linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Hugh Dickins <hughd@google.com>
To: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>,
	 Andrew Morton <akpm@linux-foundation.org>,
	 Baolin Wang <baolin.wang@linux.alibaba.com>,
	linux-fsdevel@vger.kernel.org,  linux-kernel@vger.kernel.org,
	linux-mm@kvack.org
Subject: 6.19 tmpfs __d_lookup() lockup
Date: Thu, 11 Dec 2025 19:56:38 -0800 (PST)	[thread overview]
Message-ID: <47e9d03c-7a50-2c7d-247d-36f95a5329ed@google.com> (raw)

Hi Al,

Though good for ordinary usage, your persistency mods for 6.19 cause
tmpfs to lockup in __d_lookup() during several fsstressing xfstests:
fairly reliably in generic/269, generic/476, generic/650, generic/750,
less reliably in generic/013, generic/585.  Typical dmesg below.
I have not spotted what's wrong: better hand over to you.

Of course, 2313598222f9 ("convert ramfs and tmpfs") (of Feb 26 2024!)
comes out as the first failing commit, no surprise there.

I did try inserting a BUG_ON(node == node->next) on line 2438 of
fs/dcache.c, just after __d_lookup's hlist_bl_for_each_entry_rcu(),
and that BUG was immediately hit (but, for all I know, perhaps that's
an unreliable asserition, perhaps it's acceptable for a race to result
in a momentary node == node->next there).

I did try hacking on xfstests common/rc, to allow ramfs where tmpfs
is enabled (and _df_device() then needs $DF_PROG -a to show ramfs).
I did not get a lockup or crash from any of them on ramfs (and 476 and
750 actually passed).  I don't draw any conclusion from that, maybe
ramfs just does not support some of the options which help fsstress
to generate the issue; but it might be useful background info.

Hoping you can soon guess what's gone wrong,
or ask for more info, thanks,
Hugh

[   54.354788] run fstests generic/269 at 2025-12-10 19:40:16
[  114.660512] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
[  114.670705] rcu: 	Tasks blocked on level-0 rcu_node (CPUs 0-7): P3185
[  114.681006] rcu: 	(detected by 6, t=15007 jiffies, g=3377, q=6940 ncpus=8)
[  114.691346] task:fsstress        state:R  running task     stack:0     pid:3185  tgid:3185  ppid:3059   task_flags:0x400140 flags:0x00080003
[  114.702226] Call Trace:
[  114.712908]  <TASK>
[  114.723528]  __schedule+0x67c/0x6c5
[  114.734245]  ? __d_lookup_rcu+0x7a/0x9c
[  114.744921]  ? irqentry_exit+0x27/0x35
[  114.755547]  ? sysvec_apic_timer_interrupt+0xa8/0xae
[  114.766300]  ? asm_sysvec_apic_timer_interrupt+0x1b/0x20
[  114.776953]  ? __d_lookup+0x37/0xad
[  114.787357]  ? __d_lookup+0x2f/0xad
[  114.797700]  ? d_lookup+0x2b/0x42
[  114.807970]  ? lookup_dcache+0x1f/0x60
[  114.818265]  ? lookup_one_qstr_excl+0x1a/0xbe
[  114.828525]  ? do_renameat2+0x1d1/0x3e4
[  114.838808]  ? getname_flags+0x4b/0x17a
[  114.849093]  ? __do_sys_rename+0x36/0x3e
[  114.859363]  ? __x64_sys_rename+0x11/0x13
[  114.869634]  ? x64_sys_call+0x34d/0xe2c
[  114.879797]  ? do_syscall_64+0x53/0x123
[  114.889820]  ? entry_SYSCALL_64_after_hwframe+0x4b/0x53
[  114.899748]  </TASK>


             reply	other threads:[~2025-12-12  3:56 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-12  3:56 Hugh Dickins [this message]
2025-12-12  5:02 ` Al Viro
2025-12-12  5:15   ` Hugh Dickins
2025-12-12  5:34   ` Al Viro
2025-12-12  5:57     ` Hugh Dickins
2025-12-12  6:30       ` Al Viro
2025-12-12  7:17         ` Hugh Dickins
2025-12-12 10:12           ` Hugh Dickins
2025-12-13  7:22             ` Al Viro
2025-12-14  3:27               ` shmem_rename() bugs (was Re: 6.19 tmpfs __d_lookup() lockup) Al Viro
2025-12-14  3:30                 ` [PATCH 1/2] shmem_whiteout(): fix regression from tree-in-dcache series Al Viro
2025-12-14  3:30                 ` [RFC][PATCH 2/2] shmem: fix recovery on rename failures Al Viro
2025-12-15  7:38                   ` Hugh Dickins
2025-12-15 11:58                   ` Christian Brauner
2025-12-15 16:03                   ` Chuck Lever
2025-12-15 16:54                     ` Al Viro
2025-12-16  6:02                 ` [git pull] shmem rename fixes Al Viro
2025-12-16  8:04                   ` pr-tracker-bot

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=47e9d03c-7a50-2c7d-247d-36f95a5329ed@google.com \
    --to=hughd@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=baolin.wang@linux.alibaba.com \
    --cc=brauner@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox