public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: Oleg Nesterov <oleg@redhat.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Anna-Maria Behnsen <anna-maria@linutronix.de>,
	Frederic Weisbecker <frederic@kernel.org>,
	John Stultz <jstultz@google.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>, Stephen Boyd <sboyd@kernel.org>,
	Eric Biederman <ebiederm@xmission.com>,
	Andrei Vagin <avagin@google.com>
Subject: Re: [patch V2 26/50] signal: Get rid of resched_timer logic
Date: Tue, 23 Apr 2024 23:18:50 +0200	[thread overview]
Message-ID: <87sezbu69x.ffs@tglx> (raw)
In-Reply-To: <20240419110632.GA3198@redhat.com>

On Fri, Apr 19 2024 at 13:06, Oleg Nesterov wrote:
> On 04/18, Oleg Nesterov wrote:
>> But this is not simple, collect_signal() checks SIGQUEUE_PREALLOC exactly
>> because (iiuc) we need to ensure that SI_TIMER didn't come from userspace.
>>
>> perhaps we should disallow SI_TIMER with _sys_private != 0 from userspace,
>> I dunno...
>>
>> And I don't really understand the "not to be passed to user" comment in
>> include/uapi/asm-generic/siginfo.h. copy_siginfo_to_user() just copies
>> the whole kernel_siginfo.
>
> OK, si_sys_private is cleared in dequeue_signal() (or in posixtimer_rearm()
> with this series).
>
> In the past SI_TIMER was defined as __SI_CODE(__SI_TIMER,-2), it was > 0,
> so it could not come from userspace (see the info->si_code >= 0 check in
> do_rt_sigqueueinfo).

Duh.

> Today SI_TIMER < 0. We could introduce SI_TIMER_KERNEL > 0 with the minimal
> changes, but this can't help because the commit 66dd34ad31e59 allows to send
> any siginfo to itself.

Well that predates the __SI_CODE() removal. So I doubt it's required
today, but what do I know about CRIU.

> Otoh, I have no idea how CRIU restores the posix timers. If a process has
> a pending blocked SI_TIMER signal, then I guess it actually needs to enqueue
> this signal at restore time, but resched_timer will be never true?

It can't restore the correct sys_si_private value because that is
nowhere exposed to user space.

There is no special treatment for SI_TIMER, so the signal restore might
just end up queueing a random extra SI_TIMER signal if there was one
pending.

I checked the CRIU source and it looks like this just "works" by
reconstructing and rearming the timer with the last expiry value. As
that is in the past it will fire immediately and queue the signal.

sys_si_private seems to be excluded from being set from user space in
compat mode, but in non-compat mode I can't find anything which prevents
that.

Let me stare more at this. 

Thanks,

        tglx

  reply	other threads:[~2024-04-23 21:18 UTC|newest]

