This is the mail archive of the libc-alpha@sources.redhat.com mailing list for the glibc 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: Problem with VMware 2.0.4 and glibc 2.2.5


On Wed, Apr 10, 2002 at 10:07:17PM +0200, Thorsten Kukuk wrote:
> On Wed, Apr 10, Jeroen Dekkers wrote:
> 
> > > No, we should only remember that nearly all programs using nice() for
> > > Linux are "broken" and that we should not break all this applications
> > > with such a change. I think this is such a dramatic change which will
> > > break many applictions, that we should introduce a new version to give
> > > the people the chance to use their old binaries until all software is
> > > fixed.
> > 
> > glibc isn't made to workaround all programmer's bugs. We never change
> > interfaces for bugfixes, if we are going to do that glibc will get
> > pretty bloated. I don't see why we should change the behaviour because
> > the amount of applications relying on a bug is higher this time. The
> > applications need to get fixed anyhow, just fix it now and recompile
> > it.
> 
> Sometimes it is better for your reputation not to break it. 

Reputation? Which reputation? Of being bloated, that reputation does
glibc already have by some people. I don't share that opinion, but if
we are going to put workarounds for every programmer bug in glibc I'm
actually going to agree with them.

> And
> please tell me, how a user should recompile his binary only software?

He should not use binary only software at all. I can't help it that an
user restricts himself by using binary only software. I don't use any
piece of software made by people who try to restrict my freedom.

> With this, we will make only sure that no ISV will develop commercial
> software for Linux, because he never knows how often he has to ship
> new software.  

If the ISV relies on bugs and doesn't read the manual he indeed has to
ship new software. Everyone has to ship new versions if there are bugs
in his software, why should it be different in this case?

> At first he need to implement a workaround for
> a buggy glibc function and then we fix it in a way which breaks his
> application again. You will not make friends with this.

No, he has to complain by the glibc developers that the glibc function
is broken instead of implementing a workaround. And you will never
make friends with non-free software, non-free software is just
anti-social.

Jeroen Dekkers
-- 
Jabber supporter - http://www.jabber.org Jabber ID: jdekkers@jabber.org
Debian GNU supporter - http://www.debian.org http://www.gnu.org
IRC: jeroen@openprojects

Attachment: msg00060/pgp00000.pgp
Description: PGP signature


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