[Openmp-commits] [openmp] ef8b3b5 - [OpenMP] Fix nvptx CUDA_VERSION conversion

[Joel E. Denny via Openmp-commits]

Author: Joel E. Denny
Date: 2021-02-19T11:09:26-05:00
New Revision: ef8b3b5ffd562955fbb78c72c7820cd9cd99d313

URL: https://github.com/llvm/llvm-project/commit/ef8b3b5ffd562955fbb78c72c7820cd9cd99d313
DIFF: https://github.com/llvm/llvm-project/commit/ef8b3b5ffd562955fbb78c72c7820cd9cd99d313.diff

LOG: [OpenMP] Fix nvptx CUDA_VERSION conversion

As mentioned in PR#49250, without this patch, ptxas for CUDA 9.1 fails
in the following two tests:

- openmp/libomptarget/test/mapping/lambda_mapping.cpp
- openmp/libomptarget/test/offloading/bug49021.cpp

The error looks like:

```
ptxas /tmp/lambda_mapping-081ea9.s, line 828; error   : Not a name of any known instruction: 'activemask'
```

The problem is that our cmake script converts CUDA version strings
incorrectly: 9.1 becomes 9100, but it should be 9010, as shown in
`getCudaVersion` in `clang/lib/Driver/ToolChains/Cuda.cpp`.  Thus,
`openmp/libomptarget/deviceRTLs/nvptx/src/target_impl.cu`
inadvertently enables `activemask` because it apparently becomes
available in 9.2.  This patch fixes the conversion.

This patch does not fix the other two tests in PR#49250.

Reviewed By: tianshilei1992

Differential Revision: https://reviews.llvm.org/D97012

Added: 
    

Modified: 
    openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt

Removed: 
    


################################################################################
diff  --git a/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt b/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt
index 5478cd3f6aea..1e64e7745582 100644
--- a/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt
+++ b/openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt
@@ -172,7 +172,15 @@ foreach(sm ${nvptx_sm_list})
     list(GET ptx_feature_list ${itr} ptx_num)
     set(cuda_flags ${sm_flags})
     list(APPEND cuda_flags -Xclang -target-feature -Xclang +ptx${ptx_num})
-    list(APPEND cuda_flags "-DCUDA_VERSION=${cuda_version}00")
+    if("${cuda_version}" MATCHES "^([0-9]+)([0-9])$")
+      set(cuda_version_major ${CMAKE_MATCH_1})
+      set(cuda_version_minor ${CMAKE_MATCH_2})
+    else()
+      libomptarget_error_say(
+        "Unrecognized CUDA version format: ${cuda_version}")
+    endif()
+    list(APPEND cuda_flags
+      "-DCUDA_VERSION=${cuda_version_major}0${cuda_version_minor}0")
 
     set(bc_files "")
     foreach(src ${cuda_src_files})