This is the mail archive of the gcc@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]

Re: gcc 3.1 and gdb 5.2? (and cygwin gcc 3.1 problem )


Yeah,I found that gcc will report link-error when I use stl in gcc3.1 on
cygwin.
----- Original Message -----
From: "Dylan Cuthbert" <dylan@q-games.com>
To: "Kai Henningsen" <kaih@khms.westfalen.de>; <gcc@gnu.org>
Sent: Monday, June 17, 2002 9:02 AM
Subject: Re: gcc 3.1 and gdb 5.2? (and cygwin gcc 3.1 problem )


>
> True true,  although I didn't say gdb was more reliable than gcc, I just
> said gcc was more complicated and more experimental because a lot more
> people are involved in it and there's a lot more source code etc.
>
> By the way, if I take the 3.1 release branch and compile it  under cygwin,
> then with certain types of code compiled with with the option "-g" and
using
> STL I get link errors related to internal STL functions being missing -
> stuff that probably should be inlined.  Has anyone else seen this?
>
> If I take the beta version from around the end of March I don't get this
> problem.
>
> ---------------------------------
> Q-Games, Dylan Cuthbert.
> http://www.q-games.com
>
> ----- Original Message -----
> From: "Kai Henningsen" <kaih@khms.westfalen.de>
> To: <gcc@gnu.org>
> Sent: Sunday, June 16, 2002 7:06 PM
> Subject: Re: gcc 3.1 and gdb 5.2?
>
>
> > dylan@q-games.com (Dylan Cuthbert)  wrote on 13.06.02 in
> <001c01c21274$fafdf990$2801a8c0@dcuthbert2k>:
> >
> > > My next stop was going to be the gdb and cygwin lists, however, gcc is
> the
> > > more complicated and experimental of the other two and I wanted to see
> if
> > > anyone else was experiencing problems on other platforms too.
> >
> > What a strange thing to say. From my experience, gcc on cygwin has been
> > stable just about forever, but gdb has a history of extreme instability
on
> > cygnus. (The first gdb I used on cygwin had a tendency to crash on about
> > 70% of all debugged-program crashes, and to not show any useful info on
> > about half the rest. Printf debugging was usually way faster. Or
> > reproducing the bug on Linux, of course. It's become way better, but I
> > still don't entirely trust that gdb.)
> >
> > MfG Kai
>


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