[Mlir-commits] [mlir] [MLIR][GPU] Add support for non-portable cluster size attribute (PR #95545)

[Guray Ozen]

================
@@ -94,7 +94,7 @@ module attributes {gpu.container_module} {
   // CHECK: [[S3:%.*]] = call ptr @mgpuModuleLoad(ptr @kernel_module_bin_cst, i64 4)
   // CHECK: [[S4:%.*]] = call ptr @mgpuModuleGetFunction(ptr [[S3]], ptr @kernel_module_kernel_kernel_name)
   // CHECK: [[S5:%.*]] = call ptr @mgpuStreamCreate()
-  // CHECK: call void @mgpuLaunchClusterKernel(ptr [[S4]], i64 2, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i32 0, ptr [[S5]], ptr [[S2]], ptr null)
+  // CHECK: call void @mgpuLaunchClusterKernel(ptr [[S4]], i64 2, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i64 1, i32 0, i1 false, ptr [[S5]], ptr [[S2]], ptr null)
----------------
grypp wrote:

need a test that's `true`

https://github.com/llvm/llvm-project/pull/95545