From: luminosity1999@gmail.com
To: david@fromorbit.com
Cc: linux-xfs@vger.kernel.org, allexjlzheng@tencent.com,
flyingpeng@tencent.com
Subject: Re: [PATCH 01/42] xfs: fix low space alloc deadlock
Date: Mon, 23 Dec 2024 17:43:12 +0800 [thread overview]
Message-ID: <20241223094312.39933-1-txpeng@tencent.com> (raw)
In-Reply-To: <20230118224505.1964941-2-david@fromorbit.com>
Suppose there are two threads doing delayed allocation and both of them
trigger the btree splitting. Due to delayed allocation, thread A holds
the lock of AGF 1, thread B holds the lock of AGF 3:
agno(low -> high): 1 3
A B
Then both of them initiate a btree split work(AS from A and BS from B),
but due to memory pressure described by commit
c85007e2e3942da1f9361e4b5a9388ea3a8dcc5b ("xfs: don't use BMBT btree
split workers for IO completion"), the two works are processed by the
rescuer thread, AS first, then BS. After trying each AG with the TRYLOCK
flag but failing all, AS retries with blocking allocation, until it
reaches AG 3, then deadlocks there: AGF 3 lock was held by thread B, but
its split work BS was queued after AS and has no chance to be processed.
Locking order is followed, but still deadlock occurs.
We didn't actually trigger such deadlock, but encountered a similar
hungtask reported by commit c85007e2e3942da1f9361e4b5a9388ea3a8dcc5b
("xfs: don't use BMBT btree split workers for IO completion"). That
commit has fixed our hungtask. But while analyzing the vmcore, we found
there are some other btree split works queued on the rescuer thread,
initiated from delayed allocation threads, each of which has held an AGF
lock. That leads us to this patch and thinking about such a case.
So is this possible? It seems currently once queued into the rescuer
thread, a split work has no way to know whether there is another
concurrent split work initiator holding the lock it requires.
next prev parent reply other threads:[~2024-12-23 9:43 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-18 22:44 [PATCH 00/42] xfs: per-ag centric allocation alogrithms Dave Chinner
2023-01-18 22:44 ` [PATCH 01/42] xfs: fix low space alloc deadlock Dave Chinner
2023-01-19 16:39 ` Allison Henderson
2024-12-23 9:43 ` luminosity1999 [this message]
2023-01-18 22:44 ` [PATCH 02/42] xfs: prefer free inodes at ENOSPC over chunk allocation Dave Chinner
2023-01-19 19:08 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 03/42] xfs: block reservation too large for minleft allocation Dave Chinner
2023-01-19 20:38 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 04/42] xfs: drop firstblock constraints from allocation setup Dave Chinner
2023-01-19 22:03 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 05/42] xfs: t_firstblock is tracking AGs not blocks Dave Chinner
2023-01-19 22:12 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 06/42] xfs: don't assert fail on transaction cancel with deferred ops Dave Chinner
2023-01-19 22:18 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 07/42] xfs: active perag reference counting Dave Chinner
2023-01-21 5:16 ` Allison Henderson
2023-02-01 19:08 ` Darrick J. Wong
2023-02-06 22:56 ` Dave Chinner
2023-01-18 22:44 ` [PATCH 08/42] xfs: rework the perag trace points to be perag centric Dave Chinner
2023-01-21 5:16 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 09/42] xfs: convert xfs_imap() to take a perag Dave Chinner
2023-02-01 19:10 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 10/42] xfs: use active perag references for inode allocation Dave Chinner
2023-01-22 6:48 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 11/42] xfs: inobt can use perags in many more places than it does Dave Chinner
2023-01-22 6:48 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 12/42] xfs: convert xfs_ialloc_next_ag() to an atomic Dave Chinner
2023-01-22 7:03 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 13/42] xfs: perags need atomic operational state Dave Chinner
2023-01-23 4:04 ` Allison Henderson
2023-01-18 22:44 ` [PATCH 14/42] xfs: introduce xfs_for_each_perag_wrap() Dave Chinner
