[llvm] [AMDGPU] Add new llvm.amdgcn.wave.shuffle intrinsic (PR #167372)

Nicolai Hähnle via llvm-commits llvm-commits at lists.llvm.org
Thu Dec 4 10:50:35 PST 2025


================
@@ -1894,6 +1894,10 @@ class GCNSubtarget final : public AMDGPUGenSubtargetInfo,
   bool requiresWaitsBeforeSystemScopeStores() const {
     return RequiresWaitsBeforeSystemScopeStores;
   }
+
+  bool supportsWaveWideBPermute() const {
+    return getGeneration() == AMDGPUSubtarget::GFX12 || isWave32();
+  }
----------------
nhaehnle wrote:

This is also true for gfx9 and earlier.

https://github.com/llvm/llvm-project/pull/167372


More information about the llvm-commits mailing list