tech-kern archive

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

Re: Status of revivesa



On Fri, Sep 26, 2008 at 07:34:45AM -0700, Brian Buhrow wrote:
>
> this tradition would hurt our standing in the community in a big way that
> couldn't be calculated.  I believe NetBSD is a niche OS, but I think it's
> used a lot in embeded environments, and my guess is that those
> manufacturers who produce those products based on NetBSD like the fact that
> binaries continue to work year after year, even as they retool for changes
> in hardware.  Other OS projects don't seem to worry much about binary
> backward compatibility and it is one of the things that sets NetBSD apart.

I link one component in my product against a third-party library -- a
*static* library -- which is supplied by its vendor only in binary form,
compiled on NetBSD 2.1.  I can get a newer build from them without paying
an exorbitant fee-- once -- so I will have to choose my timing carefully.

Thank goodness NetBSD 2 didn't have threads, is all I can say -- amazingly,
with the compatibility code in libc and the kernel turned on, this actually
works.

There is 3rd-party code out there of more direct use to end-users (one
prominent example is the Perforce client, particularly the visual client)
which is only ever rebuilt on NetBSD intermittently, if at all.

I would like to see revivesa get finished and checked in.  Holding up 5.0
is not helping anyone, and I for one am more than a little annoyed about
the protracted "code freeze" and its gradual thaw-without-any-announcement;
it complicates my own release engineering task enormously.

Thor


Home | Main Index | Thread Index | Old Index