@kernellogger Oh yeah, I know. It may very well have been fixed and not backported. (As an aside, it bugs me when people* report a bug & then don't push the backports through where they need to be rebased)
I empathise with Greg pushing back on people, as it's far from good use of maintainer time to go figure it out, when the reporter has the environment needed to test a more recently kernel sitting in front of them.
I do at the same time get the flip side of it being frustrating to deal with the gap between a vendor tree & mainline - 5.15 being a pretty good choice as they go..
*by people, I mean active kernel developers