[Mlir-commits] [mlir] [MLIR] Adding XeGPU 2d block operators (PR #84692)

Chao Chen llvmlistbot at llvm.org
Tue Mar 12 09:30:54 PDT 2024


================
@@ -30,4 +30,106 @@ class XeGPUTypeDef<string name, string typeMnemonic, list<Trait> traits = [],
   let mnemonic = typeMnemonic;
 }
 
+def XeGPU_TensorDesc: XeGPUTypeDef<"TensorDesc", "tensor_desc",
+        [ShapedTypeInterface], "::mlir::TensorType"> {
+  let summary = "TensorDesc describing regions of interested data.";
+  let description = [{
+    TensorDesc is a type designed to describe regions of the interested data as well as some 
+    features that are unique to Intel hardware. Different with the builtin tensor type in MLIR, 
+    it essentially only contains the meta data, and doesn't hold the data by itself. It is designed 
+    to mainly support 2d block load/store and DPAS (matrix multiplication instruction) on Intel GPU. 
+    It encodes the following information:
+
+    * shape:  the sizes/shape of the intereted data block, e.g., 8x16 means 8 rows
+              and each row contains 16 continious data element. The rows could be
+              either continuous or not, depends on whether the encoding attribute
----------------
chencha3 wrote:

@adam-smnk Thanks, they are fixed now.

https://github.com/llvm/llvm-project/pull/84692


More information about the Mlir-commits mailing list