Bug#995827: llvm-toolchain-13: FTBFS on armel, mipsel, mips64el: undefined references to __atomic_foo symbols

[Nicholas D Steeves]

Sylvestre Ledru <sylvestre at debian.org> writes:

> Hello,
>
> Le 12/10/2021 à 03:51, Nicholas D Steeves a écrit :
[snip]
>>> Le 06/10/2021 à 17:05, Simon McVittie a écrit :
>>>>> /usr/bin/ld: projects/compiler-rt/lib/scudo/standalone/CMakeFiles/clang_rt.scudo_standalone-dynamic-mipsel.dir/wrappers_c.cpp.o: in function `scudo::atomic_u64::Type scudo::atomic_load<scudo::atomic_u64>(scudo::atomic_u64 const volatile*, scudo::memory_order)':
>>>>> /<<PKGBUILDDIR>>/compiler-rt/lib/scudo/standalone/atomic_helpers.h:66: undefined reference to `__atomic_load_8'
>>>>> /usr/bin/ld: /<<PKGBUILDDIR>>/compiler-rt/lib/scudo/standalone/atomic_helpers.h:66: undefined reference to `__atomic_load_8'
>>> Yeah, I saw, it guess it is a -latomic missing again :)
>>>
>> 
>> I've subscribed to this bug, since I just noticed llvm-toolchain-13 was
>> available, and irony-mode users always want to use the latest Clang.
>> Most irony-mode users are on amd64, so I'm thinking about switching now
>> (for the greatest good), but I'm wondering this: Given that you know
>> what the cause is, do you think this bug will soon be resolved?
>> 
>
>
> I think i fixed that one but not sure yet if there is other pending issue
> See:
> https://buildd.debian.org/status/package.php?p=llvm-toolchain-13
> (it looks good for now)
>

Thank you Sylvestre!  Much appreciated :-)  I think the following two RC
bugs may be new since our last correspondence: #996632 and #996448

By the way, am I correct in understand that the focus for bookwork will
be on llvm-toolchain-13 (or possibly 14)?

Regards,
Nicholas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 857 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/pkg-llvm-team/attachments/20211026/348d24c4/attachment.sig>