Re: [RFC] x86-64: Use SSE for copy_page and clear_page

[Nick Piggin]

On Wed, 2005-06-01 at 17:22 +1000, michael@xxxxxxxxxxxxxxx wrote:
> Andi Kleen <ak@xxxxxx> writes:
> 

> > fork is only a corner case. The main case is a process allocating
> > memory using brk/mmap and then using it.
> 
> Key point: "using it". This normally involves writes to memory. Most
> applications don't commonly read memory that they haven't previously
> written to. (valgrind et al call that behaviour a "bug" :).

Then in that case you have doubled your memory bandwidth
requirement for those cachelines.

> 
> Given that, I'd say you really don't want the page zero routines
> touching the cache.
> 

The principle of locality-of-data (ie. the reason why caches
even work) says that you do ;)

Clearly some things benefit from not going through the cache.
But I don't think we should fundamentally change behaviour of
this *just* because it is worth a percent on kernel compiles.

Also, I think that trends in CPU design (more cache, further
from memory, multiple CPUs & cores) should favour stores
going to cache rather than straight to memory... But I'm
just speculating.

-- 
SUSE Labs, Novell Inc.



Send instant messages to your online friends http://au.messenger.yahoo.com 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/