Closed Bug 1467153 Opened 8 years ago Closed 7 years ago

Builds with GCC 7 toolchain fail

Categories

(Firefox Build System :: General, enhancement)

3 Branch
enhancement
Not set
normal

Tracking

(Not tracked)

RESOLVED WONTFIX

People

(Reporter: marco, Unassigned)

References

Details

Currently they fail at: > [task 2018-06-06T14:43:20.864Z] 14:43:20 INFO - TEST-UNEXPECTED-FAIL | check_stdcxx | We do not want these libstdc++ symbol versions to be used: > [task 2018-06-06T14:43:20.864Z] 14:43:20 INFO - _ZNSsC1ERKSsmRKSaIcE@GLIBCXX_3.4.23
That is: std::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned long, std::allocator<char> const&) Which is a little weird, this constructor shouldn't normally be a problem. Unless we can no longer use the old-abi basic_string? Or is somebody using the new-abi one inadvertently?
Maybe :glandium can help?
Flags: needinfo?(mh+mozilla)
Need something like what was removed in https://hg.mozilla.org/mozilla-central/rev/6803c5e3f34b , but for that function. That won't get you much further, though, because the resulting binaries will have a dependency on the GCC_7_0 symbol version from libgcc... that one I haven't looked into yet.
Flags: needinfo?(mh+mozilla)
We don't care about this anymore for code coverage builds, should we wontfix or do you want to keep this open?
Flags: needinfo?(mh+mozilla)
I guess we don't care anymore, since we're shipping with clang, and use libstdc++ 4.9. We might need to care if we end up using a more recent libstdc++ with clang, but until then...
Status: NEW → RESOLVED
Closed: 7 years ago
Flags: needinfo?(mh+mozilla)
Resolution: --- → WONTFIX
Version: Version 3 → 3 Branch
See Also: → 1536848
No longer blocks: 1410217
You need to log in before you can comment on or make changes to this bug.