Re: [RFC][PATCH v2 5/5] mutex: Give spinners a chance to spin_on_owner if need_resched() triggered while queued

[Daniel J Blueman <daniel@numascale.com>]

On Thursday, 6 February 2014 22:10:01 UTC+8, Peter Zijlstra  wrote:
 > On Wed, Feb 05, 2014 at 04:44:34PM -0500, Waiman Long wrote:
 > > I have an alternative way of breaking out of the MCS lock waiting 
queue when
 > > need_resched() is set. I overload the locked flag to indicate a 
skipped node
 > > if negative.
 >
 > I'm not quite seeing how it works (then again, I've not really read the
 > patch carefully).
 >
 > Suppose you break out; at that point you get queued and go to sleep.
 > Suppose you got woken up while you MCS entry is still 'pending' and
 > magically win the race and acquire the lock.
 >
 > At that point your MCS entry can be re-used while its still part of the
 > list.
 >
 > Its a fantastically small race window, but I don't see anything that
 > makes it impossible.
 >
 > > I run the patch through the AIM7 high-systime workload on a
 > > 4-socket server and it seemed to run fine.
 >
 > How do people run this AIM7 piece of shit? I let it run for over an hour
 > and it generated exactly 0 numbers, it just sits there eating cpu-time
 > and creating a racket from my pantry.

Without any better advice, I was building the OSDL AIM7 [1], tweaking 
DISKDIR in data/reaim.config and running (eg on a 384-core setup):
$ src/reaim -c data/reaim.config -f data/workfile.compute -i 16 -e 384

Thanks,
   Daniel

[1] http://sourceforge.net/projects/re-aim-7/
-- 
Daniel J Blueman
Principal Software Engineer, Numascale