Thread overview: 91+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-10 22:46 [patch V2 00/50] posix-timers: Cure inconsistencies and the SIG_IGN mess Thomas Gleixner
2024-04-10 22:46 ` [patch V2 01/50] selftests/timers/posix_timers: Simplify error handling Thomas Gleixner
2024-04-12  7:35   ` Anna-Maria Behnsen
2024-04-10 22:46 ` [patch V2 02/50] selftests/timers/posix_timers: Add SIG_IGN test Thomas Gleixner
2024-04-10 22:46 ` [patch V2 03/50] selftests/timers/posix_timers: Validate signal rules Thomas Gleixner
2024-04-10 22:46 ` [patch V2 04/50] selftests/timers/posix-timers: Validate SIGEV_NONE Thomas Gleixner
2024-04-10 22:46 ` [patch V2 05/50] selftests/timers/posix-timers: Validate timer_gettime() Thomas Gleixner
2024-04-10 22:46 ` [patch V2 06/50] selftests/timers/posix-timers: Validate overrun after unblock Thomas Gleixner
2024-04-10 22:46 ` [patch V2 07/50] posix-cpu-timers: Split up posix_cpu_timer_get() Thomas Gleixner
2024-04-12  7:35   ` Anna-Maria Behnsen
2024-04-12 18:25   ` Eric W. Biederman
2024-04-12 19:48     ` Thomas Gleixner
2024-04-16 14:44   ` Oleg Nesterov
2024-04-17  9:21     ` Anna-Maria Behnsen
2024-04-17 11:08       ` Oleg Nesterov
2024-04-17 20:33   ` Frederic Weisbecker
2024-04-10 22:46 ` [patch V2 08/50] posix-cpu-timers: Save interval only for armed timers Thomas Gleixner
2024-04-11 14:25   ` Anna-Maria Behnsen
2024-04-11 22:00     ` Thomas Gleixner
2024-04-10 22:46 ` [patch V2 09/50] posix-cpu-timers: Handle interval timers correctly in timer_get() Thomas Gleixner
2024-04-17 22:50   ` Frederic Weisbecker
2024-04-10 22:46 ` [patch V2 10/50] posix-cpu-timers: Handle SIGEV_NONE " Thomas Gleixner
2024-04-12 18:40   ` Eric W. Biederman
2024-04-12 19:49     ` Thomas Gleixner
2024-04-10 22:46 ` [patch V2 11/50] posix-cpu-timers: Handle SIGEV_NONE timers correctly in timer_set() Thomas Gleixner
2024-04-11 15:48   ` Anna-Maria Behnsen
2024-04-11 22:02     ` Thomas Gleixner
2024-04-17 23:04       ` Frederic Weisbecker
2024-04-10 22:46 ` [patch V2 12/50] posix-cpu-timers: Replace old expiry retrieval in posix_cpu_timer_set() Thomas Gleixner
2024-04-15 14:03   ` Anna-Maria Behnsen
2024-04-10 22:46 ` [patch V2 13/50] posix-cpu-timers: Do not arm SIGEV_NONE timers Thomas Gleixner
2024-04-15 14:03   ` Anna-Maria Behnsen
2024-04-10 22:46 ` [patch V2 14/50] posix-cpu-timers: Use @now instead of @val for clarity Thomas Gleixner
2024-04-10 22:46 ` [patch V2 15/50] posix-cpu-timers: Remove incorrect comment in posix_cpu_timer_set() Thomas Gleixner
2024-04-10 22:46 ` [patch V2 16/50] posix-cpu-timers: Simplify posix_cpu_timer_set() Thomas Gleixner
2024-04-15 15:28   ` Anna-Maria Behnsen
2024-04-10 22:46 ` [patch V2 17/50] posix-timers: Retrieve interval in common timer_settime() code Thomas Gleixner
2024-04-15 15:43   ` Anna-Maria Behnsen
2024-04-10 22:46 ` [patch V2 18/50] posix-timers: Clear overrun in common_timer_set() Thomas Gleixner
2024-04-10 22:46 ` [patch V2 19/50] posix-timers: Convert timer list to hlist Thomas Gleixner
2024-04-10 22:46 ` [patch V2 20/50] posix-timers: Consolidate timer setup Thomas Gleixner
2024-04-16 16:12   ` Anna-Maria Behnsen
2024-04-23 19:38     ` Thomas Gleixner
2024-04-10 22:46 ` [patch V2 21/50] posix-cpu-timers: Make k_itimer::it_active consistent Thomas Gleixner
2024-04-17 10:11   ` Anna-Maria Behnsen
2024-04-10 22:46 ` [patch V2 22/50] posix-timers: Consolidate signal queueing Thomas Gleixner
2024-04-10 22:46 ` [patch V2 23/50] signal: Remove task argument from dequeue_signal() Thomas Gleixner
2024-04-18 14:18   ` Oleg Nesterov
2024-04-10 22:46 ` [patch V2 24/50] signal: Replace BUG_ON()s Thomas Gleixner
2024-04-18 14:37   ` Oleg Nesterov
