[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20241115115401.3701-1-hdanton@sina.com>
Date: Fri, 15 Nov 2024 19:54:01 +0800
From: Hillf Danton <hdanton@...a.com>
To: syzbot <syzbot+852e935b899bde73626e@...kaller.appspotmail.com>
Cc: Frederic Weisbecker <frederic@...nel.org>,
linux-kernel@...r.kernel.org,
syzkaller-bugs@...glegroups.com,
tglx@...utronix.de
Subject: Re: [syzbot] [kernel?] WARNING in posixtimer_send_sigqueue
On Thu, 14 Nov 2024 17:49:23 +0100 Frederic Weisbecker <frederic@...nel.org>
Le Wed, Nov 13, 2024 at 04:52:22PM -0800
> > syzbot found the following issue on:
> >
> > HEAD commit: 929beafbe7ac Add linux-next specific files for 20241108
> > git tree: linux-next
> > console+strace: https://syzkaller.appspot.com/x/log.txt?x=10f714e8580000
> > kernel config: https://syzkaller.appspot.com/x/.config?x=75175323f2078363
> > dashboard link: https://syzkaller.appspot.com/bug?extid=852e935b899bde73626e
> > compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
> > syz repro: https://syzkaller.appspot.com/x/repro.syz?x=14f714e8580000
> > C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1657b0c0580000
> >
> > Downloadable assets:
> > disk image: https://storage.googleapis.com/syzbot-assets/9705ecb6a595/disk-929beafb.raw.xz
> > vmlinux: https://storage.googleapis.com/syzbot-assets/dbdd1f64b9b8/vmlinux-929beafb.xz
> > kernel image: https://storage.googleapis.com/syzbot-assets/3f70d07a929b/bzImage-929beafb.xz
> >
> > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > Reported-by: syzbot+852e935b899bde73626e@...kaller.appspotmail.com
> >
> > ------------[ cut here ]------------
> > WARNING: CPU: 0 PID: 5854 at kernel/signal.c:2008
> > posixtimer_send_sigqueue+0x9da/0xbc0 kernel/signal.c:2008
>
> That's because prepare_signal() does not only return false when the signal is
> ignored but also when the task group is exiting. It's possible that the task
> enters in do_exit() with pending signal and then the timer is reset and a new
> signal is queued before the sighand dies.
>
> This should fix it:
>
> diff --git a/kernel/signal.c b/kernel/signal.c
> index cbf70c808969..10b464b9d91f 100644
> --- a/kernel/signal.c
> +++ b/kernel/signal.c
> @@ -2003,9 +2003,15 @@ void posixtimer_send_sigqueue(struct k_itimer *tmr)
> if (!prepare_signal(sig, t, false)) {
> result = TRACE_SIGNAL_IGNORED;
>
> - /* Paranoia check. Try to survive. */
> - if (WARN_ON_ONCE(!list_empty(&q->list)))
> + if (!list_empty(&q->list)) {
> + /*
> + * If task group is exiting with the signal already pending,
> + * wait for __exit_signal() to do its job. Otherwise if
> + * ignored, it's not supposed to be queued. Try to survive.
> + */
> + WARN_ON_ONCE(!(t->signal->flags & SIGNAL_GROUP_EXIT));
> goto out;
> + }
>
> /* Periodic timers with SIG_IGN are queued on the ignored list */
> if (tmr->it_sig_periodic) {
>
Test Frederic's patch on top of the next tree.
#syz test
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2004,9 +2004,15 @@ void posixtimer_send_sigqueue(struct k_i
if (!prepare_signal(sig, t, false)) {
result = TRACE_SIGNAL_IGNORED;
- /* Paranoia check. Try to survive. */
- if (WARN_ON_ONCE(!list_empty(&q->list)))
+ if (!list_empty(&q->list)) {
+ /*
+ * If task group is exiting with the signal already pending,
+ * wait for __exit_signal() to do its job. Otherwise if
+ * ignored, it's not supposed to be queued. Try to survive.
+ */
+ WARN_ON_ONCE(!(t->signal->flags & SIGNAL_GROUP_EXIT));
goto out;
+ }
/* Periodic timers with SIG_IGN are queued on the ignored list */
if (tmr->it_sig_periodic) {
--
Powered by blists - more mailing lists