svn commit: r200447 - in head: share/man/man9 sys/kern sys/sys

[Attilio Rao]

2009/12/14 John Baldwin <jhb at freebsd.org>:
> On Saturday 12 December 2009 4:31:07 pm Attilio Rao wrote:
>> Author: attilio
>> Date: Sat Dec 12 21:31:07 2009
>> New Revision: 200447
>> URL: http://svn.freebsd.org/changeset/base/200447
>>
>> Log:
>>   In current code, threads performing an interruptible sleep (on both
>>   sxlock, via the sx_{s, x}lock_sig() interface, or plain lockmgr), will
>>   leave the waiters flag on forcing the owner to do a wakeup even when if
>>   the waiter queue is empty.
>>   That operation may lead to a deadlock in the case of doing a fake wakeup
>>   on the "preferred" (based on the wakeup algorithm) queue while the other
>>   queue has real waiters on it, because nobody is going to wakeup the 2nd
>>   queue waiters and they will sleep indefinitively.
>>
>>   A similar bug, is present, for lockmgr in the case the waiters are
>>   sleeping with LK_SLEEPFAIL on.  In this case, even if the waiters queue
>>   is not empty, the waiters won't progress after being awake but they will
>>   just fail, still not taking care of the 2nd queue waiters (as instead the
>>   lock owned doing the wakeup would expect).
>>
>>   In order to fix this bug in a cheap way (without adding too much locking
>>   and complicating too much the semantic) add a sleepqueue interface which
>>   does report the actual number of waiters on a specified queue of a
>>   waitchannel (sleepq_sleepcnt()) and use it in order to determine if the
>>   exclusive waiters (or shared waiters) are actually present on the lockmgr
>>   (or sx) before to give them precedence in the wakeup algorithm.
>>   This fix alone, however doesn't solve the LK_SLEEPFAIL bug. In order to
>>   cope with it, add the tracking of how many exclusive LK_SLEEPFAIL waiters
>>   a lockmgr has and if all the waiters on the exclusive waiters queue are
>>   LK_SLEEPFAIL just wake both queues.
>>
>>   The sleepq_sleepcnt() introduction and ABI breakage require
>>   __FreeBSD_version bumping.
>
> Hmm, do you need an actual count of waiters or would a 'sleepq_empty()'
> (similar to turnstile_empty()) method be sufficient?

I need the count in order to fix properly LK_SLEEPFAIL case (the idea
is: track exclusive waiters with LK_SLEEPFAIL on; if the number is
equal to the actual sleepers on the queue then wake up both queues,
otherwise nobody is going to take care of the shared waiters queue).

Attilio


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