Ionen Wolkens wrote: > On Wed, Feb 14, 2024 at 08:00:08AM -0600, Dale wrote: >> Howdy, >> >> This has been failing for a while now, month or two. I did a few >> searches but what I did find shows a similar problem was fixed ages >> ago. The link is about protobuf but seems related. Link further down. >> I'm wondering if it may be a USE flag or something triggering this since >> it can't find anything more recent. It also fails on a previous version >> as well. The error bit is about midways. I tried to get either side of >> it. > >> ld.lld: error: undefined symbol: AbslInternalSpinLockWake_lts_20230802 >>>>> referenced by spinlock_wait.h:85 >> (/usr/include/absl/base/internal/spinlock_wait.h:85) >>>>> >> CMakeFiles/xprotocol_plugin.dir/message_field_chain.cc.o:(void >> absl::lts_20230802::base_internal::CallOnceImpl<void >> (*)(google::protobuf::FieldDescriptor const*), >> google::protobuf::FieldDescriptor const*>(std::atomic<unsigned int>*, >> absl::lts_20230802::base_internal::SchedulingMode, void >> (*&&)(google::protobuf::FieldDescriptor const*), >> google::protobuf::FieldDescriptor const*&&)) >> collect2: error: ld returned 1 exit status >> Link mentioned above, dated a while back and claims upstream fixed. >> >> https://github.com/protocolbuffers/protobuf/issues/12292 > Sounds like https://bugs.gentoo.org/912797 which is fixed but only in > ~testing dev-db/mysql-8.0.34. Albeit as far as I'm aware this shouldn't > affect stable 8.0.32-r2 unless you're using ~testing protobuf-23 (aka > mixing stable and ~testing). Not that I tried building it.
That was it. I keyworded the .34 version and it compiled just fine. I thought I had tried that already but maybe I did before it was fixed. I do run some things unstable, KDE for example, but try to keep things mostly stable, especially those things that are needed to boot and such. Anyway, now I have a clean updated system again. :-D Thanks much. Dale :-) :-)
