[Mlir-commits] [mlir] [mlir][xegpu] Relax rank restriction of TensorDescType (PR #145916)

[Chao Chen]

================
@@ -446,4 +446,115 @@ gpu.module @test_kernel   {
   }
 }
 
+// -----
+#l = #xegpu.layout<inst_data = [8,32,16]>
+gpu.module @test_kernel {
+  // CHECK-LABEL: test_3d_block_tensor_desc
+  // CHECK-SAME: [[arg0:%.+]]: memref<1024x1024x1024xf16>, [[arg1:%.+]]: memref<1024x1024x1024xf16>, [[arg2:%.+]]: memref<1024x1024x1024xf16>
+  gpu.func @test_3d_block_tensor_desc(%A: memref<1024x1024x1024xf16>, %B: memref<1024x1024x1024xf16>, %C: memref<1024x1024x1024xf16>) {
+    //CHECK: [[c24:%.*]] = arith.constant 24 : index
+    //CHECK: [[c8:%.*]] = arith.constant 8 : index
+    //CHECK: [[c16:%.*]] = arith.constant 16 : index
+    //CHECK: [[c0:%.*]] = arith.constant 0 : index
+    //CHECK: [[c32:%.*]] = arith.constant 32 : index
+    %c0 = arith.constant 0 : index
+    %c32 = arith.constant 32 : index
+    %c1024 = arith.constant 1024 : index
+
+    //CHECK: [[block_id_x:%.*]] = gpu.block_id  x
+    //CHECK: [[m:%.*]] = arith.muli [[block_id_x]], [[c32]] : index
+    %block_id_x = gpu.block_id x
+    %m = arith.muli %block_id_x, %c32 : index
----------------
chencha3 wrote:

good catch. fixed it

https://github.com/llvm/llvm-project/pull/145916