]> git.ipfire.org Git - thirdparty/kernel/linux.git/commit
exit: prevent preemption of oopsing TASK_DEAD task
authorJann Horn <jannh@google.com>
Mon, 11 May 2026 15:55:11 +0000 (08:55 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Mon, 11 May 2026 15:55:11 +0000 (08:55 -0700)
commitc1fa0bb633e4a6b11e83ffc57fa5abe8ebb87891
tree477cd51b4a91a61c009d26b026e856224eebb4c3
parent5d6919055dec134de3c40167a490f33c74c12581
exit: prevent preemption of oopsing TASK_DEAD task

When an already-exiting task oopses, make_task_dead() currently calls
do_task_dead() with preemption enabled.  That is forbidden:
do_task_dead() calls __schedule(), which has a comment saying "WARNING:
must be called with preemption disabled!".

If an oopsing task is preempted in do_task_dead(), between becoming
TASK_DEAD and entering the scheduler explicitly, bad things happen:
finish_task_switch() assumes that once the scheduler has switched away
from a TASK_DEAD task, the task can never run again and its stack is no
longer needed; but that assumption apparently doesn't hold if the dead
task was preempted (the SM_PREEMPT case).

This means that the scheduler ends up repeatedly dropping references on
the dead task's stack, which can lead to use-after-free or double-free
of the entire task stack; in other words, two tasks can end up running
on the same stack, resulting in various kinds of memory corruption.

(This does not just affect "recursively oopsing" tasks; it is enough to
oops once during task exit, for example in a file_operations::release
handler)

Fixes: 7f80a2fd7db9 ("exit: Stop poorly open coding do_task_dead in make_task_dead")
Cc: stable@kernel.org
Signed-off-by: Jann Horn <jannh@google.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
kernel/exit.c