On Sat, Sep 05, 2015 at 11:30:44AM -0700, Jordan Justen wrote: > From: Francisco Jerez <[email protected]> > > Fixes > arb_shader_image_load_store/execution/load-from-cleared-image.shader_test > > Cc: Chris Wilson <[email protected]> > Cc: Jason Ekstrand <[email protected]> > Tested-by: Jordan Justen <[email protected]> > --- > RE: i965: Perform an explicit flush after doing _mesa_meta_pbo_TexSubImage > > curro has some concerns about potential perf impact by this and > wanted it to be checked on small-core w/CPU bound apps. > Unfortunately, he is on vacation now.
One thing that may help (other than having a bitfield of valid ImageUnits being more cache friendly) is that intel_update_state() need only invalidate state. Moving the heavyweight resolves/flushes to just prior to running the pipeline should help reduce the profile of the fairly frequently called mesa_update_state(). -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ mesa-dev mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/mesa-dev
