[Mlir-commits] [mlir] [mlir] Added `Convergent` trait that matches LLVM's semantics (PR #152358)

[Guray Ozen]

================
@@ -570,8 +572,9 @@ def NVVM_Barrier0Op : NVVM_Op<"barrier0"> {
   }];
 }
 
-def NVVM_BarrierOp : NVVM_Op<"barrier", [AttrSizedOperandSegments]> {
-  let arguments = (ins     
+def NVVM_BarrierOp : NVVM_Op<"barrier",
+  [Convergent, AttrSizedOperandSegments]> {
----------------
grypp wrote:

This instruction can synchronize different thread counts. Could you clarify what `convergent` refers to in this context? I may be missing the nuance—does it mean a convergent warp or a convergent CTA?

```
nvvm.barrier 32
nvvm.barrier 128
```


https://github.com/llvm/llvm-project/pull/152358