wireless project status report

Hi,

first of all thanks to everyone who reported back (PR, here or private)
that the latest round of getting stability changes in had helped them.
Special thanks to people who took them and merged them privately to
older branches or who closed PRs themselves no longer being able to
reproduce problems!

I kept debugging 11n.  In order to get some extra information I turned
debugfs on for iwlwifi which was leading me straight into panics (great
start; we had to debug this last year before);  traced the LinuxKPI
problems and opened a review (D45755).

I also finished locally all the vendor imports and sent the summary off
for review;  I ended up redoing the ath1*k bits and split them up to
allow individual driver updates more easily in the future.


The last days I had a need for some time AFK which also freed the mind a
bit and slowed things down.

Today I decided to start clearing up bits I can push without interfering
or depending on the driver updates being pushed.

Small bits went into review the last hours. Having them out of my tree
will also help seeing other changes without noise in between some better.

One starting stack of changes (there are more to come for that stack) are
related to [contig]malloc(9) and LinuxKPI.  The first change I put up was
my initial "test pilot" chnging skbuff;  hopefully this will also allow us
soon to get rid of the TUNABLE needed for rtw88/rtw89 and possibly
others and let the drivers lower the limit themselves as needed
(a bit less hassle and less chances for missing it for everyone).
In case anyone wants to try them, they are D45812 and D45813 on reviews.

There's also D45756 which is a change to "full offload scans" which may
interest people who want to give it a try.

More next week.  Lots of health,
Bjoern

-- 
Bjoern A. Zeeb                                                     r15:7