Speeding up the page allocator
Speeding up the page allocator
Posted Feb 27, 2009 1:08 UTC (Fri) by jzbiciak (guest, #5246)In reply to: Speeding up the page allocator by jzbiciak
Parent article: Speeding up the page allocator
Ah, it appears AMD K7 and beyond go one better and have a "streaming store" that doesn't even do a cache allocate. Nice.
Here's the MMX and AMD optimized copies and fills the kernel currently uses. I can't imagine they'd settle for a crappy loop here, and it looks like some thought was put into these.
http://lxr.linux.no/linux+v2.6.28.7/arch/x86/lib/mmx_32.c
On regular x86, they do indeed use "rep stosl". (I guess the AT&T syntax spells it "stosl" instead of "stosd"?) See around like 92.
http://lxr.linux.no/linux+v2.6.28.7/arch/x86/include/asm/...
Rampant speculation is fun and all, but I suspect Arjan actually measured these. :-) (Or, at least the ones in the MMX file.)