[clang] 34447ef - [clang][test] Specify value of `-fopenmp=libomp` for test. (#150301)

[via cfe-commits]

Author: Jordan Rupprecht
Date: 2025-07-23T15:46:49-05:00
New Revision: 34447efd4614cfc8f17aa5cb7305f76ebaf6eaf5

URL: https://github.com/llvm/llvm-project/commit/34447efd4614cfc8f17aa5cb7305f76ebaf6eaf5
DIFF: https://github.com/llvm/llvm-project/commit/34447efd4614cfc8f17aa5cb7305f76ebaf6eaf5.diff

LOG: [clang][test] Specify value of `-fopenmp=libomp` for test. (#150301)

`libomp` is the default value when unconfigured in cmake, but llvm can
be configured to have `libgomp` be the default instead. Explicitly
specify this value so the test does not fail when it assumes libomp is
always the default.

Added: 
    

Modified: 
    clang/test/Driver/offload-target.c

Removed: 
    


################################################################################
diff  --git a/clang/test/Driver/offload-target.c b/clang/test/Driver/offload-target.c
index 123ecd3eb830e..23a2cf26eb192 100644
--- a/clang/test/Driver/offload-target.c
+++ b/clang/test/Driver/offload-target.c
@@ -10,7 +10,7 @@
 // RUN: | FileCheck %s -check-prefix=CUDA
 // CUDA: "nvptx64-nvidia-cuda" - "clang", inputs: ["[[INPUT:.+]]"], output: "[[NV_OBJ:.+]]"
 
-// RUN: %clang -### --offload-targets=amdgcn-amd-amdhsa,nvptx64-nvidia-cuda -fopenmp \
+// RUN: %clang -### --offload-targets=amdgcn-amd-amdhsa,nvptx64-nvidia-cuda -fopenmp=libomp \
 // RUN:   -Xarch_amdgcn --offload-arch=gfx90a -Xarch_nvptx64 --offload-arch=sm_89 \
 // RUN:   -nogpulib -nogpuinc %s -ccc-print-bindings 2>&1 \
 // RUN: | FileCheck %s -check-prefix=OPENMP