On Thu, Jun 29, 2023 at 10:00 AM Mark Brown <broonie@xxxxxxxxxx> wrote: > > On Thu, Jun 29, 2023 at 04:51:34PM +0100, Lee Jones wrote: > > On Thu, 29 Jun 2023, Mark Brown wrote: > > > > The thing that's causing a lot of the issues here is that you're only > > > applying the serieses en masse, blocking things on getting absolutely > > > everything lined up (including this time over a merge window). I really > > > don't understand why you feel you're forced to batch everything together > > > like this. > > > https://lore.kernel.org/all/CAL_Jsq+Z64tuMO8a2Y=2GrXZ8q0L4Z2avCiphsn0HOOC71Dzjg@xxxxxxxxxxxxxx/ > > That says it's a bit unusual to use a separate branch with a PR, it > doesn't say anything about it being the end of the world to pick up > parts of a series that are ready without picking up the whole lot, > especially when we're getting to the merge window. There's some risk the core part could change affecting the sub components after you pick it up, or that the author just abandons getting the core part upstream and we're left with a dead driver. I'm fine if the sub-components are picked up first (and land in next first), but the MFD binding picked up first is a problem usually. And that has happened multiple times. Also, it does mean the MFD branch will have binding warnings, but that's Lee's problem if he cares (probably not). Rob
