Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 22 Apr 2024 22:27:13 +0000 (UTC)
From:      "Bjoern A. Zeeb" <bz@FreeBSD.org>
To:        FreeBSD wireless mailing list <wireless@freebsd.org>
Subject:   wireless status report
Message-ID:  <15s1sn63-r54r-s128-212s-6933s78s7745@SerrOFQ.bet>

next in thread | raw e-mail | index | archive | help
Hello everyone,

one line summary:
lots of things open to finish up before going back to other code work.



Some of this was in previous emails during the week:

- the github branch got pushed and iwlwifi d3 build was enabled.
- I also just merged all LinuxKPI prerequisites to stable/14 to be there
   before 14.1-R starts and to avoid many MFCs later.  I will not push
   the driver updates for 14.1.

If you want to/can test the github branch on main please checkout
https://github.com/freebsd/wireless/tree/bz_wifi_drv_update

In case you would love to test on stable/14 please drop me an email and
I'll create a branch for that with the changes.

- I not spent any more time on the USB debug console (still haven't
   recovered my HP xhci root bridge either).

- Looked into suspend/resume:
   (1) what we do in net80211 is not "state save" so similar problems to
   what we've been working around that firmware state gets left behind.
   Same problem which went unaddressed for 10-15-ish years and along with
   other firmware related bits need to be sorted soon.
   (2) in FreeBSD to all I know we only have the "bus paths" for
   suspend/resume where-as I see more hooks in the Linux drivers.

   I have a crude hack for a prototype very specific to iwlwifi to "fish
   out" the 802.11 structures and call into that from the PCI path but
   that's nothing which is really generally doable so I keep pondering
   alternate code paths and was trying to figure out if we can hook them
   up internally (if we do have enough compat glue).  It is both the code
   path but also access to the right "argument" then which is
   problematic.

   Also thanks to Jonathan Vasquez  for testing the up/down path but as it
   stands taking down the VAP (wlan interface destroy) and re-creating that
   seems to be the only option for now.

- Started to do an internal amd64 release build as to add firmware to
   the install media so WiFi can come up even if the firmware is not in
   src;  this also triggered me at looking to update the ports and fwget
   for when the driver updates will go in.  Doing a full matrix of which
   FreeBSD version needs which firmware version from ports long-term is
   going to be a bit support nightmare but we'll see.

- Spent some time on various reviews in Phabricator, also LinuxKPI.

- Started to update/comment on the last remaining reviews from Februar
   D43966, D43967 in order to sort these out.

- stable/13 did not get the "EPOCH" changes before 13.3-R and I am
   trying to figure out if they are save to MFC or parts of them as
   they already create conflicts when merging newer bits.  But I hope
   that apart from that I'll manage to get stable/13 aligned with 14 and
   main during the week as well.

- Some personal Email handling, some of which now also ended up (or more
   hopefully will) on already open PRs if they are not "fixable" right
   away.  One related to 8xxx/9xxx cards which still have a "takedown"
   problem it seems.


- no news on vendor imports

- Mark offered to have a look at the missing VM/Linux Page changes
   needed to move mt76 forward some time the next weeks.


Lots of health,
Bjoern

-- 
Bjoern A. Zeeb                                                     r15:7



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?15s1sn63-r54r-s128-212s-6933s78s7745>