On 19.02.2009 14:38, seth vidal wrote:
On Thu, 2009-02-19 at 06:38 +0100, Kevin Kofler wrote:
> Colin Walters wrote:
>> I forgot to point out in an earlier message that there is precedent
>> for setting up barriers for low-level software, namely the Fedora 3rd
>> party kernel driver policy:
>>
https://fedoraproject.org/wiki/KernelDriverPolicy
> I actually still don't see why we're banning separate kernel module packages
> even if they're under a GPLv2-compatible license. Useful stuff like kqemu
> with no license or patent issues is being forced into RPM Fusion purely
> because of the combination of the above policy and the ban on
> separately-packaged kernel modules. This also causes problems with getting
> the modules out at the same time as the new kernel which would just not
> happen if they were in Fedora and got pushed out with new kernels in
> grouped updates. The #1 rationale for banning kernel modules entirely
> was "FESCo was rejecting most requests anyway, so it won't change
much",
> but the option of changing that fact, or even getting rid of the FESCo
> approval requirement entirely, never even got considered.
I think they're banned b/c a kernel module package means we're not doing
the proper job of getting the module in upstream.
And because maintaining kernel modules we want to ship(¹) as patch
within the regular Fedora kernel SRPM is likely easier and better for
everyone (²).
CU
knurd
(¹) there are a lot external GPLed modules we don't want to ship for
various reasons
(²) that's the long story short and without all those boring details;
and of course having modules upstream is the best for everyone