From: "Marek Marczykowski-Górecki" <marmarek@invisiblethingslab.com>
To: stable@vger.kernel.org
Cc: regressions@lists.linux.dev, Alasdair Kergon <agk@redhat.com>,
Mike Snitzer <snitzer@kernel.org>,
Mikulas Patocka <mpatocka@redhat.com>,
dm-devel@redhat.com
Subject: Intermittent storage (dm-crypt?) freeze - regression 6.4->6.5
Date: Sat, 21 Oct 2023 05:38:58 +0200 [thread overview]
Message-ID: <ZTNH0qtmint/zLJZ@mail-itl> (raw)
[-- Attachment #1: Type: text/plain, Size: 7203 bytes --]
Hi,
Since updating from 6.4.13 to 6.5.5 occasionally I hit a storage
subsystem freeze - any I/O ends up frozen. I'm not sure what exactly
triggers the issue, but often it happens when doing some LVM operations
(lvremove, lvrename etc) on a dm-thin volume together with bulk data
copy to/from another LVM thin volume with ext4 fs.
The storage stack I use is:
nvme -> dm-crypt (LUKS) -> dm-thin (LVM thin pool) -> ext4
And this whole thing running in a (PV) dom0 under Xen, on Qubes OS 4.2 to be
specific.
I can reproduce the issue on at least 3 different machines. I did tried
also 6.5.6 and the issue is still there. I haven't checked newer
versions, but briefly reviewed git log and haven't found anything
suggesting a fix to similar issue.
I managed to bisect it down to this commit:
commit 5054e778fcd9cd29ddaa8109077cd235527e4f94
Author: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon May 1 09:19:17 2023 -0400
dm crypt: allocate compound pages if possible
It was reported that allocating pages for the write buffer in dm-crypt
causes measurable overhead [1].
Change dm-crypt to allocate compound pages if they are available. If
not, fall back to the mempool.
[1] https://listman.redhat.com/archives/dm-devel/2023-February/053284.html
Suggested-by: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>
TBH, I'm not sure if the bug is in this commit, or maybe in some
functions it uses (I don't see dm-crypt functions directly involved in
the stack traces I collected). But reverting this commit on top of 6.5.6
seems to fix the issue.
I tried also CONFIG_PROVE_LOCKING, but it didn't show any issue.
I managed to collect "blocked tasks" dump via sysrq below. Few more can
be found at https://github.com/QubesOS/qubes-issues/issues/8575
[ 4246.558313] sysrq: Show Blocked State
[ 4246.558388] task:journal-offline state:D stack:0 pid:8098 ppid:1 flags:0x00000002
[ 4246.558407] Call Trace:
[ 4246.558414] <TASK>
[ 4246.558422] __schedule+0x23d/0x670
[ 4246.558440] schedule+0x5e/0xd0
[ 4246.558450] io_schedule+0x46/0x70
[ 4246.558461] folio_wait_bit_common+0x13d/0x350
[ 4246.558475] ? __pfx_wake_page_function+0x10/0x10
[ 4246.558488] folio_wait_writeback+0x2c/0x90
[ 4246.558498] mpage_prepare_extent_to_map+0x15c/0x4d0
[ 4246.558512] ext4_do_writepages+0x25f/0x770
[ 4246.558523] ext4_writepages+0xad/0x180
[ 4246.558533] do_writepages+0xcf/0x1e0
[ 4246.558543] ? __seccomp_filter+0x32a/0x4f0
[ 4246.558554] filemap_fdatawrite_wbc+0x63/0x90
[ 4246.558567] __filemap_fdatawrite_range+0x5c/0x80
[ 4246.558578] file_write_and_wait_range+0x4a/0xb0
[ 4246.558588] ext4_sync_file+0x88/0x380
[ 4246.558598] __x64_sys_fsync+0x3b/0x70
[ 4246.558609] do_syscall_64+0x5c/0x90
[ 4246.558621] ? exit_to_user_mode_prepare+0xb2/0xd0
[ 4246.558632] entry_SYSCALL_64_after_hwframe+0x6e/0xd8
