Re: [PATCH 5.1 003/115] HID: input: make sure the wheel high resolution multiplier is set

From: Greg Kroah-Hartman
Date: Tue Jun 18 2019 - 13:55:03 EST

On Tue, Jun 18, 2019 at 11:22:55AM -0600, James Feeney wrote:
> Uhm - could someone please "clue me in" here?
> When I look into:
> 'move all the pending queues back to their "real" places'

No need to worry about that patch, that was done because my scripts
normally assume specific directory locations of the patch queues, and I
had to do a kernel release that did not include the existing pending

> I can find both the "d43c17ead879ba7c076dc2f5fd80cd76047c9ff4" patch, "HID: input: make sure the wheel high resolution multiplier is set" and the "39b3c3a5fbc5d744114e497d35bf0c12f798c134" patch, "HID: input: fix assignment of .value".
> I take this to mean that these patches are "in the stable-queue". But then, these patches are not "in the kernel".


> So then, how do these patches go from being "in the stable-queue" to being "in the kernel"?

I apply them when I do the release in a few hours/days.

> To the "uninitiated" and "naive", as I am, to outward appearances, the
> patches are "just sitting there". How do the patches get selected for
> inclusion into the "next" kernel revision?

I already selected them, sent emails saying they were selected and to
what specific branches they were selected to. Then when the -rc
releases happen so that people can do one final round of testing and
object if I messed anything up, they get sent out again (which you
responded to here.)

If all goes well, when the "deadling" passes (usually 2 days +-2 days
depending on stuff), I'll do a realease and apply the patches "for real"
to the different kernel branches and cut a release.

Then I start all over again...

I understand that seeing a git tree of patches in a quilt series is odd,
but it is very powerful and works very very well for what we do here.

Does that help?

greg k-h