From: Mel Gorman <mgorman@suse.de>
To: Michal Hocko <mhocko@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Al Viro <viro@zeniv.linux.org.uk>,
Johannes Weiner <hannes@cmpxchg.org>,
Rik van Riel <riel@redhat.com>, Neil Brown <neilb@suse.de>,
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
Sage Weil <sage@inktank.com>, Mark Fasheh <mfasheh@suse.com>,
linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] mm: Use GFP_KERNEL allocation for the page cache in page_cache_read
Date: Wed, 18 Mar 2015 14:38:47 +0000 [thread overview]
Message-ID: <20150318143847.GC3087@suse.de> (raw)
In-Reply-To: <1426687766-518-1-git-send-email-mhocko@suse.cz>
On Wed, Mar 18, 2015 at 03:09:26PM +0100, Michal Hocko wrote:
> page_cache_read has been historically using page_cache_alloc_cold to
> allocate a new page. This means that mapping_gfp_mask is used as the
> base for the gfp_mask. Many filesystems are setting this mask to
> GFP_NOFS to prevent from fs recursion issues. page_cache_read is,
> however, not called from the fs layer so it doesn't need this
> protection. Even ceph and ocfs2 which call filemap_fault from their
> fault handlers seem to be OK because they are not taking any fs lock
> before invoking generic implementation.
>
> The protection might be even harmful. There is a strong push to fail
> GFP_NOFS allocations rather than loop within allocator indefinitely with
> a very limited reclaim ability. Once we start failing those requests
> the OOM killer might be triggered prematurely because the page cache
> allocation failure is propagated up the page fault path and end up in
> pagefault_out_of_memory.
>
> Use GFP_KERNEL mask instead because it is safe from the reclaim
> recursion POV. We are already doing GFP_KERNEL allocations down
> add_to_page_cache_lru path.
>
> Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> Signed-off-by: Michal Hocko <mhocko@suse.cz>
I'm very far behind after LSF/MM so do not know where this came out of
but it loses addressing restriction hints from the driver such as
drivers/gpu/drm/gma500/gem.c: mapping_set_gfp_mask(r->gem.filp->f_mapping, GFP_KERNEL | __GFP_DMA32);
It also loses mobility hints for fragmentation avoidance.
fs/inode.c: mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE);
If users of mapping_set_gfp_mask are now being ignored then it should at
least trigger a once-off warning that the flags are being ignored so
it's obvious if a recursion does occur and cause problems.
--
Mel Gorman
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Mel Gorman <mgorman@suse.de>
To: Michal Hocko <mhocko@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Al Viro <viro@zeniv.linux.org.uk>,
Johannes Weiner <hannes@cmpxchg.org>,
Rik van Riel <riel@redhat.com>, Neil Brown <neilb@suse.de>,
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
Sage Weil <sage@inktank.com>, Mark Fasheh <mfasheh@suse.com>,
linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] mm: Use GFP_KERNEL allocation for the page cache in page_cache_read
Date: Wed, 18 Mar 2015 14:38:47 +0000 [thread overview]
Message-ID: <20150318143847.GC3087@suse.de> (raw)
In-Reply-To: <1426687766-518-1-git-send-email-mhocko@suse.cz>
On Wed, Mar 18, 2015 at 03:09:26PM +0100, Michal Hocko wrote:
> page_cache_read has been historically using page_cache_alloc_cold to
> allocate a new page. This means that mapping_gfp_mask is used as the
> base for the gfp_mask. Many filesystems are setting this mask to
> GFP_NOFS to prevent from fs recursion issues. page_cache_read is,
> however, not called from the fs layer so it doesn't need this
> protection. Even ceph and ocfs2 which call filemap_fault from their
> fault handlers seem to be OK because they are not taking any fs lock
> before invoking generic implementation.
>
> The protection might be even harmful. There is a strong push to fail
> GFP_NOFS allocations rather than loop within allocator indefinitely with
> a very limited reclaim ability. Once we start failing those requests
> the OOM killer might be triggered prematurely because the page cache
> allocation failure is propagated up the page fault path and end up in
> pagefault_out_of_memory.