[ 4246.558644] RIP: 0033:0x7710cf124d0a
[ 4246.558654] RSP: 002b:00007710ccdfda40 EFLAGS: 00000293 ORIG_RAX: 000000000000004a
[ 4246.558668] RAX: ffffffffffffffda RBX: 000064bb92f67e60 RCX: 00007710cf124d0a
[ 4246.558679] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000028
[ 4246.558691] RBP: 000064bb92f72670 R08: 0000000000000000 R09: 00007710ccdfe6c0
[ 4246.558702] R10: 00007710cf0adfee R11: 0000000000000293 R12: 000064bb92505940
[ 4246.558713] R13: 0000000000000002 R14: 00007ffc05649500 R15: 00007710cc5fe000
[ 4246.558728] </TASK>
[ 4246.558836] task:lvm state:D stack:0 pid:7835 ppid:5665 flags:0x00004006
[ 4246.558852] Call Trace:
[ 4246.558857] <TASK>
[ 4246.558863] __schedule+0x23d/0x670
[ 4246.558874] schedule+0x5e/0xd0
[ 4246.558884] io_schedule+0x46/0x70
[ 4246.558894] dm_wait_for_bios_completion+0xfc/0x110
[ 4246.558909] ? __pfx_autoremove_wake_function+0x10/0x10
[ 4246.558922] __dm_suspend+0x7e/0x1b0
[ 4246.558932] dm_internal_suspend_noflush+0x5c/0x80
[ 4246.558946] pool_presuspend+0xcc/0x130 [dm_thin_pool]
[ 4246.558968] dm_table_presuspend_targets+0x3f/0x60
[ 4246.558980] __dm_suspend+0x41/0x1b0
[ 4246.558991] dm_suspend+0xc0/0xe0
[ 4246.559001] dev_suspend+0xa5/0xd0
[ 4246.559011] ctl_ioctl+0x26e/0x350
[ 4246.559020] ? __pfx_dev_suspend+0x10/0x10
[ 4246.559032] dm_ctl_ioctl+0xe/0x20
[ 4246.559041] __x64_sys_ioctl+0x94/0xd0
[ 4246.559052] do_syscall_64+0x5c/0x90
[ 4246.559062] ? do_syscall_64+0x6b/0x90
[ 4246.559072] ? do_syscall_64+0x6b/0x90
[ 4246.559081] ? xen_pv_evtchn_do_upcall+0x54/0xb0
[ 4246.559093] entry_SYSCALL_64_after_hwframe+0x6e/0xd8
[ 4246.559104] RIP: 0033:0x7f1cb77cfe0f
[ 4246.559112] RSP: 002b:00007fff870f2560 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[ 4246.559141] RAX: ffffffffffffffda RBX: 00005b8d13c16580 RCX: 00007f1cb77cfe0f
[ 4246.559152] RDX: 00005b8d144a2180 RSI: 00000000c138fd06 RDI: 0000000000000003
[ 4246.559164] RBP: 00005b8d144a2180 R08: 00005b8d132b1190 R09: 00007fff870f2420
[ 4246.559175] R10: 0000000000000000 R11: 0000000000000246 R12: 000000000000000c
[ 4246.559186] R13: 00005b8d132aacf0 R14: 00005b8d1324414d R15: 00005b8d144a21b0
[ 4246.559199] </TASK>
[ 4246.559207] task:kworker/u8:3 state:D stack:0 pid:8033 ppid:2 flags:0x00004000
[ 4246.559222] Workqueue: writeback wb_workfn (flush-253:4)
[ 4246.559238] Call Trace:
[ 4246.559244] <TASK>
[ 4246.559249] __schedule+0x23d/0x670
[ 4246.559260] schedule+0x5e/0xd0
[ 4246.559270] io_schedule+0x46/0x70
[ 4246.559280] folio_wait_bit_common+0x13d/0x350
[ 4246.559290] ? __pfx_wake_page_function+0x10/0x10
[ 4246.559302] mpage_prepare_extent_to_map+0x309/0x4d0
[ 4246.559314] ext4_do_writepages+0x25f/0x770
[ 4246.559324] ext4_writepages+0xad/0x180
[ 4246.559334] do_writepages+0xcf/0x1e0
[ 4246.559344] ? find_busiest_group+0x42/0x1a0
[ 4246.559354] __writeback_single_inode+0x3d/0x280
[ 4246.559368] writeback_sb_inodes+0x1ed/0x4a0
[ 4246.559381] __writeback_inodes_wb+0x4c/0xf0
[ 4246.559393] wb_writeback+0x298/0x310
[ 4246.559403] wb_do_writeback+0x230/0x2b0
[ 4246.559414] wb_workfn+0x5f/0x260
[ 4246.559424] ? _raw_spin_unlock+0xe/0x30
[ 4246.559434] ? finish_task_switch.isra.0+0x95/0x2b0
[ 4246.559447] ? __schedule+0x245/0x670
[ 4246.559457] process_one_work+0x1df/0x3e0
[ 4246.559466] worker_thread+0x51/0x390
