide. Timers 1
and 2 are 32-bits wide.
Access Method
(Size: 32 bits) T0C_2: [0xFED00000] + 104h
Default: 00F00000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IRC
Bit Default &
Description
Range Access
00f00000h
31:0 Interrupt Route Capability (IRC): Indicates support for IRQ20, 21, 22, 23
RO
21.9.3.9 Timer 0 Comparator Value Register - Lower 32 Bits (T0CV_1)—Offset

108h
Access Method
(Size: 32 bits) T0CV_1: [0xFED00000] + 108h
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
CV
Bit Default &

Description
Range Access
FFFFFFFFh Comparator Value (CV): Reads return the current value of the lower 32 bits of the
31:0
RW comparator. Writes load the new value to the lower 32 bits of the comparator.
21.9.3.10 Timer 0 Comparator Value Register - Upper 32 Bits (T0CV_2)—Offset

10Ch
Access Method
Type: Memory Mapped I/O Register T0CV_2: [0xFED00000] + 10Ch
(Size: 32 bits)
Default: FFFFFFFFh

31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
CV
Bit Default &
Description
Range Access
FFFFFFFFh Comparator Value (CV): Reads return the current value of the upper 32 bits of the
31:0
RW comparator. Writes load the new value to the upper 32 bits of the comparator.

Offset 120h
Access Method
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FID
FE
TVS
TS
IE
IT
RSV3
RSV2
PIC
TYP
RSV1
IR
T32M
Bit Default &
Description
Range Access
0b
RO
0b
15 FSB Interrupt Delivery (FID): Not Supported.
RO
0b
RO
13:9
no effect.
0b
8 Timer 32-bit Mode (T32M): Not applicable since Timer 1 is a 32-bit timer.
RO
0b
RO
6
WO for Timer 0 if it is set to periodic mode.
0b
RO
4
RO this timer's interrupt. This bit is set for timer 0, and cleared for timers 1 and 2
3 timer to generate a periodic interrupt. This bit is RW for timer 0, and RO for timers 1
RO and 2.
RW an interrupt.

Bit Default &

Description
Range Access
1
active.
0b
RO

Offset 124h
Access Method
Default: 00F00000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
IRC
Bit Default &

Description
Range Access
00f00000h
31:0 Interrupt Route Capability (IRC): Indicates support for IRQ20, 21, 22, 23
RO
21.9.3.13 Timer 1 Comparator Value Register (T1CV_1)—Offset 128h

Access Method
Type: Memory Mapped I/O Register T1CV_1: [0xFED00000] + 128h
(Size: 32 bits)
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
CV
Bit Default &

Description
Range Access
FFFFFFFFh Comparator Value (CV): Reads return the current value of the 32 bits of the
31:0
RW comparator. Writes load the new value to the 32 bits of the comparator.

Offset 140h
Access Method
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
FID
IR
FE
TVS
TS
IE
IT
RSV3
T32M
RSV2
PIC
TYP
RSV1
Bit Default &
Description
Range Access
0b
RO
0b
15 FSB Interrupt Delivery (FID): Not Supported.
RO
0b
RO
13:9
no effect.
0b
8 Timer 32-bit Mode (T32M): Not applicable since Timer 2 is a 32-bit timer.
RO
0b
RO
6
WO for Timer 0 if it is set to periodic mode.
0b
RO
4
RO this timer's interrupt.
3
RO timer to generate a periodic interrupt.
RW an interrupt.
1
active.
0b
RO

Offset 144h
Access Method
Type: Memory Mapped I/O Register T2C_2: [0xFED00000] + 144h
(Size: 32 bits)
Default: 00F00800h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0
IRC

Bit Default &

Description
Range Access
00f00800h
31:0 Interrupt Route Capability (IRC): Indicates support for IRQ11, 20, 21, 22, 23
RO
21.9.3.16 Timer 2 Comparator Value Register (T2CV_1)—Offset 148h

Access Method
Type: Memory Mapped I/O Register T2CV_1: [0xFED00000] + 148h
(Size: 32 bits)
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
CV
Bit Default &
Description
Range Access
FFFFFFFFh Comparator Value (CV): Reads return the current value of the 32 bits of the
31:0
RW comparator. Writes load the new value to the 32 bits of the comparator.
21.9.4 References
IA-PC HPET (High Precision Event Timers) Specification, Revision 1.0a.
21.10 Real Time Clock (RTC)

The SoC contains a Motorola MC146818B-compatible real-time clock with 242 bytes of
battery-backed RAM. The real-time clock performs two key functions—keeping track of
the time of day and storing system data, even when the system is powered down. The
RTC operates on an external 32.768 KHz crystal and a 3.3 V battery. An option is also
provided to internally generate the 32.768 KHz clock.
The RTC supports two lockable memory ranges. By setting bits in the configuration
space, two 8-byte ranges can be locked to read and write accesses. This prevents
unauthorized reading of passwords or other system security information.
The RTC supports a date alarm that allows for scheduling a wake up event up to 30
days in advance.



Table 143. RTC Signals

Direction/
Type
I/O
RTCX1 Crystal Input 1: This signal is connected to the 32.768 KHz crystal.
Analog
I/O
RTCX2 Crystal Input 2: This signal is connected to the 32.768 KHz crystal.
Analog
I RTC Reset: When asserted, this signal resets register bits in the RTC
RTCRST_B
CMOS3.3 well.
I/O External Capacitor Connection: Connect to an edge cap (0.1uF ) on

IVCCRTCEXT
Analog the motherboard to ground, inductance has to be < 1nH.
RTC Internal Clock Select: Used to select between the oscillator

clock from the external 32.768 KHz crystal or an internally generated
I
RTC_EXT_CLK_EN_B 32.768 KHz clock.
CMOS3.3
0 = External 32.768 KHz oscillator
1 = Internal 32.768 KHz Clock (Default)
21.10.2 Features
The Real Time Clock (RTC) module provides a battery backed-up date and time keeping
device. Three interrupt features are available: time of day alarm with once a second to
once a month range, periodic rates of 122 ms to 500 ms, and end of update cycle
notification. Seconds, minutes, hours, days, day of week, month, and year are counted.
The hour is represented in twelve or twenty-four hour format, and data can be
represented in BCD or binary format. The design is meant to be functionally compatible
with the Motorola MS146818B. The time keeping comes from a 32.768 KHz oscillating
source, which is divided to achieve an update every second. The lower 14 bytes on the
lower RAM block have very specific functions. The first ten are for time and date
information. The next four (0Ah to 0Dh) are registers, which configure and report RTC
functions. A host-initiated write takes precedence over a hardware update in the event
of a collision.
21.10.2.1 Update Cycles

An update cycle occurs once a second, if the B.SET bit is not asserted and the divide
chain is properly configured. During this procedure, the stored time and date are
incremented, overflow checked, a matching alarm condition is checked, and the time
and date are rewritten to the RAM locations. The update cycle starts at least 488 ms
after A.UIP is asserted, and the entire cycle does not take more than 1984 ms to
complete. The time and date RAM locations (00h to 09h) are disconnected from the
external bus during this time.
21.10.2.2 Interrupts
The real-time clock interrupt is internally routed within the SoC both to the I/O APIC
and the 8259. It is mapped to interrupt vector 8. This interrupt does not leave the SoC,
nor is it shared with any other interrupt. However, the High Performance Event Timers
can also be mapped to IRQ8#; in this case, the RTC interrupt is blocked.
21.10.2.3 Lockable RAM Ranges

The RTC battery-backed RAM supports two 8-byte ranges that can be locked via the
RTC Configuration register. When the locking bits are set, the corresponding range in
the RAM is not readable or writable. A write cycle to those locations has no effect. A
read cycle to those locations does not return the location’s actual value (resultant value
is undefined).

Once a range is locked, the range can be unlocked only by a hard reset, which invokes
the BIOS and allows it to relock the RAM range.
21.10.3 Register Map

Figure 54. RTC Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
Memory
Space
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem) Legacy PCI
Header
D:31,F:0
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
D:21
SPI1 F:1

Fixed IO
D:31,F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7
21.10.4 I/O Registers

The RTC internal registers and RAM are organized as two banks of 128 bytes each,
called the standard and extended banks.

The first 14 bytes of the standard bank contain the RTC time and date information
along with four registers, A - D, that are used for configuration of the RTC. The
extended bank contains a full 128 bytes of battery backed SRAM. All data movement
between the host CPU and the RTC is done through registers mapped to the standard I/
O space.
Note: It is not possible to disable the extended bank.
Note: I/O Locations 70h and 71h are used for data movement to and from the standard bank.
Locations 72h and 73h used for data movement to and from the extended bank. All of
these I/O locations also have alias I/O locations, as indicated in Table 144. Index
addresses above 127 are not valid.
Note: Writes to 74h do not affect the NMI Enable bit of 70h
Table 144. I/O Registers Alias Locations

Register Original I/O Location Alias I/O Location
Real-Time Clock (Standard RAM)

70h 74h
Index Register
Real-Time Clock (Standard RAM)

71h 75h
Target Register
Extended RAM Index Register 72h 76h
Extended RAM Target Register 73h 77h
21.10.5 Indexed Registers

The RTC contains two sets of indexed registers, which are accessed using the two
separate Index and Target registers (70/71h or 72/73h).
Table 145. Indexed Registers

Start End Name
00h 00h Seconds
01h 01h Seconds Alarm
02h 02h Minutes
03h 03h Minutes Alarm
04h 04h Hours
05h 05h Hours Alarm
06h 06h Day of Week
07h 07h Day of Month
08h 08h Month
09h 09h Year
0Ah 0Ah Register A
0Bh 0Bh Register B
0Ch 0Ch Register C
0Dh 0Dh Register D
0Eh 7Fh 114 Bytes of User RAM

21.10.5.1 Offset 0Ah: Register A

This register is in the RTC well, and is used for general configuration of the RTC
functions.
Access Method
Type: RTC Indexed Register

