[PATCH] D29647: [OpenMP] Extend CLANG target options with device offloading kind.

[Hal Finkel via Phabricator via cfe-commits]

hfinkel added inline comments.


================
Comment at: test/Driver/openmp-offload.c:614
+/// Check -march propagates compute capability to device offloading toolchain.
+// RUN:   %clang -### -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -save-temps -no-canonical-prefixes -march=sm_35 %s 2>&1 \
+// RUN:   | FileCheck -check-prefix=CHK-COMPUTE-CAPABILITY %s
----------------
gtbercea wrote:
> hfinkel wrote:
> > How does this work on x86 where the host also uses -march?
> That's where the additional flag comes in. The new flag *should* be used instead. 
Ah, okay. Please just propose a patch which adds a new flag. This workaround is a bit strong for my tastes.


Repository:
  rL LLVM

https://reviews.llvm.org/D29647