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

Charitha Saumya llvmlistbot at llvm.org
Wed Jun 18 11:54:22 PDT 2025


================
@@ -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 {
----------------
charithaintc wrote:

no need for a gpu.module or gpu.func for testing purposes. we just use func.func. we use gpu.module if some gpu op is necessary  for the test case (like thread ids etc). is that the case here? 

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


More information about the Mlir-commits mailing list