(Size: 8 bits)
Default: xxxxxxxxb
7 4 0
x xxx xxxx
UIP
DV
RS
Bit Default &
Description
Range Access
xb Update in progress (UIP): When set, an update is in progress. When cleared, the
7 update cycle will not start for at least 488 µs. The time, calendar, and alarm information
RW in RAM is always available when this bit is cleared.
Division Chain Select: Controls the divider chain for the oscillator; not affected by
RSMRST# or any other reset signal.
000b: Invalid
001b: Invalid
xb
6: 4 010b: Normal Operation
RW 011b: Bypass 5 Stages (Test Mode Only)
100b: Bypass 10 Stages (Test Mode Only)
101b: Bypass 15 Stages (Test Mode Only)
110b: Divider Reset
111b: Divider Reset
Rate Select: Selects one of 13 taps of the 15 stage divider chain. The selected tap can
generate a periodic interrupt when B.PIE bit is set. Otherwise this tap sets C.PF.
0000b: Interrupt Never Toggles

0001b: 3.90625 ms
0010b: 7.8125 ms
0011b: 122.070 μs
0100b: 244.141 μs
0101b: 488.281 μs
xb
3: 0 0110b: 976.5625 μs
RW 0111b: 1.953125 ms
1000b: 3.90625 ms
1001b: 7.8125 ms
1010b: 15.625 ms
1011b: 31.25 ms
1100b: 62.5 ms
1101b: 125 ms
1110b: 250 ms
1111b: 500 ms
21.10.5.2 Offset 0Bh: Register B - General Configuration

This register resides in the resume well.
Access Method

(Size: 8 bits)
Default: x0x00xxxb

7 4 0
x 0 x 0 0 x x x
DM
HF
SET
PIE
AIE
UIE
SQWE
DSE
Bit Default &
Description
Range Access
xb Set Clock (SET): When cleared, an update cycle occurs once each second. If set, a
7 current update cycle will abort and subsequent update cycles will not occur until SET is
RW returned to zero. When set, SW may initialize time and calendar bytes safely.
0b Periodic Interrupt Enable (PIE): When set, and C.PF is set, an interrupt is
6 generated.
RW
xb
5 Alarm Interrupt Enable (AIE): When set, and C.AF is set, an interrupt is generated.
RW
0b Update-ended Interrupt Enable (UIE): When set and C.UF is set, an interrupt is
4 generated.
RW
0b
3 Square Wave Enable (SQWE): Not implemented.
RW
xb Data Mode (DM): When set, represents binary representation. When cleared, denotes
2 BCD.
RW
xb Hour Format (HF): When set, twenty-four hour mode is selected. When cleared,
1 twelve-hour mode is selected. In twelve hour mode, the seventh bit represents AM
RW (cleared) and PM (set).
xb
0 Daylight Savings Enable (DSE): Not implemented
RW
21.10.5.3 Offset 0Ch: Register C - Flag Register

All bits in this register are cleared when this register is read.
Access Method

(Size: 8 bits)
Default: 00x00000b
7 4 0
0 0 x 0 0000
PF
AF
UF
RSV1
IRQF
Bit Default &

Description
Range Access
0b Interrupt Request Flag (IRQF): This bit is an AND of the flag with its corresponding
7
RC interrupt enable in register B, and causes the RTC Interrupt to be asserted.
0b
6 Periodic Interrupt Flag (PF): Set when the tap as specified by A.RS is one.
RC
xb
5 Alarm Flag (AF): Set after all Alarm values match the current time.
RC
0b
4 Update-ended Flag (UF): Set immediately following an update cycle for each second.
RC

Bit Default &

Description
Range Access
0b
3: 0 Reserved (RSV1)
RO
21.10.5.4 Offset 0Dh: Register D - Flag Register

Access Method

(Size: 8 bits)
Default: 1xxxxxxxb
7 4 0
1 x xxxxxx
RSV1
VRT
DA
Bit Default &
Description
Range Access
0b Valid RAM and Time Bit (VRT): This bit should always be written as a 0 for write
7
RC cycle, however it will return a 1 for read cycles.
0b
6 Reserved (RSV1): This bit always returns a 0 and should be set to 0 for write cycles.
RC
Date Alarm (DA): These bits store the date of month alarm value. If set to 000000,
xb then a don’t care state is assumed. If the date alarm is not enabled, these bits will
5: 0 return zeros to mimic the functionality of the Motorola 146818B. These bits are not
RC
affected by any reset assertion.
21.10.6 References
Accessing the Real Time Clock Registers and the NMI Enable Bit:
ftp://download.intel.com/design/intarch/PAPERS/321088.pdf
21.11 Interrupt Decoding & Routing

The Legacy Bridge provides registers that are used to decode and route interrupts
received from devices within the SoC.
The interrupt decoder is responsible for receiving interrupt messages from other
devices in the SoC and decoding them for consumption by the interrupt router, the
8259 PICs and/or the I/O APIC.
The interrupt router is responsible for mapping each incoming interrupt to the
appropriate PIRQx, for consumption by the 8259 PICs and/or I/O APIC.
21.11.1 Features
21.11.1.1 Interrupt Decoder

The interrupt decoder receives interrupt messages from devices in the SoC. These
interrupts can be split into two primary groups:
• For consumption by the interrupt router
• For consumption by the 8259 PIC

21.11.1.1.1 For Consumption by the Interrupt Router
When a PCI-mapped device in the SoC asserts or de-asserts an INT[A:D] interrupt, an

interrupt message is sent to the decoder. This message is decoded to indicate to the
interrupt router which specific interrupt is asserted or de-asserted and which device the
INT[A:D] interrupt originated from.
21.11.1.1.2 For Consumption by the 8259 PIC
When a device in the SoC asserts or deasserts a legacy interrupt (IRQ), an interrupt
message is sent to the decoder. This message is decoded to indicate to the 8259 PIC
which specific interrupt (IRQ[3, 4, 5, 6, 7, 13, 14 or 15]) was asserted or deasserted.
21.11.1.2 Interrupt Router

The interrupt router aggregates the INT[A:D] interrupts for each PCI-mapped device in
the SoC, received from the interrupt decoder. It then maps these aggregated interrupts
to 8 PCI-based interrupts: PIRQ[A:H]. This mapping is configured using the 4 Interrupt
Queue Agent Registers: IRQAGENT0, IRQAGENT1, IRQAGENT2 and IRQAGENT3.
Table 146. IRQAGENT Description

IRQAGENT Description Single-Function/ Multi-Function
0 Remote Management Unit Single-Function (Supports INTA only)
PCIe*
1 Multi-Function (Supports INTA, INTB, INTC & INTD)
D:23
2 Reserved Single-Function (Supports INTA only)
IO Fabric
3 Multi-Function (Supports INTA, INTB, INTC & INTD)
D:20 & D21
PCI based interrupts PIRQ[A:H] are then available for consumption by either the 8259
PICs or the IO-APIC, depending on the configuration of the 8 PIRQx Routing Control
Registers: PIRQA, PIQRB, PIRQC, PIRQD, PIRQE, PIRQF, PIRQG, PIRQH.
When the PCI based interrupts are consumed by the IO-APIC, a fixed routing scheme is
used where interrupts PIRQ[A:H] are routed to IO-APIC interrupts IRQ[16:23].
21.11.1.2.1 Routing PCI Based Interrupts to 8259 PIC
The interrupt router can be programmed to allow PIRQA-PIRQH to be routed internally

to the 8259 as ISA compatible interrupts IRQ 3–7, 9–12 & 14–15. The assignment is
programmable through the 8 PIRQx Routing Control Registers: PIRQA, PIQRB, PIRQC,
PIRQD, PIRQE, PIRQF, PIRQG, PIRQH. See Section 21.3 for register details. One or
more PIRQs can be routed to the same IRQ input. If ISA Compatible Interrupts are not
required, the Route registers can be programmed to disable steering.
The PIRQx# lines are defined as active low, level sensitive. When a PIRQx# is routed to
specified IRQ line, software must change the IRQ's corresponding ELCR bit to level
sensitive mode. The SoC internally inverts the PIRQx# line to send an active high level
to the PIC. When a PCI interrupt is routed onto the PIC, the selected IRQ can no longer
be used by an active high device (through SERIRQ). However, active low interrupts can
share their interrupt with PCI interrupts.
21.12 8259 Programmable Interrupt Controllers (PIC)

The SoC provides an ISA-compatible programmable interrupt controller (PIC) that
incorporates the functionality of two, cascaded 8259 interrupt controllers.

21.12.1 Features
In addition to providing support for ISA compatible interrupts, this interrupt controller
can also support PCI based interrupts (PIRQs) by mapping the PCI interrupt onto a
compatible ISA interrupt line. Each 8259 controller supports eight interrupts, numbered
0–7. Table 147 shows how the controllers are connected.
Note: The SoC does not implement any external PIRQ# signals. The PIRQs referred to in this
section originate from the interrupt routing unit.
Table 147. Interrupt Controller Connections

8259
8259 Connected Pin / Function
Input
0 Internal Timer / Counter 0 output or HPET Timer #0
1 Reserved
2 Slave controller INTR output
3 IRQ3 via PIRQx

Master
4 IRQ4 via PIRQx
5 IRQ5 via PIRQx
6 IRQ6 via PIRQx
7 IRQ7 via PIRQx
0 Inverted IRQ8# from internal RTC or HPET Timer #1
1 IRQ9 via SCI or PIRQx
2 IRQ10 via SCI or PIRQx
3 IRQ11 via SCI or PIRQx or HPET Timer #2

Slave
4 IRQ12 via PIRQx
5 Reserved
6 IRQ14 via PIRQx
7 IRQ15 via PIRQx
The SoC cascades the slave controller onto the master controller through master
controller interrupt input 2. This means there are only 15 possible interrupts for the
SoC PIC.
Interrupts can be programmed individually to be edge or level, except for IRQ0, IRQ2
and IRQ8#.
Note: Active-low interrupt sources (such as a PIRQ#) are inverted inside the SoC. In the
following descriptions of the 8259s, the interrupt levels are in reference to the signals
at the internal interface of the 8259s, after the required inversions have occurred.
Therefore, the term “high” indicates “active,” which means “low” on an originating
PIRQ#.
21.12.1.1 Interrupt Handling
21.12.1.1.1 Generating Interrupts
The PIC interrupt sequence involves three bits, from the IRR, ISR, and IMR, for each
interrupt level. These bits are used to determine the interrupt vector returned, and
status of any other pending interrupts. Table 148 defines the IRR, ISR, and IMR.