>
> Use GFP_KERNEL mask instead because it is safe from the reclaim
> recursion POV. We are already doing GFP_KERNEL allocations down
> add_to_page_cache_lru path.
>
> Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
> Signed-off-by: Michal Hocko <mhocko@suse.cz>
I'm very far behind after LSF/MM so do not know where this came out of
but it loses addressing restriction hints from the driver such as
drivers/gpu/drm/gma500/gem.c: mapping_set_gfp_mask(r->gem.filp->f_mapping, GFP_KERNEL | __GFP_DMA32);
It also loses mobility hints for fragmentation avoidance.
fs/inode.c: mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE);
If users of mapping_set_gfp_mask are now being ignored then it should at
least trigger a once-off warning that the flags are being ignored so
it's obvious if a recursion does occur and cause problems.
--
Mel Gorman
SUSE Labs
next prev parent reply other threads:[~2015-03-18 14:38 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-18 14:09 [PATCH] mm: Use GFP_KERNEL allocation for the page cache in page_cache_read Michal Hocko
2015-03-18 14:09 ` Michal Hocko
2015-03-18 14:32 ` Rik van Riel
2015-03-18 14:32 ` Rik van Riel
2015-03-18 14:37 ` Michal Hocko
2015-03-18 14:37 ` Michal Hocko
2015-03-18 14:38 ` Mel Gorman [this message]
2015-03-18 14:38 ` Mel Gorman
2015-03-18 14:43 ` Michal Hocko
2015-03-18 14:43 ` Michal Hocko
2015-03-18 14:44 ` Rik van Riel
2015-03-18 14:44 ` Rik van Riel
2015-03-18 14:55 ` Michal Hocko
2015-03-18 14:55 ` Michal Hocko
2015-03-19 7:14 ` Dave Chinner
2015-03-19 7:14 ` Dave Chinner
2015-03-19 11:11 ` [PATCH] mm: Use GFP_KERNEL allocation for the page cache inpage_cache_read Tetsuo Handa
2015-03-19 11:11 ` Tetsuo Handa
2015-03-19 12:44 ` [PATCH] mm: Use GFP_KERNEL allocation for the page cache in page_cache_read Michal Hocko
2015-03-19 12:44 ` Michal Hocko
2015-03-20 3:48 ` Dave Chinner
2015-03-20 3:48 ` Dave Chinner
2015-03-20 13:14 ` Michal Hocko
2015-03-20 13:14 ` Michal Hocko
2015-03-20 22:51 ` Dave Chinner
2015-03-20 22:51 ` Dave Chinner
2015-03-23 13:02 ` Michal Hocko
2015-03-23 13:02 ` Michal Hocko
2015-03-26 9:53 ` Michal Hocko
2015-03-26 9:53 ` Michal Hocko
2015-03-26 21:43 ` Dave Chinner
2015-03-26 21:43 ` Dave Chinner
2015-03-30 8:22 ` Michal Hocko
2015-03-30 8:22 ` Michal Hocko
2015-03-31 21:46 ` Dave Chinner
2015-03-31 21:46 ` Dave Chinner
2015-04-07 12:16 ` Michal Hocko
2015-04-07 12:16 ` Michal Hocko
2015-03-18 15:45 ` Michal Hocko
2015-03-18 15:45 ` Michal Hocko
2015-03-18 21:38 ` NeilBrown
2015-03-19 13:55 ` Michal Hocko
2015-03-19 13:55 ` Michal Hocko
2015-03-19 14:27 ` Michal Hocko
2015-03-19 14:27 ` Michal Hocko
2015-03-20 3:57 ` Dave Chinner
2015-03-20 3:57 ` Dave Chinner
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=20150318143847.GC3087@suse.de \
--to=mgorman@suse.de \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mfasheh@suse.com \
--cc=mhocko@suse.cz \
--cc=neilb@suse.de \
--cc=penguin-kernel@I-love.SAKURA.ne.jp \
--cc=riel@redhat.com \
--cc=sage@inktank.com \
--cc=viro@zeniv.linux.org.uk \
/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.