[arch-dev-public] Driver Maintainers

[Roman Kyrylych]

2007/5/12, Aaron Griffin <aaronmgriffin at gmail.com>:
> On 5/11/07, James <iphitus at gmail.com> wrote:
> > On 5/12/07, Aaron Griffin <aaronmgriffin at gmail.com> wrote:
> > > On 5/11/07, Roman Kyrylych <roman.kyrylych at gmail.com> wrote:
> > > > 2007/5/11, Aaron Griffin <aaronmgriffin at gmail.com>:
> > > > > To get the ball rolling, perhaps we could start a simple wiki page
> > > > > where a user can come in and list the hardware and arch packages they
> > > > > use that may be unique - we can figure out a plan of attack while
> > > > > gathering this data.
> > > >
> > > > Yeah, I've brought this on some time ago too after finding that dead
> > > > page on devwiki.
> > > > Also, it will greatly help in cases when confirmation of a bug is needed.
> > >
> > > Started - as it says, it's a stub.  I just threw out ideas - no need
> > > to discuss it, if you think my ideas are dumb, go ahead and change it.
> > >  I'll do some procmail magic tonight and start sending mails to myself
> > > with this.
> > >
> > > http://wiki.archlinux.org/index.php/Driver_Testing
> >
> > Oh cool :)
> >
> > How about we take the suggestion a little bit further, create an
> > [arch-testing] mailing list, where we can send a mail, so that people
> > who use a package, can know it's in testing.
> >
> > Currently there's no notification if something goes into testing, so
> > it nearly always goes in without people knowing. If people knew, then
> > they'd be able to selectively take things from testing, pacman -S
> > testing/package and give us feedback.
>
> That would be useful, and probably very easy to do (arch-commits does
> the exact same thing, but doesn't respect tags... I'd have to figure
> out how to handle that...)
>
> > If not, we could at least extend that wiki page to more than drivers.
>
> The problem is that hardware related stuff has it's own separate set
> of kinks, to the point where I'd like to see some extra level of
> information here.  For example, the new xf86-video-intel may work on
> most cards, but tester Joe has an old card that seems to segfault.
>
> Package notification for testing though is a good idea, but I think
> the actual hardware issue is a different beast, and it'd be nice to
> know who is testing and what hardware they have.
>

How about extending that page for purposes of kernel package testing as well?
This will include chipset, built-in sound & network and PCI, PCI-X
(mostly RAID controllers) and ISA cards that all are supported by
modules included in kernel package.

-- 
Roman Kyrylych (Роман Кирилич)