Fix error: enumeration value 'OperandC' not handled in switch

When building with gcc 13 and clang 14 the following warning, that is treated as error, occurs: error: enumeration value 'OperandC' not handled in switch [-Werror,-Wswitch]
intel · Jan 23, 2025 · fd4c1b6 · fd4c1b6
1 parent fe29adf
commit fd4c1b6
Show file tree

Hide file tree

Showing 3 changed files with 23 additions and 11 deletions.
diff --git a/third_party/intel/lib/Dialect/TritonIntelGPU/IR/Dialect.cpp b/third_party/intel/lib/Dialect/TritonIntelGPU/IR/Dialect.cpp
@@ -245,18 +245,21 @@ DpasEncodingAttr::getDPASRepetitions(ArrayRef<int64_t> shape,
             std::max<int64_t>(1, shape[rank - 1] / (shapePerWarp[rank - 1] *
                                                     warpsPerCTA[rank - 1]))};
   } break;
+  case OpIdx::OperandC: {
+    auto shapePerWarp = getShapeC();
+    int64_t numRepBatch =
+        rank == 3 ? std::max<int64_t>(1, shape[0] /
+                                             (shapePerWarp[0] * warpsPerCTA[0]))
+                  : 1;
+    return {numRepBatch,
+            std::max<int64_t>(1, shape[rank - 2] / (shapePerWarp[rank - 2] *
+                                                    warpsPerCTA[rank - 2])),
+            std::max<int64_t>(1, shape[rank - 1] / (shapePerWarp[rank - 1] *
+                                                    warpsPerCTA[rank - 1]))};
+  } break;
   }
 
-  auto shapePerWarp = getShapeC();
-  int64_t numRepBatch =
-      rank == 3
-          ? std::max<int64_t>(1, shape[0] / (shapePerWarp[0] * warpsPerCTA[0]))
-          : 1;
-  return {numRepBatch,
-          std::max<int64_t>(1, shape[rank - 2] / (shapePerWarp[rank - 2] *
-                                                  warpsPerCTA[rank - 2])),
-          std::max<int64_t>(1, shape[rank - 1] / (shapePerWarp[rank - 1] *
-                                                  warpsPerCTA[rank - 1]))};
+  llvm_unreachable("unexpected opIdx");
 }
 
 unsigned DpasEncodingAttr::getTotalElemsPerThreadForOperand(
@@ -279,6 +282,8 @@ unsigned DpasEncodingAttr::getTotalElemsPerThreadForOperand(
     // dpas operands scalar are evenly sharded to each work item.
     return (totalElem / threadsPerWar) * product<int64_t>(rep);
   } break;
+  case OpIdx::OperandC:
+    llvm_unreachable("unexpected OperandC");
   }
   llvm_unreachable("unexpected opIdx");
 }
@@ -350,8 +355,9 @@ DpasEncodingAttr::getSizePerThreadForOperand(int kWidth, OpIdx opIdx) const {
     return {shapeB[rank - 2] / threadsPerWarp[0],
             shapeB[rank - 1] / threadsPerWarp[1] * repCluster[rank - 1]};
   } break;
+  default:
+    llvm_unreachable("unexpected opIdx");
   }
-  llvm_unreachable("unexpected opIdx");
 }
 
 SmallVector<unsigned> DpasEncodingAttr::getContigPerThread() const {

diff --git a/third_party/intel/lib/TritonIntelGPUToLLVM/ConvertLayoutOpToLLVM.cpp b/third_party/intel/lib/TritonIntelGPUToLLVM/ConvertLayoutOpToLLVM.cpp
@@ -397,6 +397,8 @@ struct ConvertLayoutOpConversion
       repInner = repetitions[1];
       repClusterOuter = repCluster[rank - 1];
     } break;
+    default:
+      llvm_unreachable("unexpected opIdx");
     }
 
     // TODO: Operands B requires extra steps to combine [8, 16] to [16, 16].

diff --git a/third_party/intel/lib/TritonIntelGPUToLLVM/LoadStoreOpToLLVM.cpp b/third_party/intel/lib/TritonIntelGPUToLLVM/LoadStoreOpToLLVM.cpp
@@ -861,6 +861,8 @@ struct LoadOpConversion
                           i32_val(outer * repOuterStride + rep * repStride));
             offsetY = i32_val(k * repKStride);
           } break;
+          default:
+            llvm_unreachable("unexpected opIdx");
           }
 
           offsetX = add(offsetX, offsetBaseX);
@@ -942,6 +944,8 @@ struct LoadOpConversion
                             k + row}] =
                       bitcast(loadVal, unpackedDPASOperandType);
                 } break;
+                default:
+                  llvm_unreachable("unexpected opIdx");
                 }
               }
         }