[llvm] [AMDGPU] Change default loop alignment for GFX9 and higher targets (PR #153065)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Mon Aug 11 20:19:51 PDT 2025


================
@@ -377,6 +378,8 @@ class AMDGPUSubtarget {
   uint64_t getExplicitKernArgSize(const Function &F, Align &MaxAlign) const;
   unsigned getKernArgSegmentSize(const Function &F, Align &MaxAlign) const;
 
+  unsigned getPrefLoopAlignment() const { return PrefLoopAlignmentLog2; }
----------------
arsenm wrote:

I'm more wondering if this is a universal property that should have always been done. The prior comment says there's no benefit pre-gfx10, so why is this now needed for gfx9?

https://github.com/llvm/llvm-project/pull/153065


More information about the llvm-commits mailing list