Kernel

[Jan Rekorajski]

[poniedziałek, 23 kwiecień 2001], Blues napisał(a):

> On Mon, 23 Apr 2001, Jan Rekorajski wrote:
> > > > May I add to the kernel.spec generating of the package from pure Linus
> > > > sources. I like the idea of kernel from rpm with initrd and so on but
> > > > sometimes the PLD kernel does't work. It would be the same phylosophy
> > > > but without all this patches.
> > > I've got better idea (I think...)
> > > Why not to add all of the patches to bcond?
> > > If it would be build without any options -> kernel with all of the patches
> > > --without patch  -> we have clean kernel.
> > And totally messy and unmaintainable spec. No.
> 
> Why?
> bcond's will be only in patch section.
> %{!?bcond_off_patch:patch0}
> %{!?bcond_off_patch:patch1}
> %{!?bcond_off_patch:patch2}
> %{!?bcond_off_patch:patch3}
> %{!?bcond_off_patch:patch4}
> %{!?bcond_off_patch:patch5}
> %{!?bcond_off_patch:patch6}
> ....
> 
> is it so difficult to maintain this?

It's unreadable, don't forget that there are also sources and nasty hacks.
Try it, it will be a mess. And remember who's maintaining this package ;>

Maybe if you do it the way test_build is done on LINUX_2_4 branch, then
I might agree.

Janek
-- 
Jan Rękorajski            |  ALL SUSPECTS ARE GUILTY. PERIOD!
baggins<at>mimuw.edu.pl   |  OTHERWISE THEY WOULDN'T BE SUSPECTS, WOULD THEY?
BOFH, MANIAC              |                   -- TROOPS by Kevin Rubio