http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55909
Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|WAITING |RESOLVED
Resolution| |INVALID
--- Comment #41 from Eric Botcazou <ebotcazou at gcc dot gnu.org> 2013-01-13
09:47:29 UTC ---
> I do apologise for consuming so much of everyone's time. My only defence being
> it wasn't terribly obvious without the comparison of assembler in #24, and an
> over reliance on the configure scripts filling in the blanks.
No need for any apology here. And you did all the real work...
> Not sure if that is a deficiency that can be tested for/warned against between
> glibc/gcc
Clearly something went wrong on the Fedora side, as I don't understand how they
can ship a glibc without TLS support in 2012. That's worth reporting to them
(if not already done). Maybe the TLS detection mechanism could be improved on
the compiler side, but that's not easy when you're not compiling in native
mode.
