<html> <head> <base href="https://bugs.freedesktop.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - 4-27% performance drop in Vulkan benchmarks" href="https://bugs.freedesktop.org/show_bug.cgi?id=109216#c9">Comment # 9</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - 4-27% performance drop in Vulkan benchmarks" href="https://bugs.freedesktop.org/show_bug.cgi?id=109216">bug 109216</a> from <a class="email" href="mailto:idr@freedesktop.org" title="Ian Romanick <idr@freedesktop.org>"> Ian Romanick</a> <pre>(In reply to Eero Tamminen from <a href="show_bug.cgi?id=109216#c5">comment #5</a>) > (In reply to Lionel Landwerlin from <a href="show_bug.cgi?id=109216#c4">comment #4</a>) > > commit 8fb8ebfbb05d3323481c8ba6d320b3a3580bad99 > > Author: Ian Romanick <<a href="mailto:ian.d.romanick@intel.com">ian.d.romanick@intel.com</a>> > > Date: Tue May 22 18:56:41 2018 -0700 > > > > intel/compiler: More peephole select > > > > Shader-db results: > > > > The one shader hurt for instructions is a compute shader that had both > > spills and fills hurt. > > I think shaders in Sacha Willems' Vulkan demos would also need to be added > to shader-db. According to INTEL_DEBUG output, Deferred Multisampling demo > doesn't have any compute shaders or spills/fills, and it regressed most. Adding the GLSL shader won't actually help this problem. We'd have to add the SPIR-V instead. The issue is the weird way that glslang generates for-loops. I have a test case that reproduces the issue in plain GLSL.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>