Table 148. Interrupt Status Registers

Bit Description
Interrupt Request Register. This bit is set on a low to high transition of the interrupt line in edge
IRR
mode, and by an active high level in level mode.
Interrupt Service Register. This bit is set, and the corresponding IRR bit cleared, when an
ISR
interrupt acknowledge cycle is seen, and the vector returned is for that interrupt.
Interrupt Mask Register. This bit determines whether an interrupt is masked. Masked interrupts
IMR
will not generate INTR.
21.12.1.1.2 Acknowledging Interrupts
The processor generates an interrupt acknowledge cycle that is translated into a

Interrupt Acknowledge Cycle by the SoC. The PIC translates this command into two
internal INTA# pulses expected by the 8259 controllers. The PIC uses the first internal
INTA# pulse to freeze the state of the interrupts for priority resolution. On the second
INTA# pulse, the master or slave sends the interrupt vector to the processor with the
acknowledged interrupt code. This code is based upon the ICW2.IVBA bits, combined
with the ICW2.IRL bits representing the interrupt within that controller.
Note: References to ICWx and OCWx registers are relevant to both the master and slave
8259 controllers.
Table 149. Content of Interrupt Vector Byte

Master, Slave Interrupt Bits [7:3] Bits [2:0]
IRQ7,15 111
IRQ6,14 110
IRQ5,13 101
IRQ4,12 100
ICW2.IVBA
IRQ3,11 011
IRQ2,10 010
IRQ1,9 001
IRQ0,8 000
21.12.1.1.3 Hardware/Software Interrupt Sequence

1. One or more of the Interrupt Request lines (IRQ) are raised high in edge mode, or
seen high in level mode, setting the corresponding IRR bit.
2. The PIC sends INTR active to the processor if an asserted interrupt is not masked.
3. The processor acknowledges the INTR and responds with an interrupt acknowledge
cycle.
4. Upon observing the special cycle, the SoC converts it into the two cycles that the
internal 8259 pair can respond to. Each cycle appears as an interrupt acknowledge
pulse on the internal INTA# pin of the cascaded interrupt controllers.
5. Upon receiving the first internally generated INTA# pulse, the highest priority ISR
bit is set and the corresponding IRR bit is reset. On the trailing edge of the first
pulse, a slave identification code is broadcast by the master to the slave on a
private, internal three bit wide bus. The slave controller uses these bits to
determine if it must respond with an interrupt vector during the second INTA#
pulse.

6. Upon receiving the second internally generated INTA# pulse, the PIC returns the
interrupt vector. If no interrupt request is present because the request was too
short in duration, the PIC returns vector 7 from the master controller.
7. This completes the interrupt cycle. In AEOI mode the ISR bit is reset at the end of
the second INTA# pulse. Otherwise, the ISR bit remains set until an appropriate
EOI command is issued at the end of the interrupt subroutine.
21.12.1.2 Initialization Command Words (ICWx)

Before operation can begin, each 8259 must be initialized. This is a four byte sequence
to ICW1, ICW2, ICW3 and ICW4. The four initialization command words are referred to
by their acronyms: ICW1, ICW2, ICW3, and ICW4.
The base address for each 8259 initialization command word is a fixed location in the I/
O memory space: 20h for the master controller, and A0h for the slave controller.
21.12.1.2.1 ICW1
A write to the master or slave controller base address with data bit 4 equal to 1 is
interpreted as a write to ICW1. Upon sensing this write, the PIC expects three more
byte writes to 21h for the master controller, or A1h for the slave controller, to complete
the ICW sequence.
A write to ICW1 starts the initialization sequence during which the following
automatically occur:
1. Following initialization, an interrupt request (IRQ) input must make a low-to-high
transition to generate an interrupt.
2. The Interrupt Mask Register is cleared.
3. IRQ7 input is assigned priority 7.
4. The slave mode address is set to 7.
5. Special mask mode is cleared and Status Read is set to IRR.
21.12.1.2.2 ICW2
The second write in the sequence (ICW2) is programmed to provide bits [7:3] of the
interrupt vector that will be released during an interrupt acknowledge. A different base
is selected for each interrupt controller.
21.12.1.2.3 ICW3
The third write in the sequence (ICW3) has a different meaning for each controller.
• For the master controller, ICW3 is used to indicate which IRQ input line is used to
cascade the slave controller. Within the SoC, IRQ2 is used. Therefore, MICW3.CCC
is set to a 1, and the other bits are set to 0s.
• For the slave controller, ICW3 is the slave identification code used during an
interrupt acknowledge cycle. On interrupt acknowledge cycles, the master
controller broadcasts a code to the slave controller if the cascaded interrupt won
arbitration on the master controller. The slave controller compares this
identification code to the value stored in its ICW3, and if it matches, the slave
controller assumes responsibility for broadcasting the interrupt vector.
21.12.1.2.4 ICW4
The final write in the sequence (ICW4) must be programmed for both controllers. At
the very least, ICW4.MM must be set to a 1 to indicate that the controllers are
operating in an Intel Architecture-based system.

21.12.1.3 Operation Command Words (OCW)

These command words reprogram the Interrupt controller to operate in various
interrupt modes.
• OCW1 masks and unmasks interrupt lines.
• OCW2 controls the rotation of interrupt priorities when in rotating priority mode,
and controls the EOI function.
• OCW3 sets up ISR/IRR reads, enables/disables the special mask mode (SMM), and
enables/disables polled interrupt mode.
21.12.1.4 Modes of Operation
21.12.1.4.1 Fully Nested Mode
In this mode, interrupt requests are ordered in priority from 0 through 7, with 0 being
the highest. When an interrupt is acknowledged, the highest priority request is
determined and its vector placed on the bus. Additionally, the ISR for the interrupt is
set. This ISR bit remains set until either: the processor issues an EOI command
immediately before returning from the service routine; or if in AEOI mode, on the
trailing edge of the second INTA#. While the ISR bit is set, all further interrupts of the
same or lower priority are inhibited, while higher levels generate another interrupt.
Interrupt priorities can be changed in the rotating priority mode.
21.12.1.4.2 Special Fully-Nested Mode
This mode is used in the case of a system where cascading is used, and the priority has
to be conserved within each slave. In this case, the special fully-nested mode is
programmed to the master controller. This mode is similar to the fully-nested mode
with the following exceptions:
• When an interrupt request from a certain slave is in service, this slave is not locked
out from the master's priority logic and further interrupt requests from higher
priority interrupts within the slave are recognized by the master and initiate
interrupts to the processor. In the normal-nested mode, a slave is masked out
when its request is in service.
• When exiting the Interrupt Service routine, software has to check whether the
interrupt serviced was the only one from that slave. This is done by sending a non-
Specific EOI command to the slave and then reading its ISR. If it is 0, a non-specific
EOI can also be sent to the master.
21.12.1.4.3 Automatic Rotation Mode (Equal Priority Devices)
In some applications, there are a number of interrupting devices of equal priority.

Automatic rotation mode provides for a sequential 8-way rotation. In this mode, a
device receives the lowest priority after being serviced. In the worst case, a device
requesting an interrupt has to wait until each of seven other devices are serviced at
most once.
There are two ways to accomplish automatic rotation using OCW2.REOI; the Rotation
on Non-Specific EOI Command (OCW2.REOI=101b) and the rotate in automatic EOI
mode which is set by (OCW2.REOI=100b).

21.12.1.4.4 Specific Rotation Mode (Specific Priority)
Software can change interrupt priorities by programming the bottom priority. For
example, if IRQ5 is programmed as the bottom priority device, then IRQ6 is the highest
priority device. The Set Priority Command is issued in OCW2 to accomplish this, where:
OCW2.REOI=11xb, and OCW2.ILS is the binary priority level code of the bottom
priority device.
In this mode, internal status is updated by software control during OCW2. However, it
is independent of the EOI command. Priority changes can be executed during an EOI
command by using the Rotate on Specific EOI Command in OCW2 (OCW2.REOI=111b)
and OCW2.ILS=IRQ level to receive bottom priority.
21.12.1.4.5 Poll Mode
Poll mode can be used to conserve space in the interrupt vector table. Multiple
interrupts that can be serviced by one interrupt service routine do not need separate
vectors if the service routine uses the poll command. Poll mode can also be used to
expand the number of interrupts. The polling interrupt service routine can call the
appropriate service routine, instead of providing the interrupt vectors in the vector
table. In this mode, the INTR output is not used and the microprocessor internal
Interrupt Enable flip-flop is reset, disabling its interrupt input. Service to devices is
achieved by software using a Poll command.
The Poll command is issued by setting OCW3.PMC. The PIC treats its next I/O read as
an interrupt acknowledge, sets the appropriate ISR bit if there is a request, and reads
the priority level. Interrupts are frozen from the OCW3 write to the I/O read. The byte
returned during the I/O read contains a 1 in Bit 7 if there is an interrupt, and the binary
code of the highest priority level in Bits 2:0.
21.12.1.4.6 Edge and Level Triggered Mode
In ISA systems this mode is programmed using ICW1.LTIM, which sets level or edge for
the entire controller. In the SoC, this bit is disabled and a register for edge and level
triggered mode selection, per interrupt input, is included. This is the Edge/Level control
Registers ELCR1 and ELCR2.
If an ELCR bit is 0, an interrupt request will be recognized by a low-to-high transition

on the corresponding IRQ input. The IRQ input can remain high without generating
another interrupt. If an ELCR bit is 1, an interrupt request will be recognized by a high
level on the corresponding IRQ input and there is no need for an edge detection. The
interrupt request must be removed before the EOI command is issued to prevent a
second interrupt from occurring.
In both the edge and level triggered modes, the IRQ inputs must remain active until
after the falling edge of the first internal INTA#. If the IRQ input goes inactive before
this time, a default IRQ7 vector is returned.
21.12.1.4.7 End of Interrupt (EOI) Operations
An EOI can occur in one of two fashions: by a command word write issued to the PIC
before returning from a service routine, the EOI command; or automatically when the
ICW4.AEOI bit is set to 1.
21.12.1.4.8 Normal End of Interrupt
In normal EOI, software writes an EOI command before leaving the interrupt service
routine to mark the interrupt as completed. There are two forms of EOI commands:
Specific and Non-Specific. When a Non-Specific EOI command is issued, the PIC clears
the highest ISR bit of those that are set to 1. Non-Specific EOI is the normal mode of
operation of the PIC within the SoC, as the interrupt being serviced currently is the

interrupt entered with the interrupt acknowledge. When the PIC is operated in modes
that preserve the fully nested structure, software can determine which ISR bit to clear
by issuing a Specific EOI.
An ISR bit that is masked is not cleared by a Non-Specific EOI if the PIC is in the special
mask mode. An EOI command must be issued for both the master and slave controller.
21.12.1.4.9 Automatic End of Interrupt Mode
In this mode, the PIC automatically performs a Non-Specific EOI operation at the
trailing edge of the last interrupt acknowledge pulse. From a system standpoint, this
mode should be used only when a nested multi-level interrupt structure is not required
within a single PIC. The AEOI mode can only be used in the master controller and not
the slave controller.
Note: Both the master and slave PICs have an AEOI bit: MICW4.AEOI and SICW4.AEOI
respectively. Only the MICW4.AEOI bit should be set by software. The SICW4.AEOI bit
should not be set by software.
21.12.1.5 Masking Interrupts
21.12.1.5.1 Masking on an Individual Interrupt Request
Each interrupt request can be masked individually by the Interrupt Mask Register
(IMR). This register is programmed through OCW1. Each bit in the IMR masks one
interrupt channel. Masking IRQ2 on the master controller masks all requests for service
from the slave controller.
21.12.1.5.2 Special Mask Mode
Some applications may require an interrupt service routine to dynamically alter the
system priority structure during its execution under software control. For example, the
routine may wish to inhibit lower priority requests for a portion of its execution but
enable some of them for another portion.
The special mask mode enables all interrupts not masked by a bit set in the Mask
register. Normally, when an interrupt service routine acknowledges an interrupt without
issuing an EOI to clear the ISR bit, the interrupt controller inhibits all lower priority
requests. In the special mask mode, any interrupts may be selectively enabled by
loading the Mask Register with the appropriate pattern.
The special mask mode is set by OCW3.ESMM=1b & OCW3.SMM=1b, and cleared
where OCW3.ESMM=1b & OCW3.SMM=0b.
21.12.1.6 Steering of PCI Interrupts

The SoC can be programmed to allow PIRQ[A:H]# to be internally routed to interrupts
3-7, 9-12, 14 or 15, through the PIRQx Route Control registers in Device 31:Function
0. One or more PIRQx# lines can be routed to the same IRQx input. The PIRQx# lines
are defined as active low, level sensitive. When a PIRQx# is routed to specified IRQ
line, software must change the corresponding ELCR1 or ELCR2 register to level
sensitive mode. The SoC will internally invert the PIRQx# line to send an active high
level to the PIC. When a PCI interrupt is routed onto the PIC, the selected IRQ can no
longer be used by an ISA device


Figure 55. 8259 Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
Memory
Space
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem) Legacy PCI
Header
D:31,F:0
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
D:21
SPI1 F:1

Fixed IO
D:31,F:0
SDIO/eMMC F:0
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7
21.12.3 I/O Registers

The interrupt controller registers are located at 20h and 21h for the master controller
(IRQ0-7), and at A0h and A1h for the slave controller (IRQ8-13). These registers have
multiple functions, depending upon the data written to them. Table 150 describes the
different register possibilities for each address.
Note: The register descriptions after Table 150 represent one register possibility.

Table 150. 8259 I/O Registers Alias Locations

Registers Original I/O Location Alias I/O Locations
24h
28h
MICW1
2Ch
MOCW2 20h 30h
34h
MOCW3
38h
3Ch
25h
MICW2 29h
2Dh
MICW3
21h 31h
MICW4
35h
MOCW1 39h
3Dh
A4h
A8h
SICW1
ACh
SoCW2 A0h B0h
B4h
SoCW3
B8h
BCh
A5h
SICW2 A9h
ADh
SICW3
A1h B1h
SICW4
B5h
SoCW1 B9h
BDh
ELCR1 4D0h N/A
ELCR2 4D1h N/A
Table 151. Summary of I/O Registers

Offset Offset Default
Register ID—Description
Start End Value
“Master Initialization Command Word 1 (MICW1)—Offset 20h” on

20h 20h 81Fh
page 908

21h 21h 63h
page 909
“Master Operational Control Word 2 (MOCW2)—Offset 24h” on

24h 24h 67h
page 909

Table 151. Summary of I/O Registers (Continued)

Offset Offset Default
Register ID—Description
Start End Value

25h 25h E7h
page 910
“Master Operational Control Word 3 (MOCW3)—Offset 28h” on

28h 28h 19Eh
page 910

29h 29h 421h
page 911
“Master Operational Control Word 1 (MOCW1)—Offset 2Dh” on

2Dh 2Dh 00h
page 912
“Slave Initialization Command Word 1 (SICW1)—Offset A0h” on

A0h A0h 81Fh
page 912

A1h A1h 63h
page 913
“Slave Operational Control Word 2 (SoCW2)—Offset A4h” on

A4h A4h 67h
page 913

A5h A5h E7h
page 914
“Slave Operational Control Word 3 (SoCW3)—Offset A8h” on

A8h A8h 19Eh
page 914

A9h A9h 421h
page 915
“Slave Operational Control Word 1 (SoCW1)—Offset ADh” on

ADh ADh 00h
page 916
4D0h 4D0h “Master Edge/Level Control (ELCR1)—Offset 4D0h” on page 916 108h
4D1h 4D1h “Slave Edge/Level Control (ELCR2)—Offset 4D1h” on page 916 14Ah
21.12.3.1 Master Initialization Command Word 1 (MICW1)—Offset 20h

A write to Initialization Command Word 1 starts the interrupt controller initialization
sequence, during which the following occurs:
• The Interrupt Mask register is cleared.
• IRQ7 input is assigned priority 7.
• The slave mode address is set to 7.
• Special Mask Mode is cleared and Status Read is set to IRR.
Once this write occurs, the controller expects writes to ICW2, ICW3, and ICW4 to
complete the initialization sequence.
Access Method
Type: I/O Register

(Size: 8 bits) MICW1: 20h
7 4 0
X X X X X X X X
MCS85
LTIM
ADI
SNGL
IC4
ICWOCWSEL

Bit Default &

Description
Range Access
X MCS85 (MCS85): These bits are MCS-85 specific, and not needed. Should be
7: 5
WO programmed to 000
X ICW/OCW Select (ICWOCWSEL): This bit must be a 1 to select ICW1 and enable the
4
WO ICW2, ICW3, and ICW4 sequence.
X
3 Edge/Level Bank Select (LTIM): Disabled. Replaced by ELCR1 and ELCR2.
WO
X
2 ADI (ADI): Should be programmed to 0.
WO
X Single or Cascade (SNGL): Must be programmed to a 0 to indicate two controllers

1
WO operating in cascade mode.
X wICW4 Write Required (IC4): This bit must be programmed to a 1 to indicate that
0
WO ICW4 needs to be programmed.

Master ICW2 is used to initialize the interrupt controller with the five most significant
bits of the interrupt vector address. The value programmed for bits[7:3] is used by the
CPU to define the base address in the interrupt vector table for the interrupt routines
associated with each IRQ on the controller. Typical ISA ICW2 values are 08h for the
master controller and 70h for the slave controller.
Access Method
Type: I/O Register

7 4 0
X X X X X X X X
IVBA
IRL
Bit Default &
Description
Range Access
X Interrupt Vector Base Address (IVBA): Bits [7:3] define the base address in the
7: 3 interrupt vector table for the interrupt routines associated with each interrupt request
WO level input.
Interrupt Request Level (IRL): When writing ICW2, these bits should all be 0. During
an interrupt acknowledge cycle, these bits are programmed by the interrupt controller
with the interrupt to be serviced. This is combined with bits [7:3] to form the interrupt
vector driven onto the data bus during the second INTA# cycle. The code is a three bit
binary code:
Code Master Interrupt Slave Interrupt
X 000 IRQ0 IRQ8
2: 0 001 IRQ1 IRQ9
WO
010 IRQ2 IRQ10
011 IRQ3 IRQ11
100 IRQ4 IRQ12
101 IRQ5 IRQ13
110 IRQ6 IRQ14
111 IRQ7 IRQ15
21.12.3.3 Master Operational Control Word 2 (MOCW2)—Offset 24h

Following a part reset or ICW initialization, the controller enters the fully nested mode
of operation. Non-specific EOI without rotation is the default. Both rotation mode and
specific EOI mode are disabled following initialization.
Access Method