2023-01-23 5:41 ` Allison Henderson
2023-02-06 23:14 ` Dave Chinner
2023-02-01 19:28 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 15/42] xfs: rework xfs_alloc_vextent() Dave Chinner
2023-02-01 19:39 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 16/42] xfs: factor xfs_alloc_vextent_this_ag() for _iterate_ags() Dave Chinner
2023-01-18 22:44 ` [PATCH 17/42] xfs: combine __xfs_alloc_vextent_this_ag and xfs_alloc_ag_vextent Dave Chinner
2023-02-01 22:25 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 18/42] xfs: use xfs_alloc_vextent_this_ag() where appropriate Dave Chinner
2023-01-18 22:44 ` [PATCH 19/42] xfs: factor xfs_bmap_btalloc() Dave Chinner
2023-01-18 22:44 ` [PATCH 20/42] xfs: use xfs_alloc_vextent_first_ag() where appropriate Dave Chinner
2023-02-01 22:43 ` Darrick J. Wong
2023-02-06 23:16 ` Dave Chinner
2023-01-18 22:44 ` [PATCH 21/42] xfs: use xfs_alloc_vextent_start_bno() " Dave Chinner
2023-02-01 22:51 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 22/42] xfs: introduce xfs_alloc_vextent_near_bno() Dave Chinner
2023-02-01 22:52 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 23/42] xfs: introduce xfs_alloc_vextent_exact_bno() Dave Chinner
2023-02-01 23:00 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 24/42] xfs: introduce xfs_alloc_vextent_prepare() Dave Chinner
2023-01-18 22:44 ` [PATCH 25/42] xfs: move allocation accounting to xfs_alloc_vextent_set_fsbno() Dave Chinner
2023-01-18 22:44 ` [PATCH 26/42] xfs: fold xfs_alloc_ag_vextent() into callers Dave Chinner
2023-01-18 22:44 ` [PATCH 27/42] xfs: move the minimum agno checks into xfs_alloc_vextent_check_args Dave Chinner
2023-01-18 22:44 ` [PATCH 28/42] xfs: convert xfs_alloc_vextent_iterate_ags() to use perag walker Dave Chinner
2023-02-01 23:13 ` Darrick J. Wong
2023-01-18 22:44 ` [PATCH 29/42] xfs: convert trim to use for_each_perag_range Dave Chinner
2023-02-01 23:15 ` Darrick J. Wong
2023-02-06 23:19 ` Dave Chinner
2023-01-18 22:44 ` [PATCH 30/42] xfs: factor out filestreams from xfs_bmap_btalloc_nullfb Dave Chinner
2023-01-18 22:44 ` [PATCH 31/42] xfs: get rid of notinit from xfs_bmap_longest_free_extent Dave Chinner
2023-01-18 22:44 ` [PATCH 32/42] xfs: use xfs_bmap_longest_free_extent() in filestreams Dave Chinner
2023-01-18 22:44 ` [PATCH 33/42] xfs: move xfs_bmap_btalloc_filestreams() to xfs_filestreams.c Dave Chinner
2023-01-18 22:44 ` [PATCH 34/42] xfs: merge filestream AG lookup into xfs_filestream_select_ag() Dave Chinner
2023-01-18 22:44 ` [PATCH 35/42] xfs: merge new filestream AG selection " Dave Chinner
2023-01-18 22:44 ` [PATCH 36/42] xfs: remove xfs_filestream_select_ag() longest extent check Dave Chinner
2023-01-18 22:45 ` [PATCH 37/42] xfs: factor out MRU hit case in xfs_filestream_select_ag Dave Chinner
2023-01-18 22:45 ` [PATCH 38/42] xfs: track an active perag reference in filestreams Dave Chinner
2023-01-18 22:45 ` [PATCH 39/42] xfs: use for_each_perag_wrap in xfs_filestream_pick_ag Dave Chinner
2023-01-18 22:45 ` [PATCH 40/42] xfs: pass perag to filestreams tracing Dave Chinner
2023-01-18 22:45 ` [PATCH 41/42] xfs: return a referenced perag from filestreams allocator Dave Chinner
2023-02-02 0:01 ` Darrick J. Wong
2023-02-06 23:22 ` Dave Chinner
2023-01-18 22:45 ` [PATCH 42/42] xfs: refactor the filestreams allocator pick functions Dave Chinner
2023-02-02 0:08 ` Darrick J. Wong
2023-02-06 23:26 ` Dave Chinner
2023-02-02 0:14 ` [PATCH 00/42] xfs: per-ag centric allocation alogrithms Darrick J. Wong
2023-02-06 23:13 ` Dave Chinner
-- strict thread matches above, loose matches on Subject: below --
2023-02-09 22:17 [PATCH v3 " Dave Chinner
2023-02-09 22:17 ` [PATCH 01/42] xfs: fix low space alloc deadlock 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=20241223094312.39933-1-txpeng@tencent.com \
--to=luminosity1999@gmail.com \
--cc=allexjlzheng@tencent.com \
--cc=david@fromorbit.com \
--cc=flyingpeng@tencent.com \
--cc=linux-xfs@vger.kernel.org \
/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