[Mlir-commits] [mlir] [MLIR][NVVM] Add support for nvvm.breakpoint Op (PR #107193)

[Pradeep Kumar]

================
@@ -2101,6 +2101,23 @@ def NVVM_WgmmaMmaAsyncOp : NVVM_Op<"wgmma.mma_async",
   }];
 }
 
+//===----------------------------------------------------------------------===//
+// NVVM breakpoint Op
+//===----------------------------------------------------------------------===//
+
+def NVVM_Breakpoint : NVVM_Op<"breakpoint"> {
----------------
schwarzschild-radius wrote:

Because it maps to the PTX `breakpoint` instruction (http://sw-mobile-docs/CUDA/GPGPU/parallel-thread-execution/index.html#miscellaneous-instructions-brkpt) and `debugtrap` looked ambigous given all other Ops maps 1-to-1 to PTX instructions

https://github.com/llvm/llvm-project/pull/107193