Re: g++ on the hurd?
On Thu, Mar 08, 2001 at 03:35:50PM +0100, Marcus Brinkmann <Marcus.Brinkmann@ruhr-uni-bochum.de> was heard to say:
> On Thu, Mar 08, 2001 at 08:59:54AM -0500, Daniel Burrows wrote:
> > I just tried to port aptitude to the Hurd (now that apt is finally
> > officially in the archive), and discovered that I had removed g++ while
> > bringing everything up-to-date a few days ago. Trying to install it revealed
> > that only the g++ dummy package is available on the Hurd; the actual compiler
> > (g++-2.95) doesn't appear to be ported yet.
>
> Ha! Something removed libstdc++2.10 (and 2.9) from the archive, and now
> dinstall thinks it is a new package and refuses to cooperate. We will have
> to wait for the ftp admins to manually take action (pray for that if you
> want).
>
> Very unfortunate. In any way, you can fetch the missing packages from
> http://incoming.debian.org/
Well, I tried..
(some of the stuff below I should probably file bugs about myself, but
I'm reporting it here so other people know what is/isn't broken)
It seems that libstdc++2.10-dev has a versioned dependency on libc6-dev.
Someone should file a bug, I guess..
I also noticed that libgtk1.2 depends on gconv-modules. On Linux, this is
provided by libc6; the Hurd doesn't appear to have it at all. (libc0.2
replaces gconv-modules, but doesn't seem to provide it) Should someone
ask the gtk maintainer to conditionalize his control file, or does libc0.2
actually provide gconv-modules?
And, while I'm listing random breakage, I tried to see if X would work on
the Hurd..no luck. It complains that /tmp/.X11-unix has "suspicious ownership";
that directory seems to be owned by root.daniel, whereas on my Linux system
the ownership is root.root. I didn't have time to investigate further,
unfortunately. What's weird is that it gets the bad ownership (root.daniel)
even if I run it as root (!!)
(and of course there are very few X programs compiled..
if I can get X working and sort out the toolchain issues, I'll start
building them.. (sawfish!) )
Also, the "console users only" mode of Xwrapper is broken. I think it
must rely on some sort of linuxism, although I don't have any idea what it
could be :-/
(what else..) I occasionally get lots of scary-looking error messages from
ext2fs about (iirc) "free inode had non-zero size.." -- could it be trying to
reallocate inodes? That sounds bad..
The default TERM is mach, not mach-color. Not sure if that's
intentional or not.
Thanks,
Daniel
--
/-------------------- Daniel Burrows <dburrows@brown.edu> --------------------\
| Whoever fights monsters should see to it that in the |
| process he does not become a monster. And when you look |
| into an abyss, the abyss also looks into you. |
| -- Friedrich Nietzsche |
\----------------- The Turtle Moves! -- http://www.lspace.org ----------------/
Reply to: