[clang] [clang] Fixing Clang HIP inconsistent order for template functions (PR #101627)
Shivam Gupta via cfe-commits
cfe-commits at lists.llvm.org
Mon Aug 12 04:02:26 PDT 2024
xgupta wrote:
First CI is failing with error -
```
clang: error: cannot find ROCm device library; provide its path via '--rocm-path' or '--rocm-device-lib-path', or pass '-nogpulib' to build without ROCm device library
clang: error: cannot find HIP runtime; provide its path via '--rocm-path', or pass '-nogpuinc' to build without HIP runtime
clang: error: cannot find HIP runtime; provide its path via '--rocm-path', or pass '-nogpuinc' to build without HIP runtime
```
I think the test run line is not following convention. You can not include hip_runtime.h, instead have to use cuda.h
so probably you need to add cuda.h in . I have tried replacing #include "hip/hip_runtime.h" with #include "../CodeGenCUDA/Inputs/cuda.h" and using run line like this way -
```// RUN: x=$( %clang_cc1 -x hip -triple amdgcn -target-cpu gfx908 -emit-llvm -fcuda-is-device %s -o - | md5sum | awk '{ print $1 }') && echo $x > %t.md5``` to pass the test case.
https://github.com/llvm/llvm-project/pull/101627
More information about the cfe-commits
mailing list