Summary from yesterdays fesco meeting

[Thorsten Leemhuis]

Am Samstag, den 18.02.2006, 12:59 -0500 schrieb Dan Williams:
> On Sat, 2006-02-18 at 15:34 +0100, Thorsten Leemhuis wrote:
> > Am Freitag, den 17.02.2006, 16:16 -0500 schrieb Dan Williams:
> > > On Fri, 2006-02-17 at 20:39 +0100, Thorsten Leemhuis wrote:
> > > >  * Kernel module standardization
> > > >   * Should archs be hardcoded with a "ExclusiveArch: i586 i686 x86_64
> > > > ppc" or similar entries? That's how it is done in beehive, but scop
> > > > doesn't like that idea to much. Warren will ask dcbw if there are
> > > > alternatives.
> > > 
> > > Warren poked me, here's my response:
> > > -------------------------------------------------------
> > >[...]
> > > a) Let packages do whatever the heck they want with their Exclusive,
> > > Exclude, BuildArch tags, including using %{ix86} as Mike suggests
> > > b) Have the buildsystem recognize kmod packages somehow (which we have
> > > to do anyway), then filter kmod packages through a "supported" list of
> > > sub-arches, including i586, i686, x86_64, ppc, athlon.  There's some
> > > support for this already in the buildsystem.
> > >[...]
> > > Let me know what you think.
> > Sound good to me. Any ideas what we need to achieve b) ? There is
> > nothing in the current kernel-module proposal that would help with that
> > (besides the "kmod" in the name).
> A custom specfile tag?  :)

:)

> Seriously though, if there are some simple rules for -kmod, like (these
> are just suggestions):
> 
> 1) The string '-kmod' MUST be the last part of the package name

That's the case already.

> 2) The package MUST BuildRequire the 'kernel' and 'kernel-devel'
> packages

Well, we of course BuildRequire 'kernel-devel' already -- mkodtool takes
care that there is a 
BuildRequire: kernel-devel-i686 = 2.6.14-1.1776_FC4
for example. But a generic 
BuildRequire: kernel-devel
in the static part of the SRPM shouldn't hurt afaics. 

Dan, just say what you prefer and we go for it. 

CU
thl
-- 
Thorsten Leemhuis <fedora at leemhuis.info>