From: Pedro Falcato <pfalcato@suse.de>
To: Benjamin Eikel <debian@eikel.org>
Cc: linux-mm@kvack.org, rcu@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
"Liam R. Howlett" <Liam.Howlett@oracle.com>,
Lorenzo Stoakes <ljs@kernel.org>,
Vlastimil Babka <vbabka@kernel.org>,
David Hildenbrand <david@kernel.org>
Subject: Re: [BUG] RCU stall in unmap_page_range during exit_mmap (6.19 regression)
Date: Tue, 17 Mar 2026 16:19:11 +0000 [thread overview]
Message-ID: <6ccygsneky6la4unzj2qd6m6o3ty3ek6xwidznzgrpdvc7tfel@qeoj7rramrmu> (raw)
In-Reply-To: <6261958.lOV4Wx5bFT@thinkpad-benjamin>
+Cc a bunch of relevant people
On Mon, Mar 16, 2026 at 09:18:37AM +0100, Benjamin Eikel wrote:
> Dear Linux kernel developers,
>
> I have reported this issue as Debian bug #1130138 [1], which has been marked
> “upstream”. I am therefore also reaching out here to make you aware of it.
> Please let me know if you need additional information.
>
> After upgrading from linux-image-6.18.15+deb14-amd64, I see this error several
> times and the system shortly freezes. It seems related to Zoom crashing first,
> but the freezes were not noticable with older Linux kernels.
Do you have a simpler repro you can share?
> Currently booted into 6.18.15+deb14-amd64 which does not exhibit this issue.
> The stalls occurred when running 6.19.6+deb14-amd64. More information about my
> system can be found in the Debian bug report.
>
Can you attach the debian kernel config? Is it possible something relevant
changed since 6.18?
> Mär 09 09:13:48 thinkpad-benjamin kernel: rcu: INFO: rcu_preempt detected
> stalls on CPUs/tasks:
> Mär 09 09:13:48 thinkpad-benjamin kernel: rcu: Tasks blocked on
> level-0 rcu_node (CPUs 0-15): P10008/1:b..l
> Mär 09 09:13:48 thinkpad-benjamin kernel: rcu: (detected by 5, t=5252
> jiffies, g=177537, q=28067 ncpus=16)
> Mär 09 09:13:48 thinkpad-benjamin kernel: task:zoom state:R
> running task stack:0 pid:10008 tgid:10008 ppid:2089
> task_flags:0x40040c flags:0x00080802
> Mär 09 09:13:48 thinkpad-benjamin kernel: Call Trace:
> Mär 09 09:13:48 thinkpad-benjamin kernel: <TASK>
> Mär 09 09:13:48 thinkpad-benjamin kernel: __schedule+0x48a/0xfc0
> Mär 09 09:13:48 thinkpad-benjamin kernel: preempt_schedule+0x32/0x50
> Mär 09 09:13:48 thinkpad-benjamin kernel: preempt_schedule_thunk+0x16/0x30
> Mär 09 09:13:48 thinkpad-benjamin kernel: _raw_spin_unlock+0x22/0x30
> Mär 09 09:13:48 thinkpad-benjamin kernel: unmap_page_range+0xbcb/0x1700
Hmm. We have a bunch of cond_resched() sprayed around this code. This _should_
work. I don't see anything particularly interesting that changed between
6.18 and 6.19.
