Summary: | dev-lang/spidermonkey built with sys-devel/clang is missing some symbols | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Oleh <moonlapse81> |
Component: | Current packages | Assignee: | Mozilla Gentoo Team <mozilla> |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | CC: | ago |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | All | ||
URL: | https://bugzilla.mozilla.org/show_bug.cgi?id=1426865 | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | |||
Bug Blocks: | 408963 | ||
Attachments: |
build log
emerge --info |
Description
Oleh
2019-03-24 12:50:22 UTC
Created attachment 570602 [details]
build log
Created attachment 570604 [details]
emerge --info
I've confirmed this too with 52.x. It seems upstream is still working on a fix, as the one(s) proposed don't suffice. I don't know the best interim solution at this point but I'll look into how we can either force gcc or warn/error-out if clang is being used. Both polkit and gjs apparently have these issues so likely the package can be considered 'broken' in all cases. (In reply to Ian Stakenvicius from comment #3) > how we can either force gcc Until more automatic solution is provided by toolchain-funcs.eclass: inherit flag-o-matic toolchain-funcs ... src_configure() { if tc-is-clang; then export CC="${CHOST}-gcc" export CXX="${CHOST}-g++" strip-unsupported-flags fi ... I have just reproduced this same issue with gcc, so avoiding clang isn't necessarily going to fix it. (In reply to Ian Stakenvicius from comment #5) > I have just reproduced this same issue with gcc, so avoiding clang isn't > necessarily going to fix it. Well give some detail on how to reproduce. (In reply to Jory A. Pratt from comment #6) > (In reply to Ian Stakenvicius from comment #5) > > I have just reproduced this same issue with gcc, so avoiding clang isn't > > necessarily going to fix it. > > Well give some detail on how to reproduce. I haven't been able to yet, rebuilds of spidermonkey with gcc7 and gcc8 have been fine. This may have been an odd distcc messup the first time around (as I was using distcc at the time). I'll report what I find if I can actually reproduce it with gcc again. Looks like it will be fixed in version 72: https://bugzilla.mozilla.org/show_bug.cgi?id=1426865#c45 Maybe the patches could apply cleanly to 60, but I have still been able to work around not having polkit. There's a backport available for esr-68.0, I dropped the patchset into the patches folder and will take a testdrive with them now. *** Bug 729904 has been marked as a duplicate of this bug. *** spidermonkey <78 is no longer in tree. The codebase has changed substantially since then, so closing. |