On Wednesday 18 April 2012, Mark Brown wrote: > On Wed, Apr 18, 2012 at 11:00:39AM +0000, Arnd Bergmann wrote: > > > 1. develop on -rc > > 2. merge with latest -next, test and make sure it works there > > 3. submit for review against -rc > > 4. have patches included in -next once reviewed, but based on -rc > > 5. when merge window opens, have patches sent for upstream inclusion > > Steps 3 and 4 should be to submit against whatever branch is appropriate > for the subsystem and driver - if people follow this process they're > going to get bounced back by a fair proportion of maintainers, -rc isn't > universally what people are looking for so people should be aware that > they need to pay attention here. > > Generally I'd say the development version is a safer bet than -rc for > most subsystems. Right. The description above was mostly done for the lpc32xx case, which is going to get merged through the arm-soc tree and that doesn't have a single development branch but instead has lots of them. For subsystems that have just one branch, I agree that it makes sense to develop against that one. Also for arm-soc, it can make sense to base on one of the existing branches, but I prefer the default to be to base on the -rc release so I can mix and match incoming branches as needed. Arnd -- To unsubscribe from this list: send the line "unsubscribe linux-input" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
