mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Catalin Marinas" <catalin.marinas@gmail.com>
To: "Pekka Enberg" <penberg@cs.helsinki.fi>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH 07/10] Remove some of the kmemleak false positives
Date: Tue, 11 Jul 2006 09:18:12 +0100	[thread overview]
Message-ID: <b0943d9e0607110118p5b9dfc28qb377b61e59ec7535@mail.gmail.com> (raw)
In-Reply-To: <84144f020607102349u62a59b26va0cec06c6d15e178@mail.gmail.com>

On 11/07/06, Pekka Enberg <penberg@cs.helsinki.fi> wrote:
> On 7/11/06, Catalin Marinas <catalin.marinas@gmail.com> wrote:
> > @@ -166,6 +166,9 @@ struct platform_device *platform_device_
> >         struct platform_object *pa;
> >
> >         pa = kzalloc(sizeof(struct platform_object) + strlen(name), GFP_KERNEL);
> > +       /* kmemleak cannot guess the object type because the block
> > +        * size is different from the object size */
> > +       memleak_typeid(pa, struct platform_object);
>
> AFAICT, we about 300 kmalloc and kzalloc calls in the kernel that
> would need this annotation.

Most of them wouldn't need this annotation. The problem here is that
the stored pointer is usually pointing to &pa->pdev.dev and "pa" is
accessed via two container_of macros. There might be a few more cases
like this which I didn't find but I don't expect a large number.

> If we really can't fix the detector to deal with these,

As I posted a few weeks ago, this cannot be done without introducing a
lot of false negatives.

> I would prefer we introduce another memory allocator
> function such as:
>
>     void *kzalloc_extra(size_t obj_size, size_t nr_extra, gfp_t flags);
>
> That would do the right thing for memleak too.

There is another case where extra bytes are added in front of the
structure for alignment (the function could have "before" and "after"
arguments). As I said above there aren't many places where kmemleak
needs this.

As Ingo said in the past, we could better go for precise type
identification/checking in kmalloc and both the kernel and kmemleak
would benefit from this  (but that's a pretty large patch).

-- 
Catalin

  reply	other threads:[~2006-07-11  8:18 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-07-10 22:09 [PATCH 00/10] Kernel memory leak detector 0.8 Catalin Marinas
2006-07-10 22:09 ` [PATCH 01/10] Base support for kmemleak Catalin Marinas
2006-07-10 22:09 ` [PATCH 02/10] Some documentation " Catalin Marinas
2006-07-10 22:09 ` [PATCH 03/10] Add the memory allocation/freeing hooks " Catalin Marinas
2006-07-11  6:17   ` Pekka Enberg
2006-07-11  7:59     ` Catalin Marinas
2006-07-10 22:10 ` [PATCH 04/10] Modules support " Catalin Marinas
2006-07-10 22:10 ` [PATCH 05/10] Add kmemleak support for i386 Catalin Marinas
2006-07-10 22:10 ` [PATCH 06/10] Add kmemleak support for ARM Catalin Marinas
2006-07-10 22:10 ` [PATCH 07/10] Remove some of the kmemleak false positives Catalin Marinas
2006-07-11  6:49   ` Pekka Enberg
2006-07-11  8:18     ` Catalin Marinas [this message]
2006-07-10 22:10 ` [PATCH 08/10] Keep the __init functions after initialization Catalin Marinas
2006-07-10 22:10 ` [PATCH 09/10] Simple testing for kmemleak Catalin Marinas
2006-07-10 22:11 ` [PATCH 10/10] Update the MAINTAINERS file " Catalin Marinas
2006-07-11 12:27 ` [PATCH 00/10] Kernel memory leak detector 0.8 Michal Piotrowski
2006-07-11 12:46   ` Michal Piotrowski
2006-07-11 12:51     ` Michal Piotrowski
2006-07-11 13:00       ` Catalin Marinas
2006-07-11 13:09         ` Michal Piotrowski
2006-07-12  9:57         ` Joseph Fannin
2006-07-12 10:38           ` Catalin Marinas
2006-07-12 10:49             ` Michal Piotrowski
2006-07-14  2:22             ` Horst von Brand
2006-07-11 12:56   ` Catalin Marinas
2006-07-11 13:17     ` Michal Piotrowski
2006-07-11 13:28       ` Catalin Marinas
2006-07-11 13:49         ` Michal Piotrowski
2006-07-11 14:02           ` Catalin Marinas
2006-07-11 15:02             ` Michal Piotrowski
2006-07-11 15:44               ` Catalin Marinas
2006-07-11 16:02                 ` Michal Piotrowski
2006-07-11 16:31                   ` Catalin Marinas
2006-07-11 17:00                     ` Michal Piotrowski
2006-07-11 21:54                       ` Catalin Marinas
2006-07-12 11:35                         ` Michal Piotrowski
2006-07-12 16:17                           ` Catalin Marinas
2006-07-12 22:55                             ` Michal Piotrowski
2006-07-13 16:35                               ` Catalin Marinas
2006-07-13 19:14                                 ` Michal Piotrowski
2006-07-14 14:53                                   ` Catalin Marinas
2006-07-12 13:19                         ` Michal Piotrowski
2006-07-12 14:37                           ` Catalin Marinas
2006-07-12 15:33                             ` Michal Piotrowski
2006-07-12 16:12                               ` Catalin Marinas
2006-07-12 10:33               ` Catalin Marinas
2006-07-12 18:46                 ` Xiaolan Zhang
2006-07-13 18:52                 ` Xiaolan Zhang
2006-07-13 19:17                   ` Michal Piotrowski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b0943d9e0607110118p5b9dfc28qb377b61e59ec7535@mail.gmail.com \
    --to=catalin.marinas@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=penberg@cs.helsinki.fi \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®