On Fri, Jul 23, 2021 at 02:54:54PM -0500, Bjorn Andersson wrote:

On Fri 23 Jul 10:54 CDT 2021, Greg Kroah-Hartman wrote:

quoted

On Fri, Jul 23, 2021 at 09:34:20AM -0500, Bjorn Andersson wrote:

quoted

On Fri 23 Jul 03:18 CDT 2021, Greg Kroah-Hartman wrote:

quoted

On Wed, Jul 21, 2021 at 03:09:37PM -0500, Bjorn Andersson wrote:

quoted

Which tree did you revert this in? 5.13.stable?)

My usb-linus branch which will go to Linus later today.  Then we can
backport the revert to older kernels as needed.

I'm not worried about the backports, I'm worried about conflicts you're
causing because you're taking a non-usb patch through the usb tree.

I was about to push a revert (to this and the other Qualcomm platforms),
but as you're taking some set of reverts through the usb tree we're just
in for a bunch of merge conflicts.

It shouldn't be a merge conflict as you can apply the same revert to
your tree now and keep on merging.  When you pick up 5.14-rc3 from Linus
it should merge "correctly", right?

I typically don't merge back the -rcs into my -next branch, is that
common practice?

I do it when Linus takes patches from my -linus branch in order to
ensure they end up in my -next branch for testing and merge issues.

But I still don't understand why you insist on driving this through your
tree. I've asked you several times to show me on the patch so I at least
can Ack it. I made a mistake, but why do you insist on keeping me - the
maintainer - out of the loop?

I had already done the revert, I wasn't trying to keep anyone out of the
loop here, sorry if it came across that way.  I just wanted to ensure
this got resolved quickly so I could move on to other issues.

This is now 1f958f3dff42 ("Revert "arm64: dts: qcom: Harmonize DWC USB3
DT nodes name"") in Linus's tree if you wish to cherry-pick it into your
tree to resolve merge issues, sorry for the confusion.

thanks,

greg k-h