lkml.org 
[lkml]   [2009]   [Dec]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
DateWed, 02 Dec 2009 09:01:14 +0200
FromPekka Enberg <>
SubjectRe: Q, slab, kmemleak_erase() and redzone?
Hi!

hooanon05@yahoo.co.jp kirjoitti:
> Pekka Enberg:
>>> No, you are absolutely correct. Can you please send an updated patch to
>>> Catalin that adds a comment on top of the cpu_cache_get() call that
>>> explains why we need it there?
>> Doh, this was supposed to be a reply to Okajima's email :-).
>
> Before I send a small patch, let me make sure about other small issues.
>
> - How heavy is 'ac = cpu_cache_get(cachep)' (which will be inserted by
> the patch)?
> It will be compiled and executed regardless CONFIG_DEBUG_KMEMLEAK, and
> it is totally meaningless when DEBUG_KMEMLEAK is disabled. Can we
> ignore this loss?

No, it won't be. The compiler should notice it's dead code and remove it
when CONFIG_KMEMLEAK is disabled.

> - Should we add a condition 'if (objp)' before calling kmemleak_erase()?
> As Catalin wrote, it may be harmless. But setting NULL is unnecessary.
> Do you accept this change too?

Yeah, I'd prefer to see the check there. While Catalin is correct,
that's not obvious from reading the code.

Pekka


\
 
 \ /
  Last update: 2009-12-02 08:03    [from the cache]
©2003-2010