From: Eric Sandeen <sandeen@redhat.com>
To: Theodore Tso <tytso@mit.edu>, Andreas Dilger <adilger@sun.com>,
David Rientjes <rientjes@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundat
Subject: Re: upcoming kerneloops.org item: get_page_from_freelist
Date: Thu, 25 Jun 2009 17:11:25 -0500 [thread overview]
Message-ID: <4A43F60D.2040801@redhat.com> (raw)
In-Reply-To: <20090625220504.GG6472@mit.edu>
Theodore Tso wrote:
> On Thu, Jun 25, 2009 at 11:26:28PM +0200, Andreas Dilger wrote:
>> IIRC there was also a good reason for this in the past, related to
>> the buffers being submitted to the block device layer, and if they
>> were allocated from the slab cache with CONFIG_DEBUG_SLAB or something
>> similar enabled the buffer would be misaligned and cause grief.
>
> So what does SLAB/SLUB/SLOB do if we create a slab cache which is a
> power of two? Can one of the allocators still return misaligned
> blocks of memory in some circumstances?
>
> - Ted
ecryptfs used to do kmalloc(PAGE_CACHE_SIZE) & virt_to_page on that, and
with SLUB + slub debug, that gave back non-aligned memory, causing
eventual corruption ...
-Eric
WARNING: multiple messages have this Message-ID (diff)
From: Eric Sandeen <sandeen@redhat.com>
To: Theodore Tso <tytso@mit.edu>, Andreas Dilger <adilger@sun.com>,
David Rientjes <rientjes@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
penberg@cs.helsinki.fi, arjan@infradead.org,
linux-kernel@vger.kernel.org, cl@linux-foundation.org,
npiggin@suse.de, linux-ext4@vger.kernel.org
Subject: Re: upcoming kerneloops.org item: get_page_from_freelist
Date: Thu, 25 Jun 2009 17:11:25 -0500 [thread overview]
Message-ID: <4A43F60D.2040801@redhat.com> (raw)
In-Reply-To: <20090625220504.GG6472@mit.edu>
Theodore Tso wrote:
> On Thu, Jun 25, 2009 at 11:26:28PM +0200, Andreas Dilger wrote:
>> IIRC there was also a good reason for this in the past, related to
>> the buffers being submitted to the block device layer, and if they
>> were allocated from the slab cache with CONFIG_DEBUG_SLAB or something
>> similar enabled the buffer would be misaligned and cause grief.
>
> So what does SLAB/SLUB/SLOB do if we create a slab cache which is a
> power of two? Can one of the allocators still return misaligned
> blocks of memory in some circumstances?
>
> - Ted
ecryptfs used to do kmalloc(PAGE_CACHE_SIZE) & virt_to_page on that, and
with SLUB + slub debug, that gave back non-aligned memory, causing
eventual corruption ...
-Eric
next prev parent reply other threads:[~2009-06-25 22:12 UTC|newest]
Thread overview: 81+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-24 15:07 upcoming kerneloops.org item: get_page_from_freelist Arjan van de Ven
2009-06-24 16:46 ` Andrew Morton
2009-06-24 16:52 ` Linus Torvalds
2009-06-24 16:55 ` Pekka Enberg
2009-06-24 16:56 ` Pekka Enberg
2009-06-24 17:00 ` Pekka Enberg
2009-06-24 17:55 ` Andrew Morton
2009-06-24 17:53 ` Pekka Enberg
2009-06-24 18:30 ` Andrew Morton
2009-06-24 18:42 ` Linus Torvalds
2009-06-24 18:44 ` Pekka Enberg
2009-06-24 18:50 ` Linus Torvalds
2009-06-24 19:12 ` Pekka J Enberg
2009-06-24 19:21 ` Linus Torvalds
2009-06-24 19:06 ` Andrew Morton
2009-06-24 19:16 ` Linus Torvalds
2009-06-24 19:36 ` Andrew Morton
2009-06-24 19:46 ` Linus Torvalds
2009-06-24 19:47 ` Linus Torvalds
2009-06-24 20:01 ` Andrew Morton
2009-06-24 20:13 ` Linus Torvalds
2009-06-24 20:40 ` Linus Torvalds
2009-06-24 22:07 ` Andrew Morton
2009-06-25 4:05 ` Nick Piggin
2009-06-25 13:25 ` Theodore Tso
2009-06-25 18:51 ` David Rientjes
2009-06-25 19:38 ` Theodore Tso
2009-06-25 19:44 ` Theodore Tso
2009-06-25 19:55 ` Andrew Morton
2009-06-25 20:11 ` Linus Torvalds
2009-06-25 20:22 ` Linus Torvalds
2009-06-25 20:36 ` David Rientjes
2009-06-25 20:51 ` Linus Torvalds
2009-06-25 22:25 ` David Rientjes
2009-06-26 8:51 ` Nick Piggin
2009-06-25 20:18 ` David Rientjes
2009-06-25 20:37 ` Theodore Tso
2009-06-25 21:05 ` Joel Becker
2009-06-25 21:05 ` Joel Becker
2009-06-25 21:26 ` Andreas Dilger
2009-06-25 21:26 ` Andreas Dilger
2009-06-25 22:05 ` Theodore Tso
2009-06-25 22:11 ` Eric Sandeen [this message]
2009-06-25 22:11 ` Eric Sandeen
2009-06-26 1:11 ` Theodore Tso
2009-06-26 5:16 ` Pekka J Enberg
2009-06-26 8:56 ` Nick Piggin
2009-06-26 8:58 ` Pekka Enberg
2009-06-26 9:07 ` Nick Piggin
2009-06-29 21:06 ` Christoph Lameter
2009-06-30 7:59 ` Nick Piggin
2009-06-26 14:41 ` Eric Sandeen
2009-06-29 21:15 ` Christoph Lameter
2009-06-29 21:20 ` Eric Sandeen
2009-06-29 22:35 ` Christoph Lameter
2009-06-25 19:55 ` Jens Axboe
2009-06-25 20:08 ` Jens Axboe
2009-06-24 21:56 ` Andrew Morton
2009-06-25 4:14 ` Nick Piggin
2009-06-25 8:21 ` David Rientjes
2009-06-29 15:30 ` Mel Gorman
2009-06-29 19:20 ` Andrew Morton
2009-06-30 11:00 ` Mel Gorman
2009-06-30 19:35 ` David Rientjes
2009-06-30 20:32 ` Mel Gorman
2009-06-30 20:51 ` David Rientjes
2009-07-01 10:22 ` Mel Gorman
2009-06-29 23:35 ` David Rientjes
2009-06-30 7:47 ` Nick Piggin
2009-06-30 8:13 ` David Rientjes
2009-06-30 8:24 ` Nick Piggin
2009-06-30 8:41 ` David Rientjes
2009-06-30 9:09 ` Nick Piggin
2009-06-30 19:47 ` David Rientjes
2009-06-30 6:27 ` Pavel Machek
2009-06-28 10:16 ` Pavel Machek
2009-06-28 18:01 ` Linus Torvalds
2009-06-28 18:27 ` Arjan van de Ven
2009-06-28 18:36 ` Linus Torvalds
2009-06-30 7:35 ` Pavel Machek
2009-06-24 18:43 ` Pekka Enberg
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=4A43F60D.2040801@redhat.com \
--to=sandeen@redhat.com \
--cc=adilger@sun.com \
--cc=akpm@linux-foundation.org \
--cc=rientjes@google.com \
--cc=torvalds@linux-foundat \
--cc=tytso@mit.edu \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.