> After all this time one does not gain stability by removing a.out. The one
> "efficiency" I can think of to dump the a.out format is kernel size
> because it doesn't need the a.out loader any more. How much does that
> take? If you don't want it, just don't compile it in your kernels.
>
> One can hide many things behind the label "for the sake of progress". I
> fail to see that any is made in this case.
I'm in this camp too. I actually have some 2.0.2x systems that are still
totally a.out except for their kernels which were compiled ELF. I have
other servers that have only recently been upgraded to ELF compatibility
and run a few ELF binaries, but are still 99% a.out. These systems aren't
broken...I see no compelling reason to upgrade all software to ELF just
because it's the new standard binary format. Anything that does get
upgraded ends up ELF...but for the scores of non-broken utilities on my
systems, there just isn't a good reason to recompile them all.
Removing a.out support will force me to stop upgrading, or more likely,
will force someone else to add it back in. Then we'll have 2
distributions of ld.so...ld.so-release, and ld.so-usable.
------------------------------------------------------------------
Jon Lewis <jlewis@fdt.net> | Unsolicited commercial e-mail will
Network Administrator | be proof-read for $199/hr.
________Finger jlewis@inorganic5.fdt.net for PGP public key_______