diff options
| author | Jonathan Gray <jsg@jsg.id.au> | 2013-03-06 13:12:45 +1100 |
|---|---|---|
| committer | Jonathan Gray <jsg@jsg.id.au> | 2013-03-06 13:12:45 +1100 |
| commit | 3a5087a161faa2a1335407fe9e6bfce8b9c2fdb8 (patch) | |
| tree | 3bec11f9d34220c560a1f5b449f5e0eede84301c /sys | |
| parent | 70063991b622074c5a5d395c3fc3057ac10818dc (diff) | |
unconditionally do the workaround to force correct ordering on gen6+
seems to fix the laggyness with firefox (but not the hangs)
Diffstat (limited to 'sys')
| -rw-r--r-- | sys/dev/pci/drm/intel_ringbuffer.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/sys/dev/pci/drm/intel_ringbuffer.c b/sys/dev/pci/drm/intel_ringbuffer.c index bfee793a71f..90ad07425ee 100644 --- a/sys/dev/pci/drm/intel_ringbuffer.c +++ b/sys/dev/pci/drm/intel_ringbuffer.c @@ -801,7 +801,10 @@ gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency) /* Workaround to force correct ordering between irq and seqno writes on * ivb (and maybe also on snb) by reading from a CS register (like * ACTHD) before reading the status page. */ +#ifdef notyet + /* XXX conditionally doing this seems to make everything laggy */ if (!lazy_coherency) +#endif intel_ring_get_active_head(ring); return intel_read_status_page(ring, I915_GEM_HWS_INDEX); } |