This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c++/16924] [3.5 Regression] dllexport attribute now causes some symbols _not_ to be exported
- From: "mark at codesourcery dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 12 Aug 2004 16:15:10 -0000
- Subject: [Bug c++/16924] [3.5 Regression] dllexport attribute now causes some symbols _not_ to be exported
- References: <20040808210637.16924.dannysmith@users.sourceforge.net>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From mark at codesourcery dot com 2004-08-12 16:15 -------
Subject: Re: [3.5 Regression] dllexport attribute now causes
some symbols _not_ to be exported
dannysmith at users dot sourceforge dot net wrote:
>------- Additional Comments From dannysmith at users dot sourceforge dot net 2004-08-11 22:55 -------
>Thanks, Mark.
>
>It does fix the regression, and cause no new failures.
>
>
Good.
>config/sh/symbian.c and mcore/mcore.c also need similar patch
>
>
Probably so.
>Should, the SYMBOL_REF_DECL also be set when encode_section_info
>overrides a prior dllimport declaration with a new declaration without
>the attribute, like so?
>
>
Yes.
I'm sorta suspicious of the NT port for needing to do this; I kinda
think that some other approach (involving SYMBOL_REF_FLAGS and overring
ASM_OUT_*) would be more reliable.
Will you please check in your patch together with mine?
Thanks,
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16924