[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#204654: glibc 2.3.2-2 breaks apache on ia64

On Mon, Sep 08, 2003 at 04:10:59AM +1000, Anthony Towns wrote:
> ] Date: Fri, 22 Aug 2003 01:41:03 +0200
> ] Cc: Matthew Wilcox <willy@debian.org>
> ] Subject: Re: Bug#204654: glibc 2.3.2-2 breaks apache on ia64
> ]
> ] Maybe it would be more helpful if you downgraded your libc6 package
> ] to the latest version known to work, compiled the nonstripped Apache
> ] binary, then upgraded again?  I think you should be able to get a
> ] useful backtrace then.
> You haven't responded, nor has anyone else, and the bug's sat idle for
> over two weeks.

Yeah, believe it or not, I actually have more important (and interesting)
things to work on.  I *really* don't enjoy working on glibc.

> > It's an ABI change that gets fixed by a recompile!  
> Voila, it's fixed, so we don't have to worry about it. I'd like it
> properly diagnosed and fixed too, but it's not RC if we can work around
> all the know cases by adding a single entry to the Conflicts: line.

Uhm, you don't seem to have thought through the consequences of this.
The ABI changed.  So when we change it *back* to something that works with
previously compiled binaries, it's entirely possible that everything
compiled since then will no longer work and have to be recompiled.
Why doesn't this scare you?

> > I wish the glibc maintainers were giving this higher priority.
> Narrow the problem down to something small and reproducible and I'm sure
> they will.

uhm, it's entirely reproducible.

"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

Reply to: