Re: [PATCH 2/4] signal: Make flush_sigqueue() use free_q to release memory

[Matthew Wilcox]

On Thu, Mar 21, 2019 at 05:45:10PM -0400, Waiman Long wrote:
> It was found that if a process had many pending signals (e.g. millions),
> the act of exiting that process might cause its parent to have a hard
> lockup especially on a debug kernel with features like KASAN enabled.
> It was because the flush_sigqueue() was called in release_task() with
> tasklist_lock held and irq disabled.

This rather apocalyptic language is a bit uncalled for.  I appreciate the
warning is scary, but all that's really happening is that the debug code
is taking too long to execute.

> To avoid this dire condition and reduce lock hold time of tasklist_lock,
> flush_sigqueue() is modified to pass in a freeing queue pointer so that
> the actual freeing of memory objects can be deferred until after the
> tasklist_lock is released and irq re-enabled.

I think this is a really bad solution.  It looks kind of generic,
but isn't.  It's terribly inefficient, and all it's really doing is
deferring the debugging code until we've re-enabled interrupts.

We'd be much better off just having a list_head in the caller
and list_splice() the queue->list onto that caller.  Then call
__sigqueue_free() for each signal on the queue.