From: Marcus Glocker <marcus@nazgul.ch>
Subject: Re: smmu(4) on QC Laptops
To: "Kirill A. Korinsky" <kirill@korins.ky>
Cc: Patrick Wildt <patrick@blueri.se>,  Mark Kettenis <mark.kettenis@xs4all.nl>, tech@openbsd.org, tobhe@openbsd.org, mlarkin@openbsd.org
Date: Sat, 2 Aug 2025 07:25:51 +0200

On Sat, Aug 02, 2025 at 06:30:30AM +0200, Marcus Glocker wrote:

> On Sat, Aug 02, 2025 at 12:38:31AM +0200, Kirill A. Korinsky wrote:
> 
> > On Fri, 01 Aug 2025 22:21:21 +0200,
> > Marcus Glocker <marcus@nazgul.ch> wrote:
> > > 
> > > @@ -315,10 +315,9 @@ softraid0 at root
> > >  scsibus2 at softraid0: 256 targets
> > >  root on sd0a (13f744812d2a8646.a) swap on sd0b dump on sd0b
> > >  qcpas0: failed to shutdown lite firmware
> > > -qwz0: wcn7850 hw2.0 fw 0x100301e1 address 00:03:7f:12:50:a1
> > > +qwz0: wcn7850 hw2.0 fw 0x100301e1 address 00:03:7f:12:73:e2
> > 
> > Sorry to ask, but how might this diff affect the hardware address?
> 
> This has nothing to do with Patrick's smmu diff.  I should have
> mentioned this.
> 
> I'm currently playing around with qwz(4), and firmware loading, and for
> some reason the last two digits of the qwz(4) hardware address are
> changing on every reboot, although I'm using the same firmware.  I
> haven't figured out yet why this happens ...

Ah, I think those devices support a "random hardware address" feature,
is what I can see in the Windows driver at least ...