On Thu, 2015-12-03 at 11:04 -0800, Matt Turner wrote: > > The improvement obtained regarding current upstream (56aff6bb4eaf) > is: > > > > total instructions in shared programs: 6819484 -> 6811698 (-0.11%) > > instructions in affected programs: 387245 -> 379459 (-2.01%) > > total loops in shared programs: 1971 -> 1971 (0.00%) > > helped: 3980 > > HURT: 0 > > GAINED: 3 > > Run the shader-db on just the program that had the problem here, add > it to your before-results, and rerun report.py.
I've rebased against new master, applied the changes you suggested, and run again the tests. That has been enough to get rid of the GAINED shaders. New results are: total instructions in shared programs: 6819828 -> 6812042 (-0.11%) instructions in affected programs: 387245 -> 379459 (-2.01%) total loops in shared programs: 1971 -> 1971 (0.00%) helped: 3980 HURT: 0 GAINED: 0 LOST: 0 I'll send a second version of the patch. J.A. _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev