[lkml]   [2009]   [May]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [patch 0/5] Support for sanitization flag in low-level page allocator
On Thu, 28 May 2009 21:36:01 +0200
Peter Zijlstra <> wrote:

> > ... and if we zero on free, we don't need to zero on allocate.
> > While this is a little controversial, it does mean that at least
> > part of the cost is just time-shifted, which means it'll not be TOO
> > bad hopefully...
> zero on allocate has the advantage of cache hotness, we're going to
> use the memory, why else allocate it.

that is why I said it's controversial.

BUT if you zero on free anyway...

And I don't think it's as big a deal as you make it.

We recycle pages in LIFO order. And L2 caches are big.

So if you zero on free, the next allocation will reuse the zeroed page.
And due to LIFO that is not too far out "often", which makes it likely
the page is still in L2 cache.

The other thing is that zero-on-allocate puts the WHOLE page in L1,
while you can study how much of that page is actually used on average,
and it'll be a percentage lower than 100%.
In fact, if it IS 100%, you shouldn't have put it in L1 because the app
does that anyway. If it is not 100% you just blew a chunk of your L1
for no value.

Don't get me wrong, I'm not arguing that zero-on-free is better, I'm
just trying to point out that the "advantage" of zero-on-allocate isn't
nearly as big as people sometimes think it is...

Arjan van de Ven Intel Open Source Technology Centre
For development, discussion and tips for power savings,

 \ /
  Last update: 2009-05-29 16:37    [W:0.158 / U:2.608 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site