lkml.org 
[lkml]   [2020]   [Aug]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: Odd-sized kmem_cache_alloc and slub_debug=Z
Hi Marco and Kees,

On Fri, Aug 07, 2020 at 08:06PM +0300, Pekka Enberg wrote:
> > Anything interesting in your .config? The fault does not reproduce
> > with 5.8.0 + x86-64 defconfig.

On Fri, Aug 7, 2020 at 8:18 PM Marco Elver <elver@google.com> wrote:
> It's quite close to defconfig, just some extra options for my test
> environment. But none that I'd imagine change this behaviour -- but
> maybe I missed something. I've attached my config. Also, just in case,
> I'm on mainline from Tuesday: 2324d50d051ec0f14a548e78554fb02513d6dcef.

Yeah, it reproduces with defconfig too, as long as you remember to
pass "slub_debug=Z"... :-/

The following seems to be the culprit:

commit 3202fa62fb43087387c65bfa9c100feffac74aa6
Author: Kees Cook <keescook@chromium.org>
Date: Wed Apr 1 21:04:27 2020 -0700

slub: relocate freelist pointer to middle of object

Reverting this commit and one of it's follow up fixes from Kees from
v5.8 makes the issue go away for me. Btw, please note that caches with
size 24 and larger do not trigger this bug, so the issue is that with
small enough object size, we're stomping on allocator metadata (I
assume part of the freelist).

- Pekka

\
 
 \ /
  Last update: 2020-08-07 21:06    [W:0.071 / U:0.324 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site