[Mlir-commits] [clang] [llvm] [mlir] [OpenMP] Reserve the main thread's warp for generic mode kernels (PR #218790)

[Joseph Huber]

================
@@ -199,6 +200,17 @@ static const omp::GV &getGridValue(const Triple &T, Function *Kernel) {
         Kernel->getFnAttribute("target-features").getValueAsString();
     if (Features.count("+wavefrontsize64"))
       return omp::getAMDGPUGridValues<64>();
+    if (Features.count("+wavefrontsize32"))
+      return omp::getAMDGPUGridValues<32>();
+
+    // Clang sets no wavefront size on OpenMP device kernels, so ask the CPU.
+    StringRef CPU = Kernel->getFnAttribute("target-cpu").getValueAsString();
+    AMDGPU::GPUKind Kind = AMDGPU::parseArchAMDGCN(CPU);
+    if (Kind == AMDGPU::GK_NONE)
+      Kind = AMDGPU::getGPUKindFromSubArch(T.getSubArch());
+    if (Kind != AMDGPU::GK_NONE &&
+        !AMDGPU::getFeatureBitset(Kind).test(AMDGPU::FEAT_SUPPORTS_WAVE32))
+      return omp::getAMDGPUGridValues<64>();
     return omp::getAMDGPUGridValues<32>();
----------------
jhuber6 wrote:

I'm not entirely convinced that this PR couldn't just be three lines of code in the offload runtime to make sure we widen the launch bounds correctly.

https://github.com/llvm/llvm-project/pull/218790