This is the mail archive of the
libstdc++@sourceware.cygnus.com
mailing list for the libstdc++ project.
RE: Compiling libstdc++-2.90.7 on Solaris 7
- To: 'Benjamin Kosnik' <bkoz at fidel dot cygnus dot com>, Michael VanLoon <MichaelV at EDIFECS dot COM>
- Subject: RE: Compiling libstdc++-2.90.7 on Solaris 7
- From: Michael VanLoon <MichaelV at EDIFECS dot COM>
- Date: Wed, 2 Feb 2000 12:20:30 -0800
- Cc: libstdc++ at sourceware dot cygnus dot com
Understand I'm not trying to pick on you guys. On the one hand, it seems
all the world is Linux, and nobody really pays enough attention to the
*BSDs. On the other hand, you have lots of work to do, and I'm sure you
just don't have time to track down the idiosyncrasies of every system. :-)
However, I understand that you have a lot of work to do and not enough time
to do it, because I am in the same place. I don't have time to track down
and fix these things myself.
I can, however, arrange access to both FreeBSD and NetBSD boxes, if someone
wanted to take a quick stab at some diagnosis of the porting issues. FWIW,
fixing these things on one version of BSD will probably go most of the
distance to fixing them on all BSDs.
-----Original Message-----
From: Benjamin Kosnik [mailto:bkoz@fidel.cygnus.com]
Sent: Wednesday, February 02, 2000 11:55 AM
To: Michael VanLoon
Cc: libstdc++@sourceware.cygnus.com
Subject: RE: Compiling libstdc++-2.90.7 on Solaris 7
If you want me to do it, I'll need a FreeBSD or NetBSD box, or access to
one. And allocated time to work on it. Since that possibility is pretty
remote, I suggest you submit fixes. . .
:(
Assuming glibc and egcs-2.95.2, the port would seemingly be minimal. I
don't understand why recent *BSD's don't really work out of the box, or
why people seem to have so many problems with the wchar_t stuff, but I
can't really do anything except wave my hands right now. . .
(hand waving)
-benjamin
> So out of curiosity, what does it take to get these sorts of things fixed
> for FreeBSD and NetBSD? Do you need someone from one of those camps to
> actually submit fixes, or is this something that the current team will
> (eventually) tend to?