On Wed, Mar 27, 2024 at 04:21:45PM +0000, Folker Schwesinger wrote: > Hi Conor and Dragan, > > thanks for your feedback! > > On Tue Mar 26, 2024 at 8:55 PM CET, Dragan Simic wrote: > > On 2024-03-26 20:46, Conor Dooley wrote: > > > On Tue, Mar 26, 2024 at 07:54:35PM +0100, Folker Schwesinger via B4 > > > Relay wrote: > > >> From: Folker Schwesinger <dev@xxxxxxxxxxxxxxxxxxxxx> > > >> > > >> Restore the behavior of the Rockchip kernel that undconditionally > > >> enables the internal strobe pulldown. > > > > > > What do you mean "restore the behaviour of the rockchip kernel"? Did > > > mainline behave the same as the rockchip kernel previously? If not, > > > using "restore" here is misleading. "Unconditionally" is also > > > incorrect, > > > because you have a property that disables it. > > Apologizes for the misleading commit message. Prior to 5.11 the Linux > kernel did not touch the pull-down registers. However, it seems the > register's (factory?) default was set to enable the pull-down. As it > was mentioned elsewhere that was the configuration recommended by > Rockchip. The 4.4 vendor (Rockchip) kernel reflects that by enabling the > pull-down in its kernel. Yeah, seems like a bit of a sticky situation. Probably the wrong polarity was chosen when the property was implemented and the property should have been the one you wanted to switch to given the default before it existed was the factory defaults. > Of course, this has nothing to do with the Linux kernel, so "restore" > was a bad choice here. > > I previously had split the driver patch into two separate patches, one > for changing the default (unconditionally at that point), the other for > adding the disable property. As both changes were minimal I decided to > squash the commits. I updated the cover letter, but forgot to update the > commit message. Sorry. No worries. Squashing them was probably the right thing to do anyway.
Attachment:
signature.asc
Description: PGP signature
