Olimex Support Forum

There are a variety of images running around, but all of them seem to have some issues right out of the box. Additionally, most of the instructions for making cards require Linux, which can be a pain if you don't have it. To streamline the process a bit, I decided to put together a Debian Ubuntu image that has working support for WPA, xfce installed, and a VNC server set to start on boot. All packages are updated as of today (25 January). Additionally, security has been improved a bit, as you are no longer signing into root all the time, but have a separate user account.

Additionally, there are two different kernels installed in the image, and they are now switchable via a script on the card. The card will boot with the "video" kernel, which enables output to the VGA port, at the expense of about 100MB of memory headed to the MALI and the CEDAR video decoder. The second kernel is the "headless" version, which gives no video output, but gives you all of the memory on the board. Your choice. You can switch between them using two scripts - "go-headless" and "go-video" in the user folder - note these need to be run with sudo (i.e., "sudo ./go-headless").

This requires a 4GB card. Do not ask if you can flash it on a 2GB card, or if I can make one for a 2GB card. If you want to use a larger card, you can use resize2fs to increase the main partition.

Included features:

wpa-supplicant (change /etc/network/interfaces to enter your ssid/key)
xfce4
tightvncserver (starts on boot)
various dev tools (python, gcc, etc)
sudo
g++
500mb swap file
custom (non-debug) kernel
swappable kernels to give more memory for those of us who need it
Support for somewhat close to 512MB memory (actually 512MB in the headless release).

Version history and changelog:
R1 (16 Dec 2012): Initial release
R2 (17 Dec 2012): Re-zipped so people can actually use it, switched to a non-debug kernel, added make (how the crap did I not think to do that the first time?!?), added sysstat, removed screensaver from xfce (was causing lock-up), updated packages, cleared apt-get cache
R3 (8 Jan 2013): Updated kernel to 3.0.57. Changed rootfs to ext4. Journaling is still enabled, as it's a really good idea on embedded systems. All packages updated.
R4 (9 Jan 2013): Recompiled kernel with correct speed for I2C. Did not install i2c-tools, but that will be in the next release, if I think about it. Also should have working VGA support now - though this may disappear going forward. Also, recompiled u-boot to give us an additional 64MB of RAM.
R5 (17 Jan 2013): Recompiled kernel with module support for various USB ethernet adapters. Updated all packages.
R6 (19 Jan 2013): Recompiled kernel with (hopefully) UVC support. Added ntp, wicd, and a few other goodies.
R7 (19 Jan 2013): Recompiled kernel / u-boot. Now have ~480MB usable RAM. Highly recommended upgrade for all users.
R8 (20 Jan 2013): Moved kernel to 3.4.24. Please report any instability. Also now offering video version (with VGA support) and headless version (no output except to serial console). Upgraded packages, blah blah.
R9 (21 Jan 2013): Rebuilt kernel after I accidentally deleted everything. Oops. Somehow USB networking support got lost along the way, but should be working now. R8 is officially deprecated, and will not be available going forward. At some point I'm going to start deprecating some of these other images, but will keep them online for now.
R10 (23 Jan 2013): Kernel changes to disable powersave for onboard wireless, and onboard wireless support built into kernel. Should see greater wireless stability, though may not be perfect. Report any wireless issues.
R11 (25 Jan 2013): We are now on Ubuntu. Specifically a build based on the linaro-developer branch. Image sizes will now be smaller downloads, owing to me building the card from scratch. Still requires a 4GB card, though. Both kernels are now included in the same release, and you can switch between them using "go-headless" and "go-video" scripts in the user folder (video kernel default). Various packages may be missing now, due to building more or less from scratch. If I'm missing one of your packages, let me know. Firefox and Filezilla are now installed by default.
R12 (27 Jan 2013): LCD support got lost along the way. My apologies, but it should be back in now. Also enabled framebuffer support, which should improve video performance, but also eats some extra memory. We are down to about 330MB now, which is not nearly so much as I'd like, but I don't see much that can be done about it, other than asking Olimex for a 1GB board. A bug was causing the board to not be able to boot, which has been fixed by setting u-boot's boot delay to zero. Not the best solution, but one that works. A few minor package updates. As always, let me know what doesn't work.
R13-14: Existed on my boards, but not on yours. Merging various changes into R15.
R15 (09 Feb 2013): Updated kernel to 3.4.29. Video kernel now compiled with Mali/Cedar support baked in, rather than as a module. Hopefully this will fix video problems. Please report.
R17 (14 Feb 2013): Added support for USB->serial adapter.
R18 (08 Mar 2013): Added FTDI support. Added wireless drivers for video kernel. Added preliminary cpufreqd support, but not working at the moment. Don't hold your breath.


Downloading image:
Get current version (R18) here. (http://files.wisch.org/olinuxino_xfce-r18.zip)

Old images: 
Get previous version (R17) here. (http://files.wisch.org/olinuxino_xfce-r17.zip)
Get previous version (R15) here. (http://files.wisch.org/olinuxino_xfce-r15.zip)
Get previous version (R12) here. (http://files.wisch.org/olinuxino_xfce-r12.zip)
Get previous version (R11) here. (http://files.wisch.org/olinuxino_xfce-r11.zip)
Get previous video version (R10) here. (http://files.wisch.org/olinuxino_xfce-r10-video.zip)
Get previous headless version (R10) here. (http://files.wisch.org/olinuxino_xfce-r10-headless.zip)
R9 images have flaky wireless, and are deprecated as well. Get R10.
R8 images are deprecated. Get R10 instead.
Get previous version (R7) here. (http://files.wisch.org/olinuxino_xfce-20130119_1.zip)
Get previous version (R6) here. (http://files.wisch.org/olinuxino_xfce-20130119.zip)
Get previous version (R5) here. (http://files.wisch.org/olinuxino_xfce-20130117.zip)
Get previous version (R4) here. (http://files.wisch.org/olinuxino_xfce-20130109.zip)
Get previous version (R3) here. (http://files.wisch.org/olinuxino_xfce-20130108.zip)
Get previous version (R2) here. (http://files.wisch.org/olinuxino_xfce.zip)

Kernels: (for those who don't want to flash the whole image every time).
R18 Video Kernel (http://files.wisch.org/3.4-video-R18.tar.gz)
R18 Headless Kernel (http://files.wisch.org/3.4-headless-R18.tar.gz)

How to install:

Windows:
Download Win32 Disk Imager (http://sourceforge.net/projects/win32diskimager/).
Insert card
Start program
Select file
Click "write"

Linux:
dd to card

Logging in
User and password are olinuxino/olinuxino. Sudo has been included, so you shouldn't have any problems becoming su. Password for VNC is "password".


Obligatory information
This is provided with absolutely zero warranty. If it screws up your card, your board, or your life, that's not my fault. I am not responsible if you put this on a mission critical application and it costs you money, or causes someone harm. This was made for my personal use, and I'm sharing it with you in case it is helpful. If you are planning on using it in an industrial application, do your own due diligence.

I'll try to fix stuff that doesn't work and update it, if you tell me what it is. That being said, I have a day job, and this is being done on the side. I'll try to be responsive, but don't expect things to be fixed tomorrow.


My XXXXXX stopped working
Tough cookies. Let me know, and I'll see if I can get a fix. I don't make any promises, but I'm trying to keep a better record of what I've changed.

Can you add XXXXXX feature?
Maybe. If it doesn't break other stuff. And if it's not too much effort. Post / send me a PM and we'll see what we can do.

Does XXXXXX work?
I have a Rev. B board and a power supply. I do not know if ethernet adapters, LCD screens, USB keyboards, daq's, etc work. Download, flash, try, report back. If it doesn't work, see previous two questions.

My video isn't working!?!
You are probably using the headless kernel. "sudo /home/olinuxino/go-video"

Why would anyone want a headless version?!?
Because many of us plan on using our boards for applications where having an extra 32-64MB of memory would be a lot more useful than video output that (literally) no one will ever look at. Because some people are planning on using video, I will attempt to maintain that branch, but it will not be my primary concern.

My board doesn't boot!
You made a backup, right? No, seriously. I don't guarantee this will work on your board. I don't guarantee that the next revision will work on your board either. There seem to be gremlins out there that are making some boards behave in what I might call a non-deterministic manner. I'll do my best to get things working for you, but again, this is not my day job, and it's on the side.

I tried to compile XXXXXX driver, but it gives me an error
You need to compile your whole kernel. There are instructions! (https://www.olimex.com/forum/index.php?topic=518.0)

Do you get paid for this?
No. So be nice to me, please.

Things I'd like to do in the future:
Fix the wireless packet dropping issues
Figure out why we have such a high IO wait rate. Seems a little excessive, even for an SD card.
Boot messages to VGA during bootup.
?

Feedback and comments welcome.

Update (17 Dec):

It appears as though XFCE's screensaver may be causing the whole system to hang. I am investigating, but if this turns out to be the case I will update the image with a non-screensaver enabled version.

Let me know if you encounter any stability problems.

Hi
   Great idea preparing ready-to-go images, I've done a little of that with the R-PI.

However after several attempts of using this image I have the same error messages with winrar, any suggestions?


!   C:\Downloads\olinuxino_xfce.zip: Unknown method in olinuxino_xfce.img
!   C:\Downloads\olinuxino_xfce.zip: No files to extract

Dangerous1

Quote from: dangerous1 on December 17, 2012, 08:36:19 PM
Hi
   Great idea preparing ready-to-go images, I've done a little of that with the R-PI.

However after several attempts of using this image I have the same error messages with winrar, any suggestions?


!   C:\Downloads\olinuxino_xfce.zip: Unknown method in olinuxino_xfce.img
!   C:\Downloads\olinuxino_xfce.zip: No files to extract

Dangerous1

The file was compressed with 7-zip in what I think should be a standard zip archive. I know it's annoying for the response to be "can you try with 7-zip and see if it works", but if you can and it does, then I'll try to re-zip with something a little more standard (like the standard Windows zip). If not, it may be a bad download.

If you could check, I would certainly appreciate it.

Thanks

Hi jwischka
   Thanks for the lightening response and the advice, yep, 7-zip worked perfectly, this is the first time I've come-across incompatability with Winrar but we learn everyday.

I will try the image and give some feedback later.

Many thanks Brian

Quote from: dangerous1 on December 17, 2012, 09:02:00 PM
Hi jwischka
   Thanks for the lightening response and the advice, yep, 7-zip worked perfectly, this is the first time I've come-across incompatability with Winrar but we learn everyday.

I will try the image and give some feedback later.

Many thanks Brian

Thanks. I'll update the link with something that will work for everyone. Appreciate the feedback.

Image (and link) updated.

Changes:
Using Dropbox instead of Google Drive (faster upload).
Used a zip format that hopefully everyone can open.
Added make (seriously... how did I not do that in R1?)
Update all packages (as of 17 Dec)
Added sysstat package
Default kernel parameters are for compiling the kernel in debug mode. No more of that. Hopefully will see some speed improvements.

Hi jwischka,

I flashed your image to an 4 gb SD card. It does not work, keeps hanging after the kernel has started.
The image from the wiki works ok.

Can you give me the error message? It's working fine on a freshly flashed card here.

Thanks

Hi jwischka
   It works very well for me, after the 7-zip thing flashed image and it worked first time.
I just added some utilities I needed and bingo.


Many thanks for taking some of the pain out it.
Dangerous1

Hello jwischka,
Thank you for you work.

Output from terminal at uart works, but how can i enable vga?
Iam new in linux

Best regards
Michael

Mike - I suspect that may be a kernel option that isn't being set properly. My first thought would be (if you're able) to compile your own Kernel, or possibly replace it with one from another image. I'm primarily running this headless, so I don't have a VGA monitor to connect it to - I just use uart and SSH.

If you aren't able to compile your own kernel, let me know and I'll see what I can do, though I may / may not be able to do a lot of debugging from where I am for the next week or so (I can build images, but not necessarily verify that they're fully working, since my board is back at home).

Hello jwischka,
thank you for your quick answer.

sorry,iam not able to compile my own kernel at the moment.

The 2th problem is, when i want to use ssh over wifi,i get this error and cant use wifi:

[....] Waiting for /dev to be fully populated...[    5.340000] usbcore: registered new interface driver rtl8192cu
[    5.370000] udevd[182]: renamed network interface wlan0 to wlan1
done.
[ ok ] Setting preliminary keymap...done.
[ ok ] Setting parameters of disc: (none).
[ ok ] Cleaning up temporary files... /tmp.
[....] Activating lvm and md swap...[    9.770000] Adding 524284k swap on /mnt/512mb.swap.  Priority:-1 extents:820 across:3792928k SS
done.
[....] Checking file systems...fsck from util-linux 2.20.1
done.
[ ok ] Mounting local filesystems...done.
[ ok ] Activating swapfile swap...done.
[ ok ] Cleaning up temporary files....
[ ok ] Setting kernel variables ...done.
[....] Configuring network interfaces...Internet Systems Consortium DHCP Client 4.2.2
Copyright 2004-2011 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/

Cannot find device "eth0"
Bind socket to interface: No such device
Failed to bring up eth0.
wpa_supplicant: /sbin/wpa_supplicant daemon failed to start
run-parts: /etc/network/if-pre-up.d/wpasupplicant exited with return code 1
Internet Systems Consortium DHCP Client 4.2.2
Copyright 2004-2011 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/

Cannot find device "wlan0"
Bind socket to interface: No such device
Failed to bring up wlan0.
done.


best regards
Michael

Michael -

I suspect the wireless interface may be different on your board for some reason. From the command prompt type:

sudo ifconfig

You should see all your interfaces listed. Then edit /etc/network/interfaces with the correct information. Let me know if you have problems.

Really good work man, keep it up!

There is some great info in this thread
so i put the image and some info about it into the wiki.

jwischka, feel free to edit the wiki or let me know if anything is wrong with the info put in and i can modify.

Thanks for the image uploaded jwischka.
I have disabled the screensavers too ( they where eating up to 50% cpu time )
Regarding the 'freezing' problem I think is related to the journaled ext3 filesystem ( iotop shows a process journald that writes frequently to the disk )
So I converted the filesistem to ext4 and disabled the journaling in the following way ( searched on google convert ext3 to ext4 )
1) insert the sdcard in a linux pc
2) check the ext3 partition: sudo fsck.ext3 /dev/mmcblk0p2
3) convert to ext4: sudo tune2fs -O extents,uninit_bg,dir_index /dev/mmcblk0p2
4) check the converted ext4 partition: sudo fsck.ext4 -yfD /dev/mmcblk0p2
5) disable journaling: sudo tune2fs -O ^has_journal /dev/mmcblk0p2
6) insert the sdcard in olinuxino and restart
( added the following line to /etc/fstab to select option noatime for root filesystem: /dev/rootfs  /  ext4  defaults,noatime,commit=60 0 1 )

I'd like to setup the italian keyboard so I followed these steps:
1) installed console-data: sudo apt-get install console-data
2) configured keyboard layout: sudo dpkg-reconfigure console-data
after this step the keyboard is correctly configured, but when I restart olinuxino the keybord reverts to us style and I have to rerun dpkg-reconfigure ( which is annoying )
I have tried with: sudo dpkg-reconfigure keyboard-configuration but it does not work

Quote from: davidefa on December 22, 2012, 01:21:32 PM
Thanks for the image uploaded jwischka.
I have disabled the screensavers too ( they where eating up to 50% cpu time )
Regarding the 'freezing' problem I think is related to the journaled ext3 filesystem ( iotop shows a process journald that writes frequently to the disk )
So I converted the filesistem to ext4 and disabled the journaling in the following way ( searched on google convert ext3 to ext4 )
1) insert the sdcard in a linux pc
2) check the ext3 partition: sudo fsck.ext3 - /dev/mmcblk0p2
3) convert to ext4: sudo tune2fs -O extents,uninit_bg,dir_index /dev/mmcblk0p2
4) check the converted ext4 partition: sudo fsck.ext4 -yfD /dev/mmcblk0p2
5) disable journaling: sudo tune2fs -O ^has_journal /dev/mmcblk0p2
6) insert the sdcard in olinuxino and restart
( added the following line to /etc/fstab to select option noatime for root filesystem: /dev/rootfs  /  ext4  defaults,noatime,commit=60 0 1 )

I'd like to setup the italian keyboard so I followed these steps:
1) installed console-data: sudo apt-get install console-data
2) configured keyboard layout: sudo dpkg-reconfigure console-data
after this step the keyboard is correctly configured, but when I restart olinuxino the keybord reverts to us style and I have to rerun dpkg-reconfigure ( which is annoying )
I have tried with: sudo dpkg-reconfigure keyboard-configuration but it does not work

The I/O wait times on this board are quite bad, no matter what file system you are using, though things are helped quite a bit by having a fast card. I haven't tried moving to ext4, but if your experience suggests it's better, I may do that for the next release. I suspect there are probably some lower level issues as well (kernel/driver problems) that are causing things to be as slow as they are when it comes to disk I/O.

As to the keyboard issues, have you tried configuring things in /etc/rc.local to reconfigure on boot?

I don't have my board with me (and won't for a week or two), but I can try to look at it when I get back.

Also, it seems you guys have made Dropbox suspend my public files for a while - which is fine, just not something I've ever had a problem with before. I've updated the link with one on my own server.

Hello jwischka,
I am not able to connect to my wlan.
iget this Error:

Listening on LPF/wlan1/48:02:2a:ea:10:83
Sending on   LPF/wlan1/48:02:2a:ea:10:83
Sending on   Socket/fallback
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 8
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 13
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 14
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 10
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 10
No DHCPOFFERS received.
Unable to obtain a lease on first try.  Exiting.
Failed to bring up wlan1.
done.

i make following changes in the /etc/networks/interfaces file:
auto lo
iface lo inet loopback

auto eth0
iface eth0 inet dhcp

auto wlan1
iface wlan1 inet dhcp
    wpa-ssid "mike"
    wpa-key-mgmt WPA-PSK
    wpa-psk *1111111111111*

please help

best regards
Michael

sudo ifconfig

What is the output?

Quote from: mike1968 on December 22, 2012, 10:44:59 PM
Hello jwischka,
I am not able to connect to my wlan.
iget this Error:

Listening on LPF/wlan1/48:02:2a:ea:10:83
Sending on   LPF/wlan1/48:02:2a:ea:10:83
Sending on   Socket/fallback
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 8
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 13
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 14
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 10
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 10
No DHCPOFFERS received.
Unable to obtain a lease on first try.  Exiting.
Failed to bring up wlan1.
done.

i make following changes in the /etc/networks/interfaces file:
auto lo
iface lo inet loopback

auto eth0
iface eth0 inet dhcp

auto wlan1
iface wlan1 inet dhcp
    wpa-ssid "mike"
    wpa-key-mgmt WPA-PSK
    wpa-psk *1111111111111*

please help

best regards
Michael

i get this output.

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

wlan1     Link encap:Ethernet  HWaddr 48:02:2a:ea:10:83
          UP BROADCAST MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:86 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

Have you tried taking the security off your network and seeing if you can connect then? Also, might check to make sure you don't have an inadvertent space after your ssid or password?

Quote from: mike1968 on December 22, 2012, 11:31:00 PM
i get this output.

lo        Link encap:Local Loopback
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:16436  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

wlan1     Link encap:Ethernet  HWaddr 48:02:2a:ea:10:83
          UP BROADCAST MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:86 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)

how can i change the subnet-mask ??

Sending on   Socket/fallback
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3  8)

Quote from: mike1968 on December 23, 2012, 12:15:43 AM
how can i change the subnet-mask ??

Sending on   Socket/fallback
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3  8)

That's not the issue - 255.255.255.255 means it's sending things out without respect to what network you are on, which is what you want - since you don't have a network yet. It's a broadcast packet, not an actual packet to be routed.

Again, did you try with no security settings? My suspicion is an incorrect password, specifically one with a space after it.

Well, I disable all security settings but nothing change.

same error massage:

istening on LPF/wlan1/48:02:2a:ea:10:83
Sending on   LPF/wlan1/48:02:2a:ea:10:83
Sending on   Socket/fallback
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 8
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 9
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 9
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 14
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 8
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 6
No DHCPOFFERS received.
Unable to obtain a lease on first try.  Exiting.
Failed to bring up wlan1.

Sorry, I should have seen this before. You don't have quotes around your SSID/password do you?

so, should look like:

wpa-ssid mike
wpa-psk thisismypassword

Quote from: mike1968 on December 23, 2012, 12:34:22 AM
Well, I disable all security settings but nothing change.

same error massage:

istening on LPF/wlan1/48:02:2a:ea:10:83
Sending on   LPF/wlan1/48:02:2a:ea:10:83
Sending on   Socket/fallback
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 3
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 8
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 9
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 9
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 14
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 8
DHCPDISCOVER on wlan1 to 255.255.255.255 port 67 interval 6
No DHCPOFFERS received.
Unable to obtain a lease on first try.  Exiting.
Failed to bring up wlan1.

Mike, the default Ethernet interface appears to be eth2.

Check out this forum entry: https://www.olimex.com/forum/index.php?topic=301.0

David

Hi is anyone else using this version 2 image and using a VGA screen?

when i installed the image i had no vga output. To get it to work i had to copy both the script file and the uImage file from a working VGA image i had.

if anyone else has VGA working let me know otherwise ill upload the files so that the VGA output can be incorporated in the next version

Updated to version 3.

Get current version (R3) here. (http://files.wisch.org/olinuxino_xfce-20130108.zip)

Changelog:
Updated kernel to 3.0.57.
Changed FS to ext4, but left journaling intact. Journaling is fairly important when you may lose power at any point in time.
Updated all packages to latest version.

Updated to version 4.

Kernel recompiled with proper I2C timings. i2c-tools are not installed, but I'll try to remember to do that on the next release.

New u-boot should give us an additional 64MB of memory to work with. I'll try to track down what's eating 80MB of ram that doesn't need to be in a future release.

Also, VGA support should be working now (had to add some things back into the kernel to get it to compile), but I may pare that down later, so no guarantees going forward.

Hello jwischka,

Now with the new image my problem with wlan is now solved and vga is running.
Thank you for your work.

How could i change between vga and 7"lcd?

Best regards michael

Quote from: mike1968 on January 13, 2013, 10:53:02 PM
Hello jwischka,

Now with the new image my problem with wlan is now solved and vga is running.
Thank you for your work.

How could i change between vga and 7"lcd?

Best regards michael

You'll need to copy a different script.bin file over to the first partition. It's a fairly simple process, if you have one that works. I think one should be in the wiki, or you can make your own.

Quote from: jwischka on December 17, 2012, 06:36:15 AM
Things I'd like to do:
Figure out how to get more free memory. We have 512MB, why is only ~320MB showing up?

A portion of the memory is reserved for the kernel and what remains is for user space.

Quote from: jwischka on January 13, 2013, 11:23:13 PM

Quote from: mike1968 on January 13, 2013, 10:53:02 PM
Hello jwischka,

Now with the new image my problem with wlan is now solved and vga is running.
Thank you for your work.

How could i change between vga and 7"lcd?

Best regards michael

You'll need to copy a different script.bin file over to the first partition. It's a fairly simple process, if you have one that works. I think one should be in the wiki, or you can make your own.

Thx lcd script works fine

Quote from: viniciusfre on January 13, 2013, 11:56:21 PM

Quote from: jwischka on December 17, 2012, 06:36:15 AM
Things I'd like to do:
Figure out how to get more free memory. We have 512MB, why is only ~320MB showing up?

A portion of the memory is reserved for the kernel and what remains is for user space.

It's actually more complicated than that. 64MB is reserved for the graphics processor, ~20MB is reserved for the LCD, and about 80MB is reserved for "VE". Additionally the u-boot also was shorting us by about 60MB, so that's been taken care of at least.

Nonetheless, 80MB is absurdly large to be reserved for the kernel. I'd like to see about cutting that down, when I have free time.

I just unpacked my A13-OLinuXino (w/o WiFi) for the first time an put on this new olinuxino_xfce-20130109.img using an 8 GB SD card.  It booted out-of-the-box!  Thanks jwischka!!  Keyboard, mouse, and VGA worked just fine, in both the initial CLI and X. 

The only thing I could not get to work is the USB/Ethernet adapter.  I uncommented the eth0 stanza in /etc/network/interface and even added an eth1 and eth2 stanza (I read on this forum that the default is eth2) - did not help.  lsusb is not installed, so I can't check that, but every unplugging and plugging in of this Ethernet adapter is faithfully reported in syslog (as "new high speed USB device number xx using sw-ehci").  lsmod shows no kernel modules at all, just the heading, which I thought was odd, as I'm used to see a ton of modules there.  ifconfig -a reports the loopback "lo" and a "tunl0" (which I have never seen before on my main system), but no eth0, eth1, or eth2.

So surely, I am doing something wrong - but what???

I have the same problem... with this version the usb-ethernet don't work. I have installed usbutils, and with lsusb i find the ethernet. It's name is ASIX Electronics.
Searching on google i find this:
http://linux.m2osw.com/asix-electronics-ax88178-driver

Now i try to install this module following this tutorial, if it work, i update this post... See you later

Quote from: Ekkehard on January 17, 2013, 01:06:23 AM
I just unpacked my A13-OLinuXino (w/o WiFi) for the first time an put on this new olinuxino_xfce-20130109.img using an 8 GB SD card.  It booted out-of-the-box!  Thanks jwischka!!  Keyboard, mouse, and VGA worked just fine, in both the initial CLI and X. 

The only thing I could not get to work is the USB/Ethernet adapter.  I uncommented the eth0 stanza in /etc/network/interface and even added an eth1 and eth2 stanza (I read on this forum that the default is eth2) - did not help.  lsusb is not installed, so I can't check that, but every unplugging and plugging in of this Ethernet adapter is faithfully reported in syslog (as "new high speed USB device number xx using sw-ehci").  lsmod shows no kernel modules at all, just the heading, which I thought was odd, as I'm used to see a ton of modules there.  ifconfig -a reports the loopback "lo" and a "tunl0" (which I have never seen before on my main system), but no eth0, eth1, or eth2.

So surely, I am doing something wrong - but what???

I believe USB ethernet is enabled in the kernel, but I may need to check to make absolutely certain. I'll try to do that later today, though things are a bit crazy at work, so it may get pushed to the back burner.

jwischka i try to recompile modules, but there are problem with kernel headers.

After make command this is the output:

olinuxino@debian:~/AX88772B_772A_760_772_178_LINUX_Driver_v4.4.1_Source$ make
make -C /lib/modules/3.0.57jwischka-20130109+/build SUBDIRS=/home/olinuxino/AX88772B_772A_760_772_178_LINUX_Driver_v4.4.1_Source modules
make: *** /lib/modules/3.0.57jwischka-20130109+/build: No such file or directory.  Stop.
make: *** [default] Error 2

it can't find /lib/modules/3.0.57jwischka-20130109+/build

lrwxrwxrwx 1 root root    32 Jan  9  2013 build -> /home/jeff/Documents/linux-sunxi
lrwxrwxrwx 1 root root    32 Jan  9  2013 source -> /home/jeff/Documents/linux-sunxi

By the way, thanks for this new release, it seems very good and i appreciate the boot time :D

hi, and thanks for all the good work.

may i request you post a .zip of the root filesystem (with any details of swap, other needed partitions.

this will avoid the need to keep flashing cards with dd when the boot partition is already there.
i am getting problems with flashing, maybe wear.

thanks!

Quote from: drspastic on January 17, 2013, 04:44:43 PM
hi, and thanks for all the good work.

may i request you post a .zip of the root filesystem (with any details of swap, other needed partitions.

this will avoid the need to keep flashing cards with dd when the boot partition is already there.
i am getting problems with flashing, maybe wear.

thanks!

I'll see what I can do. FYI, SD cards have built in controllers which should control the wear leveling issue, and in any event, cards are, for the most part, supposed to be able to experience thousands (or more precisely tens of thousands) of write cycles before failing. What issues are you experiencing, exactly?

Hi maurosk,

since the MOD-ENC28J60 UEXT module still does not work, i also tried a USB ethernet adapter with an ASIX chipset as a workaround.
You can find the module under: Device Drivers -> Network device support -> USB Network Adapters -> ASIX AX88xxx. So there is no need to build it on your own.
But one side note!
I did NOT get the build in module working with the 3.0.52 kernel or 3.0.57 kernel so i ended up using the 3.4 kernel branch. Now it works fine. But maybe this is just a problem with my specific vendor version of this adapter.
If you really want to build the module on your own, you have to edit $KDIR which refers to the path of the kernel source directory.


Kind regards
Leo

Hi Leo, thanks for hint, how i can add ASIX modules in jwischka kernel? Can you explain me how to recompile this kernel? What i need? Thanks in advance

Quote from: maurosk on January 17, 2013, 05:29:37 PM
Hi Leo, thanks for hint, how i can add ASIX modules in jwischka kernel? Can you explain me how to recompile this kernel? What i need? Thanks in advance

You'll need to fork the kernel source from the linux-sunxi branch, for starters. After that, it's basically configuration. You don't need to recompile my kernel - you can compile your own! :)

Yes i know this, but your kernel is good, it need just ethernet mod.. i try to recompile my own kernel, but it isn't good like yours, that's why i want to recompile yours :D

Quote from: maurosk on January 17, 2013, 05:39:38 PM
Yes i know this, but your kernel is good, it need just ethernet mod.. i try to recompile my own kernel, but it isn't good like yours, that's why i want to recompile yours :D

Let me boot into Linux and check...

Quote from: maurosk on January 17, 2013, 05:39:38 PM
Yes i know this, but your kernel is good, it need just ethernet mod.. i try to recompile my own kernel, but it isn't good like yours, that's why i want to recompile yours :D

Adding various usb support as a module. Recompiled, rebuilding, updating, and will re-upload.

Thanks jwischka, i hope to return the favor :D

Quote from: maurosk on January 17, 2013, 05:57:51 PM
Thanks jwischka, i hope to return the favor :D

I just hope the kernel works... :)

Quote from: jwischka on January 17, 2013, 05:54:20 PM
Adding various usb support as a module. Recompiled, rebuilding, updating, and will re-upload.

Or just send him your .config  ;)
Even with a new kernel he needs the kernel sources to build the asix module against it.

Quote from: leo on January 17, 2013, 06:00:51 PM

Quote from: jwischka on January 17, 2013, 05:54:20 PM
Adding various usb support as a module. Recompiled, rebuilding, updating, and will re-upload.

Or just send him your .config  ;)
Even with a new kernel he needs the kernel sources to build the asix module against it.

I built all the modules. I'll try to tar the lot up, in addition to doing another full blown image with updated goodies...

good jwischka, i'm look eagerly your new upload :D

Yup - I'll second that motion.  Would be a shame if we all had to re-invent the wheels that jwischka has already invented AND trued so beautifully :-)  For starters, a .config file would be much appreciated.

Taken care of - you guys are just too fast for me :-)

And yes, ASIX Electronics is the manufacturer of chip that's in my Ethernet adapter (SABRENT NT-USB20) as well.

Quote from: Ekkehard on January 17, 2013, 06:08:44 PM
Yup - I'll second that motion.  Would be a shame if we all had to re-invent the wheels that jwischka has already invented AND trued so beautifully :-)  For starters, a .config file would be much appreciated.

Taken care of - you guys are just too fast for me :-)

And yes, ASIX Electronics is the manufacturer of chip that's in my Ethernet adapter (SABRENT NT-USB20) as well.

Link to kernel/modules/firmware (http://files.wisch.org/jwischka_20130117.tar)
.config (http://files.wisch.org/.config)

Will update with a new image soonish.

It works, good job :D

R5 released.

Get current version (R5) here. (http://files.wisch.org/olinuxino_xfce-20130117.zip)

Updated kernel with usb networking support as modules, updated packages.

So let me put this straight:

• Yesterday afternoon I started to evaluate a new processor board
• I put on a nicely packaged piece of OSS, but ran into a problem
• I posted the problem here on this forum the same evening - turns out, I was asking for an unsupported feature - so, in my book, a wish, not a bug report
• The very next morning the author of the OSS posts a new release with the wish included, that actually works (minor detail - I know)
• What do I call that?  Freaking A-M-A-Z-I-N-G

Yes, jwischka, you are freaking A-M-A-Z-I-N-G!!!  Can't thank you enough!!!!  You saved me oodles of time!!

Hello guys,
I have here A13 and I cannot successfully login into system. I prepared SD card with your R5 system. No problem, after power up the kernel start and then I am prompted to insert username/password in console. But when I enter olinuxino/olinuxino then there is flashed very quickly welcome text "...ABSOLUTLY NO WARRANTY, etc" and then the screen is cleared and I see just login prompt again (debian login:).

Do you have any advice how to solve this my problem? I tested two 4GB SDcard with my coleague for sure, but same trouble. Basic Debian system without X from your website I have successfully tested with 2GB SDcard - login and using system in console without problems. VGA monitor is used. 

Jaromir
   

Have you tried the R4 image? If not, can you? I didn't fully test the R5 card, so there may be something up. Unfortunately there is no power at my building right now, so I can't test it for you. :/

Quote from: atal on January 18, 2013, 02:43:14 PM
Hello guys,
I have here A13 and I cannot successfully login into system. I prepared SD card with your R5 system. No problem, after power up the kernel start and then I am prompted to insert username/password in console. But when I enter olinuxino/olinuxino then there is flashed very quickly welcome text "...ABSOLUTLY NO WARRANTY, etc" and then the screen is cleared and I see just login prompt again (debian login:).

Do you have any advice how to solve this my problem? I tested two 4GB SDcard with my coleague for sure, but same trouble. Basic Debian system without X from your website I have successfully tested with 2GB SDcard - login and using system in console without problems. VGA monitor is used. 

Jaromir
   

Thanks for your fast reply. We tested R4 with same troubles yesterday.

Quote from: atal on January 18, 2013, 02:50:56 PM
Thanks for your fast reply. We tested R4 with same troubles yesterday.

I would suggest trying to log in as root, but I've forgotten the password (at least, I'm not sure enough to know without testing).

You might see if you can chroot from a Linux box to the card, then create a new user.

It sounds like the login is succeeding, but then rapidly returning you to the login screen. Do you have a network connection? If so you could try to vnc into the board, or ssh. Can you see if you get the same result from the UART console?

OK, I tried your suggestions. At first I tried to connect via Putty (we have LAN adapter) - no response. Then I modified the forgotten password for root manually in password and shadow file. Same behaving after login root attempt - quick flash of text and login screen showed again.

For complete info our configuration: A13, USB mouse, USB keyboard, LAN adapter, WiFi module on board, 12V power, VGA monitor, 4GB SDcard.

For today it is finish, have a nice weekend.

What happened with the official (Olimex) image on SD card?

Note: has different username/password when you try it!

Quote from: jwischka on January 17, 2013, 05:08:14 PM

Quote from: drspastic on January 17, 2013, 04:44:43 PM
hi, and thanks for all the good work.

may i request you post a .zip of the root filesystem (with any details of swap, other needed partitions.

this will avoid the need to keep flashing cards with dd when the boot partition is already there.
i am getting problems with flashing, maybe wear.

thanks!

I'll see what I can do. FYI, SD cards have built in controllers which should control the wear leveling issue, and in any event, cards are, for the most part, supposed to be able to experience thousands (or more precisely tens of thousands) of write cycles before failing. What issues are you experiencing, exactly?

the problems i have are cards writing ok a few times then refusing to complete dd with i/o error. im using sandisk and trancsend cards in various adapters. always get very warm on dd.

also my reason to want to download rootfs is its faster and easier unless a user is inexperianced in using gparted etc. also there is no point rewriting an identical boot sector.

by the way i just blew your v4 image to a 8gb sandisk HC card and it booted faster than any image yet. pity i just logged on here and found there is already a v5. now downloading that. pity v4 didnt have any of the wifi tools installed. have you considered using the madwifi package? it has pretty wide support for many different cards.

atal,
I used/tested R5 quite extensively yesterday and found no problems with it at all (very minor issues fixed in less than a minute see P.S.).  What you describe sounds like an exit or Ctr-D equivalent.  Could have any number of causes.  Since you're using the virgin root file system that I've been using, software/configuration issues are mostly excluded.  Have you tried logging in through other means such as a serial port or Ethernet?  Since you made it through the boot process OK, defective RAM or insufficient power are less likely put still on the table.

P.S.  jwischka, regarding your root file system, you may want to include your hostname (debian) with IP address 127.0.1.1 (or any other loopback address) in the /etc/hosts file - makes sudo stop complaining.  Also, now that you made the Ethernet adapter work, it might be prudent to uncomment the eth0 stanza in /etc/network/interfaces.  Finally, as soon as there is an Ethernet connection (wired or wireless), file sharing is likely going to happen, and then I think ntp would be a good idea to install to get the correct time stamps.  Again, all of these were fixed within a minute, so not a problem.

Jwishka..great job :o
Is it posible to ad uvc camera support in your next version?

Today i downloaded all filles discript in the wiki  to compile my own kernel.
Ubuntu is running onmy netbook but compilation faild.
So i have to spend the weekend to figure out whats wrong.

Best regards
Michael

Post a few (if need be, many) lines showing what failed.  Someone else may recognise the problem.

I've started building it all, hit some issues but now have them all fixed, in case I can help.

Say which Ubuntu etc you're building on.  I'm using Linux Mint 14 and had to install a few things that slightly surprised me.

Good suggestions, all. I will try to add and put up a new version tonight, though due to (still) having no power, I'm at a friend's house with slow internet. Sigh...

John, I'm building with Kubuntu 12.10.

Uploading new image, should be available shortly.

Added ntp, fixed various things. Upgraded packages. Installed wicd.

Added, hopefully, uvc support via modules in the kernel.

I have to be honest, things were a little weird when I was transferring stuff over to the card, and as I don't have my board with me, I wasn't able to test. So this might not work. Or it might. As always, use at your own risk.

Madwifi is going to be more than I want to mess with tonight, so that will have to wait until a later build. Likewise with a rootfs, I'll need to develop a good procedure. Of course, with this, there are kernel updates, so a rootfs isn't all that you need. Besides, you could always mount the image, since you know the card geometry.

At any rate, once things are up I will update this post and OP with a link - I have no idea how slow this internet is, so it could be in the morning. Yay DSL.

Edit 1: GASP! 6 hours. At home it takes about 16 minutes... sigh...

R6 is available. (http://files.wisch.org/olinuxino_xfce-20130119.zip)

Recommended update for all users: R7

Recompiled kernel / uboot. We now can actually *use* all (or almost all) of our 512mb of memory, which is clever. There is still some reserved for the Mali, but we're talking ~480 usable, which is a lot better than before. Seeing as how I/O wait is such a performance issue this should speed things up considerably.

You can get the image here (http://files.wisch.org/olinuxino_xfce-20130119_1.zip)

Not to keep posting, but....

Is there any interest in a headless version of this? Meaning a kernel where there is no support for the Mali? Even with the Mali modularized, it's still taking about 32MB of memory, which, depending on how you are planning on using this thing, could be better used to do things like increase performance. I personally only interact with the board via VNC/ssh, so I would prefer the extra memory, but it seems that most of you interact via VGA/LCD. Thoughts?

For ease of testing mine is headless.

Also, might reduce what's going on and who knows might even boot!

I've built U-Boot with md5sum support and can see the kernel is loaded correctly from my SD card so it's not the card to blame for the failure to boot.  Progress!

Looks ever more like a hardware fault.

John

I'll package / upload a headless kernel for you to test... will pm link. If it doesn't work, I have an idea about disabling swap emulation, since that's around where it seems to fail on your board. We'll see, I suppose...

Thanks, Jeff.

I have a kernel built (as in, the make ran quite a long time and went OK) but diverted to U-Boot to get md5sum in.  Kernel and testing "soon" but I'd be pleased if you can do one as my first may well not work LOL

John

Quote from: JohnS on January 19, 2013, 10:28:42 PM
Thanks, Jeff.

I have a kernel built (as in, the make ran quite a long time and went OK) but diverted to U-Boot to get md5sum in.  Kernel and testing "soon" but I'd be pleased if you can do one as my first may well not work LOL

John

PM sent. Mali support removed, and a lot of other things modularized. Hopefully will work.

Did not mean to create a wrong impression... 

I was using the VGA and USB keyboard/mouse only for the first time bring-up - kind of as a safety net while I was getting familiar with the board (and for the first time around the Ethernet adapter didn't work).  Ultimately, this board will not be connected to my KVNM switch anymore and will simply be sitting on its merry IP address for me to ssh into - for the occasional maintenance.  In fact, most applications I can think of for this board would be headless...

So that's a "me too", I'd much prefer the additional memory over the ability to hook it up to the KVM switch.

I'll try to put up a headless image tomorrow - also looking to see if we can move to 3.4 without too many problems.

>>> IT BOOTS!! <<<

Fantastic.  Thanks so much.

So... I wonder what was stopping it?  Any ideas how I home in on it?

Please post/email your kernel config so I can build it myself.

John

hi jwischka;

Touchscreen working in R7 image?

Quote from: JohnS on January 20, 2013, 10:09:35 AM
>>> IT BOOTS!! <<<

Fantastic.  Thanks so much.

So... I wonder what was stopping it?  Any ideas how I home in on it?

Please post/email your kernel config so I can build it myself.

John

Prayer? You can get the kernel config (I think) here (http://files.wisch.org/.config-3.0.57-headless). I'd start by trying to add the mali support back in. That would be one of the biggest drains on the power, obviously, but it's internal to the a13, so seems kind of weird. It could also be a memory issue, as the mali is taking 32-64mb of your ram. Have you done a full memtest?

Quote from: Eos on January 20, 2013, 12:27:18 PM
hi jwischka;

Touchscreen working in R7 image?

No idea, since I don't have one. If you test and report back, I'll try to fix it if it isn't.

I have a headless 3.4 kernel working well, but having trouble getting VGA output to work - will probably try to get both working today, if I have time, and then see about posting some updates.

v7 working great, you have been very busy with this. if you do stop vga support i will be dissapointed as i am looking forward to more updates. i think i sorted my dd problem out; involves seting the card back to single fat32 partition first, remove then reinsert.

all my usb devices are now listed by name correctly, and you seem to have got rid of a problem with the commandline where going through previous commends with up and down keys corrupted the line.
maybe this was to do with the mali?

do you think this will every get as far as being a live installer? i mean a bigger image with everything lurking, and a menu driven first boot in vga console to select/deselect packages and modules. would be a nice universal distro.

ok i have a few problems, as i say v7 is booting well and x works great on vga, usb devices recognised and identified.
however my network usb devices dont load their modules, even after installing manually via dpkg their respective deb firmware packages. i checked the files had appeared and they were all correct in /lib/firmware.

any ideas? i just need to get network!

Quote from: jwischka on January 20, 2013, 02:50:05 PMHave you done a full memtest?

Not full, as I think uboot lives somewhere in it, but other than that, yes.  With its extended test.

Also, Linux stayed up all night.

Quote from: drspastic on January 20, 2013, 05:22:03 PM
ok i have a few problems, as i say v7 is booting well and x works great on vga, usb devices recognised and identified.
however my network usb devices dont load their modules, even after installing manually via dpkg their respective deb firmware packages. i checked the files had appeared and they were all correct in /lib/firmware.

any ideas? i just need to get network!

Have you tried to modprobe them? Which specific device are you using and I can make sure there is kernel support compiled.

R8 posted.

A few things to note:

1) We have moved to 3.4 from 3.0. This means that things may not work. More to the point, it means your things may not work. I will not know if your things do not work unless you tell me.

2) I have split development into two paths: video and headless. Headless does not have video output on the VGA port, or on the LCD. Don't ask why your video is not working if you downloaded the headless installation. My application is primarily headless, so if the video stops working, I may not know (though I'll try to check). Headless also may be updated more frequently.

3) I'll try to optimize the 3.4 kernel over the next few releases - I basically just took the 3.0 configs and built them. Actually, I tried to start with a fresh 3.4 config, but quickly realized I couldn't find the settings that gave more optimized memory, so I went back to the old configs. Please report any stability issues.

Links:

Get current video version (R8) here. (http://files.wisch.org/olinuxino_xfce-r8-video.zip)
Get current headless version (R8) here. (http://files.wisch.org/olinuxino_xfce-r8-headless.zip)

Note: I'm having a hard time telling whether wireless is working in R8 or not. I think it is, but can't quite be sure. Please test. Also, I decided to delete everything that I've done. Not the smartest move, I'll agree, but accidents happen. Blah. Hopefully will be back up to speed soonish.

Great news! I deleted all my source files on accident :/

Hey, it was time to make my own github repo, right?

Bleh. Hopefully will be back up to speed soonish.

Must have been deleted part way through my download :(

I still have your r7 (3.0.57) headless config if you'd like it back...

John

Quote from: jwischka on January 21, 2013, 12:12:31 AM

Quote from: drspastic on January 20, 2013, 05:22:03 PM
ok i have a few problems, as i say v7 is booting well and x works great on vga, usb devices recognised and identified.
however my network usb devices dont load their modules, even after installing manually via dpkg their respective deb firmware packages. i checked the files had appeared and they were all correct in /lib/firmware.

any ideas? i just need to get network!

Have you tried to modprobe them? Which specific device are you using and I can make sure there is kernel support compiled.

modprobe shows nothing, the mods i was expecting in the kernel was rt2800 wich as far as i know supports the rt3070 usb wifi chipset i use, and is included in kernels 3 and up. i guess you stripped them to save space.
rt73 should work too. as for my other devices, i need the zd1201 wich was included in your earlier images, i just had to add the firmware. finally the pegasus driver that should kick my smc 2206 usb-eth into life.

bad luck about your deleting, but you have been putting out images like a race car! time to sit down and have a cold beer maybe. i never got a chance to try your latest.

stirling work you have done, thanks a million and best wishes.

I think I should b able to recover most things by the end of the day. I'll check into the networking issues.

Quote from: drspastic on January 21, 2013, 12:53:52 PM

Quote from: jwischka on January 21, 2013, 12:12:31 AM

Quote from: drspastic on January 20, 2013, 05:22:03 PM
ok i have a few problems, as i say v7 is booting well and x works great on vga, usb devices recognised and identified.
however my network usb devices dont load their modules, even after installing manually via dpkg their respective deb firmware packages. i checked the files had appeared and they were all correct in /lib/firmware.

any ideas? i just need to get network!

Have you tried to modprobe them? Which specific device are you using and I can make sure there is kernel support compiled.

modprobe shows nothing, the mods i was expecting in the kernel was rt2800 wich as far as i know supports the rt3070 usb wifi chipset i use, and is included in kernels 3 and up. i guess you stripped them to save space.
rt73 should work too. as for my other devices, i need the zd1201 wich was included in your earlier images, i just had to add the firmware. finally the pegasus driver that should kick my smc 2206 usb-eth into life.

bad luck about your deleting, but you have been putting out images like a race car! time to sit down and have a cold beer maybe. i never got a chance to try your latest.

stirling work you have done, thanks a million and best wishes.

Ok. R9 released. Your networking should be working again.

Rebuilt 3.4 kernel. Hopefully one that works this time. And one that I won't delete. If there are issues, we can deal with it. Things may be a bit slower than they were under 3.0 performance wise, but I am thinking we will get more improvements over time in 3.4 than 3.0. Also, I haven't really gotten around to tweaking things yet.

As of now, R8 is deprecated, and will not be available.

Links:
R9 video (http://files.wisch.org/olinuxino_xfce-r9-video.zip)
R9 headless (http://files.wisch.org/olinuxino_xfce-r9-headless.zip)

Hi, I got bad CRC when I inflate the source, is the file zipped on Windows?

Quote from: elwiss on January 22, 2013, 05:55:03 AM
Hi, I got bad CRC when I inflate the source, is the file zipped on Windows?

Yes. Which image did you try? Both appear to have uploaded fine.

The video version.

Quote from: elwiss on January 22, 2013, 06:00:22 AM
The video version.

A couple of things:

1) Can you try redownloading to make sure it isn't a fluke
2) Can you describe how you are unzipping the file?

I'll try to look at it in the morning if you still have problems after a redownload.

Yeah I started a new download couple of minutes ago.

To unzip:
unzip archive.zip

Quote from: elwiss on January 22, 2013, 06:06:32 AM
Yeah I started a new download couple of minutes ago.

To unzip:

Code Select Expand

unzip archive.zip

Hmm... wouldn't think that would cause problems. If it doesn't work I'll pull a new image tomorrow.

Well it did it again, i'll check if the problem happen in Windows. I'll post result later.

Unziped correctly unsing 7zip. Must be something wrong with Linux or something...

I get the same unzip problem ONLY on one Linux system.  It's a bug somewhere - libs or kernel or ...

unzips fine on 2.6.x kernel.  Fails on my dev system 3.5.0

I've ben unzipping on one system, then using tar which works OK on both.

(Problem is on my ToDo list to investigate/report.)

John

Quote from: elwiss on January 22, 2013, 07:28:05 AM
Unziped correctly unsing 7zip. Must be something wrong with Linux or something...

Hrm. Odd - but you did get it unzipped?

Interesting... 7-zip works for me, too, whereas unzip does not.

An occasional other program also fails on 3.5.0 so I think it may be a (dynamic?) lib.

John

Quote from: JohnS on January 22, 2013, 03:31:18 PM
Interesting... 7-zip works for me, too, whereas unzip does not.

An occasional other program also fails on 3.5.0 so I think it may be a (dynamic?) lib.

John

Does ark (or similar) work? Basically, is it only unzip that is failing? This seems to be a weird problem.

I cant write the image on the sd card. I am getting this error:

I'm using 4GB sd card (and Windows 7) :)

Quote from: shinobi on January 22, 2013, 04:35:48 PM
I cant write the image on the sd card. I am getting this error:

I'm using 4GB sd card (and Windows 7) :)

Looks like your 4GB card isn't really 4GB. I don't suppose you have an 8gb one to play with?

What brand are you using?

Quote from: jwischka on January 22, 2013, 04:38:48 PM
Looks like your 4GB card isn't really 4GB. I don't suppose you have an 8gb one to play with?

No I dont.

Quote from: jwischka on January 22, 2013, 04:38:48 PM
What brand are you using?

Kingston :)

Quote from: shinobi on January 22, 2013, 05:05:35 PM

Quote from: jwischka on January 22, 2013, 04:38:48 PM
Looks like your 4GB card isn't really 4GB. I don't suppose you have an 8gb one to play with?

No I dont.

Quote from: jwischka on January 22, 2013, 04:38:48 PM
What brand are you using?

Kingston :)

Hm. A lot of Kingston cards are known to be fake, even when purchased from reputable dealers. (http://www.bunniestudios.com/blog/?p=918) It might be worth checking to see if yours is legit.

It's also not as easy, but you could flash the standard 2GB image (from olimex), then extend the file system to your 4gb card, then do some quick package updates and install my kernel, which I could post separately. Somewhat of a pain, but workable, if you want to try it (I can give you instructions)

So everyone is aware, it seems as though there may be a bit of a stability issue with regard to the on-board wireless in R9. I am testing an R10 build at the moment, and also checking to see whether the problem is only on the on-board, or also on separate USB adapters. The wireless on the board has always been flaky, so it's not terribly surprising, and there are a couple of culprits that I'm looking into. Hopefully this means an R10 build will drop later today, but I'll probably not put one out until I can track down the source of the issues.

This could also mean returning to the 3.0.57 branch, depending on whether we can track this down.

Quote from: shinobi on January 22, 2013, 05:05:35 PM

Quote from: jwischka on January 22, 2013, 04:38:48 PM
Looks like your 4GB card isn't really 4GB. I don't suppose you have an 8gb one to play with?

No I dont.

Quote from: jwischka on January 22, 2013, 04:38:48 PM
What brand are you using?

Kingston :)

Any chance you were writing the IMG to a partition on the SD instead of to the whole SD?

That's not actually how windiskimager works - it's done at a drive level. To make sure, you could go into diskpart and do a "clean" command on the drive, but it shouldn't be necessary. I suspect that there may literally be not enough space on the card, though that would be somewhat odd...

Quote from: JohnS on January 22, 2013, 05:22:05 PM

Quote from: shinobi on January 22, 2013, 05:05:35 PM

Quote from: jwischka on January 22, 2013, 04:38:48 PM
Looks like your 4GB card isn't really 4GB. I don't suppose you have an 8gb one to play with?

No I dont.

Quote from: jwischka on January 22, 2013, 04:38:48 PM
What brand are you using?

Kingston :)

Any chance you were writing the IMG to a partition on the SD instead of to the whole SD?

Quote from: JohnS on January 22, 2013, 04:38:48 PM
Any chance you were writing the IMG to a partition on the SD instead of to the whole SD?

Not sure. I am using win32diskimager. I think it writes to the whole card.

Quote from: jwischka on January 22, 2013, 04:38:48 PM
Hm. A lot of Kingston cards are known to be fake, even when purchased from reputable dealers. (http://www.bunniestudios.com/blog/?p=918) It might be worth checking to see if yours is legit.

Just tried another 4gb kingston card. Same result.

Quote from: jwischka on January 22, 2013, 05:08:53 PM
It's also not as easy, but you could flash the standard 2GB image (from olimex), then extend the file system to your 4gb card, then do some quick package updates and install my kernel, which I could post separately. Somewhat of a pain, but workable, if you want to try it (I can give you instructions)

No, no thanks but this is too complicated for me.

Maybe I should buy 8gb card :)

That is probably the easiest solution - then you can extend the FS out. Personally I'm running on a 16gb card, and they're cheap enough now on Amazon that there's not much of a reason to stay with 4GB, unless you're developing an industrial product where every dollar counts.

Quote from: shinobi on January 22, 2013, 05:31:00 PM

Quote from: JohnS on January 22, 2013, 04:38:48 PM
Any chance you were writing the IMG to a partition on the SD instead of to the whole SD?

Not sure. I am using win32diskimager. I think it writes to the whole card.

Quote from: jwischka on January 22, 2013, 04:38:48 PM
Hm. A lot of Kingston cards are known to be fake, even when purchased from reputable dealers. (http://www.bunniestudios.com/blog/?p=918) It might be worth checking to see if yours is legit.

Just tried another 4gb kingston card. Same result.

Quote from: jwischka on January 22, 2013, 05:08:53 PM
It's also not as easy, but you could flash the standard 2GB image (from olimex), then extend the file system to your 4gb card, then do some quick package updates and install my kernel, which I could post separately. Somewhat of a pain, but workable, if you want to try it (I can give you instructions)

No, no thanks but this is too complicated for me.

Maybe I should buy 8gb card :)

Update:

R10 images, complete with kernels that have wifi which stays on for longer than 30 minutes, are ripped and ready to upload. Unfortunately my internet connection is not, but hopefully will be fixed in the morning.

For R11 I am considering moving the base rootfs to something cobbled together from the linaro developer edition, which I think may provide slightly speedier performance and boots. If anyone has any thoughts on the matter, this would be the time to share them.

Hello


First of all, thanks for all your work creating images for us :D


Second, will your future images be > 4Gb ?


I guess I'll have to buy a bigger card then  ( that and the USB keyb ).

I plan to keep the images at 4GB, and let people extend the filesystem for larger cards.

Quote from: buzibus on January 23, 2013, 05:30:03 AM
Hello


First of all, thanks for all your work creating images for us :D


Second, will your future images be > 4Gb ?


I guess I'll have to buy a bigger card then  ( that and the USB keyb ).

Since version 8 and now 9 I get a message like this:




Yes I'm using a Kingston 4Gb too.


Maybe its better to get a 8Gb file and keep the 4Gb for other purposes.

Somewhat odd - the card I have been pulling images from hasn't changed, so I'm not sure why it is a few sectors longer than other cards.

Is this a brand new card? Have you used it for anything? I'm wondering if there may be some issues with bad blocks marked on the card?

Quote from: buzibus on January 23, 2013, 05:44:06 AM
Since version 8 and now 9 I get a message like this:




Yes I'm using a Kingston 4Gb too.


Maybe its better to get a 8Gb file and keep the 4Gb for other purposes.

I'm using 8GB cards but have a 4GB and it is 7737344 blocks of 512 bytes so again it would not fit.  The Olimex image fitted.

It wouldn't be much work for anyone with a smaller card to tweak the image but it's a bit of a pain.  (On Linux some loopback mounts make it pretty easy.  No idea on Windoze!)

Hey jwischka,

It worked for me, Thanks dude........ :)

thought of hosting files as a torrent? it might get download speed up. i will happily seed latest images.
im now downloading 9 and will post a report later.
im glad it wasnt just me being stupid with r8 networking!

Quote from: drspastic on January 23, 2013, 01:49:40 PM
thought of hosting files as a torrent? it might get download speed up. i will happily seed latest images.
im now downloading 9 and will post a report later.
im glad it wasnt just me being stupid with r8 networking!

No - I haven't thought of hosting it as a torrent, but I have thought about researching why Apache seems to be capping download speeds. The server is in a rack that has a solid 100MBit up, so the internet pipe itself isn't the problem.

That sounds like a weekend project, though...

What bus clock did you set for the i2c interfaces? I'm trying to connect an UEXT MOD-IO2 board and the communication doesn't work.

I have used the lastest firmware for that board, but still no luck, At the moment I use R5 of your image.

Quote from: Bernd on January 23, 2013, 04:21:54 PM
What bus clock did you set for the i2c interfaces? I'm trying to connect an UEXT MOD-IO2 board and the communication doesn't work.

I have used the lastest firmware for that board, but still no luck, At the moment I use R5 of your image.

Should be 100kHz - I'll check and make sure, but I'm fairly certain I reset that when things got borked. R10 images are up, but I'll verify the clock speed before posting the links.

Quote from: jwischka on January 23, 2013, 04:51:02 PM

Quote from: Bernd on January 23, 2013, 04:21:54 PM
What bus clock did you set for the i2c interfaces? I'm trying to connect an UEXT MOD-IO2 board and the communication doesn't work.

I have used the lastest firmware for that board, but still no luck, At the moment I use R5 of your image.

Should be 100kHz - I'll check and make sure, but I'm fairly certain I reset that when things got borked. R10 images are up, but I'll verify the clock speed before posting the links.

Thanks. I will install the R10 image and then I can be sure that the clock speed isn't the problem. :-) 100kHz schould be fine.

R10 images posted.

Kernel recompiled to disable power save for wireless modules. Also recompiled with on-board wireless support built in. Wireless stability should be improved from R9, and R9 images are now deprecated. Please report any wireless stability issues

With regard to I2C, clock speed is confirmed to be 100kHz - i2cdetect works fine, so it shouldn't be a kernel problem.

Get current video version (R10) here. (http://files.wisch.org/olinuxino_xfce-r10-video.zip)
Get current headless version (R10) here. (http://files.wisch.org/olinuxino_xfce-r10-headless.zip)

Report any issues. R11 may be a linaro build instead of Debian - we'll see.

Quote from: jwischka on January 23, 2013, 05:59:20 AM
Somewhat odd - the card I have been pulling images from hasn't changed, so I'm not sure why it is a few sectors longer than other cards.

Is this a brand new card? Have you used it for anything? I'm wondering if there may be some issues with bad blocks marked on the card?

The problem might reside in the fact that storage equipment manufacturers often abuse the meaning of the SI Gigabyte .

For them a GB is 10^9 and not 1073741824 bytes as the SI defines (1024*1024*1024). So 4 GB becomes around 3.7 GB. Take a few more bytes for volume information and the available space goes down to 3.6GB ( that's what windows reports ).

Anyway I've bought a 8GB card to be able to use your images as downloaded.

You should look into creating torrents for this, you might save yourself some bandwidth :)


Downloading R10 video version right now.


Thank you or your work.

What it fundamentally comes down to is that the cards are of difference sizes - significantly, in this case. When I pull the image, it reads my entire card, whatever sectors are actually there, which it seems are more sectors than on some other people's cards. The ideal thing to do would be to maintain some 2gb cards and let everyone extend to 4 - but that's a pain for them, and I don't have a 2gb card around - and on top of that, the images are hard to fit on 2gb anyway.

We're not actually charged for bandwidth on a per-byte basis, so it's not a big deal... just have to stay under 50mbps 95% of the time, which isn't too hard to do for our application.

Quote from: buzibus on January 23, 2013, 10:34:04 PM

Quote from: jwischka on January 23, 2013, 05:59:20 AM
Somewhat odd - the card I have been pulling images from hasn't changed, so I'm not sure why it is a few sectors longer than other cards.

Is this a brand new card? Have you used it for anything? I'm wondering if there may be some issues with bad blocks marked on the card?

The problem might reside in the fact that storage equipment manufacturers often abuse the meaning of the SI Gigabyte .

For them a GB is 10^9 and not 1073741824 bytes as the SI defines (1024*1024*1024). So 4 GB becomes around 3.7 GB. Take a few more bytes for volume information and the available space goes down to 3.6GB ( that's what windows reports ).

Anyway I've bought a 8GB card to be able to use your images as downloaded.

You should look into creating torrents for this, you might save yourself some bandwidth :)


Downloading R10 video version right now.


Thank you or your work.

magnet:?xt=urn:btih:52f6a7d1e5dad3e058ca50a3e83f8d2d9678dced&dn=olinuxino%5Fxfce-r10-video.zip&tr=http%3A%2F%2Fannounce.torrentsmd.com%3A8080%2Fannounce

it took a long time to download the last image so im posting a magnet so it can be downloaded from here as well. i have unlimited upload and can give about 1.5 M/s

if this is naughty in any way, then scrub the link, otherwise im happy to keep a torrent of the latest and greatest image alive 24/7 from here in bulgaria.

I've got no problems with it.

Quote from: drspastic on January 23, 2013, 11:24:31 PM
magnet:?xt=urn:btih:52f6a7d1e5dad3e058ca50a3e83f8d2d9678dced&dn=olinuxino%5Fxfce-r10-video.zip&tr=http%3A%2F%2Fannounce.torrentsmd.com%3A8080%2Fannounce

it took a long time to download the last image so im posting a magnet so it can be downloaded from here as well. i have unlimited upload and can give about 1.5 M/s

if this is naughty in any way, then scrub the link, otherwise im happy to keep a torrent of the latest and greatest image alive 24/7 from here in bulgaria.

When i go in /dev i was ttyS0 ttyS1 ttyS2 and ttyS3. As far as i understood that ttyS0 is uart1. Since olinuxino-wifi supports 2 UART. What ttyS[1-2-3] corresponds to?

One more thing:  i could not type "|" this symbol in R9 image.   

Quote from: ozbekmert on January 24, 2013, 12:29:12 AM
When i go in /dev i was ttyS0 ttyS1 ttyS2 and ttyS3. As far as i understood that ttyS0 is uart1. Since olinuxino-wifi supports 2 UART. What ttyS[1-2-3] corresponds to?

One more thing:  i could not type "|" this symbol in R9 image.

a) Different consoles.

b) You might need to change your keyboard mapping - I'm not sure what the default is, but I'm guessing 104 key english, which might not be what you are using.

can someone please scrub my torrent post! the image does not seem to be ok. burned to card fine but the directories in second patition are corrupt. i will download again and test/repost. thanks and sorry

can you post  the checksum of v10 please

Finally :) I'm running the R10 Video version and it's looking good but:

- what's is the root password for su ? I've tried olinuxino, password and root... none worked :(
- shouldn't "shutdown -h now" work? how do I make sure I don't corrupt this image?
- I'm going to try to have WPA2 working... is this possible?


I'm uploading this image (r10) to https://mega.co.nz/ and I'll share the link when its done, should be faster ( it took me nearly 4h at +-74KBs :''( )

Quote from: drspastic on January 24, 2013, 01:28:22 AM
can you post  the checksum of v10 please

SHA1: C7C981CB554B5D369E88FD859B01828245C10DAF

Quote from: buzibus on January 24, 2013, 02:30:04 AM

Finally :) I'm running the R10 Video version and it's looking good but:

- what's is the root password for su ? I've tried olinuxino, password and root... none worked :(
- shouldn't "shutdown -h now" work? how do I make sure I don't corrupt this image?
- I'm going to try to have WPA2 working... is this possible?


I'm uploading this image (r10) to https://mega.co.nz/ and I'll share the link when its done, should be faster ( it took me nearly 4h at +-74KBs :''( )

1) You can change the password via sudo if you need to. Or better yet, just use sudo. I think I remember the su pw, but I'm not sure to be honest. The point is that it shouldn't be posted, or easily guessed, thereby creating a security risk. Thus it isn't.

2) Yes, it should work. Since we are using a journalled file system, you are relatively safe in killing power to the board, but there is always the chance of corruption.

3) Ok... so maybe I should look into why Apache is being slow. Perhaps this evening. Or tomorrow. Or sometime.  Since I have direct vpn access to the machine, I never use http for transfers. I'll look into it.

oops I forgot about sudo before shutdown and su -  Doh

but "sudo shutdown -r now" isn't causing a restart but a complete shutdown


ive gotten the keyboard set to my local def and now I've to get wpa2 working... now that sounds like its going to be a bigger challenge :D


(upload to mega is still at 66% :| )



thanks for your help jwischka

Not a problem - WPA2 should work fine - I am running it here, and it should be as simple as changing the SSID/key in /etc/network/interfaces

You can do sudo reboot, and that will get you a restart.

Looking into why apache is being annoying, but no love yet.

Quote from: buzibus on January 24, 2013, 03:07:27 AM

oops I forgot about sudo before shutdown and su -  Doh

but "sudo shutdown -r now" isn't causing a restart but a complete shutdown


ive gotten the keyboard set to my local def and now I've to get wpa2 working... now that sounds like its going to be a bigger challenge :D


(upload to mega is still at 66% :| )



thanks for your help jwischka

weird... wlan2 does not appear anymore if I type ipconfig -a

idk what I did, the settings in /etc/network/interfaces appear correct.... dunno :(

I'm a bit tired, I think im going to sleep and try again in the morning



so... here's a FASTER download link for the R10 video image:

https://mega.co.nz/#!QNtWXbQZ!cdw3zQ_ZgdeE5By_ogGgECKq9486CtG65NT7yYqhDRQ

The revision Rxx are built for A13 board revision B.
Did anybody test it on the board revision C? Does it work?

Thanks.

Yup, I've got a Rev.C Board and it works great!

Thank you very much! Wireless seems very stable (built in).

One question - it seems to be missing sun4i_gpio? I've also got some troubles getting an USB webcam to work, but I haven't done much in this direction yet.

Gpio may have been lost when I deleted everything. Will look into it and recompile. There may not be webcam support at the moment either, but I can add it if necessary. I hope.

Quote from: vstef_is on January 24, 2013, 10:50:33 AM
Yup, I've got a Rev.C Board and it works great!

Thank you very much! Wireless seems very stable (built in).

One question - it seems to be missing sun4i_gpio? I've also got some troubles getting an USB webcam to work, but I haven't done much in this direction yet.

Thanks jwischka my board wifi is working.


Idk what happened but today when I booted the card Wifi was working without having to change anything. just worked whe turned on.


I just hope nothing is wrong with the board because yesterday Wifi wasn't working in Debian or Android


Has anyone tested the download link for the image I posted earlier ?

its getting near to perfection. i stuck the .deb firmware on for my rt3070usb and internet is fast and stable.
i really like this one!

but where is the audio? i cant find any audio devices. pity because i intend using audio in/out for amateur radio as a data modem. i loaded fldigi which usually finds any audio device but it listed nothing.

also, i wondered if you disabled the gpu, if so what do i need to do to get faster graphics. i so far only tested graphics on youtube in iceweasel and bzflags tank game and its very jumpy. i assume that when i get the audio up this will become an issue with the spectrum display and waterfall charts i need for the ham radio stuff. im assuming the gpu should deal with a lot of the fast fourier transforms.

otherwise the v10 image is crisp and fast, the best yet by far. and thanks for the checksum, my first download was very corrupted.

well its now 2am and time for bed. spent 4 hours seeing what goodies works from synaptic. chromium browser dont by the way but that sucks anyhow.

its a buzz to see this board browsing the net properly. more power to you!

I'll check on the audio and video issues... they may require a kernel adjustment.

Glad it's working well for you.

Quote from: drspastic on January 25, 2013, 02:03:41 AM
its getting near to perfection. i stuck the .deb firmware on for my rt3070usb and internet is fast and stable.
i really like this one!

but where is the audio? i cant find any audio devices. pity because i intend using audio in/out for amateur radio as a data modem. i loaded fldigi which usually finds any audio device but it listed nothing.

also, i wondered if you disabled the gpu, if so what do i need to do to get faster graphics. i so far only tested graphics on youtube in iceweasel and bzflags tank game and its very jumpy. i assume that when i get the audio up this will become an issue with the spectrum display and waterfall charts i need for the ham radio stuff. im assuming the gpu should deal with a lot of the fast fourier transforms.

otherwise the v10 image is crisp and fast, the best yet by far. and thanks for the checksum, my first download was very corrupted.

well its now 2am and time for bed. spent 4 hours seeing what goodies works from synaptic. chromium browser dont by the way but that sucks anyhow.

its a buzz to see this board browsing the net properly. more power to you!

i will have a poke about tomorrow just to see if i can rule out my usual stupidity. i just tried freetuxtv and the video quality was quite smooth and watchable. also the deinterlace settings worked (i assume this must mean the gpu works after all?) im not quite sure the appropriate commands to enquire yet as to the status of gpu and audio devs. thats tomorows lesson. i just looked through /dev and it does look rather empty regards sound stuff i expect to see.

anyone else experiencing problems in xfce ?

im connecting with VNC and in Wicd network manager it says: no networks found

if I try to start the browser I get "failed to execute default browser" and "input/output error"

I've checked and the browser is "debian sensible" on default applications


EDIT: got wireless to work in xfce...had to change interface to wlan2 (doh again)

Quote from: buzibus on January 25, 2013, 03:19:38 AM
anyone else experiencing problems in xfce ?

im connecting with VNC and in Wicd network manager it says: no networks found

if I try to start the browser I get "failed to execute default browser" and "input/output error"

I've checked and the browser is "debian sensible" on default applications


EDIT: got wireless to work in xfce...had to change interface to wlan2 (doh again)

apt-get iceweasel

Quote from: drspastic on January 25, 2013, 03:55:25 AM
apt-get iceweasel

first first I had thought sensible-browser was a browser by itself, then i was trying chromium but unfortunately it just spelt out "segmentation fault"


thanks im posting from my olinuxino board now :) - i have not enough gui linux experience for this, im more of a command line kind of person hehehe ;)

Quote from: drspastic on January 25, 2013, 02:30:19 AM
i will have a poke about tomorrow just to see if i can rule out my usual stupidity. i just tried freetuxtv and the video quality was quite smooth and watchable. also the deinterlace settings worked (i assume this must mean the gpu works after all?) im not quite sure the appropriate commands to enquire yet as to the status of gpu and audio devs. thats tomorows lesson. i just looked through /dev and it does look rather empty regards sound stuff i expect to see.

I believe the Mali should be enabled, at least as a module. If support is not compiled into the kernel at all, you get the headless kernel with the full 512 of memory. I do recall at some point I added various audio drivers, and had some compile issues on some of them, so then took them out. It's possible that I just broke something along the way, so I'll try to look at that tomorrow.

My board does not have super functional VGA, so it's hard(er) for me to debug video problems. I suspect some of the video issues you're seeing on YouTube probably boils down to 1) I don't think we have full acceleration support in the sunxi kernel for the Mali yet, and 2) decoding FLV's (or WebM) is just a reasonably processor intensive task in general, so it may be tough for the A13 to keep up with, even given the decoding.

Given the lack of people commenting on Debian/Ubuntu issue, I will probably move to Ubuntu in the next build because it is my personal preference. In that vein, does anyone have any packages that they would specifically like to see in the base release?

hmmm so if mali 400 aint fully implemented in the kernel then i guess allwinner are trying to push us all to android....no thanks allwinner.

as for ubuntu, i used it for many years but gave up when it started getting really fat and lazy like windows.
i imagine switching now will put you back a few paces, it it worth it? besides debian is a very standard base with lots of support and more likely to fit with industrial use.

a good one to play with if you want to expand is slack. just for pure stability.

both deb and buntu have a really fat bag of apps available , i hear rumours that ubuntu changed something with the latest releases that make package management incompatible with earlier staff and deb.
maybe that wont affect arm repos. im no expert.

for my penny, you invested so much time getting so far with deb v10 image it might be best to stay debian.

and hope allwinner cough up some more code for the mali400

ps. how about trying to shrink the system into a true lightweight like the many sub 200meg distros? what is it that makes them not need the extra gigabytes? the board dont have much hardware to support.

It's easy to make it smaller: just leave out all the things people asked for above.  Which would be silly!

To get a small Linux build your own  :)

Quote from: drspastic on January 25, 2013, 09:44:17 AM
hmmm so if mali 400 aint fully implemented in the kernel then i guess allwinner are trying to push us all to android....no thanks allwinner.

as for ubuntu, i used it for many years but gave up when it started getting really fat and lazy like windows.
i imagine switching now will put you back a few paces, it it worth it? besides debian is a very standard base with lots of support and more likely to fit with industrial use.

a good one to play with if you want to expand is slack. just for pure stability.

both deb and buntu have a really fat bag of apps available , i hear rumours that ubuntu changed something with the latest releases that make package management incompatible with earlier staff and deb.
maybe that wont affect arm repos. im no expert.

for my penny, you invested so much time getting so far with deb v10 image it might be best to stay debian.

and hope allwinner cough up some more code for the mali400

The main reason to go to Ubuntu over Debian would be faster updates (which means, perhaps, better support for armhf, faster). Most of the "slowness" people complain about with Ubuntu is related to Unity, and is a non-factor on this board. At this point, I have a Ubuntu image running on a separate card, so it really wouldn't be a huge deal to go that direction.

The main reason to stay with Debian is support for older processors (ARM6, like in the Raspberry Pi). That's not really a huge factor for us.

Package availability is pretty similar, and I would see it being about the same going forward, as both distros have plans to support the armhf platform going forward. Which is good for us.

Quote from: drspastic on January 25, 2013, 09:49:39 AM
ps. how about trying to shrink the system into a true lightweight like the many sub 200meg distros? what is it that makes them not need the extra gigabytes? the board dont have much hardware to support.

Quote from: JohnS on January 25, 2013, 10:38:03 AM
It's easy to make it smaller: just leave out all the things people asked for above.  Which would be silly!

To get a small Linux build your own  :)

^^ what he said. Take out XFCE. Take out swap. Don't compile in support for USB networking, USB video cameras, audio devices, etc. Don't include GCC, python, or development tools, etc. Which is what you would do if you were trying to make a deployable image in an industrial environment - literally rip everything away. But if you're trying to make something generic that comes up and boots into a nice, pretty environment so that people can use it without having problems... you need some of that other stuff.

More to the point, though, I'm wondering if there is some junk on my memory card that is being read in the DD that doesn't really need to be there - stuff that could be perhaps "fixed" by a clean card, and drop the compressed size of the image. I may try to build an entirely new card today from scratch and see how that goes.

i suppose the dev stuff does eat a lot but true we will need it to be able to progress in our individual end applications. i think i have got stuck into an unhealthy pattern of thinking after years of using tiny, include everything, run in ram distros. looking back i see that on my laptop my 'tiny 130M distro' after i added the dev stuff (160M) and then another 500M that seems to have fell in a memory hole somewhere, it aint that small any more.

now about the sound, do you think i will be able to get it running without any fiddling with kernels?
i will do some reading up on the interweb and see if theres anything, but please tell me if im wasting my time.

the linaro image did seem to work well but i kicked it aside as i couldnt get the net working. i will bitch no more about ubuntu. its been a while since i last used it, although i do use lupu 5.28 which is based on it and runs well in minimal systems.

most of the ham radio apps i need are available in ubuntu and debian, and i will look forward to '11' whatever flavour you make it.

I'm looking into it now. Audio support is compiled into the kernel, but as a module so I think that should work. If you can't get it working then I will try to compile built-in and we can see how that goes. It also looks the CEDAR support was pulled out somewhere, which could explain why some of the video performance was bad. I'm recompiling a new kernel with that built-in as we speak.

I think I'm going to build an image from linaro-nano, which should be quite lean - the default boot takes about 3 seconds, though obviously it will take longer when we add in some other toys.

GPIO will be back in the next kernel... forgot to add that in after the delete.

Quote from: drspastic on January 25, 2013, 04:40:20 PM
i suppose the dev stuff does eat a lot but true we will need it to be able to progress in our individual end applications. i think i have got stuck into an unhealthy pattern of thinking after years of using tiny, include everything, run in ram distros. looking back i see that on my laptop my 'tiny 130M distro' after i added the dev stuff (160M) and then another 500M that seems to have fell in a memory hole somewhere, it aint that small any more.

now about the sound, do you think i will be able to get it running without any fiddling with kernels?
i will do some reading up on the interweb and see if theres anything, but please tell me if im wasting my time.

the linaro image did seem to work well but i kicked it aside as i couldnt get the net working. i will bitch no more about ubuntu. its been a while since i last used it, although i do use lupu 5.28 which is based on it and runs well in minimal systems.

most of the ham radio apps i need are available in ubuntu and debian, and i will look forward to '11' whatever flavour you make it.

Rather than less I was going to ask for more.  Actually, sort of the opposite (read on...)

With chasing my AXP209 problem I've not kept up with recent releases but what I mean above is:
as well as more can we have less?  That is, the less command :)

John

Quote from: JohnS on January 25, 2013, 04:58:36 PM
Rather than less I was going to ask for more.  Actually, sort of the opposite (read on...)

With chasing my AXP209 problem I've not kept up with recent releases but what I mean above is:
as well as more can we have less?  That is, the less command :)

John

Heh... yeah, I'll add it in.

jwischka,

packages that I'd like to see?

• As I said earlier, with a networked nix box, time stamps are close to mandatory, so ntp would be nice
• I personally like sshfs to mount my *real* disk during development so I don't wear out the SD card
• Yeah - less would be more ;)
• Maybe g++?
• How much room is left to fill 4G?

I will also use apache2 and a bunch of simple web pages as the ultimately only way to communicate with the OlinuXino in my application.  But I won't favor including it in the standard package.  After all, once networked, apt-get install is rather easy to use.

You said you had performance issues with apache?  It served my (admittedly simple) web pages with rather good performance - no complaints here...

NTP will be on, and less as well. Installing sshfs right now. g++ has been on all the images.

There should be plenty of room, so we can add whatever, really. I agree, though, that apache is probably something best left for others to install.

It's a speed issue - for some reason on my main webserver apache seems to be capping download speeds at ~300KB/s. Which is annoying. Haven't had a ton of time to debug it, but this is the first time I've ever really seen the issue with apache.

Quote from: Ekkehard on January 25, 2013, 06:39:18 PM
jwischka,

packages that I'd like to see?

• As I said earlier, with a networked nix box, time stamps are close to mandatory, so ntp would be nice
• I personally like sshfs to mount my *real* disk during development so I don't wear out the SD card
• Yeah - less would be more ;)
• Maybe g++?
• How much room is left to fill 4G?

I will also use apache2 and a bunch of simple web pages as the ultimately only way to communicate with the OlinuXino in my application.  But I won't favor including it in the standard package.  After all, once networked, apt-get install is rather easy to use.

You said you had performance issues with apache?  It served my (admittedly simple) web pages with rather good performance - no complaints here...

Dear jwischka thank you for this amazing work.  I have a question regarding booting. I am using headless version (R10) and I realize that the only way to boot normal the board is to have connected the serial cable.  Is this a normal behavior? This is a problem because i want to use the board on embedded system and there is no way to connect the serial port every time. Do you know who I can change this behavior
Thank you

You should be able to connect via SSH or VNC, once the board is booted and has a network connection - but the board should boot without a serial connection.

Quote from: skenter on January 25, 2013, 11:58:12 PM
Dear jwischka thank you for this amazing work.  I have a question regarding booting. I am using headless version (R10) and I realize that the only way to boot normal the board is to have connected the serial cable.  Is this a normal behavior? This is a problem because i want to use the board on embedded system and there is no way to connect the serial port every time. Do you know who I can change this behavior
Thank you

Yes i know, but i can't undrestand why is not bootin with out the serial. With out it is geting on SUNXI

Quote from: jwischka on January 26, 2013, 12:05:30 AM
You should be able to connect via SSH or VNC, once the board is booted and has a network connection - but the board should boot without a serial connection.

Quote from: skenter on January 25, 2013, 11:58:12 PM
Dear jwischka thank you for this amazing work.  I have a question regarding booting. I am using headless version (R10) and I realize that the only way to boot normal the board is to have connected the serial cable.  Is this a normal behavior? This is a problem because i want to use the board on embedded system and there is no way to connect the serial port every time. Do you know who I can change this behavior
Thank you

How do you know it's getting stuck on sunxi? I'm not entirely sure what you mean. Do you mean that it's not getting past uboot?

Quote from: skenter on January 26, 2013, 12:09:32 AM
Yes i know, but i can't undrestand why is not bootin with out the serial. With out it is geting on SUNXI

jwischka,

you said g++ is always installed.  You sure?
I just did
$ g++ --version
-bash: g++: command not found
$
on a brand-new headless R10 image. gcc is there, just g++ isn't.  Not sure how important it is for this community - it is 19.1 MB additional space...  My code won't compile without it, but of course I don't mind at all installing it myself.  Just wanted to be sure we're on the same page with respect to it being pre-installed.

Hrm. I was certain it was on there. Anyway, it is (was) on R11, which I checked when you asked. Of course, now I'm rebuilding R11 from scratch, because I'm having problems getting VNC to load properly. Sigh...

I also have it on my own personal build (which is on a separate card) so I might have been thinking of that. Hard to keep all these cards straight. ;)

Quote from: Ekkehard on January 26, 2013, 12:40:05 AM
jwischka,

you said g++ is always installed.  You sure?
I just did

Code Select Expand

$ g++ --version
-bash: g++: command not found
$
on a brand-new headless R10 image. gcc is there, just g++ isn't.  Not sure how important it is for this community - it is 19.1 MB additional space...  My code won't compile without it, but of course I don't mind at all installing it myself.  Just wanted to be sure we're on the same page with respect to it being pre-installed.

Thank you for your reply. Here is what I am trying, when I have connected the serial port of the board to my laptop terminal and I am power up the board everything works fine. Debian start and I can use vnc. Then I am disconnecting the serial port and power up the board again. After couples o minutes, that I am realize tha there is no ip connectivity and when I am using serial to get at the board I am getting only the sun5i# prompt.

Quote from: jwischka on January 26, 2013, 12:31:26 AM
How do you know it's getting stuck on sunxi? I'm not entirely sure what you mean. Do you mean that it's not getting past uboot?

Quote from: skenter on January 26, 2013, 12:09:32 AM
Yes i know, but i can't undrestand why is not bootin with out the serial. With out it is geting on SUNXI

I've verified that this is an issue (in R11 as well) - I'll go back and see if I can find what the problem is
and get back to you on it, but may wait until Sunday or Monday. Obv. I'll have to get that fixed.

Quote from: skenter on January 26, 2013, 12:45:12 AM
Thank you for your reply. Here is what I am trying, when I have connected the serial port of the board to my laptop terminal and I am power up the board everything works fine. Debian start and I can use vnc. Then I am disconnecting the serial port and power up the board again. After couples o minutes, that I am realize tha there is no ip connectivity and when I am using serial to get at the board I am getting only the sun5i# prompt.

Quote from: jwischka on January 26, 2013, 12:31:26 AM
How do you know it's getting stuck on sunxi? I'm not entirely sure what you mean. Do you mean that it's not getting past uboot?

Quote from: skenter on January 26, 2013, 12:09:32 AM
Yes i know, but i can't undrestand why is not bootin with out the serial. With out it is geting on SUNXI

R11 released: (copied from the original post)

R11 (25 Jan 2013): We are now on Ubuntu. Specifically a build based on the linaro-developer branch. Image sizes will now be smaller downloads, owing to me building the card from scratch. Still requires a 4GB card, though. Both kernels are now included in the same release, and you can switch between them using "go-headless" and "go-video" scripts in the user folder (video kernel default). Various packages may be missing now, due to building more or less from scratch. If I'm missing one of your packages, let me know. Firefox and Filezilla are now installed by default.

Additionally, there are two different kernels installed in the image, and they are now switchable via a script on the card. The card will boot with the "video" kernel, which enables output to the VGA port, at the expense of about 100MB of memory headed to the MALI and the CEDAR video decoder. The second kernel is the "headless" version, which gives no video output, but gives you all of the memory on the board. Your choice. You can switch between them using two scripts - "go-headless" and "go-video" in the user folder - note these need to be run with sudo (i.e., "sudo ./go-headless").

Get previous version (R11) here. (http://files.wisch.org/olinuxino_xfce-r11.zip)


Known bug: you do have to have a serial connection for some reason. I'm looking into it - I think it's a u-boot problem, actually, but we'll see. May be Monday before it's done.

Thank you jwischka, I have spend a lot of time trying to figure out what is happening.

Quote from: skenter on January 26, 2013, 12:45:12 AM
Thank you for your reply. Here is what I am trying, when I have connected the serial port of the board to my laptop terminal and I am power up the board everything works fine. Debian start and I can use vnc. Then I am disconnecting the serial port and power up the board again. After couples o minutes, that I am realize tha there is no ip connectivity and when I am using serial to get at the board I am getting only the sun5i# prompt.

Quote from: jwischka on January 26, 2013, 12:31:26 AM
How do you know it's getting stuck on sunxi? I'm not entirely sure what you mean. Do you mean that it's not getting past uboot?

Quote from: skenter on January 26, 2013, 12:09:32 AM
Yes i know, but i can't undrestand why is not bootin with out the serial. With out it is geting on SUNXI

i saw the bug report and decided to save my card another write cycle.... i still cant work out how to get audio on the last debian image, although you no longer support the debian base, i still quite like this image and would happily stick with it, even just as a failsafe backup card. is it possible to issue a patch for the mali and sound? i did all sorts of searches for the module i need for audio and google gave me nothing i could use. i feel stupid. im sure its nothing that special.
if not i will have to wait and hope for the linaro image to start taking shape.
i suffer the human condition: scared of change, impatient, and ignorant.
all hail the aliens!

ps; is there a way for us who appreciate your work to issue some kind of recompense for your hard work so far? if not then maybe olimex could open some kind of donation system seeing as you are doing so much to promote the platform. maybe next time i order i can give them a few extra bucks towards sending you the latest a10 offering when it comes out.

Hello jwischka,

I use (R10) version video, vga is working well  :-*
But when i want to use my lcd, nothing is on the screen.
I change the script.bin (use the script.bin for lcd from the wiki)
Whats wrong?
An older image works with lcd

Best regards
michael

Hello jwischka,

i want to disable serial console(ttyS0) because i am going to use uart1. I googled lots of stuff including changing /etc/inittab ... Everything come up with re-compiling kernel. Could you help me with this? I am new at this field.

You need to make some or all of these changes:
1. stop bootstrap messages
2. stop debug messages as the kernel boots / shuts down
3. stop the login prompt
4. er, anything I've forgotten

#1 build your own u-boot (or find out if it can be told to be quiet - I've not looked, bu try the doc!)
#2 I think you can do this by tweaking the command line passed to the kernel, which now says something like console=ttyS0
#3 yes, change inittab (look for the line about getty and tty1)
#4 be sure to post what you find LOL

John

Quote from: skenter on January 26, 2013, 08:46:19 AM
Thank you jwischka, I have spend a lot of time trying to figure out what is happening.

Quote from: skenter on January 26, 2013, 12:09:32 AM
Yes i know, but i can't undrestand why is not bootin with out the serial. With out it is geting on SUNXI

Ok. I know why this is happening, and I've come up with a fix for it. I don't really like the fix for it, but I think for most people it won't really matter. Essentially the u-boot output is getting looped around to the u-boot input when the serial console is not plugged in (I'm not sure why this is happening at the moment, but I know that it is, and is causing the problem). Which means that it's pausing the boot, which is very clever. Setting the bootwait to zero fixes this, but means you can't pause u-boot. So there's that. The next release (R12, uploading soon) will contain a zero-wait default, so everyone should see faster boot times and booting when you don't have a serial console plugged in - but you may also not be able to pause the boot. Pick your poison.

Quote from: drspastic on January 27, 2013, 05:17:32 PM
i saw the bug report and decided to save my card another write cycle.... i still cant work out how to get audio on the last debian image, although you no longer support the debian base, i still quite like this image and would happily stick with it, even just as a failsafe backup card. is it possible to issue a patch for the mali and sound? i did all sorts of searches for the module i need for audio and google gave me nothing i could use. i feel stupid. im sure its nothing that special.
if not i will have to wait and hope for the linaro image to start taking shape.
i suffer the human condition: scared of change, impatient, and ignorant.
all hail the aliens!

The kernel should be independent of the debian/ubuntu issues - you should be able to grab the kernel from the latest releas(es) and put it on the old image, and all should be fine. If you're interested, I can pack the kernel/modules up for download and you can get them separately, and give you instructions for replacement, if necessary. Likewise you could always go back to the previous kernels if they actually worked.

The newest video kernels should have Mali and sound support - if you try them out and they don't, let me know and I'll see if I can help you debug it.

Quote from: drspastic on January 27, 2013, 05:24:12 PM
ps; is there a way for us who appreciate your work to issue some kind of recompense for your hard work so far? if not then maybe olimex could open some kind of donation system seeing as you are doing so much to promote the platform. maybe next time i order i can give them a few extra bucks towards sending you the latest a10 offering when it comes out.

Thank you for the thought - I do appreciate it, but I'm really not doing it for money. I'd love if olimex sent me a developer version of the A10s micro board, and I'd probably consider doing releases for other boards they sent me, but it's certainly not something I expect. If you really do feel like it's been beneficial and would like to donate something, I would ask that you find a local organization in your area that feeds hungry people and give the money to them - I am very fortunate to have more than I need, but try to spend time working with and helping people who have real trouble putting food in their mouths. Even if they are on the other side of the world, their needs are more important than my beer fund. Again, I really do appreciate the thought, and I would be honored if you gave the price of a meal to a local organization that feeds the hungry.

Quote from: mike1968 on January 27, 2013, 08:31:17 PM
Hello jwischka,

I use (R10) version video, vga is working well  :-*
But when i want to use my lcd, nothing is on the screen.
I change the script.bin (use the script.bin for lcd from the wiki)
Whats wrong?
An older image works with lcd

Best regards
michael

It looks like LCD support may have accidentally been removed. I've added it back in in R12, along with some framebuffer support, which takes the memory down a bit on the video side of things. I'm open to removing it, if people don't use the frame buffer parts of things. I run headless, so I don't care. Let me know what seems to work best for you (and let me know if it doesn't work in the most recent release (R12).

There's also been some traffic on the sunxi github about LCD modules not loading properly. I haven't been paying attention to it, but if this still doesn't work let me know and I'll actually read the emails and see if it applies to our situation.

Quote from: ozbekmert on January 27, 2013, 09:12:08 PM
Hello jwischka,

i want to disable serial console(ttyS0) because i am going to use uart1. I googled lots of stuff including changing /etc/inittab ... Everything come up with re-compiling kernel. Could you help me with this? I am new at this field.

Do you need help recompiling the kernel? You should be able to do it with a uEnv.txt file or boot.scr. I can definitely give you pointers on recompiling the kernel though.


I am currently finalizing the image and making sure most stuff works. I'll post again when it's finished uploading.

R12 is up.

Should have LCD support back in (sorry!). If it doesn't work, let me know.

Bootwait has been changed to "fix" a problem wherein the board wouldn't start if the serial console wasn't attached. This is not a good fix, and it's not one that I really like, but it is one that really works. I'm open to better ways of fixing it, if someone has one.

Also added some framebuffer support, which has taken us back down into the ~330MB total available memory range on the video kernel. For those of you who use video, let me know what your primary usage is - if most of you don't want/need the CEDAR acceleration, you could have 80MB of memory back. I never hook up a monitor, so it's no skin off my back either way. If we had 1GB instead of 512MB to play with, none of this would really matter, but that's how life goes.

A couple of package updates, added aptitude.

Get the image (R12) here. (http://files.wisch.org/olinuxino_xfce-r12.zip)

thanks jwischka for making new releases

and the DL now is much faster - thanks for that too - no more waiting 3h to get the file :)

No problem. I'm not sure what changed on the server (or, perhaps I should say, I didn't change anything), but it has seemed faster today in download times. Let's hope it's permanent!

Quote from: buzibus on January 29, 2013, 03:20:57 AM
thanks jwischka for making new releases

and the DL now is much faster - thanks for that too - no more waiting 3h to get the file :)

hi, tanks for the offer yes it would be great if you post the kernel and modules so i can mix and match. it would also help with the learning curve. maybe include source code too, if its not to big, or put in a seperate link. its easier to learn if i see all the building blocks in front of me.

For source you'll want to just clone into the github - then I can post .config files if you want. There are a few guides on the forum about building kernels that help things along.

I probably don't have the old versions of kernels (everything pre-10, probably) but will try to post them along with instructions on changing them out later on.

Quote from: drspastic on January 29, 2013, 08:45:08 PM
hi, tanks for the offer yes it would be great if you post the kernel and modules so i can mix and match. it would also help with the learning curve. maybe include source code too, if its not to big, or put in a seperate link. its easier to learn if i see all the building blocks in front of me.

i look forward to it.

something strange something good something bad!

im still running R10. last night i went into root to view the contents of the boot partition, i didnt alter anything, just view. logged out. this morning i turn on and find i cant run startx from user account??????

so i try from root and x starts fine.... the missus reports she cant get online from the laptop so i go to my router and reconnect from seamonkey on the a13, internet back on.....the net does this here often, and i test loading a page.   AUDIO IS WORKING!!!!!! im amazed.

so the audio drivers only work from root account. any ideas how this can be? is it permissions on the device? how do i fix it?

also im stuck as to how to get the user account to run x again.... last night from root i installed abiword to view a text file. when i logged on today i noticed i didnt close some apps (a terminal window, and the mounted primary partition in file manager) i closed them down and logged out, tried again as user but no change.

ideas please!

aha! added olinuxino to audio in /etc/group

now i have fldigi running as a powerful smart modem on my yaesu ft817. what an upgrade!

now if only i can get this exact same setup running on the imx233 i could fit it inside the radio and add a tv port at the rear. this could be the best upgrade for the '817 in 15 years.

im looking at the size of the a13 micro but the different hardware/lack of tv out says NO!

I tried to write the R12 image on my 4GB µSD and it says that there's not enough space (about 300 blocks of 512 are missing) Is there a way to modify this? My SD card is an EMTEC 4GB class 4

yo might want to try writing the image to another device that is larger such as usb stick and then manually copy the contents of the second partition to a newly created ext4 partition on your target. i have done this a few times on cards that wont play nicely. you should find that your first attempt to dd your target card did write the first boot partition ok, then it spazzed out on the second when it ran out of space. so leave that first one be and gparted the second to fill the rest of the card.

I used an 8GB card LOL

Quote from: jibaine on February 02, 2013, 03:18:33 PM
I tried to write the R12 image on my 4GB µSD and it says that there's not enough space (about 300 blocks of 512 are missing) Is there a way to modify this? My SD card is an EMTEC 4GB class 4

ps. dont forget if you do this you will have to change permissions of /home/olinuxino/ back to olinuxino or you wont be able to startx or access your files. easy, sudo startx then right click properties of /home/olinuxino

was cedar enabled on the r10 release? is there a command line way to varify this and/or test it? would it help performance to allocate more ram to it at startup?

Hi jwischka , thank you for the modification on -boot's, boot delay to zero. I want to used relase R10 which is on debian but i don't know where to make this modification to boot delay. Is this a script.bin modification ?

Thank you

Quote from: jwischka on January 28, 2013, 12:31:07 AM
R12 is up.

Should have LCD support back in (sorry!). If it doesn't work, let me know.

Bootwait has been changed to "fix" a problem wherein the board wouldn't start if the serial console wasn't attached. This is not a good fix, and it's not one that I really like, but it is one that really works. I'm open to better ways of fixing it, if someone has one.

Also added some framebuffer support, which has taken us back down into the ~330MB total available memory range on the video kernel. For those of you who use video, let me know what your primary usage is - if most of you don't want/need the CEDAR acceleration, you could have 80MB of memory back. I never hook up a monitor, so it's no skin off my back either way. If we had 1GB instead of 512MB to play with, none of this would really matter, but that's how life goes.

A couple of package updates, added aptitude.

Get the image (R12) here. (http://files.wisch.org/olinuxino_xfce-r12.zip)

Sorry all - have been travelling and just got back in yesterday.

Quote from: drspastic on February 04, 2013, 04:51:20 PM
was cedar enabled on the r10 release? is there a command line way to varify this and/or test it? would it help performance to allocate more ram to it at startup?

I'm not sure. You can check the dmesg log and look for something taking 80MB of memory - if so, it should be activated.

Quote from: skenter on February 05, 2013, 11:19:36 AM
Hi jwischka , thank you for the modification on -boot's, boot delay to zero. I want to used relase R10 which is on debian but i don't know where to make this modification to boot delay. Is this a script.bin modification ?

Thank you

If you have a serial console, stop the boot (in u-boot) and type:

setenv bootdelay 0
saveenv

jwischka is possible to enable output on VGA during boot on R12?

Quote from: maurosk on February 05, 2013, 06:24:50 PM
jwischka is possible to enable output on VGA during boot on R12?

Should be a script.bin issue - you do have the video kernel enabled, right?

Yes i'm capable to start X, but during the boot i have black screen :D

Oh - so you're wanting debug messages during boot, before the login prompt comes up? I'll have to look in to that.

Yes, if possible i want enable the output of the "boot sequence" on the vga, or bettere is enabling a bootsplash :D

I know there is an option for a boot image in the kernel compile, but I'm not quite sure how to make that work. I'll have to look at the other, but it's not the highest priority at the moment, so may be for a future release.

Ok, tnx jwischka, you are very available!! The boot splash isn't priority i know, but is very beautifull :D.
You think is difficult implement the output on VGA of the boot sequence?

Quote from: maurosk on February 05, 2013, 06:59:00 PM
Ok, tnx jwischka, you are very available!! The boot splash isn't priority i know, but is very beautifull :D.
You think is difficult implement the output on VGA of the boot sequence?

Dunno... Haven't tried to do it yet :).

Ok, tnx a lot for your time :D

Quote from: jwischka on February 05, 2013, 03:13:50 PM
Sorry all - have been travelling and just got back in yesterday.

Quote from: drspastic on February 04, 2013, 04:51:20 PM
was cedar enabled on the r10 release? is there a command line way to varify this and/or test it? would it help performance to allocate more ram to it at startup?

I'm not sure. You can check the dmesg log and look for something taking 80MB of memory - if so, it should be activated.

Quote from: skenter on February 05, 2013, 11:19:36 AM
Hi jwischka , thank you for the modification on -boot's, boot delay to zero. I want to used relase R10 which is on debian but i don't know where to make this modification to boot delay. Is this a script.bin modification ?

Thank you

If you have a serial console, stop the boot (in u-boot) and type:

setenv bootdelay 0
saveenv

[90168.410000] Normal free:3056kB min:2884kB low:3604kB high:4324kB active_anon:83920kB inactive_anon:86332kB active_file:153544kB inactive_file:123604kB unevictable:0kB isolated(anon):40kB isolated(file):4kB present:520192kB mlocked:0kB dirty:5968kB writeback:0kB mapped:74508kB shmem:4972kB slab_reclaimable:15784kB slab_unreclaimable:6400kB kernel_stack:1512kB pagetables:2652kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:28 all_unreclaimable? no

86332kB, 83920kB, 74508kB... theres a few things about 80m. honestly im not sure what to look for.
i think i will need to do something about this (r10) kernel as i just realised that the sun4igpio is not (as far as i can tell) included, and i dont know how to proceed. i need to get those gpio working to do some control interfacing.

i just compiled thelinkbox repeater control software and was about to brainstorm how to adapt it to switch the radio transmitter on at the right moment when i realised gpio dont work at all.

(i reserve the right to have been stupid again and missed something obvious)

Yeah, that's not the part you need. GPIO is broker in R10, so that's a given, and fixable. I'll post a new kernel and instructions on how to install it when I can, but it will probably be tomorrow morning (eastern time, USA).

Hopefully that will knock out both your problems in one move.

Quote from: drspastic on February 08, 2013, 01:34:34 AM

Quote from: jwischka on February 05, 2013, 03:13:50 PM
Sorry all - have been travelling and just got back in yesterday.

Quote from: drspastic on February 04, 2013, 04:51:20 PM
was cedar enabled on the r10 release? is there a command line way to varify this and/or test it? would it help performance to allocate more ram to it at startup?

I'm not sure. You can check the dmesg log and look for something taking 80MB of memory - if so, it should be activated.

Quote from: skenter on February 05, 2013, 11:19:36 AM
Hi jwischka , thank you for the modification on -boot's, boot delay to zero. I want to used relase R10 which is on debian but i don't know where to make this modification to boot delay. Is this a script.bin modification ?

Thank you

If you have a serial console, stop the boot (in u-boot) and type:

setenv bootdelay 0
saveenv

[90168.410000] Normal free:3056kB min:2884kB low:3604kB high:4324kB active_anon:83920kB inactive_anon:86332kB active_file:153544kB inactive_file:123604kB unevictable:0kB isolated(anon):40kB isolated(file):4kB present:520192kB mlocked:0kB dirty:5968kB writeback:0kB mapped:74508kB shmem:4972kB slab_reclaimable:15784kB slab_unreclaimable:6400kB kernel_stack:1512kB pagetables:2652kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:28 all_unreclaimable? no

86332kB, 83920kB, 74508kB... theres a few things about 80m. honestly im not sure what to look for.
i think i will need to do something about this (r10) kernel as i just realised that the sun4igpio is not (as far as i can tell) included, and i dont know how to proceed. i need to get those gpio working to do some control interfacing.

i just compiled thelinkbox repeater control software and was about to brainstorm how to adapt it to switch the radio transmitter on at the right moment when i realised gpio dont work at all.

(i reserve the right to have been stupid again and missed something obvious)

thats great. im now going through the guide to compiling kernel to make sdcard just for practice. headbanging stuff. you sure have some patience. i need to get myself to the point where i can understand what you have done and sort things out myself instead of bothering you every other day.
i will be sitting at the screen waiting for your next output. definitely try and write a guide if you get time.

Are you sure that MALI and CEDAR are needed?

I have disabled them and am still getting the VGA output and more free RAM.
Since, MAIL and CEDAR implementations are yet to be "found out", there is no need to add their drivers support.

Quote from: nvd on February 08, 2013, 04:36:27 PM
Are you sure that MALI and CEDAR are needed?

I have disabled them and am still getting the VGA output and more free RAM.
Since, MAIL and CEDAR implementations are yet to be "found out", there is no need to add their drivers support.

They may not be needed for basic output, but there may be some benefit in video performance over not having the drivers installed. Full hardware acceleration has not been implemented yet, AFAIK, but what level of performance boost there is ... is unknown.

Quote from: drspastic on February 08, 2013, 12:42:51 PM
thats great. im now going through the guide to compiling kernel to make sdcard just for practice. headbanging stuff. you sure have some patience. i need to get myself to the point where i can understand what you have done and sort things out myself instead of bothering you every other day.
i will be sitting at the screen waiting for your next output. definitely try and write a guide if you get time.

You can download an R13 video kernel here (http://files.wisch.org/3.4-video-R13.tar.gz). It should have GPIO support, and CEDAR/Mali are enabled.

To install, unzip/tar the files to a folder. You'll need to copy the directory to your /lib/modules folder, and then you'll need to mount the first partition on the card and copy the new uImage over:

sudo su
cp -a 3.4.24-jwischka-3.4-video-20130127-R13+ /lib/modules/
mount /dev/mmcblk0p1 /mnt
cp uImage /mnt
sync
umount /mnt
reboot

After that you should be on R13 - "uname -a" will confirm.

As far as compiling the kernel... where are you getting stuck? If there's a part of the other guides that's unclear I can try to improve them by focusing on it.

You had mentioned in a message compiling a new kernel on the board itself. I'm not really sure that's something you want to do, for a couple of reasons. 1) the source code is large (~1GB), which means you're going to eat up a large portion (possibly) of your card. 2) compile times, even on my laptop (Core i5, SSD, plenty of ram) are on the order of 10+ minutes, meaning a compile on the board itself would be ... well, long. and 3) compiling on an SD card in general isn't the best idea, due to the disc wear.

In the kernel file, I've included my config file. Just copy config-video to .config in the directory where your source code is and type "make ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- uImage and it should build the same kernel file you downloaded (assuming your source is up to date, and you've made the changes for the i2c bus...)

Let me know where you're getting stuck, and I'll see what I can do.

Sorry to ask but which changes for the i2c bus?

John

Quote from: JohnS on February 08, 2013, 06:50:01 PM
Sorry to ask but which changes to the i2c bus?

John

Default frequency in the sunxi code is 400/200/200 (kHz). They need to be changed to 100 for it to work properly.

Thanks.

http://pastie.org/6098799 is my errors. it did take a long time to compile but i dont have a powerful machine or any machine that uses a physical hard disk im running on an 8gig card and everything since installing git and trying to compile has eaten much of that. still not sure how to clear the unused stuff out and free some gigs. now running git gc and its also taking a long time. i can wait.

im sure cedar support must help by giving hardware video decoding. otherwise its little brain has to do it in software.
as for mali, i think it must do something to improve 3d. i know nothing of the details but there are open source projects to implement mali400 support. i like bzflags and want more speed to not get blown up ;)

i will try to bung r13 kernel in tomorrow on a clone of my main 8gig card. see how the wind blows.

If it's a missing file the general idea is to find which package provides it and install it.

John

Did this happen with my .config file as well?

Quote from: drspastic on February 09, 2013, 12:52:54 AM
http://pastie.org/6098799 is my errors. it did take a long time to compile but i dont have a powerful machine or any machine that uses a physical hard disk im running on an 8gig card and everything since installing git and trying to compile has eaten much of that. still not sure how to clear the unused stuff out and free some gigs. now running git gc and its also taking a long time. i can wait.

im sure cedar support must help by giving hardware video decoding. otherwise its little brain has to do it in software.
as for mali, i think it must do something to improve 3d. i know nothing of the details but there are open source projects to implement mali400 support. i like bzflags and want more speed to not get blown up ;)

i will try to bung r13 kernel in tomorrow on a clone of my main 8gig card. see how the wind blows.

not tried yet

r13 kernel in, no change to 3d or video performance though. :(

will have to wait for olinuxino quad core?

Quote from: drspastic on February 09, 2013, 05:44:39 PM
r13 kernel in, no change to 3d or video performance though. :(

will have to wait for olinuxino quad core?

The A31? I would imagine that as (if) the Mali becomes better supported, you may see more improvement. I would also bet that you're running into memory problems. Are you trying Youtube videos, or what? I can hook up my cubieboard and see if it does better (since it has 2x the ram).

Quote from: jwischka on February 09, 2013, 05:49:25 PM

Quote from: drspastic on February 09, 2013, 05:44:39 PM
r13 kernel in, no change to 3d or video performance though. :(

will have to wait for olinuxino quad core?

The A31? I would imagine that as (if) the Mali becomes better supported, you may see more improvement. I would also bet that you're running into memory problems. Are you trying Youtube videos, or what? I can hook up my cubieboard and see if it does better (since it has 2x the ram).

youtube is unwatchable, and i know that with my bottom of the pile 500mhz android phone plays it smooth and my now rather dated samsung omnia 1 can pump decent video into the telly with a Marvell PXA312.

is it the linux overhead thats dragging the a13 performance down? i know nothing about gpu's so what part of the chain is NOT getting the apparently rather swift mali400 to do its stuff? is it the module that drives it? or does it need propriety drivers to function above basic usage like the nvidia stuff one has to stick onto an i86 system to make the graphics card do any fun stuff.

Shouldn't be overhead. Let me do sine testing and see if I can figure out what's going on.

Quote from: drspastic on February 09, 2013, 08:53:22 PM

Quote from: jwischka on February 09, 2013, 05:49:25 PM

Quote from: drspastic on February 09, 2013, 05:44:39 PM
r13 kernel in, no change to 3d or video performance though. :(

will have to wait for olinuxino quad core?

The A31? I would imagine that as (if) the Mali becomes better supported, you may see more improvement. I would also bet that you're running into memory problems. Are you trying Youtube videos, or what? I can hook up my cubieboard and see if it does better (since it has 2x the ram).

youtube is unwatchable, and i know that with my bottom of the pile 500mhz android phone plays it smooth and my now rather dated samsung omnia 1 can pump decent video into the telly with a Marvell PXA312.

is it the linux overhead thats dragging the a13 performance down? i know nothing about gpu's so what part of the chain is NOT getting the apparently rather swift mali400 to do its stuff? is it the module that drives it? or does it need propriety drivers to function above basic usage like the nvidia stuff one has to stick onto an i86 system to make the graphics card do any fun stuff.

Quote from: drspastic on February 09, 2013, 08:53:22 PM
youtube is unwatchable, and i know that with my bottom of the pile 500mhz android phone plays it smooth and my now rather dated samsung omnia 1 can pump decent video into the telly with a Marvell PXA312.

is it the linux overhead thats dragging the a13 performance down? i know nothing about gpu's so what part of the chain is NOT getting the apparently rather swift mali400 to do its stuff? is it the module that drives it? or does it need propriety drivers to function above basic usage like the nvidia stuff one has to stick onto an i86 system to make the graphics card do any fun stuff.

Ok... I haven't hooked this up to a monitor yet (don't have a VGA one handy), but can you try manually loading the Mali driver (which, being compiled as a module, isn't automatically loaded - something I can fix, probably)?

from a command prompt:

sudo modprobe mali

Does that seem to make any difference?

EDIT:

Also CEDAR?

sudo modprobe sun5i_cedar

Or possibly try R14, which should have video baked in (http://files.wisch.org/3.4-video-R14.tar.gz), rather than as a module.

Edit again: poo... looks like 3.4.29 has been merged in, which is what I get for not doing a pull. So in short, check this and see if it works, but expect an R15 on the way today or tomorrow.

R15 released.

Image (http://files.wisch.org/olinuxino_xfce-r15.zip)
Video Kernel (http://files.wisch.org/3.4-video-R15.tar.gz)
Headless Kernel (http://files.wisch.org/3.4-headless-R15.tar.gz)

Kernels updated to 3.4.29. Mali/Cedar now baked in (instead of a module) on video kernel.

Hello,

I've been away for a while, going to try R15 now.


Why 3 file releases now?


R10 was working fine until I had to reboot my router but Olinuxino didn't get an IP when the router came up ( no idea why, all the other systems were fine, including my trusty beaglebone ).

As I didnt had any screen or Uart cable nearby it stood there more than one week without reconnecting :(

Three files do that the guys who are still using Debian can just put the kernels on.

R10 still has the bug where you need the UART cable. If you flash R15 that should boot for you.

Quote from: buzibus on February 10, 2013, 03:03:34 AM

Hello,

I've been away for a while, going to try R15 now.


Why 3 file releases now?


R10 was working fine until I had to reboot my router but Olinuxino didn't get an IP when the router came up ( no idea why, all the other systems were fine, including my trusty beaglebone ).

As I didnt had any screen or Uart cable nearby it stood there more than one week without reconnecting :(

Hello,

Iam using r11 and it works well for me.

But i could not load "usbserial" modul

Please help

Best regards michael

many thanks for the 3 release system, it was no fun having to start from scratch each time a new build came out. will try 15 later and   let you know if built in kernel smoothes video any. no time at the moment to play compiling but would be good to see the config list of modules in/out/loadable you used

Quote from: jwischka on February 09, 2013, 11:21:00 PM

Quote from: drspastic on February 09, 2013, 08:53:22 PM
youtube is unwatchable, and i know that with my bottom of the pile 500mhz android phone plays it smooth and my now rather dated samsung omnia 1 can pump decent video into the telly with a Marvell PXA312.

is it the linux overhead thats dragging the a13 performance down? i know nothing about gpu's so what part of the chain is NOT getting the apparently rather swift mali400 to do its stuff? is it the module that drives it? or does it need propriety drivers to function above basic usage like the nvidia stuff one has to stick onto an i86 system to make the graphics card do any fun stuff.

Ok... I haven't hooked this up to a monitor yet (don't have a VGA one handy), but can you try manually loading the Mali driver (which, being compiled as a module, isn't automatically loaded - something I can fix, probably)?

from a command prompt:

sudo modprobe mali

Does that seem to make any difference?

EDIT:

Also CEDAR?

sudo modprobe sun5i_cedar

olinuxino@debian:~$ sudo modprobe mali
[sudo] password for olinuxino:
ERROR: could not insert 'mali': Bad address
olinuxino@debian:~$ sudo modprobe CEDAR
FATAL: Module CEDAR not found.
olinuxino@debian:~$

Quote from: drspastic on February 10, 2013, 04:18:26 PM

Quote from: jwischka on February 09, 2013, 11:21:00 PM

Quote from: drspastic on February 09, 2013, 08:53:22 PM
youtube is unwatchable, and i know that with my bottom of the pile 500mhz android phone plays it smooth and my now rather dated samsung omnia 1 can pump decent video into the telly with a Marvell PXA312.

is it the linux overhead thats dragging the a13 performance down? i know nothing about gpu's so what part of the chain is NOT getting the apparently rather swift mali400 to do its stuff? is it the module that drives it? or does it need propriety drivers to function above basic usage like the nvidia stuff one has to stick onto an i86 system to make the graphics card do any fun stuff.

Ok... I haven't hooked this up to a monitor yet (don't have a VGA one handy), but can you try manually loading the Mali driver (which, being compiled as a module, isn't automatically loaded - something I can fix, probably)?

from a command prompt:

sudo modprobe mali

Does that seem to make any difference?

EDIT:

Also CEDAR?

sudo modprobe sun5i_cedar

olinuxino@debian:~$ sudo modprobe mali
olinuxino@debian:~$ sudo modprobe CEDAR
FATAL: Module CEDAR not found.
olinuxino@debian:~$

Won't have to do that (and can't do that) now that they are built in. The modules don't exist anymore.

Quote from: mike1968 on February 10, 2013, 12:00:10 PM
Hello,

Iam using r11 and it works well for me.

But i could not load "usbserial" modul

Please help

Best regards michael

Does USB-serial not work? In other words, it may be compiled into the kernel itself, and you may not need to load it.

Also, do you see the same behavior on R15?

R15; assaultcube now renders and doesnt crash! not playable but a good test, so something changed for the better.

Quote from: drspastic on February 10, 2013, 06:00:43 PM
R15; assaultcube now renders and doesnt crash! not playable but a good test, so something changed for the better.

Youtube?

nah, it plays music if the video is static ie photo of album cover, and then only if i dont use the mouse!

anyone else reading this get youtube to stream ok?

Quote from: drspastic on February 10, 2013, 08:45:40 PM
nah, it plays music if the video is static ie photo of album cover, and then only if i dont use the mouse!

anyone else reading this get youtube to stream ok?

Can you run top while you're trying to play to see if it's a memory or CPU issue (or likely both).

Quote from: jwischka on February 10, 2013, 04:20:16 PM

Quote from: drspastic on February 10, 2013, 04:18:26 PM

Quote from: jwischka on February 09, 2013, 11:21:00 PM

Quote from: drspastic on February 09, 2013, 08:53:22 PM
youtube is unwatchable, and i know that with my bottom of the pile 500mhz android phone plays it smooth and my now rather dated samsung omnia 1 can pump decent video into the telly with a Marvell PXA312.

is it the linux overhead thats dragging the a13 performance down? i know nothing about gpu's so what part of the chain is NOT getting the apparently rather swift mali400 to do its stuff? is it the module that drives it? or does it need propriety drivers to function above basic usage like the nvidia stuff one has to stick onto an i86 system to make the graphics card do any fun stuff.

Ok... I haven't hooked this up to a monitor yet (don't have a VGA one handy), but can you try manually loading the Mali driver (which, being compiled as a module, isn't automatically loaded - something I can fix, probably)?

from a command prompt:

sudo modprobe mali

Does that seem to make any difference?

EDIT:

Also CEDAR?

sudo modprobe sun5i_cedar

olinuxino@debian:~$ sudo modprobe mali
olinuxino@debian:~$ sudo modprobe CEDAR
FATAL: Module CEDAR not found.
olinuxino@debian:~$

Won't have to do that (and can't do that) now that they are built in. The modules don't exist anymore.

Quote from: mike1968 on February 10, 2013, 12:00:10 PM
Hello,

Iam using r11 and it works well for me.

But i could not load "usbserial" modul

Please help

Best regards michael

Does USB-serial not work? In other words, it may be compiled into the kernel itself, and you may not need to load it.

Also, do you see the same behavior on R15?

R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Quote from: jwischka on February 10, 2013, 09:29:58 PM

Quote from: drspastic on February 10, 2013, 08:45:40 PM
nah, it plays music if the video is static ie photo of album cover, and then only if i dont use the mouse!

anyone else reading this get youtube to stream ok?

Can you run top while you're trying to play to see if it's a memory or CPU issue (or likely both).

cpu is about 95% eaten by gtk-gnash with the remainder taken by xorg mostly. i think she just runs out of steam. maybe i shouls try xvesa?

Quote from: drspastic on February 11, 2013, 01:35:39 AM

Quote from: jwischka on February 10, 2013, 09:29:58 PM

Quote from: drspastic on February 10, 2013, 08:45:40 PM
nah, it plays music if the video is static ie photo of album cover, and then only if i dont use the mouse!

anyone else reading this get youtube to stream ok?

Can you run top while you're trying to play to see if it's a memory or CPU issue (or likely both).

cpu is about 95% eaten by gtk-gnash with the remainder taken by xorg mostly. i think she just runs out of steam. maybe i shouls try xvesa?

Couldn't hurt. Might try using a lighter weight browser and see if that helps too...

im going away for a few days, when im back i will try compile your R16 for myself then start again with a basic filesystem like the old debian No-x and build it back up.

would there be any problems you can see with using the first partition as-is R16 and very minimal filesystem in #2 with a script to switch root to filesystem on an external 60Gb usb drive?

that would give me more space to compile stuff in and no problems with wear but im not sure on the data throughput of 480M/s usb2 against that of the micro sd HC card.
even if its slower i can still leave it to compile stuff overnight.

all the best and thanks for all the hard work

Quote from: drspastic on February 11, 2013, 02:14:26 AM
im going away for a few days, when im back i will try compile your R16 for myself then start again with a basic filesystem like the old debian No-x and build it back up.

would there be any problems you can see with using the first partition as-is R16 and very minimal filesystem in #2 with a script to switch root to filesystem on an external 60Gb usb drive?

that would give me more space to compile stuff in and no problems with wear but im not sure on the data throughput of 480M/s usb2 against that of the micro sd HC card.
even if its slower i can still leave it to compile stuff overnight.

all the best and thanks for all the hard work

Shouldn't be any problem. The throughput over the USB bus would actually be roughly equivalent to my UHS-I card, I think - I see about 30MB/s on reads, about 10-20MB/s on writes (sequential, of course).

You might need to make sure there's support in the kernel for external usb HDs ,but I think there is - just haven't tested it.

i just plugged it in and it automounts ok...my dodgy edit worked but not unmounting very well. should add sync i think i set it up async  ::)

it currently has a tweaked version of puppy lucid live on it for i86 but thats in a squash filesystem so dumping a rootfs for a13 wont clash at all.

Quote from: jwischka on February 11, 2013, 01:22:00 AM

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Jwischka thank you for the kernel.... but sorry...didn`t work

Still no   /dev/ttyUSBx
What did i wrong???

Best regards
Michael

Not sure. Can you send me a (product) link to what you're trying to use?

Quote from: mike1968 on February 11, 2013, 09:05:58 PM

Quote from: jwischka on February 11, 2013, 01:22:00 AM

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Jwischka thank you for the kernel.... but sorry...didn`t work

Still no   /dev/ttyUSBx
What did i wrong???

Best regards
Michael

Quote from: jwischka on February 11, 2013, 09:14:58 PM
Not sure. Can you send me a (product) link to what you're trying to use?