[ 4246.559475] ? __pfx_worker_thread+0x10/0x10
[ 4246.559484] kthread+0xe5/0x120
[ 4246.559495] ? __pfx_kthread+0x10/0x10
[ 4246.559504] ret_from_fork+0x31/0x50
[ 4246.559514] ? __pfx_kthread+0x10/0x10
[ 4246.559523] ret_from_fork_asm+0x1b/0x30
[ 4246.559536] </TASK>
--
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next reply other threads:[~2023-10-21 3:39 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-21 3:38 Marek Marczykowski-Górecki [this message]
2023-10-21 7:48 ` Intermittent storage (dm-crypt?) freeze - regression 6.4->6.5 Bagas Sanjaya
2023-10-29 6:23 ` Linux regression tracking #update (Thorsten Leemhuis)
2023-10-23 20:59 ` Mikulas Patocka
2023-10-25 3:10 ` Marek Marczykowski-Górecki
2023-10-25 3:22 ` Marek Marczykowski-Górecki
2023-10-25 10:13 ` Mikulas Patocka
2023-10-27 17:32 ` Mikulas Patocka
2023-10-28 9:23 ` Matthew Wilcox
2023-10-28 15:14 ` Mike Snitzer
2023-10-29 11:15 ` Marek Marczykowski-Górecki
2023-10-29 20:02 ` Vlastimil Babka
2023-10-30 7:37 ` Mikulas Patocka
2023-10-30 8:37 ` Vlastimil Babka
2023-10-30 11:22 ` Mikulas Patocka
2023-10-30 11:30 ` Vlastimil Babka
2023-10-30 11:37 ` Mikulas Patocka
2023-10-30 12:25 ` Jan Kara
2023-10-30 13:30 ` Marek Marczykowski-Górecki
2023-10-30 14:08 ` Mikulas Patocka
2023-10-30 15:56 ` Jan Kara
2023-10-30 16:51 ` Marek Marczykowski-Górecki
2023-10-30 17:50 ` Mikulas Patocka
2023-10-31 3:48 ` Marek Marczykowski-Górecki
2023-10-31 14:01 ` Jan Kara
2023-10-31 15:42 ` Marek Marczykowski-Górecki
2023-10-31 17:17 ` Mikulas Patocka
2023-10-31 17:24 ` Mikulas Patocka
2023-11-02 0:38 ` Marek Marczykowski-Górecki
2023-11-02 9:28 ` Mikulas Patocka
2023-11-02 11:45 ` Marek Marczykowski-Górecki
2023-11-02 17:06 ` Mikulas Patocka
2023-11-03 15:01 ` Marek Marczykowski-Górecki
2023-11-03 15:10 ` Keith Busch
2023-11-03 16:15 ` Marek Marczykowski-Górecki
2023-11-03 16:54 ` Keith Busch
2023-11-03 20:30 ` Marek Marczykowski-G'orecki
2023-11-03 22:42 ` Keith Busch
2023-11-04 9:27 ` Mikulas Patocka
2023-11-04 13:59 ` Keith Busch
2023-11-06 7:10 ` Christoph Hellwig
2023-11-06 14:59 ` [PATCH] swiotlb-xen: provide the "max_mapping_size" method Mikulas Patocka
2023-11-06 15:16 ` Keith Busch
2023-11-06 15:30 ` Mike Snitzer
2023-11-06 17:12 ` [PATCH v2] " Mikulas Patocka
2023-11-07 4:18 ` Stefano Stabellini
2023-11-08 7:31 ` Christoph Hellwig
2023-11-06 7:08 ` Intermittent storage (dm-crypt?) freeze - regression 6.4->6.5 Christoph Hellwig
2023-11-02 12:21 ` Jan Kara
2023-11-01 1:27 ` Ming Lei
2023-11-01 2:14 ` Marek Marczykowski-Górecki
2023-11-01 2:15 ` Marek Marczykowski-Górecki
2023-11-01 2:35 ` Marek Marczykowski-Górecki
2023-11-01 3:24 ` Ming Lei
2023-11-01 10:15 ` Hannes Reinecke
2023-11-01 10:26 ` Jan Kara
2023-11-01 11:23 ` Ming Lei
2023-11-02 14:02 ` Keith Busch
2023-11-01 12:16 ` Mikulas Patocka
2023-10-30 11:28 ` Jan Kara
2023-10-30 11:49 ` Mikulas Patocka
2023-10-30 12:11 ` Jan Kara
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=ZTNH0qtmint/zLJZ@mail-itl \
--to=marmarek@invisiblethingslab.com \
--cc=agk@redhat.com \
--cc=dm-devel@redhat.com \
--cc=mpatocka@redhat.com \
--cc=regressions@lists.linux.dev \
--cc=snitzer@kernel.org \
--cc=stable@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;
as well as URLs for NNTP newsgroup(s).