Re: [PATCH 1/2] drm/i915: Prepare for multiple GTs

2022-01-12 Thread Andi Shyti
Hi Dale, thanks for looking into this patch, > > + /* > > +* i915->gt[0] == &i915->gt0 > > +*/ > > +#define I915_MAX_GT 4 > > + struct intel_gt *gt[I915_MAX_GT]; > > + > > > It would be nice if I915_MAX_GT was defined in a more basic header file so > that the definition of I915_MAX_

Re: [PATCH 1/2] drm/i915: Prepare for multiple GTs

2022-01-11 Thread Stimson, Dale B
Hi Andi, On 2022-01-11 14:15:51, Andi Shyti wrote: > > From: Tvrtko Ursulin > > On a multi-tile platform, each tile has its own registers + GGTT > space, and BAR 0 is extended to cover all of them. > > Up to four gts are supported in i915->gt[], with slot zero > shadowing the existing i915->gt

[PATCH 1/2] drm/i915: Prepare for multiple GTs

2022-01-11 Thread Andi Shyti
From: Tvrtko Ursulin On a multi-tile platform, each tile has its own registers + GGTT space, and BAR 0 is extended to cover all of them. Up to four gts are supported in i915->gt[], with slot zero shadowing the existing i915->gt0 to enable source compatibility with legacy driver paths. A for_each