getrusage: use sig->stats_lock rather than lock_task_sighand()
authorOleg Nesterov <oleg@redhat.com>
Mon, 22 Jan 2024 15:50:53 +0000 (16:50 +0100)
committerAndrew Morton <akpm@linux-foundation.org>
Thu, 8 Feb 2024 05:20:32 +0000 (21:20 -0800)
commitf7ec1cd5cc7ef3ad964b677ba82b8b77f1c93009
treece218b54749d63d25ef3cb706b5e23b776c13000
parentdaa694e4137571b4ebec330f9a9b4d54aa8b8089
getrusage: use sig->stats_lock rather than lock_task_sighand()

lock_task_sighand() can trigger a hard lockup. If NR_CPUS threads call
getrusage() at the same time and the process has NR_THREADS, spin_lock_irq
will spin with irqs disabled O(NR_CPUS * NR_THREADS) time.

Change getrusage() to use sig->stats_lock, it was specifically designed
for this type of use. This way it runs lockless in the likely case.

TODO:
- Change do_task_stat() to use sig->stats_lock too, then we can
  remove spin_lock_irq(siglock) in wait_task_zombie().

- Turn sig->stats_lock into seqcount_rwlock_t, this way the
  readers in the slow mode won't exclude each other. See
  https://lore.kernel.org/all/20230913154907.GA26210@redhat.com/

- stats_lock has to disable irqs because ->siglock can be taken
  in irq context, it would be very nice to change __exit_signal()
  to avoid the siglock->stats_lock dependency.

Link: https://lkml.kernel.org/r/20240122155053.GA26214@redhat.com
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reported-by: Dylan Hatch <dylanbhatch@google.com>
Tested-by: Dylan Hatch <dylanbhatch@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
kernel/sys.c