Date: Tue, 21 May 2024 21:33:28 +0000 (UTC) From: "Bjoern A. Zeeb" <bz@FreeBSD.org> To: freebsd-wireless@freebsd.org Subject: wireless porject status report Message-ID: <1284rs3q-n083-9q42-67n7-02994p1nr3o8@SerrOFQ.bet>
next in thread | raw e-mail | index | archive | help
XB!
Hello everyone,
yesterday was a public holiday, so update on Tuesday.
- Finished updating drivers to Linux 6.9.
And implemented the missing code (I could spot) for LinuxKPI 802.11.
- Finished updating the fwget/ports updates as a follow-up.
+ I have not pushed any of that yet.
- Just got review for the latest ports bits so firmware can finally go
in, fwget will follow (at this point likely with or without an updated
review) [D44945, D44918]
- I have review for one of the two non-80211 LinuxKPI changes I had put
out [D45181, D45180].
- I'll likely put the chanctx one in [D45185] even if there's no review
on LinuxKPI 802.11 code much anymore.
- Dealt with changes from others; also for 802.11.
- I prepared a vendor tree for vendor/Linux/iwlwifi to match our
previous imports; I sent out email but go not feedback back so far;
I'll likely follow-up on git@ again as sorting this out is one of the
major blocks of putting the update into main now.
- I also believe I found a solution for the mac80211 suspend/resume
problem which can be implemented without massively changing each
(Linux based) driver or the framework ("maintainance costs") but may
even help us to bring in new features longer-term from the LinuxKPI
compat layer into net80211 (WoWLAN for example). Testing still needs
to happen if the proof of concept would work but it requires us
writing some new code.
- I also believe to have a solution for PR 275255 ; I tested the updated
code with modern chipsets and got it working there again without
firmware crashes. I was about to test an 8xxx card (in an arm64
machine) as I ... (see next item); I'll send a cleaned up version for
main and stable/14 out on the PR hopefully tomorrow to test.
- found some problems with omissions in LinuxKPI which hit on arm64 and
started to fix them only to hit the nseg=1 busdma bounce problem
again. Sent out emails to parties talked to in the past with the
goal to finally (after 2-ish years) find a solution which will work
for everyone and could go into the tree. I believe people also hit
the problem with native code outside of wireless (after reading up
in my archives). I know there are people out there looking for
wireless STA support on arm64 so 'watch this space -- or your inbox'.
- I have not done any further PR bashing the week trying to sort out
some of the old stuff in the database.
I'll likely work may way backwards through this stack; once that is all
sorted I can hopefully also finally go back to work on the "new" stuff
everyone wants.
/bz
--
Bjoern A. Zeeb r15:7
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1284rs3q-n083-9q42-67n7-02994p1nr3o8>