Re: [GIT PULL] omap fixes for v2.6.32-rc3

[Felipe Contreras <felipe.contreras@gmail.com>]

On Fri, Oct 9, 2009 at 5:48 PM, Greg KH <gregkh@suse.de> wrote:
> On Fri, Oct 09, 2009 at 03:08:00PM +0300, Felipe Contreras wrote:
>> On Fri, Oct 9, 2009 at 2:26 AM, Tony Lindgren <tony@atomide.com> wrote:
>> > * Felipe Contreras <felipe.contreras@gmail.com> [091008 15:20]:
>>
>> <snip/>
>>
>> >> > Here you can see Uwe Kleine-K??nig sending the original patch without
>> >> > even CC'ing the mmc list:
>> >> > http://marc.info/?l=linux-kernel&m=124820861213849&w=2
>> >> >
>> >> > And it was unintentionally broken by this one:
>> >> > http://marc.info/?l=linux-mmc&m=125249763422227&w=2
>> >>
>> >> Er, actually that one was ok, the problem was introduced only in the
>> >> final version:
>> >> http://marc.info/?l=linux-mmc&m=125366315417006&w=2
>> >
>> > Right, looks like Roger forgot to Cc linux-mmc.
>> >
>> >> > I don't see what we are waiting for, the code is clearly broken, even
>> >> > the compiler warns that nobody is using omap_hsmmc_probe().
>> >
>> > I've sent Roger's patch to Andrew & linux-mmc so hopefully it will get
>> > merged soon:
>> >
>> > http://www.mail-archive.com/linux-mmc@vger.kernel.org/msg00539.html
>>
>> I don't see what's specific about mmc in that patch, nor omap. It's
>> just a fix for an obvious mistake that must be picked ASAP and can't
>> possibly be NAK'ed.
>>
>> There's a tree for trivial fixes:
>> git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial
>>
>> And one for includecheck:
>> git://git.kernel.org/pub/scm/linux/kernel/git/jaswinder/linux-2.6
>>
>> What's the procedure for obvious fixes? I'm CC'ing Greg KH and Andrew
>> Morton because I think they might interested in fixing these kinds of
>> issues quickly in the future.
>
> What kind of issues?  Why would I be interested, where is the problem?

The obvious-brain-dead-duh kind of issues, like this one.

The problem is that OMAP devices (like beagleboard) are not booting
correctly right now because of a wrong merge. It has been identified,
tested, and acked, but nobody has picked it up for a pull request, so
it's not clear it will be on -rc4.

To me it's not clear who should push the patch, it seems it doesn't
belong on linux-omap, so Tony is pushing it through linux-mmc, which I
don't think is the right place. They should be handling mmc-related
issues, not obvious breakage.

In order to keep the engines oiled I think there must be a process to
flag obvious generic breakage so it's immediately picked; maybe an
'obvious-fixes' tree, or a 'simple-fixes' that has 'trivial',
'includecheck', and similar, or maybe nothing needs to be done. Up to
you to decide.

Cheers.

-- 
Felipe Contreras