Re: [Intel-gfx] [PATCH v2] drm/i915: Prune gen8_gt_irq_handler

2018-02-19 Thread Chris Wilson
Quoting Mika Kuoppala (2018-02-19 13:59:43) > Chris Wilson writes: > > > The compiler is not automatically caching the i915->regs address inside > > a register and emitting a load for every mmio access. For simple > > functions like gen8_gt_irq_handler that are already using the raw > > accessors

Re: [Intel-gfx] [PATCH v2] drm/i915: Prune gen8_gt_irq_handler

2018-02-19 Thread Mika Kuoppala
Chris Wilson writes: > The compiler is not automatically caching the i915->regs address inside > a register and emitting a load for every mmio access. For simple > functions like gen8_gt_irq_handler that are already using the raw > accessors, we can open-code them for substantial savings: > > add

[Intel-gfx] [PATCH v2] drm/i915: Prune gen8_gt_irq_handler

2018-02-19 Thread Chris Wilson
The compiler is not automatically caching the i915->regs address inside a register and emitting a load for every mmio access. For simple functions like gen8_gt_irq_handler that are already using the raw accessors, we can open-code them for substantial savings: add/remove: 0/0 grow/shrink: 0/2 up/d