Search Criteria
Package Details: clang-prefixed-release 19.1.6-1
Package Actions
Git Clone URL: | https://aur.archlinux.org/clang-prefixed-release.git (read-only, click to copy) |
---|---|
Package Base: | clang-prefixed-release |
Description: | Up to date official clang releases installed at /opt/clang/latest to avoid system wide usage/impact |
Upstream URL: | https://llvm.org/ |
Licenses: | custom:Apache 2.0 with LLVM Exception |
Submitter: | sirspudd |
Maintainer: | sirspudd |
Last Packager: | sirspudd |
Votes: | 4 |
Popularity: | 0.029581 |
First Submitted: | 2023-01-26 02:02 (UTC) |
Last Updated: | 2024-12-17 20:16 (UTC) |
Dependencies (18)
- clang (llvm-rocm-gitAUR, llvm-gitAUR, clang-minimal-gitAUR, clang17-binAUR) (make)
- cmake (cmake-gitAUR) (make)
- git (git-gitAUR, git-glAUR) (make)
- jsoncpp (jsoncpp-cmake-gitAUR, jsoncpp-cmakeAUR, jsoncpp-gitAUR) (make)
- libc++ (libc++-msanAUR, libc++-modulesAUR) (make)
- libedit (make)
- libffi (libffi-gitAUR) (make)
- libunwind (libunwind-carbonAUR, libunwind-gitAUR) (make)
- libxml2 (libxml2-gitAUR, libxml2-2.9AUR) (make)
- lld (llvm-rocm-gitAUR, llvm-gitAUR) (make)
- ncurses (ncurses-gitAUR) (make)
- ninja (ninja-kitwareAUR, ninja-memAUR, ninja-fuchsia-gitAUR, ninja-gitAUR, ninja-jobserverAUR) (make)
- ocl-icd (opencl-icd-loaderAUR, khronos-ocl-icd-loaderAUR, khronos-ocl-icd-gitAUR) (make)
- opencl-headers (opencl-headers-gitAUR) (make)
- python (python37AUR, python311AUR, python310AUR) (make)
- swig (swig-gitAUR) (make)
- z3 (z3-gitAUR) (make)
- python-psutil (check)
Latest Comments
1 2 3 Next › Last »
sirspudd commented on 2024-06-27 06:29 (UTC) (edited on 2024-06-27 06:29 (UTC) by sirspudd)
@mazieres: Sorry, just read through to completion; I have no clue how/when the internal libunwind is linked against vs an external unwind.
I agree with you it is getting built, but can't tell you why if this version is being linked against it is failing.
sirspudd commented on 2024-06-27 06:26 (UTC)
@mazieres: thank you, added libunwind as a dependency to the PKGBUILD, shouldn't fail there anymore
mazieres commented on 2024-06-27 03:21 (UTC) (edited on 2024-06-27 03:21 (UTC) by mazieres)
@sirspudd Sorry, I had to dig in a bit further. It seems internal tools are failing to find libunwind. This is what I get when building 18.1.8 on a system that's just been updated with pacman -Syu on June 26, 2024 but still has clang-prefixed-release 18.1.3-1 installed:
It seems like the source tree contains it own libunwind, but this error is occurring before libunwind is built. I don't understand how these tools are getting linked. When I delete them from _build/bin/ and run ninja -v it doesn't print anything about how the tools are being linked.
Any ideas?
sirspudd commented on 2024-06-27 01:55 (UTC)
@mazieres
it builds fine for me locally and you did not actually include the error from the compiler which would help me in remotely advising you/debugging the failure point.
mazieres commented on 2024-06-27 01:33 (UTC)
18.1.3 built fine for me, but now on 18.1.8 I get a build error:
[835/8900] Building CXX object projects/compiler-rt/lib/msan/CMakeFiles/clang_rt.msan-x86_64.dir/msan_interceptors.cpp.o [836/8900] Building CXX object projects/compiler-rt/lib/asan/CMakeFiles/RTAsan.x86_64.dir/asan_interceptors.cpp.o ninja: build stopped: subcommand failed.
sirspudd commented on 2024-01-30 21:49 (UTC)
Cool; I have verified 18.1 rc1 builds with gcc (without libc) and builds with clang 17 (it does not build with the clang in the Arch repos)
shardik commented on 2023-11-29 18:32 (UTC)
I ran in the same error, failing to build 17.0.5 with clang 16. Then I tried to build 17.0.1 with clang 16, wich works. So you can build 17.0.1 with system clang (16), and then, with 17.0.1 installed and added to $PATH, build 17.0.5.
sirspudd commented on 2023-11-22 20:05 (UTC)
@lahwaacz there is a convenient bool you can now toggle to have clang 17.0.5 break with with gcc (13) or the system clang (16). I would recommend the clang-github-bin package which bypasses this bootstrap problem. I nearly added it as a dependency to building this package, but this feels a little user hostile to do this behind the scenes.
sirspudd commented on 2023-11-22 19:06 (UTC)
Ok, I have confirmed that compiling with gcc explodes for its own reasons; I went through the easy fixes, but there is some legitimate unhappiness there which makes proceeding hard
1 2 3 Next › Last »