On 11/10/2017 5:15 PM, Chris Wilson wrote:
Quoting Patchwork (2017-11-11 01:03:20)
== Series Details ==
Series: series starting with [1/2] drm/i915: Clear per-engine fault register as early as possible
URL : https://patchwork.freedesktop.org/series/33649/
State : success
BAT results arrived before patches, yay!
Patch 1 looks ok for the minimal fix. I think we want to juggle around
the suspend/resume i915_check_and_clear_faults() some more, but for now
that looks fine. R-b.
Patch 2 still has a lack of statics on the non-exported functions. But
the names are indeed clearer. I would make the if-else-chain:
if (GEN >= 8)
gen8_check_and_clear_faults();
else if (GEN >= 6)
gen6_check_and_clear_faults();
else
return;
You also want to move the POSTING_READ since it is still chasing the
wrong register on gen8, no?
I left the POSTING_READ because it's the RCS reg, which is same as gen8,
but then the comment above it is no longer valid anyway (the first patch
is addressing that)... v3 soon.
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
