Messages in this thread | | | Date | Wed, 26 Feb 2014 20:41:35 +0900 | From | Mark Brown <> | Subject | Re: [PATCH V1 2/3] MFD: da9052: Add new DA9053 BC chip variant |
| |
On Wed, Feb 26, 2014 at 09:32:03AM +0000, Lee Jones wrote:
> > It can be worth doing anyway with a subsystem that's actively developed > > since sometimees the dependencies are the other way - the APIs in Linus' > > tree may have gone away.
> Then what happens if the tree that your patch finally gets sucked into > is pulled before the one you've written your code against? I'd say > basing your code on -next is generally a bad idea.
Submitters should defintely at the very least be checking against the subsystem tree (since that's what they're really submitting against) - it's always annoying when someone submits code that either doesn't apply or fails to build with the current tree. [unhandled content-type:application/pgp-signature] | |