[Mlir-commits] [mlir] [MLIR][XEGPU] Add blocking support for scatter ops (PR #144766)

[Jianhui Li]

================
@@ -355,4 +350,100 @@ gpu.module @kernel  attributes {spirv.target_env = #spirv.target_env<#spirv.vce<
     xegpu.store_nd %2, %3: vector<8x32xf32>, !xegpu.tensor_desc<8x32xf32, #t>
     gpu.return
   }
-}
\ No newline at end of file
+}
+
+// -----
+gpu.module @test_kernel {
----------------
Jianhui-Li wrote:

The test doesn't require gpu op, but currently almost all test file uses this style.  We may consider changing it for all test cases in a separate PR.  

https://github.com/llvm/llvm-project/pull/144766