[boost] [config][android] GCC-5 and 6 failures.

[John Maddock via Boost]

There are some runtime (segfault) failures running config_test on
andoid, see for example:
http://www.boost.org/development/tests/develop/developer/CrystaX-apilevel-21-armeabi-v7a-llvm-libc++-config-gcc-5-config_test-variants_.html

Can anyone with access to this platform (or preferably the test runner)
see if defining BOOST_NO_CXX11_THREAD_LOCAL fixes the issue? That would
be my hunch for the problem anyway.

Many thanks, John.




---
This email has been checked for viruses by AVG.
http://www.avg.com


_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

[Dmitry Moskalchuk via Boost]

Hi John,

Sure I can check that. However, as far as I see, this error has nothing
to Boost. It's internal error in our (CrystaX) clang implementation,
which is subject for fix on our side.

Do you still want to workaround it in Boost::Config? If yes, I can
adjust configuration and re-run tests, but please be little more
specific then - where and what exactly should I change to re-run tests
(I can deduce it on my own, but better would be if you'd say exactly to
avoid potential mistakes)?
--
Dmitry Moskalchuk

Post by John Maddock via Boost
There are some runtime (segfault) failures running config_test on
http://www.boost.org/development/tests/develop/developer/CrystaX-apilevel-21-armeabi-v7a-llvm-libc++-config-gcc-5-config_test-variants_.html
Can anyone with access to this platform (or preferably the test
runner) see if defining BOOST_NO_CXX11_THREAD_LOCAL fixes the issue?
That would be my hunch for the problem anyway.
Many thanks, John.
---
This email has been checked for viruses by AVG.
http://www.avg.com
_______________________________________________
http://lists.boost.org/mailman/listinfo.cgi/boost

[John Maddock via Boost]

Post by Dmitry Moskalchuk via Boost
Hi John,
Sure I can check that. However, as far as I see, this error has nothing
to Boost. It's internal error in our (CrystaX) clang implementation,
which is subject for fix on our side.

In which case, I'm inclined to ignore it.

Thanks, John.

Post by Dmitry Moskalchuk via Boost
Do you still want to workaround it in Boost::Config? If yes, I can
adjust configuration and re-run tests, but please be little more
specific then - where and what exactly should I change to re-run tests
(I can deduce it on my own, but better would be if you'd say exactly to
avoid potential mistakes)?
_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

---
This email has been checked for viruses by AVG.
http://www.avg.com


_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost