| |||||||||||||
Forking the ARM kernel?Forking the ARM kernel?Posted Jun 3, 2011 14:21 UTC (Fri) by linusw (subscriber, #40300)Parent article: Forking the ARM kernel?
One thing that seems hard to counter is getting already forked drivers out of the kernel (c.f. drivers/mmc/msm_sdcc.* which is a clone of mmci.*), and feeling helpless about that whole business. I have toyed with the idea of proposing patches deleting the inferior forks, but that feels hostile, at the same time asking kindly top merge support into the existing driver doesn't seem to work either.
(Log in to post comments)
Forking the ARM kernel? Posted Jun 4, 2011 10:16 UTC (Sat) by arnd (subscriber, #8866) [Link]
Outright removing would be rather hostile indeed, but maybe moving these drivers to staging is an acceptable way. This way, the driver can still stay around until we're sure that there are no direct regressions and we can remove the staging driver once the main driver is known to be working correctly.
Forking the ARM kernel? Posted Jun 5, 2011 8:44 UTC (Sun) by linusw (subscriber, #40300) [Link]
That's a pretty good idea actually, I'll try that whenever I find some time!
Forking the ARM kernel? Posted Jun 9, 2011 21:36 UTC (Thu) by Wol (guest, #4433) [Link]
And if, having found a vendor who actually works with the kernel devs, you leave their driver in main and move competing drivers to staging regardless of quality (okay, if they work with the kernel devs chances are their drivers are best, anyway), you send a clear message to all the other companies that it pays to play nice ...
Cheers,
|
|||||||||||||