[Mlir-commits] [mlir] [MLIR][NVVM] Update dot.accumulate NVVM Ops (PR #140518)

Durgadoss R llvmlistbot at llvm.org
Fri May 23 01:26:46 PDT 2025


================
@@ -3533,35 +3533,35 @@ def NVVM_Tcgen05StOp : NVVM_Op<"tcgen05.st", [NVVMRequiresSMa<[100, 101]>]> {
 }
 
 //===----------------------------------------------------------------------===//
-// NVVM dot.accumulate.4way Op
+// NVVM dot.accumulate Ops
 //===----------------------------------------------------------------------===//
 
-def DotAccumulate4WayS8 : I32EnumAttrCase<"S8", 1, "s8">;
-def DotAccumulate4WayU8 : I32EnumAttrCase<"U8", 0, "u8">;
+def DotAccumulateSigned : I32EnumAttrCase<"SIGNED", 1, "signed">;
+def DotAccumulateUnsigned : I32EnumAttrCase<"UNSIGNED", 0, "unsigned">;
 
-def DotAccumulate4WayType : I32EnumAttr<"DotAccumulate4WayType",
-                              "NVVM DotAccumulate4WayType",
-                              [DotAccumulate4WayS8, DotAccumulate4WayU8]> {
+def DotAccumulateType : I32EnumAttr<"DotAccumulateType",
+                              "NVVM DotAccumulateType",
+                              [DotAccumulateSigned, DotAccumulateUnsigned]> {
   let cppNamespace = "::mlir::NVVM";
   let genSpecializedAttr = 0;
 }
 
-def DotAccumulate4WayTypeAttr : EnumAttr<NVVM_Dialect, DotAccumulate4WayType, "dot_accumulate_4way_type"> {
+def DotAccumulateTypeAttr : EnumAttr<NVVM_Dialect, DotAccumulateType, "dot_accumulate_type"> {
   let assemblyFormat = "`<` $value `>`";
 }
 
 def NVVM_DotAccumulate4WayOp : NVVM_Op<"dot.accumulate.4way"> {
-  let summary = "Four-way byte dot product-accumulate instruction.";
+  let summary = "Four-way byte dot product-accumulate instruction";
   let description = [{
     Performs a four-way byte dot-product which is accumulated in a 32-bit
     result.
     Operand `a` and `b` are vectors of 4 bytes between which the dot product is 
     computed.
     The `a_type` and `b_type` attributes specify the type of the elements in `a`
     and `b` respectively.
-    If `a_type` or `b_type` is `s8`, then the elements in the corresponding 
+    If `a_type` or `b_type` is `s`, then the elements in the corresponding 
     vector are sign-extended to 32-bit before the dot product is computed.
-    If `a_type` or `b_type` is `u8`, then the elements in the corresponding 
+    If `a_type` or `b_type` is `u`, then the elements in the corresponding 
----------------
durga4github wrote:

nit: rename these too

https://github.com/llvm/llvm-project/pull/140518


More information about the Mlir-commits mailing list