Type: I/O Register MOCW2: 24h

(Size: 8 bits)
7 4 0
0 0 1 X X X X X
OCW2S
ILS
Bit REOI
Default &
Description
Range Access
Rotate and EOI Codes (REOI): R, SL, EOI - These three bits control the Rotate and
End of Interrupt modes and combinations of the two. A chart of these combinations is
listed above under the bit definition.
000 - Rotate in Auto EOI Mode (Clear)
001 - Non-specific EOI command
001b 010 - No Operation
7: 5 011 - *Specific EOI Command
WO
100 - Rotate in Auto EOI Mode (Set)
101 - Rotate on Non-Specific EOI Command
110 - *Set Priority Command
111 - *Rotate on Specific EOI Command
*L0 - L2 Are Used
X
4: 3 OCW2 Select (OCW2S): When selecting OCW2, bits 4:3 = 00
WO
Interrupt Level Select (L2, L1, L0) (ILS): L2, L1, and L0 determine the interrupt
level acted upon when the SL bit is active. A simple binary code, outlined above, selects
the channel for the command to act upon. When the SL bit is inactive, these bits do not
X have a defined function; programming L2, L1 and L0 to 0 is sufficient in this case.
2: 0 Bits Interrupt Level Bits Interrupt Level
WO 000 IRQ0/8 100 IRQ4/12
001 IRQ1/9 101 IRQ5/13
010 IRQ2/10 110 IRQ6/14
011 IRQ3/11 111 IRQ7/15

Access Method
Type: I/O Register

7 4 0
X X X X X X X X
MBZ1
MBZ
CCC
Bit Default &

Description
Range Access
X
7: 3 MBZ (MBZ): These bits must be programmed to zero.
WO
X Cascaded Controller Connection (CCC): This bit must always be programmed to a 1

2 to indicate the slave controller for interrupts 8 15 is cascaded on IRQ2.
WO
X
1: 0 MBZ (MBZ1): These bits must be programmed to zero.
WO
21.12.3.5 Master Operational Control Word 3 (MOCW3)—Offset 28h

Access Method

Type: I/O Register MOCW3: 28h

(Size: 8 bits)
7 4 0
0 0 1 X X X 1 0
RESERVED
O3S
SMM
ESMM
PMC
RRC
Bit Default &
Description
Range Access
0b
7 RESERVED (RESERVED): Must be 0.
RO
Special Mask Mode (SMM): If this bit is set, the Special Mask Mode can be used by an
0b interrupt service routine to dynamically alter the system priority structure while the
6
WO routine is executing, through selective enabling/ disabling of the other channel's mask
bits. Bit 6, the ESMM bit, must be set for this bit to have any meaning.
1b Enable Special Mask Mode (ESMM): When set, the SMM bit is enabled to set or reset
5 the Special Mask Mode. When cleared, the SMM bit becomes a don't care.
WO
X
4: 3 OCW3 Select (O3S): When selecting OCW3, bits 4:3 = 01
WO
Poll Mode Command (PMC): When cleared, poll command is not issued. When set,
X the next I/O read to the interrupt controller is treated as an interrupt acknowledge
2 cycle. An encoded byte is driven onto the data bus, representing the highest priority
WO
level requesting service.
Register Read Command (RRC): These bits provide control for reading the ISR and
Interrupt IRR. When bit 1=0, bit 0 will not affect the register read selection. Following
ICW initialization, the default OCW3 port address read will be read IRR. To retain the
10b current selection (read ISR or read IRR), always write a 0 to bit 1 when programming
1: 0 this register. The selected register can be read repeatedly without reprogramming
WO OCW3. To select a new status register, OCW3 must be reprogrammed prior to
attempting the read.
00 No Action 01 No Action
10 Read IRQ Register 11 Read IS Register

Access Method
Type: I/O Register MICW4: 29h

(Size: 8 bits)
7 4 0
X X X 0 0 0 0 1
MSBM
SFNM
BUF
AOEI
MM
MBZ
Bit Default &

Description
Range Access
X
WO
0b Special Fully Nested Mode (SFNM): Should normally be disabled by writing a 0 to

4
WO this bit. If SFNM=1, the special fully nested mode is programmed.
0b Buffered Mode (BUF): Must be cleared for non-buffered mode. Writing 1 will result in
3
WO undefined behavior.

Bit Default &

Description
Range Access
0b Master/Slave in Buffered Mode (MSBM): Not used. Should always be programmed

2
WO to 0.
0b Automatic End of Interrupt (AOEI): This bit should normally be programmed to 0.

1 This is the normal end of interrupt. If this bit is 1, the automatic end of interrupt mode
WO is programmed.
1b Microprocessor Mode (MM): This bit must be written to 1 to indicate that the
0 controller is operating in an Intel Architecture-based system. Writing 0 will result in
21.12.3.7 Master Operational Control Word 1 (MOCW1)—Offset 2Dh

Access Method
Type: I/O Register

(Size: 8 bits) MOCW1: 2Dh
7 4 0
0 0 0 0 0 0 0 0
IRM
Bit Default &
Description
Range Access
Interrupt Request Mask (IRM): When a 1 is written to any bit in this register, the
00h corresponding IRQ line is masked. When a 0 is written to any bit in this register, the
7: 0 corresponding IRQ mask bit is cleared, and interrupt requests will again be accepted by
RW the controller. Masking IRQ2 on the master controller will also mask the interrupt
requests from the slave controller.
21.12.3.8 Slave Initialization Command Word 1 (SICW1)—Offset A0h

.A write to Initialization Command Word 1 starts the interrupt controller initialization
sequence, during which the following occurs:
• The Interrupt Mask register is cleared.
• IRQ7 input is assigned priority 7.
• The slave mode address is set to 7.
• Special Mask Mode is cleared and Status Read is set to IRR.
Once this write occurs, the controller expects writes to ICW2, ICW3, and ICW4 to
complete the initialization sequence.
Access Method
Type: I/O Register SICW1: A0h

(Size: 8 bits)
7 4 0
X X X X X X X X
MCS85
ADI
ICWOCWSEL
LTIM
SNGL
IC4

Bit Default &

Description
Range Access
X MCS85 (MCS85): These bits are MCS-85 specific, and not needed. Should be
7: 5
WO programmed to 000
X ICW/OCW select (ICWOCWSEL): This bit must be a 1 to select ICW1 and enable the
4
WO ICW2, ICW3, and ICW4 sequence.
X
3 Edge/Level Bank Select (LTIM): Disabled. Replaced by ELCR1 and ELCR2.
WO
X
2 ADI (ADI): Should be programmed to 0.
WO
X Single or Cascade (SNGL): Must be programmed to a 0 to indicate two controllers

1
WO operating in cascade mode.
X wICW4 Write Required (IC4): This bit must be programmed to a 1 to indicate that
0
WO ICW4 needs to be programmed.

Slave ICW2 is used to initialize the interrupt controller with the five most significant bits
of the interrupt vector address. The value programmed for bits[7:3] is used by the CPU
to define the base address in the interrupt vector table for the interrupt routines
associated with each IRQ on the controller. Typical ISA ICW2 values are 08h for the
master controller and 70h for the slave controller.
Access Method
Type: I/O Register

(Size: 8 bits) SICW2: A1h
7 4 0
X X X X X X X X
IVBA
IRL
Bit Default &
Description
Range Access
X Interrupt Vector Base Address (IVBA): Bits [7:3] define the base address in the
7: 3 interrupt vector table for the interrupt routines associated with each interrupt request
WO level input.
Interrupt Request Lever (IRL): When writing ICW2, these bits should all be 0.
During an interrupt acknowledge cycle, these bits are programmed by the interrupt
controller with the interrupt to be serviced. This is combined with bits [7:3] to form the
interrupt vector driven onto the data bus during the second INTA# cycle. The code is a
three bit binary code:
Code Master Interrupt Slave Interrupt
X 000 IRQ0 IRQ8
2: 0 001 IRQ1 IRQ9
WO
010 IRQ2 IRQ10
011 IRQ3 IRQ11
100 IRQ4 IRQ12
101 IRQ5 IRQ13
110 IRQ6 IRQ14
111 IRQ7 IRQ15
21.12.3.10 Slave Operational Control Word 2 (SoCW2)—Offset A4h

Following a part reset or ICW initialization, the controller enters the fully nested mode
of operation. Non-specific EOI without rotation is the default. Both rotation mode and
specific EOI mode are disabled following initialization.
Access Method

Type: I/O Register SoCW2: A4h

(Size: 8 bits)
7 4 0
0 0 1 X X X X X
OCW2S
ILS
Bit REOI
Default &
Description
Range Access
Rotate and EOI Codes (REOI): R, SL, EOI - These three bits control the Rotate and
End of Interrupt modes and combinations of the two. A chart of these combinations is
listed above under the bit definition.
000 - Rotate in Auto EOI Mode (Clear)
001 - Non-specific EOI command
001b 010 - No Operation
7: 5 011 - *Specific EOI Command
WO
100 - Rotate in Auto EOI Mode (Set)
101 - Rotate on Non-Specific EOI Command
110 - *Set Priority Command
111 - *Rotate on Specific EOI Command
*L0 - L2 Are Used
X
4: 3 OCW2 Select (OCW2S): When selecting OCW2, bits 4:3 = 00
WO
Interrupt Level Select (L2, L1, L0) (ILS): L2, L1, and L0 determine the interrupt
level acted upon when the SL bit is active. A simple binary code, outlined above, selects
the channel for the command to act upon. When the SL bit is inactive, these bits do not
X have a defined function; programming L2, L1 and L0 to 0 is sufficient in this case.
2: 0 Bits Interrupt Level Bits Interrupt Level
WO 000 IRQ0/8 100 IRQ4/12
001 IRQ1/9 101 IRQ5/13
010 IRQ2/10 110 IRQ6/14
011 IRQ3/11 111 IRQ7/15

Access Method
Type: I/O Register

(Size: 8 bits) SICW3: A5h
7 4 0
X X X X X X X X
MBZ1
MBZ
CCC
Bit Default &

Description
Range Access
X
WO
X Cascaded Controller Connection (CCC): This bit must always be programmed to a 1

2 to indicate the slave controller for interrupts 8 15 is cascaded on IRQ2.
WO
X
1: 0 MBZ (MBZ1): These bits must be programmed to zero.
WO
21.12.3.12 Slave Operational Control Word 3 (SoCW3)—Offset A8h

