Re: [Intel-gfx] [PATCH] drm/i915: Kill the active list spinlock
On Wed, 4 Aug 2010 14:09:45 +0100, Chris Wilson wrote: > This spinlock only served debugging purposes in a time when we could not > be sure of the mutex ever being released upon a GPU hang. As we now > should be able rely on hangcheck to do the job for us (and that error > reporting should not itself require the struct mutex) we can kill the > incomplete and misleading attempt at protection. > > Signed-off-by: Chris Wilson So, when am I getting hangcheck resets on Ironlake? Hmm? But yeah, this was a hack during initial bringup of GEM and I'll be glad to see it go away... once you rebase so it doesn't conflict. pgpVq8hin3dS5.pgp Description: PGP signature ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH] drm/i915: Kill the active list spinlock
On Fri, 06 Aug 2010 14:41:02 -0700, Eric Anholt wrote: > On Wed, 4 Aug 2010 14:09:45 +0100, Chris Wilson > wrote: > > This spinlock only served debugging purposes in a time when we could not > > be sure of the mutex ever being released upon a GPU hang. As we now > > should be able rely on hangcheck to do the job for us (and that error > > reporting should not itself require the struct mutex) we can kill the > > incomplete and misleading attempt at protection. > > > > Signed-off-by: Chris Wilson > > So, when am I getting hangcheck resets on Ironlake? Hmm? I suppose you'll want a more sophisticated watchdog using GPU timers, as well ;-) Yes, we should sort out the reset bits for ILK and the rest. -- Chris Wilson, Intel Open Source Technology Centre ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx