================
@@ -0,0 +1,7 @@
+/// Some target-specific options are ignored for GPU, so %clang exits with
code 0.
+// DEFINE: %{gpu_opts} = --cuda-gpu-arch=sm_60
--cuda-path=%S/Inputs/CUDA/usr/local/cuda --no-cuda-version-check
+// DEFINE: %{check} = %clang -### -c %{gpu_opts} -mcmodel=medium %s
+// RUN: %{check} -fbasic-block-sections=all
+
+// REDEFINE: %{gpu_opts} = -x hip --rocm-path=%S/Inputs/rocm -nogpulib
----------------
jhuber6 wrote:
Should probably include `-nogpuinc` as well. Best way to avoid spurious
failures due to lack of a local CUDA / ROCm installation. Maybe in the future
LLVM based offloading won't depend on so much external stuff.
https://github.com/llvm/llvm-project/pull/79222
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
