Updates will be applied April 15th at 12pm EDT (UTC-0400). GitLab could be a little slow between 12 - 12:45pm EDT.

Commit 3b754215 authored by Brad King's avatar Brad King

VS: Improve workaround for CUDA -Xcompiler placement bug

In commit v3.9.0-rc1~431^2~6 (VS: Place CUDA host compiler options in
proper project file fields, 2017-03-07) we worked around a bug in the
CUDA VS integration by dropping `AdditionalCompilerOptions`.  However,
this silently drops `-Xcompiler=` options given by the user that don't
map to one of CudaCompile's dedicated settings.  Improve the workaround
to instead put the remaining `AdditionalCompilerOptions` into the
`AdditionalOptions` field behind `-Xcompiler=` ourselves.
parent f2059585
......@@ -2523,9 +2523,16 @@ bool cmVisualStudio10TargetGenerator::ComputeCudaOptions(
// `CUDA 8.0.targets` places these before nvcc! Just drop whatever
// did not parse and hope it works.
// `CUDA 8.0.targets` places AdditionalCompilerOptions before nvcc!
// Pass them through -Xcompiler in AdditionalOptions instead.
if (const char* acoPtr = cudaOptions.GetFlag("AdditionalCompilerOptions")) {
std::string aco = acoPtr;
if (!aco.empty()) {
aco = this->LocalGenerator->EscapeForShell(aco, false);
cudaOptions.AppendFlag("AdditionalOptions", "-Xcompiler=" + aco);
......@@ -15,6 +15,10 @@ set(CMAKE_CUDA_STANDARD 11)
add_library(CUDASeparateLibA STATIC file1.cu file2.cu file3.cu)
target_compile_options(CUDASeparateLibA PRIVATE -Xcompiler=-bigobj)
#Having file4/file5 in a shared library causes serious problems
#with the nvcc linker and it will generate bad entries that will
#cause a segv when trying to run the executable
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment