bjope wrote:
> > Ok, seems like the only problem is the MemProf unit test
> > compiler-rt/lib/memprof/tests/
> > So if I manually disable that I don't see any other failures with this
> > patch.
>
> I think any components that uses freshly built clang (stage1) could
> potentially run into this problem again, the correct way to solve it is to
> provide a compatible C++ Library to the freshly built clang when it built
> compiler-rt and related tests. For example, use -D with
> `RUNTIMES_${target}_CMAKE_CXX_FLAGS` and `RUNTIMES_${target}_CMAKE_CXX_FLAGS`
> to pass `--gcc-toolchain` (it might be deprecated and there is an updated
> flag for it) to the runtimes build. This will probably allow stage1 clang to
> use gcc's stdlib like your bootstrap clang.
Thanks! Yes, I think we hadn't considered that some things we set up for the
regular build wasn't forwarded to the compiler-rt test.
When setting
`-DRUNTIMES_${target}_COMPILER_RT_TEST_COMPILER_CFLAGS=--gcc-toolchain=<path-to-toolchain>`
the problem @mikaelholmen mentioned disappeared. I haven't tried adding it to
CMAKE_CXX_FLAGS, but maybe that is a better option.
https://github.com/llvm/llvm-project/pull/65823
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
