Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 25 Feb 2020 09:21:22 -0700
From:      Ian Lepore <ian@freebsd.org>
To:        mike@karels.net, freebsd-stable@freebsd.org
Subject:   Re: [FreeBSD-Announce] FreeBSD 12.0 end-of-life
Message-ID:  <1ff5990a0ee5c40a1f6fb693083b2eb96028cfdc.camel@freebsd.org>
In-Reply-To: <202002250101.01P11ncG090399@mail.karels.net>
References:  <202002250101.01P11ncG090399@mail.karels.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 2020-02-24 at 19:01 -0600, Mike Karels wrote:
> > From mike  Sun Feb 23 17:24:54 2020
>                                           
> 
> Gerard E. Seibert wrote:
> > On Sun, 23 Feb 2020 16:27:40 -0600, Mike Karels stated:
> > > In this case (a USB failure), I bisected the problem some months ago.
> > > The offending commit was an ACPI update.  I have not yet "downgraded"
> > > to 11.3, but I will when I have enough time.
> > Obviously, I am not an expert here, but why can't the update to ACPI be
> > reversed and why is it only affecting some systems. And why do you have
> > to downgrade to 11.3? What are you running now?
> 
> The update to ACPI was pulled from upstream, and IIRC it included multiple
> changes.  There have been many updates to ACPI since.  I don't have any
> idea which part of the change caused the problem, or what it does; no idea
> how it affects USB on some systems.  I'm running 12.1 now, but USB is
> problematical (produces an error message once a second, and my UPS control
> doesn't seem to work).
> 
> 		Mike
> 

There may have been two changes working together to create the problem.
At some time last year, changes were made to the usb drivers to be
sensitive to acpi data of some sort (maybe something about hubs, I
forget the details).  That alone may have been fine on most systems,
but then a later update of acpi code started causing problems which in
the past might not have affected usb but now does.

It's just a vague theory, based on my memory of some change happening
last year that injected awareness of acpi data into the usb driver
subsystem.  I wonder if it would be possible to add some sort of
tunable to disable that awareness, if nothing else just to see if it's
the cause of any usb trouble.

-- Ian





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1ff5990a0ee5c40a1f6fb693083b2eb96028cfdc.camel>