Re: [RESEND] shm: shm exit scalability fixes

[Andrew Morton <akpm@linux-foundation.org>]

On Tue, 17 Jun 2014 12:27:44 -0500 Jack Miller <millerjo@us.ibm.com> wrote:

> [ RESEND note: Adding relevant CCs, fixed a couple of typos in commit message,
> patches unchanged. Original intro follows. ]
> 
> All -
> 
> This is small set of patches our team has had kicking around for a few versions
> internally that fixes tasks getting hung on shm_exit when there are many
> threads hammering it at once.
> 
> Anton wrote a simple test to cause the issue:
> 
> http://ozlabs.org/~anton/junkcode/bust_shm_exit.c
> 
> Before applying this patchset, this test code will cause either hanging
> tracebacks or pthread out of memory errors.
> 
> After this patchset, it will still produce output like:
> 
> root@somehost:~# ./bust_shm_exit 1024 160
> ...
> INFO: rcu_sched detected stalls on CPUs/tasks: {} (detected by 116, t=2111
> jiffies, g=241, c=240, q=7113)
> INFO: Stall ended before state dump start
> ...
> But the task will continue to run along happily, so we consider this an
> improvement over hanging, even if it's a bit noisy.
> 
> I didn't author these patches, but I'd be happy to take any feedback and
> address any issues.

As it stands, these patches will be recorded as authored by yourself. 
You can override this by putting

From: Someone Else <email-address>

at the start of the changelog.  Please let me know if you'd like me to
make such changes.


I agree that adding 8/16 bytes to the task_struct (if CONFIG_SYSVIPC)
is regrettable.  But the poor thing is a dumping ground anyway :( I
suppose one could replace this and sysvsem with a

	struct ipc_stuff *ipc_stuff;	/* NULL if I never used IPC */

but there are probably lots of similar changes we could (and one day
may) make to the task_struct.

I queued the patches, tagged for 3.16-rc1 - hopefully Davidlohr and
Manfred will be able to find the time to go over them closely.