Hi Jonathan,
On Fri, Jul 21, 2023 at 07:05:58AM -0700, Jonathan Cavitt wrote:
> Refactor i915_coherent_map_type to be GT-centric rather than
> device-centric. Each GT may require different coherency
> handling due to hardware workarounds.
[...]
> -enum i915_map_type i915_coherent_map_type(struct drm_i915_private *i915,
> +enum i915_map_type i915_coherent_map_type(struct intel_gt *gt,
> struct drm_i915_gem_object *obj,
> bool always_coherent)
> {
> /*
> * Wa_22016122933: always return I915_MAP_WC for MTL
> */
> - if (i915_gem_object_is_lmem(obj) || IS_METEORLAKE(i915))
> + if (i915_gem_object_is_lmem(obj) || IS_METEORLAKE(gt->i915))
> return I915_MAP_WC;
> - if (HAS_LLC(i915) || always_coherent)
> + if (HAS_LLC(gt->i915) || always_coherent)
> return I915_MAP_WB;
> else
> return I915_MAP_WC;
this doesn't fully look right to me as gt-centric stuff need to
be in the gt/ directory.
As for this patch, here we don't need any reference to the gt.
You could eventually add some wrapper inside gt/ that perform the
per 'gt' checks and at the end call this function here.
Andi
