Amanda-Users

Re: Glibc

2009-02-25 16:25:51
Subject: Re: Glibc
From: stan <stanb AT panix DOT com>
To: "Dustin J. Mitchell" <dustin AT zmanda DOT com>
Date: Wed, 25 Feb 2009 16:23:30 -0500
On Wed, Feb 25, 2009 at 04:01:20PM -0500, Dustin J. Mitchell wrote:
> On Wed, Feb 25, 2009 at 3:40 PM, stan <stanb AT panix DOT com> wrote:
> > Now I am worried as it unpacked, I noted a number of files in a directory
> > that seemed to pertain to specific machine types. Now I see that the list of
> > supported hardware/OS combinations is very limited. I need to make this
> > work on HP-UX 10.20 on PA-RISC, and Solaris 5, and 8 on SPARC. It does not
> > appear that these hardware/software combinations are supported for glibc.
> 
> I know folks who have gotten glib installed on HPUX 11, but not 10.
> It's available from SunFreeWare for Solaris 8, but I don't know about
> 5.  I don't know anything about PA-RISC.

HP-UX on PA_RISC, or Itanium? 
> 
> > I don't understand what glibc is, clearly. I thought that it was the libc
> > that is used on Linux distributions, but that must not be the case. What
> > does it bring to the party that Amanda needs?
> 
> That's a common misconception:
>  glibc -- GNU implementation of libc
>  glib -- basic C utility library developed for the GNOME project
> glib provides a whole slew of useful low- and mid-level utilities that
> are invaluable for use in Amanda.  It has a nice list of basic data
> structures (linked lists, hash tables, dynamically resizable arrays),
> some nice string utilities, a limited XML parser, and a basic
> implementation of classes and objects.  It has replaced a lot of
> pretty bug-ridden handrolled implementations in Amanda.

Ah, well maybe I am trying to compile the wrong thing. I downloaded glibc
from GNU, so we may not be in as bad a shape as I think. I did have xacc
(Gnucash now) working on HP-UX 10.20 many years  ago, and it depends on
glib, now that you remind me what it is.

-- 
One of the main causes of the fall of the roman empire was that, lacking
zero, they had no way to indicate successful termination of their C
programs.

<Prev in Thread] Current Thread [Next in Thread>