On Tue, 13 Jun 2006 20:39:08 +0100
John Haxby <[EMAIL PROTECTED]> wrote:

> Lee Revell wrote:
> > I think it's a problem that FC5 does not have an RPM package containing
> > these modules.
> >   
> I think it's a question of cost of maintenance.   FC5 is currently on 
> 2.6.12.20 (although you might be forgiven fo thinking is actually 
> 2.6.16-1.2133_FC5).   /proc/asound says that we have 1.0.11rc2 which is 
> exactly what's in 2.6.16.   Someone could, I'm sure, package up a newer 
> version of ALSA and put in in a repo somewhere or even submit it to 
> Fedora Extras or Livna.
> 
> Of course, the difficulty with that is that replacing files in an RPM is 
> well-nigh impossible.   RPM simply doesn't allow you to corrupt an 
> installed RPM.    However, someone could take the kernel source RPM, 
> combine it with the latest ALSA build and create a new kernel RPM.
> 
> That's difficult though.   It'll break other stuff.
> 
> On the other hand, perhaps someone would volunteer to de-couple the ALSA 
> modules from the rest of the kernel in the kernel build so that the ALSA 
> modules can be updated and then we'll get to be able to install newer 
> versions of ALSA without waiting for Linus to pick up the next version 
> of ALSA.
> 
> Whatever happens, it's quite a lot of work for someone.  And while 
> there's a fair amount of noise on the alsa-user list there's a limited 
> audience for all that work.   Is it worth the hassle?   Put a bug in 
> bugzilla.redhat.com and see what happens.   Or even better patch the 
> kernel build and attach that to the bug report.   (Not you Lee, I know 
> you have better things to exercise your not inconsiderable talents on.)
> 
> The alternative for the snd-riptide module under Fedora is to wait for 
> 2.6.17 to finalise and then for FC5 to pick up that kernel.   We're on 
> 2.6.17-rc6 it can't be much longer before 2.6.17 final hits the streets 
> and then there'll be a QA delay before FC5 produces a 2.6.17-based RPM 
> and we should then be fine.
> 
> jch
> 

You are "aiming too low".

The true answers are:

1) drivers running in user space;
2) stable, documented and adhered to ABI, so a driver compiled
once by anybody for i386 (the biggest common denominator for IA31/IA64,
AMD/Intel) will run as binary image with any kernel of any IA31/IA64
distribution.


_______________________________________________
Alsa-user mailing list
Alsa-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/alsa-user

Reply via email to