On Mon, May 5, 2014 at 3:51 AM, Peter Robinson <pbrobinson(a)gmail.com> wrote:
> I have had erratic results with recent Fedora 20 kernels and a
Beagle Bone
> Black, with some kernels failing to even boot. The latest kernel -
> 3.14.2-200.fc20.armv7hl - seems to be in generally good shape. Ethernet, USB
> and the XFCE desktop all seem to be working well. However, one regression
> seems to have occurred. With older kernels I got the following:
Interesting on the desktop side of things, I was certain I would need
a patch for the panel support for it to work. What xorg conf do you
have?
> modprobe cpufreq-cpu0
> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies
> 300000 600000 800000 1000000
>
> which I think is the correct set of frequencies for this board. Certainly
> the highest speed should be 1GHz. With3.14.2-200.fc20.armv7hl I get the
> following:
>
> modprobe cpufreq-cpu0
> cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies
> 275000 500000 600000 720000
Yes, in <= 3.13 we needed patches for the BBBlack but with 3.14 it
mostly works fine without but it seems the bits for 1ghz haven't
landed upstream. Can you file a RHBZ against the kernel for this
regression, reply here for the number and I'll sort out a patch.
On mainline it got held up behind the: OPP modifier
http://comments.gmane.org/gmane.linux.ports.arm.kernel/309466
Regards,
--
Robert Nelson
http://www.rcn-ee.com/