[Mlir-commits] [llvm] [mlir] [OpenMPIRBuilder] Do not call host runtime for GPU teams codegen (PR #79984)

[Sergio Afonso]

================
@@ -0,0 +1,25 @@
+// RUN: mlir-translate -mlir-to-llvmir %s | FileCheck %s
+
+// The aim of the test is to check the LLVM IR codegen for the device
+// for omp teams construct
+
+module attributes {dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<"dlti.alloca_memory_space", 5 : ui32>>, llvm.data_layout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-p7:160:256:256:32-p8:128:128-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5-G1-ni:7:8:9", llvm.target_triple = "amdgcn-amd-amdhsa", omp.is_gpu = true, omp.is_target_device = true, omp.target = #omp.target<target_cpu = "gfx90a", target_features = "">} {
+  llvm.func @foo(i32)
+  llvm.func @omp_target_teams_shared_simple(%arg0 : i32)  {
+    omp.teams {
+      llvm.call @foo(%arg0) : (i32) -> ()
+      omp.terminator
+    }
+  llvm.return
+  }
+}
+
+// CHECK-LABEL: @omp_target_teams_shared_simple
+// CHECK-SAME: (i32 [[ARG0:%.+]])
+// CHECK: call void @[[OUTLINED_FN:.*]](
+// CHECK-NOT: call void {{.+}} @__kmpc_fork_teams(ptr @{{.+}}, i32 1, ptr [[OUTLINED_FN:.+]], ptr [[STRUCT_ARG:.*]])
----------------
skatrak wrote:

I think we don't really need to check the arguments or return type here. Otherwise a change in the arguments or return type while keeping the call would make this test pass as well.

```suggestion
// CHECK-NOT: call {{.+}} @__kmpc_fork_teams
```

https://github.com/llvm/llvm-project/pull/79984