From: Ian Romanick <ian.d.roman...@intel.com> No shader-db changes on any Gen6+ platform.
All of the shaders with cycles hurt by more than ~2% are from Master of Orion. All of the shaders have instructions helped. It looks like the pass enables some control flow to be converted to bcsels, then the scheduler does dumb things. These are new shaders (just added before doing this shader-db run), so there's probably some low-hanging fruit. Iron Lake total instructions in shared programs: 8207949 -> 8207310 (<.01%) instructions in affected programs: 84465 -> 83826 (-0.76%) helped: 114 HURT: 26 helped stats (abs) min: 2 max: 18 x̄: 7.71 x̃: 9 helped stats (rel) min: 0.17% max: 13.73% x̄: 2.41% x̃: 1.05% HURT stats (abs) min: 2 max: 20 x̄: 9.23 x̃: 8 HURT stats (rel) min: 0.70% max: 2.48% x̄: 1.66% x̃: 1.61% 95% mean confidence interval for instructions value: -5.84 -3.29 95% mean confidence interval for instructions %-change: -2.19% -1.12% Instructions are helped. total cycles in shared programs: 187478874 -> 187491364 (<.01%) cycles in affected programs: 506734 -> 519224 (2.46%) helped: 100 HURT: 40 helped stats (abs) min: 2 max: 76 x̄: 22.68 x̃: 16 helped stats (rel) min: 0.02% max: 6.16% x̄: 0.87% x̃: 0.63% HURT stats (abs) min: 4 max: 1402 x̄: 368.95 x̃: 30 HURT stats (rel) min: 0.33% max: 23.12% x̄: 5.36% x̃: 1.43% 95% mean confidence interval for cycles value: 28.54 149.89 95% mean confidence interval for cycles %-change: 0.09% 1.74% Cycles are HURT. GM45 total instructions in shared programs: 5047454 -> 5047096 (<.01%) instructions in affected programs: 46751 -> 46393 (-0.77%) helped: 63 HURT: 13 helped stats (abs) min: 2 max: 29 x̄: 7.59 x̃: 9 helped stats (rel) min: 0.17% max: 13.73% x̄: 2.74% x̃: 1.04% HURT stats (abs) min: 2 max: 20 x̄: 9.23 x̃: 8 HURT stats (rel) min: 0.66% max: 2.35% x̄: 1.58% x̃: 1.52% 95% mean confidence interval for instructions value: -6.49 -2.93 95% mean confidence interval for instructions %-change: -2.80% -1.19% Instructions are helped. total cycles in shared programs: 128068304 -> 128075484 (<.01%) cycles in affected programs: 324560 -> 331740 (2.21%) helped: 53 HURT: 23 helped stats (abs) min: 6 max: 76 x̄: 32.57 x̃: 32 helped stats (rel) min: 0.08% max: 4.74% x̄: 1.04% x̃: 0.79% HURT stats (abs) min: 4 max: 1400 x̄: 387.22 x̃: 40 HURT stats (rel) min: 0.56% max: 19.94% x̄: 5.05% x̃: 1.49% 95% mean confidence interval for cycles value: 7.43 181.52 95% mean confidence interval for cycles %-change: -0.23% 1.83% Inconclusive result (%-change mean confidence interval includes 0). Signed-off-by: Ian Romanick <ian.d.roman...@intel.com> --- src/intel/compiler/brw_nir.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/intel/compiler/brw_nir.c b/src/intel/compiler/brw_nir.c index 04d399fd0b4..2eba6a42266 100644 --- a/src/intel/compiler/brw_nir.c +++ b/src/intel/compiler/brw_nir.c @@ -588,8 +588,8 @@ brw_nir_optimize(nir_shader *nir, const struct brw_compiler *compiler, (nir->info.stage == MESA_SHADER_TESS_CTRL || nir->info.stage == MESA_SHADER_TESS_EVAL); OPT(nir_opt_peephole_select, 0, is_vec4_tessellation, false); - if (compiler->devinfo->gen >= 6) - OPT(nir_opt_peephole_select, 1, is_vec4_tessellation, true); + OPT(nir_opt_peephole_select, 1, is_vec4_tessellation, + compiler->devinfo->gen >= 6); OPT(nir_opt_intrinsics); OPT(nir_opt_algebraic); -- 2.14.4 _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev