lkml.org 
[lkml]   [2014]   [Jan]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] mm: slub: fix page->_count corruption (again)
On 01/28/2014 03:29 PM, Andrew Morton wrote:
> On Tue, 28 Jan 2014 15:17:22 -0800 Dave Hansen <dave@sr71.net> wrote:
> This code is borderline insane.

No argument here.

> Yes, struct page is special and it's worth spending time and doing
> weird things to optimise it. But sheesh.
>
> An alternative is to make that cmpxchg quietly go away. Is it more
> trouble than it is worth?

It has measurable performance benefits, and the benefits go up as the
cost of en/disabling interrupts goes up (like if it takes you a hypercall).

Fengguang, could you run a set of tests for the top patch in this branch
to see if we'd be giving much up by axing the code?

https://github.com/hansendc/linux/tree/slub-nocmpxchg-for-Fengguang-20140128

I was talking with one of the distros about turning it off as well.
They mentioned that they saw a few performance regressions when it was
turned off. I'll share details when I get them.



\
 
 \ /
  Last update: 2014-01-29 01:21    [W:0.086 / U:0.952 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site