[Server-devel] [XSCE] Re: a kernel for our solidrun protos?

Thu Jan 2 00:00:43 EST 2014

You could probably guide me a little as to what info, tests, etc would be
helpful to you.

Right now XSCE server software does not load, because some of the kernel
networking userland (rtnetfilter) interfaces are not enabled.  I probably
will need to do more of that sort of thing to get hostapd, and
infrastructure mode running.  But I know very little about the things you
might be interested in.

On Wed, Jan 1, 2014 at 11:17 PM, T Gillett <tgillett at gmail.com> wrote:

> Hi George et al
>
> I am interested to know details of the on board wifi sub-system in the
> Cubox.
>
> I got some general info on it from the forum a while ago, but not exactly
> would be used in production.
>
> If you gather any detailed information on it as you work with the
> prototypes it would be good if you can share it here.
>
> Thanks
> Terry
>
>
> On Thu, Jan 2, 2014 at 6:03 AM, George Hunt <georgejhunt at gmail.com> wrote:
>
>> Hi Jon,
>>
>> On our skype call yesterday, I learned that Tim probably will not get his
>> Cubox's in time (we need at least 2 for the Cubox to be a viable test
>> case). So my sense of urgency is much diminished.
>>
>> Progress report -- mostly as a log because the hardware seems really
>> promising and I expect to be coming back to  Cubox later in the year. When
>> I was doing yum updates of the rootfs, top reported 100% utilization for
>> the yum task, but also 75% idle on the summary panel.
>>
>> I was abe to get uEnv.txt to change the commandline for the kernel, as
>> verified by "cat  /proc/cmdline".  But still could not get the frame buffer
>> and XFCE desktop settings to work together for the GUI console (640x480
>> resolution was usable, but a little awkward). I did a little googling, and
>> after trying your video command line args, I tried
>> "video=mxcfb0:dev=hdmi,1024x768M at 60m" . I gave up on this task because
>> ssh into the device is what I usually do anyway.
>>
>> On the Cubox, I groupinstalled "Development Tools", and compiled the
>> modeules, and did a "make install". This operation did not set up
>> /lib/modules/<kernelid> properly, so I reached the extent of my comfort
>> area. I was not able to get the esata enclosure (startech model:sat2510u2e)
>> to be recognized. No idea of how much word is involved there.
>>
>> I could go back and try to get Rabeeh's kernel to load and run. It's
>> about 30Mb, whereas mine turned out to be only 4MB or so.  I guess he put
>> everything in the kernel.  I suspect the the load adress needs to be
>> jiggled for his kernel to run.
>>
>> I used uEnv.txt to modify the u-boot environment variable: mmcargs, and
>> mount an external usb SATA drive as the rootfs. (so the load process took
>> the kernel from mmcblk0p1, and mounted /dev/sda1 as /.
>>
>> As I was trying to load the XSCE software, I found that I needed a second
>> ethernet adapter, but neither the Cubox, or the trimslice which I use as
>> tool, seems to have drivers for the AX88772 USB to 10/100 Fast
>> Ethernet/HomePNA/HomePlug controller. I'm sure there are other adapters
>> which the kernel can handle, so I stopped on this track.
>>
>> In a way, I'm relieved that all of this doesn't need to be done and
>> tested before we leave on Jan 17.  But I'm hopeful about Cubox.  I'm just
>> not sure about the industrial design, the packaging, the way to make the
>> interconnects reliable and robust.  I found the console usb serial
>> connector interferes with the USB connector above it, and makes the console
>> connection problematic.
>>
>> We'll need to find lots of solutions before we can have 100's of these in
>> remote places.
>>
>> George
>>
>>
>>
>>
>>
>> On Tue, Dec 31, 2013 at 11:43 AM, Jon Nettleton <jon.nettleton at gmail.com>wrote:
>>
>>> Okay sounds like progress and we just need some fine tuning.  Freescale
>>> just release their 3.10.17-beta source today and I have ported my work over
>>> and fixed a few bugs along the way.  You can force the framebuffer
>>> resolution with a video=mxcfb,1280x720 at 60 or whatever resolution you
>>> want on the kernel commandline.
>>>
>>> Unfortunately the rest of my attention for the night is being taken by
>>> my spinach and artichoke dip, and shrimp and pineapple wrapped with bacon.
>>>  The New Year's festivities are in full effect here in Europe already.
>>>  Ping me tomorrow and I will have time to get you booted and everything
>>> running.
>>>
>>> Happy New Years
>>>
>>> -Jon
>>>
>>>
>>> On Tue, Dec 31, 2013 at 5:31 PM, George Hunt <georgejhunt at gmail.com>wrote:
>>>
>>>> Jon,
>>>>
>>>> Got the rootfs up and running.  The problem I was having last night
>>>> turned out to be stale and incorrect UUID's in fstab.
>>>>
>>>> Now because I don't have hdmi monitor, (using hdmi to vga) I need to
>>>> specify the monitor settings available to X11. The current default of
>>>> 640x400 doesn't work we with many of the screens in xfce, which is the
>>>> default for the rootfs I grabbed.
>>>>
>>>> Also I'm not recognizing the esata  enclosure/drive. I think I should
>>>> switch to the kernel that Rabeeh made available as the hummingbird def may
>>>> not have included drivers required.
>>>>
>>>>
>>>> On Mon, Dec 30, 2013 at 11:29 PM, George Hunt <georgejhunt at gmail.com>wrote:
>>>>
>>>>> After a few false starts:
>>>>>
>>>>>    - Putting the microsd card upside down does not work.
>>>>>    - The microusb console connector is really intermittent (does not
>>>>>    seat well with my connector body).
>>>>>    - My hdmi to vga adapter causes the monitor to jitter
>>>>>
>>>>> I was not able to get the stock u-boot that Rabeeh provides to boot.
>>>>> I downloaded the git repo, and selected the i4 make version.
>>>>>
>>>>> Then for drill, I compiled his git repo kernel, and put it on
>>>>> partition 1. I untared the rootfs onto a second partition.
>>>>>
>>>>> At this point it's not clear that the kernel is able to get into the
>>>>> rootfs, though the console messages say that it has discovered the mmcblk
>>>>> device, and that it has mmcblk0p1, and mmcblk0p2.
>>>>>
>>>>> I may need to change the kernel command line (I got to this point
>>>>> without any modification of partition 1).
>>>>>
>>>>> More play tomorrow
>>>>>
>>>>>
>>>>>
>>>>> On Sun, Dec 29, 2013 at 10:45 AM, Jon Nettleton <
>>>>> jon.nettleton at gmail.com> wrote:
>>>>>
>>>>>> George,
>>>>>>
>>>>>> Start out running the u-boot and kernel that Rabeeh has provided,
>>>>>> that is based on the current "stable" kernel sources provided by Freescale.
>>>>>>  I have a 3.10 based kernel that is more performant but I have not run it
>>>>>> on the i4 hardware yet.  Getting a baseline on the standard we can then
>>>>>> better diagnose bugs or performance issues one way or the other with the
>>>>>> new kernel.  For the second partition feel free to untar the F18 armhfp
>>>>>> base image.  Changing the kernel command line can be done using a uEnv.txt
>>>>>> file on the first partition of the SDHC port.  If we need anything more
>>>>>> advanced I can write a u-boot boot script, or custom u-boot to handle
>>>>>> loading things.  We may want a "fallback" kernel or something similar.
>>>>>>
>>>>>> Let me know if you have any questions or problems.
>>>>>>
>>>>>> Jon
>>>>>>
>>>>>
>>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.laptop.org/pipermail/server-devel/attachments/20140102/e0554235/attachment-0001.html>