Re: [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks()

From	Nikolay Borisov <kernel@kyup.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks()
Date	2016-02-25 10:10 +0100
Message-ID	<r605R-1As-23@gated-at.bofh.it> (permalink)
References	<r5L6N-7D3-5@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 02/24/2016 07:09 PM, Konstantin Khlebnikov wrote:
> This might be unexpected but pages allocated for sbi->s_buddy_cache are
> charged to current memory cgroup. So, GFP_NOFS allocation could fail if
> current task has been killed by OOM or if current memory cgroup has no
> free memory left. Block allocator cannot handle such failures here yet.
> 
> Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>

Adding new users of GFP_NOFAIL is deprecated. Where exactly does the
block allocator fail, I skimmed the code and failing ext4_mb_load_buddy
seems to be handled at all call sites. There are some BUG_ONs but from
the comments there I guess they should occur when we try to find a page
and not allocate a new one?

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

[PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks() Konstantin Khlebnikov <khlebnikov@yandex-team.ru> - 2016-02-24 18:10 +0100
  Re: [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks() Michal Hocko <mhocko@kernel.org> - 2016-02-25 10:10 +0100
    Re: [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks() Nikolay Borisov <kernel@kyup.com> - 2016-02-25 10:20 +0100
      Re: [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks() Michal Hocko <mhocko@kernel.org> - 2016-02-25 11:30 +0100
  Re: [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks() Nikolay Borisov <kernel@kyup.com> - 2016-02-25 10:10 +0100

csiph-web