Summary: | =sys-libs/compiler-rt-14.0.0 fails to configure | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Dan Goodliffe <gentoo> |
Component: | Current packages | Assignee: | LLVM support project <llvm> |
Status: | RESOLVED INVALID | ||
Severity: | normal | CC: | jstein, mgorny, sam |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
See Also: | https://bugs.gentoo.org/show_bug.cgi?id=691544 | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
build.log
CMakeOutput.log CMakeError.log |
Description
Dan Goodliffe
2022-03-25 16:49:04 UTC
Created attachment 767805 [details]
build.log
Please include: -- Configuring incomplete, errors occurred! See also "/tmp/portage/sys-libs/compiler-rt-14.0.0/work/compiler-rt-14.0.0_build/CMakeFiles/CMakeOutput.log". See also "/tmp/portage/sys-libs/compiler-rt-14.0.0/work/compiler-rt-14.0.0_build/CMakeFiles/CMakeError.log". * ERROR: sys-libs/compiler-rt-14.0.0::gentoo failed (configure phase): also, try without distcc? CMake Warning at cmake/Modules/CompilerRTUtils.cmake:287 (message): LLVM source tree not found at "/tmp/portage/sys-libs/compiler-rt-14.0.0/work/llvm". You are not using the monorepo layout. This configuration is DEPRECATED. Call Stack (most recent call first): CMakeLists.txt:70 (load_llvm_config) CMake Warning at cmake/Modules/CompilerRTUtils.cmake:315 (message): Consulting llvm-config for the LLVM source path as a fallback. This behavior will be removed in the future. Call Stack (most recent call first): CMakeLists.txt:70 (load_llvm_config) -- Using LLVM source path (/dev/null) from llvm-config CMake Warning at cmake/Modules/CompilerRTUtils.cmake:352 (message): llvm-config finding testingsupport failed with status 1 Call Stack (most recent call first): CMakeLists.txt:70 (load_llvm_config) The path to work is correct, but there is no llvm directory in there: ~ ✔ l /tmp/portage/sys-libs/compiler-rt-14.0.0/work total 0 drwxr-xr-x 3 portage portage 80 Mar 14 09:44 cmake/ drwxr-xr-x 11 portage portage 360 Mar 25 09:25 compiler-rt/ drwxr-xr-x 3 portage portage 140 Mar 25 09:25 compiler-rt-14.0.0_build/ drwxr-xr-x 2 portage portage 60 Mar 25 09:25 llvm-gentoo-patchset-14.0.0/ I suspect it's a series of failures stemming from that one, but that's just experience and guesswork at this point. Ah yes, disabling distcc does fix the problem, even though those warnings still occur. Do you still want those related CMake logs from when distcc is enabled? No great surprise, compiler-rt-sanitizers has exactly the same problem/solution. CMake warnings are usually noise, but could you upload both of those logs anyway? Id like to see the result with distcc and see what it's misfetscting. FWIW there's a few issues with CMake and this, it's to do with it misdetecting distcc log output as compiler output IIRC. There is a bug upstream (I'll link it here later) about it. Created attachment 767811 [details]
CMakeOutput.log
Created attachment 767812 [details]
CMakeError.log
Oh this problem is obvious when you figure it out. All the llvm/clang tools attempt to update in one go... so it gets through a lot of them and then borks at compiler-rt. The problem is that distccd isn't configured with with clang-14 in its list of executable commands. Add the v14 binaries and the whole problem goes away. testingsupport warnings are normal and not be worried about. Do you have any further problems or is everything good now? It's all good now thanks... sorry for the distraction. It'd be nice if that problem stuck out in a slightly more obvious way mind. Just to be clear, am I guessing right that the failure is because distcc is emitting some warnings that are confusing CMake checks? (In reply to Michał Górny from comment #12) > Just to be clear, am I guessing right that the failure is because distcc is > emitting some warnings that are confusing CMake checks? yes, it's https://gitlab.kitware.com/cmake/cmake/-/issues/22573 I think The fact that this still happens despite the cmake patch that gentoo provides for this in tree (as discussed in https://bugs.gentoo.org/691544) suggests that something else is going on here. (In reply to Tom Gillespie from comment #14) > The fact that this still happens despite the cmake patch that gentoo > provides for this in tree (as discussed in https://bugs.gentoo.org/691544) > suggests that something else is going on here. As far as I'm aware, it's just that there's more areas where the checks fail with distcc. I don't think it's anything deeper. |