On Patchwork for managing linux-cxl

On Patchwork for managing linux-cxl

[Dan Williams]

At the beginning of the v6.1 development cycle I stepped back a bit from
the linux-cxl mailing list to tackle a long standing concern with the
way DAX page references were managed [1]. It was relevant for CXL since safe
/ managed CXL-hotplug needs the guarantee that device-removal properly
waits for all users of device's memory to either release their pins, or
otherwise assure the revocation of their access.

In the meantime the CXL patch backlog grew. I am circling back to it
now, but it strikes me that linux-cxl@vger.kernel.org is busier than
nvdimm@lists.linux.dev. It is busier to the point where it likely needs
more formal tracking and patch status visibility for contributors.

To that end I am going to start using the CXL Patchwork instance beyond
the automation that tracks superseding and application to cxl/next, but
this will be a "learn as we go" type process.

The current accounts marked as "maintainers" are visible on the profile
page [2]. If you want your account added to that list and you have a
"Reviewed-by:" credit in drivers/cxl/ just let me know and I'll add you.
That mainly gives you the right to change the status of a patch and mark
yourself (or be marked by others) as the delegate for a given series.

There is some cleanup needed, the ndctl patches currently only get
automation support on the nvdimm patchwork instance, so there are
patches that are merged upstream that are still shown in the 'New'
state. The qemu patches I have marked as "Handled Elsewhere" for now,
but if it would help to track their status on the CXL patchwork just let
me know.

[1]: https://lore.kernel.org/linux-mm/166329930818.2786261.6086109734008025807.stgit@dwillia2-xfh.jf.intel.com/
[2]: https://patchwork.kernel.org/project/cxl/