code review 7424054: runtime: fix deadlock

runtime: fix deadlock
The deadlock episodically occurs on misc/cgo/test/TestCthread.
The problem is that starttheworld() leaves some P's with local work
without M's. Then all active M's enter into syscalls, but reject to
wake another M's due to the following check (both in entersyscallblock() and in retake()):
if(p->runqhead == p->runqtail &&
        runtime·atomicload(&runtime·sched.nmspinning) +
        runtime·atomicload(&runtime·sched.npidle) > 0)
        continue;

[rsc, 2013-03-04 20:04:02]

please hg mail if this is ready for review
thanks.

[dvyukov, 2013-03-04 21:10:39]

sigh!

On Mon, Mar 4, 2013 at 10:04 PM,  <rsc@golang.org> wrote:
> please hg mail if this is ready for review
> thanks.
>
>
> https://codereview.appspot.com/7424054/

[dvyukov, 2013-03-04 21:17:50]

Yes, it is ready for review.
I can't hg mail right now.


On Monday, March 4, 2013, Dmitry Vyukov wrote:

> sigh!
>
> On Mon, Mar 4, 2013 at 10:04 PM,  <rsc@golang.org <javascript:;>> wrote:
> > please hg mail if this is ready for review
> > thanks.
> >
> >
> > https://codereview.appspot.com/7424054/
>

[dvyukov, 2013-03-04 21:18:12]

Hello rsc@golang.org (cc: golang-dev@googlegroups.com),

I'd like you to review this change to
https://dvyukov%40google.com@code.google.com/p/go/

[dvyukov, 2013-03-07 10:24:41]

ping

[rsc, 2013-03-07 13:48:41]

LGTM

[dvyukov, 2013-03-07 17:40:10]

*** Submitted as https://code.google.com/p/go/source/detail?r=a346bcf77992 ***

runtime: fix deadlock
The deadlock episodically occurs on misc/cgo/test/TestCthread.
The problem is that starttheworld() leaves some P's with local work
without M's. Then all active M's enter into syscalls, but reject to
wake another M's due to the following check (both in entersyscallblock() and in
retake()):
if(p->runqhead == p->runqtail &&
        runtime·atomicload(&runtime·sched.nmspinning) +
        runtime·atomicload(&runtime·sched.npidle) > 0)
        continue;

R=rsc
CC=golang-dev
https://codereview.appspot.com/7424054

[albert.strasheim, 2013-03-08 05:21:32]

On 2013/03/07 17:40:10, dvyukov wrote:
> *** Submitted as https://code.google.com/p/go/source/detail?r=a346bcf77992 ***
> runtime: fix deadlock

testing seems to indicate: Fixes issue 4998.