On Wed, Nov 17, 2021 at 02:16:13PM +0000, Chris Wilson wrote:
Quoting Andi Shyti (2021-11-17 13:34:56)
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index a9727447c0379..4bfedc04f5c70 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -936,7 +936,7 @@ hsw_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) static void gen9_wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal) {
const struct sseu_dev_info *sseu = &i915->gt.info.sseu;
const struct sseu_dev_info *sseu = &i915->gt0.info.sseu;This feels like it should be pulling from uncore->gt, since the MCR is across an uncore.
Overall though, rather than introduce bare &i915->gt0, how about pulling in the patch for to_gt(i915)?
agreed, why are we going a different route rather than using the to_gt() patches Michal prepared? If the to_gt() is not what we want, then please sync with Michal on the proper direction
Lucas De Marchi