Access Method

Type: I/O Register SoCW3: A8h

(Size: 8 bits)
7 4 0
0 0 1 X X X 1 0
RESERVED
O3S
SMM
ESMM
PMC
RRC
Bit Default &
Description
Range Access
0b
7 RESERVED (RESERVED): Must be 0.
RO
Special Mask Mode (SMM): If this bit is set, the Special Mask Mode can be used by an
0b interrupt service routine to dynamically alter the system priority structure while the
6
WO routine is executing, through selective enabling/ disabling of the other channel's mask
bits. Bit 6, the ESMM bit, must be set for this bit to have any meaning.
1b Enable Special Mask Mode (ESMM): When set, the SMM bit is enabled to set or reset
5 the Special Mask Mode. When cleared, the SMM bit becomes a don't care.
WO
X
4: 3 OCW3 Select (O3S): When selecting OCW3, bits 4:3 = 01
WO
Poll Mode Command (PMC): When cleared, poll command is not issued. When set,
X the next I/O read to the interrupt controller is treated as an interrupt acknowledge
2 cycle. An encoded byte is driven onto the data bus, representing the highest priority
WO
level requesting service.
Register Read Command (RRC): These bits provide control for reading the ISR and
Interrupt IRR. When bit 1=0, bit 0 will not affect the register read selection. Following
ICW initialization, the default OCW3 port address read will be read IRR. To retain the
10b current selection (read ISR or read IRR), always write a 0 to bit 1 when programming
1: 0 this register. The selected register can be read repeatedly without reprogramming
WO OCW3. To select a new status register, OCW3 must be reprogrammed prior to
attempting the read.
00 No Action 01 No Action
10 Read IRQ Register 11 Read IS Register

Access Method
Type: I/O Register SICW4: A9h

(Size: 8 bits)
7 4 0
X X X 0 0 0 0 1
MSBM
SFNM
BUF
AOEI
MM
MBZ
Bit Default &

Description
Range Access
X
WO
0b Special Fully Nested Mode (SFNM): Should normally be disabled by writing a 0 to

4
WO this bit. If SFNM=1, the special fully nested mode is programmed.
0b Buffered Mode (BUF): Must be cleared for non-buffered mode. Writing 1 will result in
3

Bit Default &

Description
Range Access
0b Master/Slave Buffered Mode (MSBM): Not used. Should always be programmed to

2
WO 0.
0b Automatic End of Interrupt (AOEI): This bit should normally be programmed to 0.

1 This is the normal end of interrupt. If this bit is 1, the automatic end of interrupt mode
WO is programmed.
1b Microprocessor Mode (MM): This bit must be written to 1 to indicate that the
0 controller is operating in an Intel Architecture-based system. Writing 0 will result in
21.12.3.14 Slave Operational Control Word 1 (SoCW1)—Offset ADh

Access Method
Type: I/O Register

SoCW1: ADh
(Size: 8 bits)
7 4 0
0 0 0 0 0 0 0 0
IRM
Bit Default &
Description
Range Access
Interrupt Request Mask (IRM): When a 1 is written to any bit in this register, the
00h corresponding IRQ line is masked. When a 0 is written to any bit in this register, the
7: 0 corresponding IRQ mask bit is cleared, and interrupt requests will again be accepted by
RW the controller. Masking IRQ2 on the master controller will also mask the interrupt
requests from the slave controller.
21.12.3.15 Master Edge/Level Control (ELCR1)—Offset 4D0h

Access Method
Type: I/O Register ELCR1: 4D0h

(Size: 8 bits)
7 4 0
X X X X X 0 0 0
ELC
RESERVED
Bit Default &

Description
Range Access
X Edge Level Control (ECL[7:3]) (ELC):: In edge mode, (bit cleared), the interrupt is
7: 3 recognized by a low to high transition. In level mode (bit set), the interrupt is
RW recognized by a high level.
0b
2: 0 RESERVED (RESERVED): Reserved.
RO
21.12.3.16 Slave Edge/Level Control (ELCR2)—Offset 4D1h

Access Method
Type: I/O Register ELCR2: 4D1h

(Size: 8 bits)

7 4 0
X X 0 X X X X 0
ELC1
ELC2
RESERVED
RESERVED1
Bit Default &
Description
Range Access
X Edge Level Control (ECL[15:14]) (ELC1): In edge mode, (bit cleared), the interrupt
7: 6 is recognized by a low to high transition. In level mode (bit set), the interrupt is
RW recognized by a high level. Bit 7 applies to IRQ15, and bit 6 to IRQ14.
0b
5 RESERVED (RESERVED): Reserved.
RO
Edge Level Control (ECL[12:9]) (ELC2):: In edge mode, (bit cleared), the interrupt
X is recognized by a low to high transition. In level mode (bit set), the interrupt is
4: 1 recognized by a high level. Bit 4 applies to IRQ12, bit 3 to IRQ11, bit 2 to IRQ10, and bit
RW
1 to IRQ9.
0b
0 RESERVED (RESERVED): Reserved.
RO

21.13 I/O APIC

The I/O Advanced Programmable Interrupt Controller (APIC) is used to support line
interrupts more flexibly than the 8259 PIC. Line interrupts are routed to it from
multiple sources, including legacy devices, via the interrupt decoder or they are routed
to it from the interrupt router in the Legacy Bridge. These line-based interrupts are
then used to generate interrupt messages targeting the local APIC in the processor.
21.13.1 Features
• 24 interrupt lines
— IRQ0-23
• Edge or level trigger mode per interrupt
• Active low or high polarity per interrupt
• Works with local APIC in processor via MSIs
• MSIs can target specific processor core
• Established APIC programming model
Figure 56. Detailed I/O APIC Block Diagram
To/From System Bus MSI’s
I/O APIC INT0

INT1
FEC0 0000h IDX MSI Machine
INT2
FEC0 0010h WDW INT3
INT4
FEC0 0040h EOI
INT5
INT6
INT7
INT8
ID VS INT9
INT10
INT11
INT[0] INT12
RTE[0]
INT13
INT14
INT15
...
INT16
INT17
INT[23] INT18
RTE[23] INT19
INT20
INT21
INT22
INT23
INT[23:0]
Note: INT13 is unavailable and is effectively tied low within the I/O APIC, INT14 & INT15 are unused in the

SoC and are tied low.
MSIs generated by the I/O APIC are sent as 32-bit memory writes to the Local APIC.
The address and data of the write transaction are used as follows.
Figure 57. MSI Address and Data
RTE[n].DSM Destination Mode

Set if RTE[n].DLM = 001b Redirection Hint
RTE[n].EDID Extended Dest. ID
RTE[n].DID Destination ID
FEEh 00b
MSI 31 : 20 19 : 12 11 : 4 3 2 1:0
Address
MSI 31:16 151413:1211 10:8 7:0
Data
0000h 00b
RTE[n].TM Trigger Mode
Delivery Status (1b)
RTE[n].DSM Destination Mode

RTE[n].DLM Delivery Mode
RTE[n].VCT Vector
Destination ID (DID) and Extended Destination ID (EDID) are used to target a specific
processor core’s local APIC.
21.13.2 Use
The I/O APIC contains indirectly accessed I/O APIC registers and normal memory
mapped registers. There are three memory mapped registers:
• Index Register (IDX)
• Window Register (WDW)
• End Of Interrupt Register (EOI)
The Index register selects an indirect I/O APIC register (ID/VS/RTE[n]) to appear in the
Window register.
The Window register is used to read or write the indirect register selected by the Index
register.
The EOI register is written to by the Local APIC in the processor. The I/O APIC
compares the lower eight bits written to the EOI register to the Vector set for each
interrupt (RTE.VCT). All interrupts that match this vector will have their RTE.RIRR
register cleared. All other EOI register bits are ignored.
21.13.3 Unsupported Modes

These delivery modes are not supported for the following reasons:

• NMI/INIT: This cannot be delivered while the CPU is in the Stop Grant state. In
addition, this is a break event for power management.
• SMI: There is no way to block the delivery of the SMI_B, except through BIOS.
• Virtual Wire Mode B: The Legacy Bridge does not support the INTR of the 8259
routed to the I/OxAPIC pin 0.

Figure 58. I/O APIC Register Map
Memory
Space
I/O APIC
Space
IDX Value
ID 0h
FEC00000h Index (IDX) VS 1h
FEC00010h Window (WDW)

RTE[0] 10h, 11h
FEC00040h End of Int (EOI) RTE[1] 12h, 13h
RTE[23] 3Eh, 3Fh
21.13.5 Memory Mapped Registers

The APIC is accessed via an indirect addressing scheme. These registers are mapped
into memory space. The registers are shown below.

Table 152. I/O APIC Memory Mapped Registers

Address Symbol Register
FEC00000h IDX Index Register
FEC00010h WDW Window Register
FEC00040h EOI End of Interrupt Register
21.13.5.1 Index Register (IDX)—Offset FEC00000h

Access Method
Type: Memory Mapped I/O Register IDX: FEC00000h

(Size: 8 bits)
Default: 00h
7 4 0
0 0 0 0 0 0 0 0
IDX
Bit Default &

Description
Range Access
0h Index (IDX): This 8-bit register selects which indirect register appears in the window
7: 0 register to be manipulated by software. Software will program this register to select the
RW desired APIC internal register.
21.13.5.2 Window Register (WDW)—Offset FEC00010h

Access Method
Type: Memory Mapped I/O Register WDW: FEC00010h

(Size: 32 bits)
Default: FFFFFFFFh
31 28 24 20 16 12 8 4 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
WDW
Bit Default &

Description
Range Access
FFFFFFFFh Window (WDW): This 32-bit register specifies the data to be read or written to the
31: 0 register pointed to by the IDX register. This register can be accessed only in DW
RW quantities.
21.13.5.3 End of Interrupt Register (EOI)—Offset FEC00040h

Access Method
Type: Memory Mapped I/O Register EOI: FEC00040h

(Size: 32 bits)
Default: 00000000h

31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
EOI
RESERVED1
Bit Default &
Description
Range Access
0b
31: 8 RESERVED (RESERVED1): Reserved.
RO
EOI (EOI): When a write is issued to this register, the IOxAPIC will check the lower 8
0h bits written to this register, and compare it with the vector field for each entry in the I/O
7: 0
RO Redirection Table. When a match is found, RTE.RIRR for that entry will be cleared. If
multiple entries have the same vector, each of those entries will have RTE.RIRR cleared.
21.13.6 Index Registers

These registers are selected with the IDX register, and read/written through the WDW
register. Accessing these registers must be done as DW requests, otherwise unspecified
behavior will result. Software should not attempt to write to reserved registers.
Reserved registers may return non-zero values when read.
Note: There is one pair of redirection (RTE) registers per interrupt line. Each pair forms a 64-
bit RTE register.
Note: Specified offsets should be placed in IDX, not added to IDX.
Table 153. Index Registers

Offset Symbol Register
00 ID Identification
01 VS Version
02-0F - Reserved
10-11 RTE0 Redirection Table 0
12-13 RTE1 Redirection Table 1
... ... ...
3E-3F RTE23 Redirection Table 23
40-FF - Reserved
21.13.6.1 Identification Register (ID)—Offset 0h

Access Method
Type: Indirect I/O APIC Register ID: 0h

(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
RSVD0
RSVD1
AID

Bit Default &

Description
Range Access
0h
31: 28 Reserved (RSVD0): Reserved.
RW
0h
27: 24 APIC Identification (AID): Software must program this value before using the APIC.
RW
0h
RW
21.13.6.2 Version Register (VS)—Offset 1h

Access Method
Type: Indirect I/O APIC Register VS: 1h

(Size: 32 bits)
Default: 00170020h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0
RSVD0
RSVD1
MRE
PRQ
VS
Bit Default &
Description
Range Access
0h
RW
17h Maximum Redirection Entries (MRE): This is the entry number (0 being the lowest
23: 16 entry) of the highest entry in the redirection table. This field is hardwired to indicate the
RO total number of interrupts.
0b Pin Assertion Register Supported (PRQ): The I/O APIC does not implement the Pin
15 Assertion Register.
RO
0h
RW
20h
7: 0 Version (VS): Identifies the implementation version as I/O APIC.
RO
21.13.6.3 Redirection Table Entry Lower (RTE[0-23]L)—Offset 10h - 3Eh

Lower 32-bits of the RTE register.
Access Method
Type: Indirect I/O APIC Register RTE[0-23]L: 10h - 3Eh

(Size: 32 bits)
Default: 00010000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MSK
RIRR
POL
DS
VCT
RSVD0
TM
DSM
DLM

Bit Default &

Description
Range Access
0h
RW
1b Mask (MSK): When set, interrupts are not delivered nor held pending. When cleared,
16
RW and edge or level on this interrupt results in the delivery of the interrupt.
0b Trigger Mode (TM): When cleared, the interrupt is edge sensitive. When set, the
15
RW interrupt is level sensitive.
Remote IRR (RIRR): This is used for level triggered interrupts; its meaning is
0b undefined for edge triggered interrupts. This bit is set when I/O APIC sends the level
14 interrupt message to the CPU. This bit is cleared when an EOI message is received that
RW matches the VCT field. This bit is never set for SMI, NMI, INIT, or ExtINT delivery
modes.
0b Polarity (POL): This specifies the polarity of each interrupt input. When cleared, the
13
RW signal is active high. When set, the signal is active low. 0: Active High 1: Active Low
0b Delivery Status (DS): This field contains the current status of the delivery of this
12 interrupt. When set, an interrupt is pending and not yet delivered. When cleared, there
RO is no activity for this entry.
0b Destination Mode (DSM): This field is used by the local APIC to determine whether it
11 is the destination of the message.
RW
Delivery Mode (DLM): This field specifies how the APICs listed in the destination field
should act upon reception of this signal. Certain Delivery Modes will only operate as
intended when used in conjunction with a specific trigger mode. These encodings are:
000: Fixed
001: Lowest Priority
0h 010: SMI - Not supported.
10: 8
RW 011: Reserved
100: NMI - Not supported.
101: INIT - Not supported.
110: Reserved
111: ExtINT
0h Vector (VCT): This field contains the interrupt vector for this interrupt. Values range
7: 0
RW between 10h and FEh.
21.13.6.4 Redirection Table Entry Upper (RTE[0-23]U)—Offset 11h - 3Fh

Upper 32-bits of the RTE register.
Access Method
Type: Indirect I/O APIC Register RTE[0-23]U: 11h - 3Fh

(Size: 32 bits)
Default: 00000000h
31 28 24 20 16 12 8 4 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
DID
EDID
RSVD
Bit Default &

Description
Range Access
0h
31: 24 Destination ID (DID): Destination ID of the local APIC.
RW
0h
23: 16 Extended Destination ID (EDID): Extended destination ID of the local APIC.
RW

Bit Default &

Description
Range Access
0h
RW

21.14 Watchdog Timer

The Watchdog timer can be used to trigger a reset in the event that the system has
become unresponsive.
21.14.1 Features
Selectable Prescaler - approximately 1 MHz (1 s to 1 s) and approximately 1 KHz (1
ms to 17 min)
• 33 MHz Clock (30 ns Clock Ticks)
• WDT Mode:
— Second stage drives WDT_TOUT high or inverts the previous value. Used only
after first timeout occurs.
— Status bit preserved in RTC well for possible error detection and correction
— Drives WDT_TOUT if OUTPUT is enabled
• Timer can be disabled (default state) or Locked (Hard Reset required to disable
WDT)
• WDT Automatic Reload of Preload value when WDT Reload Sequence is performed
Note: WDT_TOUT is not available as a top-level SoC output.
21.14.2 Use
This Watchdog timer provides a resolution that ranges from 1 s to ~17 minutes. The
timer uses a 35-bit down-counter.
The counter is loaded with the value from the 1st Preload register. The timer is then
enabled and it starts counting down. The time at which the WDT first starts counting
down is called the first stage. If the host fails to reload the WDT before the 35-bit down
counter reaches zero the WDT generates an internal interrupt within the WDT.
After the internal interrupt is generated when the first stage has counted down to zero,
the WDT loads the value from the 2nd Preload register into the WDT's 35-bit Down-
Counter and starts counting down. The WDT is now in the second stage. If the host still
fails to reload the WDT before the second timeout, the WDT drives the WDT_TOUT
signal high and sets the timeout bit (WDT_TIMEOUT). This bit indicates that the System
has become unstable. The WDT_TOUT signal is held high until the system is Reset or
the WDT times out again (Depends on WDT Timeout Configuration). The process of
reloading the WDT involves the following sequence of writes:
• Write 80 to offset WDTBA + 0Ch
• Write 86 to offset WDTBA + 0Ch
• Write 1 to WDT_RELOAD in Reload Register.
The same process is used for setting the values in the preload registers. The only
difference exists in step 3. Instead of writing a '1' to the WDT_RELOAD, you write the
desired preload value into the corresponding Preload register. This value is not loaded
into the 35-bit down counter until the next time the WDT reenters the stage. For
example, if Preload Value 2 is changed, it is not loaded into the 35-bit down counter
until the next time the WDT enters the second stage.
Note: The WDT output, WDT_TOUT, is not available as a top-level SoC output pin


Figure 59. Watchdog Timer Register Map
PCI Space
CPU
Core
Host Bridge
D:0,F:0
Memory
Space
PCI
CAM
(I/O)
Bus 0
PCI
ECAM
(Mem) Legacy PCI
Header
D:31,F:0
RP0 F:0
PCIe*
D:23
SPI0 F:0 RP0 F:1

IO Fabric
D:21
SPI1 F:1
WDTBA
I2C*/GPIOF:2 IO Space
Legacy Bridge
D:31,F:0
WDT IO
SDIO/eMMC F:0 Registers
HSUART0 F:1
IO Fabric D:20
USB Device F:2

EHCI Host F:3
OCHI Host F:4
HSUART1 F:5
MAC0 F:6
MAC1 F:7
21.14.4 I/O Mapped Registers
Table 154. Summary of I/O Registers—WDTBA

Offset Default
Start Value
0h 0h “Preload Value 1 Register 0 (PV1R0)—Offset 0h” on page 928 FFh
2h 2h “Preload Value 1 Register 2 (PV1R2)—Offset 2h” on page 929 0Fh

Table 154. Summary of I/O Registers—WDTBA (Continued)

Offset Default
Start Value
6h 6h “Preload Value 2 Register 2 (PV2R2)—Offset 6h” on page 930 0Fh
Ch Ch “Reload Register 0 (RR0)—Offset Ch” on page 930 00h
Dh Dh “Reload Register 1 (RR1)—Offset Dh” on page 931 00h
10h 10h “WDT Configuration Register (WDTCR)—Offset 10h” on page 931 00h
18h 18h “WDT Lock Register (WDTLR)—Offset 18h” on page 932 00h
21.14.4.1 Preload Value 1 Register 0 (PV1R0)—Offset 0h

Access Method
Type: I/O Register PV1R0: [WDTBA] + 0h
(Size: 8 bits)
WDTBA Type: PCI Configuration Register (Size: 32 bits)

WDTBA Reference: [B:0, D:31, F:0] + 84h
Default: FFh
7 4 0
1 1 1 1 1 1 1 1
PV1
Bit Default &

Description
Range Access
Preload Value 1[7:0] (PV1): This register is used to hold the bits 7 down to 0 of the
Preload Value 1 for the Watch Dog Timer. The Value in the Preload Register is
FFh Automatically transferred into the 35-bit down counter every time the WDT enters the
7:0
RW first stage. The value loaded into the preload register needs to be one less than the
intended period. This is because the timer makes use of zero-based counting (i.e. zero
is counted as part of the decrement).

Access Method
(Size: 8 bits)

Default: FFh
7 4 0
1 1 1 1 1 1 1 1
PV1

Bit Default &

Description
Range Access
Preload Value 1[15:8] (PV1): This register is used to hold the bits 15 down to 8 of
the Preload Value 1 for the Watch Dog Timer. The Value in the Preload Register is
7:0

Access Method
(Size: 8 bits)

Default: 0Fh
7 4 0
0 0 0 0 1 1 1 1
PV1
RSV
Bit Default &

Description
Range Access
0b
RO
Fh Automatically transferred into the 35-bit down counter every time the WDT enters the
3:0

Access Method
(Size: 8 bits)

Default: FFh
7 4 0
1 1 1 1 1 1 1 1
PV2
Bit Default &

Description
Range Access
Preload Value 2[7:0] (PV2): This register is used to hold the bits 7 down to 0 of the
Preload Value 2 for the Watch Dog Timer. The Value in the Preload Register is
7:0


Access Method
(Size: 8 bits)

Default: FFh
7 4 0
1 1 1 1 1 1 1 1
PV2
Bit Default &
Description
Range Access
7:0

Access Method
(Size: 8 bits)

Default: 0Fh
7 4 0
0 0 0 0 1 1 1 1
RSV
PV2
Bit Default &

Description
Range Access
0b
RO
Fh Automatically transferred into the 35-bit down counter every time the WDT enters the
3:0
21.14.4.7 Reload Register 0 (RR0)—Offset Ch

Access Method
Type: I/O Register RR0: [WDTBA] + Ch
(Size: 8 bits)


Default: 00h
7 4 0
0 0 0 0 0 0 0 0
WDT_RLD0
Bit Default &
Description
Range Access
0b WDT Reload 0 (WDT_RLD0): The reload sequence is only necessary for the Reload
7:0
WO register and Preload Value registers and is not used in Free Running mode.
21.14.4.8 Reload Register 1 (RR1)—Offset Dh

Access Method
Type: I/O Register RR1: [WDTBA] + Dh
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
WDT_RDL
RSV
WDT_TOUT
Bit Default &
Description
Range Access
0b
RO
WDT Timeout (WDT_TOUT): This bit is located in the RTC Well and its value is not
lost if the host resets the system. It is set to 1 if the host fails to reset the WDT before
0b the 35-bit Down-Counter reaches zero for the second time in a row. This bit is cleared
1
RW/1C by performing the Register Unlocking Sequence followed by a 1 to this bit.
0 = Normal (Default)
1 = System has become unstable.
0b WDT Reload (WDT_RDL): To prevent a timeout the host must perform the Register
0
RW Unlocking Sequence followed by a 1 to this bit.
21.14.4.9 WDT Configuration Register (WDTCR)—Offset 10h

Access Method
Type: I/O Register WDTCR: [WDTBA] + 10h
(Size: 8 bits)

Default: 00h

7 4 0
0 0 0 0 0 0 0 0
WDT_PRE_SEL
RSVD
WDT_RESET_SEL
RSV2
WDT_TOUT_EN
WDT_RESET_EN
Bit Default &
Description
Range Access
0b
RO
WDT Timeout Output Enable (WDT_TOUT_EN): This bit indicates whether or not
0b the WDT toggles the external WDT_TOUT signal if the WDT times out.
5
RW 0 = Enabled (Default)
1 = Disabled
WDT Reset Enable (WDT_RESET_EN): When this bit is enable (set to 1), it allows
0b internal reset to be trigger when WDT timeout in the second stage. It either trigger
4 COLD or WARM reset depend on WDT_RESET_SEL bit.
RW 0 = Disable internal reset (Default)
1 = Enable internal COLD or WARM reset.
WDT Reset Select (WDT_RESET_SEL): This determines which reset to be triggered

0b when WDT_RESET_EN is set.
3
RW 0 = Cold Reset (Default)
1 = Warm Reset
WDT Prescaler Select (WDT_PRE_SEL): The WDT provides two options for
prescaling the main Down Counter. The preload values are loaded into the main down
counter right justified. The prescaler adjusts the starting point of the 35-bit down
counter.
0b 0 = The 20-bit Preload Value is loaded into bits 34:15 of the main down counter. The
2
RW resulting timer clock is the PCI Clock (33 MHz) divided by 2^15. The approximate clock
generated is 1 KHz, (1 ms to 10 min). (Default)
1 = The 20-bit Preload Value is loaded into bits 24:05 of the main down counter. The
resulting timer clock is the PCI Clock (33 MHz) divided by 2^5. The approximate clock
generated is 1 MHz, (1 us to 1sec)
0b
RO
21.14.4.10 WDT Lock Register (WDTLR)—Offset 18h

Access Method
Type: I/O Register WDTLR: [WDTBA] + 18h
(Size: 8 bits)

Default: 00h
7 4 0
0 0 0 0 0 0 0 0
RSV
WDT_ENABLE
WDT_TOUT_CNF
WDT_LOCK

Bit Default &

Description
Range Access
0b
RO
WDT Timeout Configuration (WDT_TOUT_CNF): This register is used to choose the

functionality of the timer.
0b 0 = Watchdog Timer Mode: When enabled (i.e. WDT_ENABLE goes from 0 to 1) the
2 timer reloads Preload Value 1 and start decrementing. (Default) Upon reaching the
RW second stage timeout the WDT_TOUT is driven high once and does not change again
until Power is cycled or a hard reset occurs.
1 = Reserved
Watchdog Timer Enable (WDT_ENABLE): The following bit enables or disables the
WDT.
0 = Disabled (Default)
1 = Enabled
0b Note: This bit cannot be modified if WDT_LOCK has been set.
1 Note: In WDT mode Preload Value 1 is reloaded every time WDT_ENABLE goes from 0 to
RW 1 or the WDT_RELOAD bit is written using the proper sequence of writes (See Register
Unlocking Sequence). When the WDT second stage timeout occurs, a reset must
happen.
Note: Software must guarantee that a timeout is not about to occur before disabling the
timer. A reload sequence is suggested.
Watchdog Timer Lock (WDT_LOCK): Setting this bit locks the values of this register
until a hard-reset occurs or power is cycled.
0b 0 = Unlocked (Default)
0
RW/O 1 = Locked
Note: Writing a 0 has no effect on this bit. Write is only allowed from 0 to 1 once. It
cannot be changed until either power is cycled or a hard reset occurs
§§



Quark x1000 Datasheet

Uploaded by

Copyright:

Available Formats

Quark x1000 Datasheet

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Quark x1000 Datasheet

Uploaded by

Copyright:

Available Formats

Intel® Quark™ SoC X1000

Document Number: 329676-005US

Intel® Quark™ SoC X1000

1.0 Introduction ............................................................................................................ 37

Intel® Quark™ SoC X1000

4.8 General Interface Timing ....................................................................................79

Intel® Quark™ SoC X1000

8.4.2.3 Core C2 State......................................................................... 110

Intel® Quark™ SoC X1000

12.5.8 Message Control Register eXtension (MCRX) (SB_ADDR_EXTN_REG)—Offset

Intel® Quark™ SoC X1000

12.7.2.23 MTRR Fixed 4KB Range 0xE0000 (MTRR_FIX4K_E0000)—Offset

Intel® Quark™ SoC X1000

12.7.3.7 Thermal Sensor Mode Register (P_CFG_B0)—Offset B0h............... 178

Intel® Quark™ SoC X1000

13.2.1 System Memory Technology Supported ................................................... 234

Intel® Quark™ SoC X1000

14.5.13 PCI Express Capabilities (CLIST_XCAP)—Offset 40h ................................. 274

Intel® Quark™ SoC X1000

15.5.8 Header Type (HEADER_TYPE)—Offset Eh ................................................. 316

Intel® Quark™ SoC X1000

15.6.24 MMC Transmit 64 Octet Good Bad Frame Counter Register

Intel® Quark™ SoC X1000

15.6.52 MMC Receive Alignment Error Frame Counter Register

Intel® Quark™ SoC X1000

15.6.81 MMC Receive IPV6 No Payload Frame Counter Register

Intel® Quark™ SoC X1000

15.6.110 Timestamp Addend Register (Register 454)

Intel® Quark™ SoC X1000

16.5.1.9 BIST (BIST)—Offset Fh ............................................................ 442

Intel® Quark™ SoC X1000

16.5.2.30 Message Control (MESSAGE_CTRL)—Offset A2h........................ 466

Intel® Quark™ SoC X1000

16.6.1.10 IN Endpoint 1 Maximum Packet Size Register

Intel® Quark™ SoC X1000

16.6.1.42 OUT Endpoint 2 Buffer Size Register (ep2_out_bufsize_udc_reg)—

Intel® Quark™ SoC X1000

16.6.2 USB EHCI ............................................................................................ 549

Intel® Quark™ SoC X1000

17.2.2.1 SD 3.0 Bus Topology ............................................................... 592

Intel® Quark™ SoC X1000

17.6.15 Block Gap Control Register (BLK_GAP_CTL)—Offset 2Ah ........................... 628

Intel® Quark™ SoC X1000

18.5 PCI Configuration Registers .............................................................................. 669

Intel® Quark™ SoC X1000

18.6.2.8 Channel 0 Source Status Address (SSTATAR0)—Offset 30h ........... 701

Intel® Quark™ SoC X1000

19.1.1 Signal Descriptions ............................................................................... 731

Intel® Quark™ SoC X1000

19.5.1.5 Standard Speed Clock SCL Low Count (IC_SS_SCL_LCNT)—Offset

Intel® Quark™ SoC X1000

20.4.9 BIST (BIST)—Offset Fh ......................................................................... 796

Intel® Quark™ SoC X1000

21.4.1.3 Interrupt Queue Agent 0 (IRQAGENT0)—Offset 3140h.................. 829

Intel® Quark™ SoC X1000

21.7.2 Features ............................................................................................. 856

Intel® Quark™ SoC X1000

21.9.3.2 General Capabilities and ID Register - Upper 32 Bits

Intel® Quark™ SoC X1000

21.12.3.6 Master Initialization Command Word 4 (MICW4)—Offset 29h ...... 911