From 248e5bf6ef4ecf292928b99bf755b26b6ebd3c6f Mon Sep 17 00:00:00 2001 From: Yaxun Liu Date: Tue, 29 Jan 2019 13:20:23 +0000 Subject: [CUDA][HIP] Do not diagnose use of _Float16 r352221 caused regressions in CUDA/HIP since device function may use _Float16 whereas host does not support it. In this case host compilation should not diagnose usage of _Float16 in device functions or variables. For now just do not diagnose _Float16 for CUDA/HIP. In the future we should have more precise check. Differential Revision: https://reviews.llvm.org/D57369 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@352488 91177308-0d34-0410-b5e6-96231b3b80d8 --- test/SemaCUDA/float16.cu | 7 +++++++ 1 file changed, 7 insertions(+) create mode 100644 test/SemaCUDA/float16.cu (limited to 'test/SemaCUDA') diff --git a/test/SemaCUDA/float16.cu b/test/SemaCUDA/float16.cu new file mode 100644 index 0000000000..a9cbe87f32 --- /dev/null +++ b/test/SemaCUDA/float16.cu @@ -0,0 +1,7 @@ +// RUN: %clang_cc1 -fsyntax-only -triple x86_64 -aux-triple amdgcn -verify %s +// expected-no-diagnostics +#include "Inputs/cuda.h" + +__device__ void f(_Float16 x); + +__device__ _Float16 x = 1.0f16; -- cgit v1.2.1