This request supersedes:
request 760715
(Show diff)
Overview
Request 760739 accepted
- Enable support for riscv64
- clang-riscv64-rv64gc.diff, riscv64-suse-linux.patch,
llvm-riscv64-fix-cffi.diff, D60657-riscv-pcrel_lo.diff: Backports from
master
- Created by Andreas_Schwab
- In state accepted
- Package maintainers: aaronpuchert and ptesarik
- Supersedes 760715
Loading...
target maintainer
Basically Ok, just a few requests:
- D60657-riscv-pcrel_lo.diff: Can you add the test case from the commit as well? We run the tests and this would allow us to catch issues.
- clang-riscv64-rv64gc.diff, riscv64-suse-linux.patch: Can you use the same format as for the other patches, i.e. include the git header?
- The memory constraints seem rather high, do we need that? I'm somewhat worried that this reduces the number of machines we can build on. Note that we scale the number of compile jobs according to the available memory, so the constraints should be determined by the memory usage of the biggest link jobs (libLLVM.so, libclang.so).
author
source maintainer
target maintainer
The testcase doesn't pass, it depends on other changes The resources are needed.
target maintainer
So you're saying that the memory is needed by the link jobs? Otherwise you should rather adapt mem_per_compile_job
.
Login required, please
login
in order to comment
Request History
Andreas_Schwab created request
- Enable support for riscv64
- clang-riscv64-rv64gc.diff, riscv64-suse-linux.patch,
llvm-riscv64-fix-cffi.diff, D60657-riscv-pcrel_lo.diff: Backports from
master
pluskalm accepted request
ok
Are you sure that this works? In a project using Clang as compiler CMake configuration fails with
There appears to be a linker bug that affects users of std::call_once in big executables