2024-04-10 22:46 ` [patch V2 25/50] signal: Confine POSIX_TIMERS properly Thomas Gleixner
2024-04-17 12:09   ` Anna-Maria Behnsen
2024-04-18 15:23   ` Oleg Nesterov
2024-04-19  5:42     ` Thomas Gleixner
2024-04-10 22:46 ` [patch V2 26/50] signal: Get rid of resched_timer logic Thomas Gleixner
2024-04-18 16:38   ` Oleg Nesterov
2024-04-18 18:18     ` Oleg Nesterov
2024-04-19 11:06       ` Oleg Nesterov
2024-04-23 21:18         ` Thomas Gleixner [this message]
2024-04-24  1:48           ` Thomas Gleixner
2024-04-25  7:22             ` Andrei Vagin
2024-04-10 22:46 ` [patch V2 27/50] posix-timers: Cure si_sys_private race Thomas Gleixner
2024-04-10 22:47 ` [patch V2 28/50] signal: Allow POSIX timer signals to be dropped Thomas Gleixner
2024-04-10 22:47 ` [patch V2 29/50] posix-timers: Drop signal if timer has been deleted or reprogrammed Thomas Gleixner
2024-04-10 22:47 ` [patch V2 30/50] posix-timers: Rename k_itimer::it_requeue_pending Thomas Gleixner
2024-04-10 22:47 ` [patch V2 31/50] posix-timers: Add proper state tracking Thomas Gleixner
2024-04-17 13:40   ` Anna-Maria Behnsen
2024-04-10 22:47 ` [patch V2 32/50] posix-timers: Make signal delivery consistent Thomas Gleixner
2024-04-18 10:29   ` Anna-Maria Behnsen
2024-04-10 22:47 ` [patch V2 33/50] posix-timers: Make signal overrun accounting sensible Thomas Gleixner
2024-04-10 22:47 ` [patch V2 34/50] posix-cpu-timers: Use dedicated flag for CPU timer nanosleep Thomas Gleixner
2024-04-10 22:47 ` [patch V2 35/50] posix-timers: Add a refcount to struct k_itimer Thomas Gleixner
2024-04-10 22:47 ` [patch V2 36/50] signal: Split up __sigqueue_alloc() Thomas Gleixner
2024-04-10 22:47 ` [patch V2 37/50] signal: Provide posixtimer_sigqueue_init() Thomas Gleixner
2024-04-10 22:47 ` [patch V2 38/50] signal: Add sys_private_ptr to siginfo::_sifields::_timer Thomas Gleixner
2024-04-10 22:47 ` [patch V2 39/50] posix-timers: Store PID type in the timer Thomas Gleixner
2024-04-10 22:47 ` [patch V2 40/50] signal: Refactor send_sigqueue() Thomas Gleixner
2024-04-10 22:47 ` [patch V2 41/50] posix-timers: Embed sigqueue in struct k_itimer Thomas Gleixner
2024-04-10 22:47 ` [patch V2 42/50] signal: Cleanup unused posix-timer leftovers Thomas Gleixner
2024-04-10 22:47 ` [patch V2 43/50] signal: Add task argument to flush_sigqueue_mask() Thomas Gleixner
2024-04-10 22:47 ` [patch V2 44/50] signal: Provide ignored_posix_timers list Thomas Gleixner
2024-04-10 22:47 ` [patch V2 45/50] posix-timers: Handle ignored list on delete and exit Thomas Gleixner
2024-04-10 22:47 ` [patch V2 46/50] signal: Handle ignored signals in do_sigaction(action != SIG_IGN) Thomas Gleixner
2024-04-10 22:47 ` [patch V2 47/50] signal: Queue ignored posixtimers on ignore list Thomas Gleixner
2024-04-10 22:47 ` [patch V2 48/50] posix-timers: Cleanup SIG_IGN workaround leftovers Thomas Gleixner
2024-04-10 22:47 ` [patch V2 49/50] alarmtimers: Remove the throttle mechanism from alarm_forward_now() Thomas Gleixner
2024-04-10 22:47 ` [patch V2 50/50] alarmtimers: Remove return value from alarm functions Thomas Gleixner
2024-04-15 12:30 ` [PATCH] posix-timers: Handle returned errors poperly in [i]timer_delete() Anna-Maria Behnsen
2024-04-15 13:00   ` Oleg Nesterov
2024-04-15 14:15     ` Anna-Maria Behnsen
2024-04-15 16:27       ` Oleg Nesterov

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=87sezbu69x.ffs@tglx \
    --to=tglx@linutronix.de \
    --cc=anna-maria@linutronix.de \
    --cc=avagin@google.com \
    --cc=ebiederm@xmission.com \
    --cc=frederic@kernel.org \
    --cc=jstultz@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=oleg@redhat.com \
    --cc=peterz@infradead.org \
    --cc=sboyd@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