> Mär 09 09:13:48 thinkpad-benjamin kernel: unmap_vmas+0xa1/0x180
> Mär 09 09:13:48 thinkpad-benjamin kernel: exit_mmap+0xe1/0x3b0
> Mär 09 09:13:48 thinkpad-benjamin kernel: __mmput+0x41/0x150
> Mär 09 09:13:48 thinkpad-benjamin kernel: do_exit+0x251/0xa70
> Mär 09 09:13:48 thinkpad-benjamin kernel: do_group_exit+0x2d/0xc0
> Mär 09 09:13:48 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:13:48 thinkpad-benjamin kernel: get_signal+0x8a4/0x920
> Mär 09 09:13:48 thinkpad-benjamin kernel:
> arch_do_signal_or_restart+0x40/0x250
> Mär 09 09:13:48 thinkpad-benjamin kernel: exit_to_user_mode_loop+0x8e/0x480
> Mär 09 09:13:48 thinkpad-benjamin kernel: ? __x64_sys_sched_setaffinity+0x6a/
> 0xb0
> Mär 09 09:13:48 thinkpad-benjamin kernel: do_syscall_64+0x20b/0x600
> Mär 09 09:13:48 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:13:48 thinkpad-benjamin kernel: ? ksys_read+0xcd/0xf0
> Mär 09 09:13:48 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:13:48 thinkpad-benjamin kernel: ? do_syscall_64+0xbe/0x600
> Mär 09 09:13:48 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:13:48 thinkpad-benjamin kernel: ? do_syscall_64+0xbe/0x600
> Mär 09 09:13:48 thinkpad-benjamin kernel: ? __x64_sys_close+0x3d/0x80
> Mär 09 09:13:48 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:13:48 thinkpad-benjamin kernel: ? do_syscall_64+0xbe/0x600
> Mär 09 09:13:48 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:13:48 thinkpad-benjamin kernel: ? exc_page_fault+0x7e/0x1a0
> Mär 09 09:13:48 thinkpad-benjamin kernel:
> entry_SYSCALL_64_after_hwframe+0x76/0x7e
> Mär 09 09:13:48 thinkpad-benjamin kernel: RIP: 0033:0x7f2876359e2d
> Mär 09 09:13:48 thinkpad-benjamin kernel: RSP: 002b:00007ffc91243ad8 EFLAGS:
> 00000246 ORIG_RAX: 00000000000000cb
> Mär 09 09:13:48 thinkpad-benjamin kernel: RAX: 0000000000000000 RBX:
> 00007f2875207700 RCX: 00007f2876359e2d
> Mär 09 09:13:48 thinkpad-benjamin kernel: RDX: 00007ffc91243ae0 RSI:
> 0000000000000080 RDI: 0000000000002718
> Mär 09 09:13:48 thinkpad-benjamin kernel: RBP: 0000000000000020 R08:
> 0000000000000001 R09: 00007f2875206de8
> Mär 09 09:13:48 thinkpad-benjamin kernel: R10: 0000000000000000 R11:
> 0000000000000246 R12: 00007ffc91243ae0
> Mär 09 09:13:48 thinkpad-benjamin kernel: R13: 00007ffc91243c80 R14:
> 0000000000000000 R15: 0000000000000020
> Mär 09 09:13:48 thinkpad-benjamin kernel: </TASK>
> Mär 09 09:13:48 thinkpad-benjamin kernel: rcu: INFO: rcu_preempt detected
> expedited stalls on CPUs/tasks: { P10008 } 5361 jiffies s: 1609 root: 0x0/T
> Mär 09 09:13:48 thinkpad-benjamin kernel: rcu: blocking rcu_node structures
> (internal RCU debug):
>
> Another one:
>
> Mär 09 09:14:51 thinkpad-benjamin kernel: rcu: INFO: rcu_preempt detected
> stalls on CPUs/tasks:
> Mär 09 09:14:51 thinkpad-benjamin kernel: rcu: Tasks blocked on
> level-0 rcu_node (CPUs 0-15): P10008/1:b..l
> Mär 09 09:14:51 thinkpad-benjamin kernel: rcu: (detected by 5, t=21007
> jiffies, g=177537, q=87903 ncpus=16)
> Mär 09 09:14:51 thinkpad-benjamin kernel: task:zoom state:R
> running task stack:0 pid:10008 tgid:10008 ppid:2089
> task_flags:0x40040c flags:0x00080802
> Mär 09 09:14:51 thinkpad-benjamin kernel: Call Trace:
> Mär 09 09:14:51 thinkpad-benjamin kernel: <TASK>
> Mär 09 09:14:51 thinkpad-benjamin kernel: __schedule+0x48a/0xfc0
> Mär 09 09:14:51 thinkpad-benjamin kernel: preempt_schedule+0x32/0x50
> Mär 09 09:14:51 thinkpad-benjamin kernel: preempt_schedule_thunk+0x16/0x30
> Mär 09 09:14:51 thinkpad-benjamin kernel: _raw_spin_unlock+0x22/0x30
> Mär 09 09:14:51 thinkpad-benjamin kernel: unmap_page_range+0xbcb/0x1700
> Mär 09 09:14:51 thinkpad-benjamin kernel: unmap_vmas+0xa1/0x180
> Mär 09 09:14:51 thinkpad-benjamin kernel: exit_mmap+0xe1/0x3b0
> Mär 09 09:14:51 thinkpad-benjamin kernel: __mmput+0x41/0x150
> Mär 09 09:14:51 thinkpad-benjamin kernel: do_exit+0x251/0xa70
> Mär 09 09:14:51 thinkpad-benjamin kernel: do_group_exit+0x2d/0xc0
> Mär 09 09:14:51 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:14:51 thinkpad-benjamin kernel: get_signal+0x8a4/0x920
> Mär 09 09:14:51 thinkpad-benjamin kernel:
> arch_do_signal_or_restart+0x40/0x250
> Mär 09 09:14:51 thinkpad-benjamin kernel: exit_to_user_mode_loop+0x8e/0x480
> Mär 09 09:14:51 thinkpad-benjamin kernel: ? __x64_sys_sched_setaffinity+0x6a/
> 0xb0
> Mär 09 09:14:51 thinkpad-benjamin kernel: do_syscall_64+0x20b/0x600
> Mär 09 09:14:51 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:14:51 thinkpad-benjamin kernel: ? ksys_read+0xcd/0xf0
> Mär 09 09:14:51 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:14:51 thinkpad-benjamin kernel: ? do_syscall_64+0xbe/0x600
> Mär 09 09:14:51 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:14:51 thinkpad-benjamin kernel: ? do_syscall_64+0xbe/0x600
> Mär 09 09:14:51 thinkpad-benjamin kernel: ? __x64_sys_close+0x3d/0x80
> Mär 09 09:14:51 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:14:51 thinkpad-benjamin kernel: ? do_syscall_64+0xbe/0x600
> Mär 09 09:14:51 thinkpad-benjamin kernel: ?
> srso_alias_return_thunk+0x5/0xfbef5
> Mär 09 09:14:51 thinkpad-benjamin kernel: ? exc_page_fault+0x7e/0x1a0
> Mär 09 09:14:51 thinkpad-benjamin kernel:
> entry_SYSCALL_64_after_hwframe+0x76/0x7e
> Mär 09 09:14:51 thinkpad-benjamin kernel: RIP: 0033:0x7f2876359e2d
> Mär 09 09:14:51 thinkpad-benjamin kernel: RSP: 002b:00007ffc91243ad8 EFLAGS:
> 00000246 ORIG_RAX: 00000000000000cb
> Mär 09 09:14:51 thinkpad-benjamin kernel: RAX: 0000000000000000 RBX:
> 00007f2875207700 RCX: 00007f2876359e2d
> Mär 09 09:14:51 thinkpad-benjamin kernel: RDX: 00007ffc91243ae0 RSI:
> 0000000000000080 RDI: 0000000000002718
> Mär 09 09:14:51 thinkpad-benjamin kernel: RBP: 0000000000000020 R08:
> 0000000000000001 R09: 00007f2875206de8
> Mär 09 09:14:51 thinkpad-benjamin kernel: R10: 0000000000000000 R11:
> 0000000000000246 R12: 00007ffc91243ae0
> Mär 09 09:14:51 thinkpad-benjamin kernel: R13: 00007ffc91243c80 R14:
> 0000000000000000 R15: 0000000000000020
> Mär 09 09:14:51 thinkpad-benjamin kernel: </TASK>
> Mär 09 09:14:52 thinkpad-benjamin kernel: rcu: INFO: rcu_preempt detected
> expedited stalls on CPUs/tasks: { P10008 } 21233 jiffies s: 1609 root: 0x0/T
> Mär 09 09:14:52 thinkpad-benjamin kernel: rcu: blocking rcu_node structures
> (internal RCU debug):
>
> Note the stall duration escalating from t=5252 jiffies (~5s) to t=21007 jiffies
> (~21s), with the same grace period (g=177537), indicating the unmap_page_range
> operation blocked for over a minute.
>
> Kind regards
> Benjamin
>
> [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1130138
>
>
>
--
Pedro
next prev parent reply other threads:[~2026-03-17 16:19 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-16 8:18 [BUG] RCU stall in unmap_page_range during exit_mmap (6.19 regression) Benjamin Eikel
2026-03-17 14:46 ` [regression] " Thorsten Leemhuis
2026-03-17 16:19 ` Pedro Falcato [this message]
2026-03-19 16:28 ` [BUG] " Benjamin Eikel
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=6ccygsneky6la4unzj2qd6m6o3ty3ek6xwidznzgrpdvc7tfel@qeoj7rramrmu \
--to=pfalcato@suse.de \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=david@kernel.org \
--cc=debian@eikel.org \
--cc=linux-mm@kvack.org \
--cc=ljs@kernel.org \
--cc=rcu@vger.kernel.org \
--cc=vbabka@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