Re: HEADSUP: cpufreq import complete, acpi_throttling changed

From: Eric Kjeldergaard <kjelderg_at_gmail.com>
Date: Tue, 1 Mar 2005 15:34:24 -0600
On Mon, 21 Feb 2005 17:49:04 -0800, Nate Lawson <nate_at_root.org> wrote:
> Eric Kjeldergaard wrote:
> > On Sun, 06 Feb 2005 13:21:32 -0800, Nate Lawson <nate_at_root.org> wrote:
> >
> >>I've finished the major work of importing cpufreq.  As part of this, the
> >>sysctls for acpi throttling have been removed.  The power_profile script
> >>has been updated, so you can use performance/economy_cpu_freq= in
> >>rc.conf to set AC on/offline cpu frequencies.  The acpi throttling
> >>support has been compiled into acpi_perf.ko so load that to get
> >>throttling.  Do a sysctl dev.cpu to get an understanding of the cpufreq
> >>sysctls.
> >
> >
> > I have been using cpufreq on and off since this got comitted and just
> > noticed a problem now.  I can't really say when it started happening,
> > but I can say that it happens with a current from about 2 hours ago.
> > What happens is when I kldload cpufreq, usb stops working.  If I have
> > a device that was already connected, it gets disconnected, I am unable
> > to connect any new devices.  It's just completely unresponsive.
> > Whatever the issue is, it does not help to kldunload cpufreq, I am
> > forced to reboot to get USB working again.  The system is an IBM
> > Thinkpad R40.  I can give you whatever relevant info you need on
> > request.
> 
> Hmm.  If you boot with cpufreq and usb loaded, does it have this
> problem?  Run the below script to send me your debug info:
> 
> http://www.root.org/~nate/freebsd/cpufreq_debug
> 

Finally got a chance to test again.  It seems to happen regardless of
whether cpufreq is loaded at boot or dynamically during runtime.  The
output of your script is attached.


-- 
If I write a signature, my emails will appear more personalised.

Received on Tue Mar 01 2005 - 20:34:59 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:29 UTC