Noticable Delays Since Beta 3 (possible cause)
davidxu at freebsd.org
Mon Oct 11 07:21:41 PDT 2004
Sam Lawrance wrote:
>On Mon, 2004-10-11 at 04:55 -0400, Robert Watson wrote:
>>On Sat, 9 Oct 2004, Sam Lawrance wrote:
>>>I'm sorry, that patch was missing paths.. this one:
>>>is much nicer.
>>Are there plans to get this patch merged for 5.3? I was chatting with
>>Brian yesterday about the accept locking issue in sofree() and he was
>>indicating that he was seeing symptoms very much like the ones described
>>in this thread (long wakeup times for keypresses, etc).
>I just noticed that the messages David sent to me and the list have yet
>to make it through to the list. One was a patch which puts the wakeup in
>the critical_exit where he suggested (but I didn't understand :). Here
Yes, I made the patch, the theory is holding spinlock increases critical
releasing a spinlock decreases it, when critical count is about to be
to zero, I know curthread has no spinlock in hand, so it is not possible
sleepq lock and sched lock LOR problem if I do a wakupe(&proc0) at the time.
I believe that it is a very short time that a thread is in critical
the time we have to swap in a thread by doing disk I/O which is slow,
of wakeup proc0 in spinlock can be ignored. I am going to sleep, and
now, if anyone has time to commit it, free to do.
More information about the freebsd-current