[haiku-development] Re: Res: Scheduler algorithm improvements

  • From: Fredrik Holmqvist <fredrik.holmqvist@xxxxxxxxx>
  • To: haiku-development@xxxxxxxxxxxxx
  • Date: Tue, 10 Nov 2009 21:14:28 +0100

2009/11/7 Axel Dörfler <axeld@xxxxxxxxxxxxxxxx>:
> A panic will point you to the "evildoer" anyway. Not sure there is a
> need for more.

Tried out the spinlock debugging, and while it panicked once during
upstart (in test_latency_FPV1) it didn't while playing audio or
running timingtest. Timingtest still said the worst jitter was 2457us,
while most was between 6-90us. I'll see if I can get
scheduling_recorder working.

/Fredrik Holmqvist, TQH

Other related posts: