Quoting Tvrtko Ursulin (2019-07-03 12:26:36) > > On 03/07/2019 10:17, Chris Wilson wrote: > > We don't care about the result of the read, so it may be garbage, we > > only care that the mmio is flushed. As such, we can forgo using an > > individual forcewake and lock around any posting-read for an engine. > > > > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > > --- > > drivers/gpu/drm/i915/gt/intel_engine.h | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h > > index 557b08b13feb..0331e9ac2485 100644 > > --- a/drivers/gpu/drm/i915/gt/intel_engine.h > > +++ b/drivers/gpu/drm/i915/gt/intel_engine.h > > @@ -51,7 +51,7 @@ struct drm_printer; > > #define ENGINE_READ16(...) __ENGINE_READ_OP(read16, __VA_ARGS__) > > #define ENGINE_READ(...) __ENGINE_READ_OP(read, __VA_ARGS__) > > #define ENGINE_READ_FW(...) __ENGINE_READ_OP(read_fw, __VA_ARGS__) > > -#define ENGINE_POSTING_READ(...) __ENGINE_READ_OP(posting_read, __VA_ARGS__) > > +#define ENGINE_POSTING_READ(...) __ENGINE_READ_OP(posting_read_fw, __VA_ARGS__) > > #define ENGINE_POSTING_READ16(...) __ENGINE_READ_OP(posting_read16, __VA_ARGS__) > > > > #define ENGINE_READ64(engine__, lower_reg__, upper_reg__) \ > > > > Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx> > > Can we apply this to all posting reads? (intel_uncore_posting_read*) I briefly considered it, but was too lazy to think beyond the current set. What gave me pause for concern was intel_gt_flush_ggtt_writes() where we have to worry about overlapping mmio access causing gen7 to explode, generalising the lock drop is tricky. -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
