[PATCH] D68873: [AMDGPU] Amend target loop unroll defaults

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

Could use a test



================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp:93
                                             TTI::UnrollingPreferences &UP) {
   UP.Threshold = 300; // Twice the default.
   UP.MaxCount = std::numeric_limits<unsigned>::max();
----------------
This would now be dead


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp:100
+  // Set more aggressive defaults for PAL shaders
+  if (TargetTriple.getOS() == Triple::AMDPAL) {
+    UP.MaxPercentThresholdBoost = 1000;
----------------
These should probably be the same for all OSes


================
Comment at: llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h:52
 
+  AMDGPUSubtarget::Generation Gen;
+
----------------
You don't need to add this field. You already have the subtarget available here, you just need to change the type


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D68873/new/

https://reviews.llvm.org/D68873