This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug c++/19159] [4.0/4.1 Regression] Undefined symbol: vtable for __cxxabiv1::__vmi_class_type_info


------- Additional Comments From dave at hiauly1 dot hia dot nrc dot ca  2005-04-05 12:06 -------
Subject: Re:  [4.0/4.1 Regression] Undefined symbol: vtable for __cxxabiv1::__vmi_class_type_info

> ------- Additional Comments From mmitchel at gcc dot gnu dot org  2005-04-05 06:49 -------
> Dave --
> 
> I suspect that there is some PCH involvement here.  In particular, the warnings
> you're seeing while compiling header_cassert are very surprising, in that the
> preprocessed source for that file does not contain the string "_S_do_it".  So, I
> think the problem is related to PCH.  To fix that, I'll need a test case that I
> can reproduce in a cross environment; i.e., preprocessed source that I can turn
> into a PCH, and then some use of the PCH that produces the warnings.
> 
> In any case, I'm going to test, and then check in, the latest patch, as lots of
> link-time warnings is better than link failures.

The warnings are from GCC:

ginyu% grep "used but never defined" *.c
decl2.c:          cp_warning_at ("inline function %qD used but never defined", decl);

You could be right that this is a PCH problem and possibly configuring
with --disable-libstdcxx-pch would help.

Dave


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19159


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]