On Sat, Aug 19, 2017 at 11:07:35AM -0700, Jason Ekstrand wrote: > In 76e2f390f9863a35, when Topi switched num_samples from 0 to 1 for > single-sampled, he accidentally switched the last parameter in the call > to miptree_create_for_teximage from 0 to 1 thinking it was num_samples > when it was actually layout_flags. Switching from 0 to 1 added the > MIPTREE_LAYOUT_ACCELERATED_UPLOAD flag which causes us to allocate a > busy BO instead of an idle one. This caused the subsequent CPU upload > to consistently stall. The end result was a 15% performance drop in the > SynMark v7 DrvRes microbenchmark. This restores the old behavior and > fixes the performance regression.
Oh man :(, huge thanks for tracking this down: Reviewed-by: Topi Pohjolainen <topi.pohjolai...@intel.com> > > Cc: Kenneth Graunke <kenn...@whitecape.org> > Cc: Topi Pohjolainen <topi.pohjolai...@intel.com> > Fixes: 76e2f390f9863a356d1419982dec705260d67eff > Bugzilla: https://bugs.freedesktop.org/102260 > Cc: mesa-sta...@lists.freedesktop.org > --- > src/mesa/drivers/dri/i965/intel_tex.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/src/mesa/drivers/dri/i965/intel_tex.c > b/src/mesa/drivers/dri/i965/intel_tex.c > index 890c82d..94a7ad3 100644 > --- a/src/mesa/drivers/dri/i965/intel_tex.c > +++ b/src/mesa/drivers/dri/i965/intel_tex.c > @@ -94,7 +94,7 @@ intel_alloc_texture_image_buffer(struct gl_context *ctx, > } else { > intel_image->mt = intel_miptree_create_for_teximage(brw, intel_texobj, > intel_image, > - 1 /* samples */); > + > MIPTREE_CREATE_DEFAULT); > if (!intel_image->mt) > return false; > > -- > 2.5.0.400.gff86faf > > _______________________________________________ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/mesa-dev _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
