> Unquestionably, Linus is prodigious! Talented, hard-working, and a veritable
> goldmine of ingenuity and inventiveness. But hey....he's just one man...and
> he can only do so much. Just how many configurations he has access to, I've
> no idea, but certainly not _all_ of them. And testing is _quite_ time-
> consuming also.
Since I started this (I think) by making a statement which was later
misinterpreted, I'll make one last-ditch attempt to stop this useless
thread...again.
Every _DEVELOPER_ (not Linus!) should _try_ to check that each file
that they are working on (not the whole kernel!), can be compiled
with various combinations of configurations (CONFIG_*). The purpose
being to eliminate _silly_ compile time errors (like missing semi-
colons).
It is often _IMPOSSIBLE_ to thoroughly test all features of kernel
code, but that is what Beta testers are for. IMHO, no patch should be
sent to Linus if the developer has doubts about it. In this case,
posting it to the list and asking for comments on your EXPERIMENTAL
code seems appropriate to me. Even "experimental" kernels should
not be subjected to code which has never been tested by anyone IMHO.
If you have a big project with a lot of code, then it makes more sense to
setup your own mailing list for this purpose. _MANY_ sysadmins will give
you a mailing list for linux development if you just ask!
- --
Andrew E. Mileski
mailto:aem@ott.hookup.net http://www.redhat.com/~aem/
Linux Plug-and-Play Project Leader http://www.redhat.com/linux-info/pnp/
PGP Public Keys: mail with 'Subject: help' to pgp-public-keys@keys.pgp.net
-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
iQCVAwUBMf4YyHiRpcRMqyR5AQGsPwP9ERSK1Nv+5wANUJZ7AGATciavcy5nqgo2
gWGI+TblSgnC+RoL/P/o9HkeTUXcy5LYy5ATsBFFIGr7fWxLgcxGCeqJxMrUf7cI
YPDjCxt6OGSUcXKjIi4Y4Hj8sPXuGYCXq43SX/n6XjVhvD5D3moZw7zaH/Zcnufw
ZBtfI/n8hkU=
=8rBj
-----END PGP SIGNATURE-----