| Issue |
82205
|
| Summary |
libc/src/__support/FPUtil/aarch64/fenv_darwin_impl.h:164:9: error: no viable conversion from '__llvm_libc_19_0_0_git::fputil::FPBits<float>' to 'float'
|
| Labels |
build-problem,
libc
|
| Assignees |
|
| Reporter |
rudkx
|
I'm building on macOS using Xcode 15.2 with my repo at commit `79709498eaa5` and hitting these errors:
```
/Users/rudkx/src/llvm-project/libc/src/__support/FPUtil/aarch64/fenv_darwin_impl.h:164:9: error: no viable conversion from '__llvm_libc_19_0_0_git::fputil::FPBits<float>' to 'float'
164 | float large_value = FPBits<float>::max_normal();
| ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~
/Users/rudkx/src/llvm-project/libc/src/__support/FPUtil/aarch64/fenv_darwin_impl.h:165:9: error: no viable conversion from '__llvm_libc_19_0_0_git::fputil::FPBits<float>' to 'float'
165 | float small_value = FPBits<float>::min_normal();
| ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~
/Users/rudkx/src/llvm-project/libc/src/__support/FPUtil/aarch64/fenv_darwin_impl.h:280:33: error: implicit conversion loses integer precision: 'const uint64_t' (aka 'const unsigned long long') to 'uint32_t' (aka 'unsigned int') [-Werror,-Wshorten-64-to-32]
280 | FEnv::set_control_word(state->ControlWord);
| ~~~~ ~~~~~~~^~~~~~~~~~~
/Users/rudkx/src/llvm-project/libc/src/__support/FPUtil/aarch64/fenv_darwin_impl.h:281:32: error: implicit conversion loses integer precision: 'const uint64_t' (aka 'const unsigned long long') to 'uint32_t' (aka 'unsigned int') [-Werror,-Wshorten-64-to-32]
281 | FEnv::set_status_word(state->StatusWord);
| ~~~~ ~~~~~~~^~~~~~~~~~
4 errors generated.
```
I configured and built like this:
```
cmake -S llvm -B assert -G Ninja -DLLVM_ENABLE_PROJECTS="clang;mlir;openmp;pstl" -DLLVM_ENABLE_ASSERTIONS=ON -DCMAKE_BUILD_TYPE=Release -DLLVM_ENABLE_RUNTIMES="libc;libunwind;libcxxabi;libcxx;compiler-rt"
cmake --build assert
```
![]()
_______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
