From: Danny Smith <dannysmith@clear.net.nz>
To: "Aaron W. LaFramboise" <aaronavay62@aaronwl.com>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>, Christopher Faylor <me@cgf.cx>
Subject: Re: [Patch} Fix dynamic casts across dll boundaries on windows targets
Date: Tue, 19 Oct 2004 10:52:00 -0000 [thread overview]
Message-ID: <000601c4b5c4$017f8b90$b64861cb@DANNY> (raw)
In-Reply-To: <4174AF4B.2080307@aaronwl.com>
Aaron W. LaFramboise wrote:
> Danny Smith wrote:
>
>> 2004-10-16 Danny Smith <>
>>
>> * config/i386/cygming.h (TARGET_OS_CPP_BUILTINS): Define
>> __GXX_MERGED_TYPEINFO_NAMES to 0.
>
> Does this pessimize the case where typeinfo symbols really are
> availible?
It pessimizes linking against static libs/objects
An example might be when a DLL exports the typeinfo
> symbols and they are linkable with auto-import. (Will auto-import
> work for this, anyway?)
>
There is a test case in the original bug report :
http://sourceforge.net/tracker/?func=detail&atid=102435&aid=568054&group
_id=2435
> Another source for experience in this area is the recent ELF
> visibility work. For example, see the secton Caveats in
> <http://www.nedprod.com/programs/gccvisibility.html>. The suggestion
> here is that if you want RTTI to work, you must take necessary steps
> to export typeinfo symbols. Is a similar approach for PE
> unreasonable?
One of the problems with RTTI vs dllimport is that dllimported objects
do not have a constant address at compile time. This creates a problem
for static initialisation of, say, another RTTI object or a vtable
that has a
dllimported typeinfo member.
Danny
>
> Aaron W. LaFramboise
next prev parent reply other threads:[~2004-10-19 10:17 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-10-16 9:47 Danny Smith
2004-10-17 22:25 ` Christopher Faylor
2004-10-19 6:21 ` Aaron W. LaFramboise
2004-10-19 10:52 ` Danny Smith [this message]
2004-10-21 3:59 ` Aaron W. LaFramboise
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='000601c4b5c4$017f8b90$b64861cb@DANNY' \
--to=dannysmith@clear.net.nz \
--cc=aaronavay62@aaronwl.com \
--cc=dannysmith@users.sourceforge.net \
--cc=gcc-patches@gcc.gnu.org \
--cc=me@cgf.cx \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).