Re: Bug about sched_4bsd?

2010/1/21 Attilio Rao <attilio_at_freebsd.org>:
> 2010/1/20 Giovanni Trematerra <giovanni.trematerra_at_gmail.com>:
>> On Mon, Jan 18, 2010 at 3:58 AM, Attilio Rao <attilio_at_freebsd.org> wrote:
>>> 2010/1/17 Kohji Okuno <okuno.kohji_at_jp.panasonic.com>:
>>>> Hello,
>>>>
>>>> Could you check sched_4bsd.patch, please?
>>>
>>> I think, instead, that what needs to happen is to have sched_switch()
>>> to do a lock handover from sleepq/turnstile spinlock to schedlock.
>>> That way, if threads are willing to contest on td_lock they will be
>>> still inhibited.
>>> I'm not sure if this patch breaks any invariant, if you may test I
>>> would appreciate:
>>> http://www.freebsd.org/~attilio/sched_4bsd_schedlock.diff
>>
>> I stressed an 8-core machine with pho's stress2 kernel stress suite and
>> your patch seems to break the invariant THREAD_LOCKPTR_ASSERT in
>> turnstile_claim:subr_turnstile.c
>
> Oh, right, I guess what we really want is to block the td_lock.
> This is the new patch:
> http://www.freebsd.org/~attilio/sched_4bsd_schedlock2.diff

FYI,
I've updated the patch with a more correct use of thread_lock_block()
and a merge of this interface with the custom interface of
thread_block_switch() from ULE.

The overall result should be better (and more correct). Please refresh
your patch. Testing with ULE should be due as well.

Thanks,
Attilio


-- 
Peace can only be achieved by understanding - A. Einstein