Quote from: mike1968 on February 11, 2013, 09:05:58 PM

Quote from: jwischka on February 11, 2013, 01:22:00 AM

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Jwischka thank you for the kernel.... but sorry...didn`t work

Still no   /dev/ttyUSBx
What did i wrong???

Best regards
Michael

It is a usb->seial adapterlike this:
http://www.exp-tech.de/Zubehoer/USB-SERIAL-CABLE-F.html
It works perfect with my raspberrypi
Best regards

What are you trying to connect to?

Quote from: mike1968 on February 11, 2013, 10:31:20 PM

Quote from: jwischka on February 11, 2013, 09:14:58 PM
Not sure. Can you send me a (product) link to what you're trying to use?

Quote from: mike1968 on February 11, 2013, 09:05:58 PM

Quote from: jwischka on February 11, 2013, 01:22:00 AM

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Jwischka thank you for the kernel.... but sorry...didn`t work

Still no   /dev/ttyUSBx
What did i wrong???

Best regards
Michael

It is a usb->seial adapterlike this:
http://www.exp-tech.de/Zubehoer/USB-SERIAL-CABLE-F.html
It works perfect with my raspberrypi
Best regards

Quote from: jwischka on February 11, 2013, 11:14:14 PM
What are you trying to connect to?

Quote from: mike1968 on February 11, 2013, 10:31:20 PM

Quote from: jwischka on February 11, 2013, 09:14:58 PM
Not sure. Can you send me a (product) link to what you're trying to use?

Quote from: mike1968 on February 11, 2013, 09:05:58 PM

Quote from: jwischka on February 11, 2013, 01:22:00 AM

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Jwischka thank you for the kernel.... but sorry...didn`t work

Still no   /dev/ttyUSBx
What did i wrong???

Best regards
Michael

It is a usb->seial adapterlike this:
http://www.exp-tech.de/Zubehoer/USB-SERIAL-CABLE-F.html
It works perfect with my raspberrypi
Best regards

Well, i have some modules with serial interfaces from my arduino-projekts like  gsm,gps,bt,rf,xbee

Ok... now that I know more of what you're trying to do, I'll see what I can figure out.

My wife is having surgery tomorrow, so I'm not sure how quickly I'll get it done, but I'll try to keep you posted as to progress. Hopefully it's something simple.

Quote from: mike1968 on February 11, 2013, 11:32:50 PM

Quote from: jwischka on February 11, 2013, 11:14:14 PM
What are you trying to connect to?

Quote from: mike1968 on February 11, 2013, 10:31:20 PM

Quote from: jwischka on February 11, 2013, 09:14:58 PM
Not sure. Can you send me a (product) link to what you're trying to use?

Quote from: mike1968 on February 11, 2013, 09:05:58 PM

Quote from: jwischka on February 11, 2013, 01:22:00 AM

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Jwischka thank you for the kernel.... but sorry...didn`t work

Still no   /dev/ttyUSBx
What did i wrong???

Best regards
Michael

It is a usb->seial adapterlike this:
http://www.exp-tech.de/Zubehoer/USB-SERIAL-CABLE-F.html
It works perfect with my raspberrypi
Best regards

Well, i have some modules with serial interfaces from my arduino-projekts like  gsm,gps,bt,rf,xbee

Does'nt it has a USBSerial Driver? Whenever I tried to connected the MOD-USB-RS232 module to the board, the device is not creating in the /dev. Why is it so?

Hello bimodh,

The problem is probably in the MOD-USB-RS232. There is a high chance that the MOD-USB-RS232 you received was shipped without the final firmware it needs. The situation is further explained here: https://www.olimex.com/forum/index.php?topic=810.msg3644#msg3644 (https://www.olimex.com/forum/index.php?topic=810.msg3644#msg3644)

Sorry for the incontinence caused.

Best regards,
Lub/OLIMEX

Hello jwischka

Thank you for the work. I am a beginner in embedded linux. What you have done is a big help for me.

I have tried R15 and R12 video images on a A13 board. Both have some problems with a boot. There is only a flashing underscore symbol "_" on a VGA screen. Maybe the reason is that i don't have serial connection? Unfortunately i didn't ordered serial cable for this board. Do you have other suggestions what is the problem?

Fortunately linaro image from wiki works.

Modified 20.02.13
I made some experiments and found that this is not a problem of the image, but the problem of my zip utility.
It's very strange but if I extract image on windows it works but when I extract by unzip or ark it doesn't.
I checked that md5sum of downloaded files in linux and windows are similar.
I use openSuse 12.2. Unzip version 6.0

I'll look into it, but video output should be enabled. I'll have to connect it up to a monitor, which I can probably do this afternoon.

It's possible it might be a script.bin issue?

Quote from: p.kartavyy on February 13, 2013, 10:45:00 AM
Hello jwischka

Thank you for the work. I am a beginner in embedded linux. What you have done is a big help for me.

I have tried R15 and R12 video images on a A13 board. Both have some problems with a boot. There is only a flashing underscore symbol "_" on a VGA screen. Maybe the reason is that i don't have serial connection? Unfortunately i didn't ordered serial cable for this board. Do you have other suggestions what is the problem?

Fortunately linaro image from wiki works.

Hope your wife is ok.

Did you have any idea to solve the problem?

Quote from: jwischka on February 11, 2013, 11:34:31 PM
Ok... now that I know more of what you're trying to do, I'll see what I can figure out.

My wife is having surgery tomorrow, so I'm not sure how quickly I'll get it done, but I'll try to keep you posted as to progress. Hopefully it's something simple.

Quote from: mike1968 on February 11, 2013, 11:32:50 PM

Quote from: jwischka on February 11, 2013, 11:14:14 PM
What are you trying to connect to?

Quote from: mike1968 on February 11, 2013, 10:31:20 PM

Quote from: jwischka on February 11, 2013, 09:14:58 PM
Not sure. Can you send me a (product) link to what you're trying to use?

Quote from: mike1968 on February 11, 2013, 09:05:58 PM

Quote from: jwischka on February 11, 2013, 01:22:00 AM

Quote from: jwischka on February 11, 2013, 12:13:24 AM

Quote from: mike1968 on February 10, 2013, 11:38:24 PM
R15 same probleme
I use a pl2303 usb to serial adapter.
With lsusb it is detected as bus 002 device 005, but there is no /dev/ttyUSBx

Going to require a new kernel. Building now and will upload in a bit. I'm trying to get by with generic drivers, rather than having to build about 400 modules, but if it doesn't work, we can go the other way too.

R16 kernels are up. Try them and see if they work for your USB->serial issues.

R16 Headless (http://files.wisch.org/3.4-headless-R16.tar.gz)
R16 Video (http://files.wisch.org/3.4-video-R16.tar.gz)

Jwischka thank you for the kernel.... but sorry...didn`t work

Still no   /dev/ttyUSBx
What did i wrong???

Best regards
Michael

It is a usb->seial adapterlike this:
http://www.exp-tech.de/Zubehoer/USB-SERIAL-CABLE-F.html
It works perfect with my raspberrypi
Best regards

Well, i have some modules with serial interfaces from my arduino-projekts like  gsm,gps,bt,rf,xbee

Quote from: mike1968 on February 13, 2013, 07:23:57 PM
Hope your wife is ok.

Did you have any idea to solve the problem?

The procedure (diagnostic) went well, but results were not very good - looks like open heart surgery this summer.

I'll try to look at it this afternoon, but I haven't had a chance yet.

Quote from: jwischka on February 13, 2013, 08:36:54 PM

Quote from: mike1968 on February 13, 2013, 07:23:57 PM
Hope your wife is ok.

Did you have any idea to solve the problem?

The procedure (diagnostic) went well, but results were not very good - looks like open heart surgery this summer.

I'll try to look at it this afternoon, but I haven't had a chance yet.

Wish your wife all the best...head up

I had a heard attack last summer..so i can believe you feel

Mike -

New kernels should take care of your problem:

R17 headless (http://files.wisch.org/3.4-headless-R17.tar.gz)
R17 video (http://files.wisch.org/3.4-video-R17.tar.gz)

Get current version (R17) here. (http://files.wisch.org/olinuxino_xfce-r17.zip)

Quote from: mike1968 on February 13, 2013, 09:06:51 PM

Quote from: jwischka on February 13, 2013, 08:36:54 PM

Quote from: mike1968 on February 13, 2013, 07:23:57 PM
Hope your wife is ok.

Did you have any idea to solve the problem?

The procedure (diagnostic) went well, but results were not very good - looks like open heart surgery this summer.

I'll try to look at it this afternoon, but I haven't had a chance yet.

Wish your wife all the best...head up

I had a heard attack last summer..so i can believe you feel

Quote from: jwischka on February 14, 2013, 06:21:05 PM
Mike -

New kernels should take care of your problem:

R17 headless (http://files.wisch.org/3.4-headless-R17.tar.gz)
R17 video (http://files.wisch.org/3.4-video-R17.tar.gz)

Get current version (R17) here. (http://files.wisch.org/olinuxino_xfce-r17.zip)

Quote from: mike1968 on February 13, 2013, 09:06:51 PM

Quote from: jwischka on February 13, 2013, 08:36:54 PM

Quote from: mike1968 on February 13, 2013, 07:23:57 PM
Hope your wife is ok.

Did you have any idea to solve the problem?

The procedure (diagnostic) went well, but results were not very good - looks like open heart surgery this summer.

I'll try to look at it this afternoon, but I haven't had a chance yet.

Wish your wife all the best...head up

I had a heard attack last summer..so i can believe you feel

Hello jeff,
My problem solved.....usb serial is running.
Thank you very much
Michael

Hello;

By way of report, I haven't had any luck yet with R17, R15, R12 on my board.

After numerous iterations, the most promising outcome has been that my OLinuXino enters a sort of boot-loop, where the screen fades in grey then goes black again every 15s, sometimes with a fine line about 1/3 of the way down the LCD.

The board's a rev.c WIFI with the 7" LCD and touchscreen, as well as the Olimex-supplied AC adapter. It's been flashed with the LCD image to internal NVRAM.

I've tried a limited variety of sizes and brands of uSD card (Adata 16GB, Kingston 4GB, Phillips 4GB). DD actually returns an out-of-room error when I try flashing the images to either 4GB card. I've tried both 'sudo dd if=yourfile.img of=/dev/sdX' as well as 'sudo dd bs=4M oflag=sync if=yourfile.img of=/dev/sdX' (where X is the letter representing the flash card in question). I 'Eject This Card Safely' from Nautilus each time as well postflash.

The board boots Android ok. I also have tried an Olimex-supplied Debian uSD card. That just boots into Android with a message about a damaged SD card. So I'm out of ideas.

Any tips? Sorry to bug jwischka and other contributors - without you, we neophytes wouldn't have an in, so thanks.

It sounds like there's some sort of kernel panic happening, perhaps. Is there any chance you have a usb-serial adapter and could get a log of the messages? It's going to be really hard to debug what's going on without that (unless I actually figure out a way to redirect it to the VGA, which is on the eventual to-do list).

Quote from: tehowe on February 17, 2013, 08:52:48 AM
Hello;

By way of report, I haven't had any luck yet with R17, R15, R12 on my board.

After numerous iterations, the most promising outcome has been that my OLinuXino enters a sort of boot-loop, where the screen fades in grey then goes black again every 15s, sometimes with a fine line about 1/3 of the way down the LCD.

The board's a rev.c WIFI with the 7" LCD and touchscreen, as well as the Olimex-supplied AC adapter. It's been flashed with the LCD image to internal NVRAM.

I've tried a limited variety of sizes and brands of uSD card (Adata 16GB, Kingston 4GB, Phillips 4GB). DD actually returns an out-of-room error when I try flashing the images to either 4GB card. I've tried both 'sudo dd if=yourfile.img of=/dev/sdX' as well as 'sudo dd bs=4M oflag=sync if=yourfile.img of=/dev/sdX' (where X is the letter representing the flash card in question). I 'Eject This Card Safely' from Nautilus each time as well postflash.

The board boots Android ok. I also have tried an Olimex-supplied Debian uSD card. That just boots into Android with a message about a damaged SD card. So I'm out of ideas.

Any tips? Sorry to bug jwischka and other contributors - without you, we neophytes wouldn't have an in, so thanks.

Probably best to use an SD card with enough room.

We keep posting the same information: many "4GB" cards are LESS than 4GB so the image does not fit.

Easy fix: use a bigger one.

If dd fails, not much point booting the card.
If dd is OK, try booting.  If that fails, look at the debug messages on uart1
Easy to record them using e.g.
screen -L /dev/ttyUSB0 115200

John

Hello all,

I also have the problem with the 4GB uSD.
Hope this helps shed some light into the problem.
After flashing the image, dd reports: no space left on device (3.9 GB copied).
About the reboot cycle. It's due to a kernel panic, indeed. There is a problem related to mounting the file system.
This is what is reported thru the serial-usb connection:


Starting kernel ...

[    0.000000] Booting Linux on physical CPU 0
[    0.000000] Initializing cgroup subsys cpuset
[    0.000000] Linux version 3.4.29-jwischka-3.4-video-20130209-R15+ (root@jeff-ThinkPad-T420) (gcc version 4.7.2 (Ubuntu/Linaro 4.7.2-1ubuntu1) ) #20 PREEMPT Sat Feb 9 17:57:12 EST 2013
[    0.000000] CPU: ARMv7 Processor [413fc082] revision 2 (ARMv7), cr=10c5387d
[    0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
[    0.000000] Machine: sun5i
[    0.000000] Memory Reserved:
[    0.000000]    SYS  : 0x43000000 - 0x4300ffff  (  64 kB)
[    0.000000]    LCD  : 0x5a000000 - 0x5bffffff  (  32 MB)
[    0.000000]    VE   : 0x44000000 - 0x48ffffff  (  80 MB)
[    0.000000] Memory policy: ECC disabled, Data cache writeback
[    0.000000] chip-id: Unknown (AW1625)
[    0.000000] On node 0 totalpages: 131072
[    0.000000] free_area_init_node: node 0, pgdat c075ae48, node_mem_map c0803000
[    0.000000]   Normal zone: 1024 pages used for memmap
[    0.000000]   Normal zone: 0 pages reserved
[    0.000000]   Normal zone: 130048 pages, LIFO batch:31
[    0.000000] pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768
[    0.000000] pcpu-alloc:

• 0
  [    0.000000] Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 130048
  [    0.000000] Kernel command line: console=ttyS0,115200 root=/dev/mmcblk0p2 rootwait loglevel=8 panic=10
  [    0.000000] PID hash table entries: 2048 (order: 1, 8192 bytes)
  [    0.000000] Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
  [    0.000000] Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
  [    0.000000] Memory: 512MB = 512MB total
  [    0.000000] Memory: 396788k/396788k available, 127500k reserved, 0K highmem
  [    0.000000] Virtual kernel memory layout:
  [    0.000000]     vector  : 0xffff0000 - 0xffff1000   (   4 kB)
  [    0.000000]     fixmap  : 0xfff00000 - 0xfffe0000   ( 896 kB)
  [    0.000000]     vmalloc : 0xe0800000 - 0xff000000   ( 488 MB)
  [    0.000000]     lowmem  : 0xc0000000 - 0xe0000000   ( 512 MB)
  [    0.000000]     pkmap   : 0xbfe00000 - 0xc0000000   (   2 MB)
  [    0.000000]     modules : 0xbf000000 - 0xbfe00000   (  14 MB)
  [    0.000000]       .text : 0xc0008000 - 0xc06cf000   (6940 kB)
  [    0.000000]       .init : 0xc06cf000 - 0xc06f6000   ( 156 kB)
  [    0.000000]       .data : 0xc06f6000 - 0xc075b900   ( 407 kB)
  [    0.000000]        .bss : 0xc075b924 - 0xc08028c0   ( 668 kB)
  [    0.000000] SLUB: Genslabs=11, HWalign=64, Order=0-3, MinObjects=0, CPUs=1, Nodes=1
  [    0.000000] NR_IRQS:96
  [    0.000000] timer0: Periodic Mode
  [    0.000000] sched_clock: 32 bits at 100 Hz, resolution 10000000ns, wraps every 4294967286ms
  [    0.000000] Console: colour dummy device 80x30
  [    0.000000] Calibrating delay loop... 1001.88 BogoMIPS (lpj=5009408)
  [    0.040000] pid_max: default: 32768 minimum: 301
  [    0.040000] Mount-cache hash table entries: 512
  [    0.040000] Initializing cgroup subsys cpuacct
  [    0.040000] Initializing cgroup subsys devices
  [    0.040000] Initializing cgroup subsys freezer
  [    0.040000] Initializing cgroup subsys blkio
  [    0.040000] CPU: Testing write buffer coherency: ok
  [    0.040000] hw perfevents: enabled with ARMv7 Cortex-A8 PMU driver, 5 counters available
  [    0.040000] Setting up static identity map for 0x404f4280 - 0x404f42d8
  [    0.040000] devtmpfs: initialized
  [    0.040000] dummy:
  [    0.040000] NET: Registered protocol family 16
  [    0.040000] hw-breakpoint: debug architecture 0x4 unsupported.
  [    0.040000] SOFTWINNER DMA Driver, (c) 2003-2004,2006 Simtec Electronics
  [    0.040000] Initialize DMAC OK
  [    0.040000] bio: create slab <bio-0> at 0
  [    0.050000] SCSI subsystem initialized
  [    0.050000] usbcore: registered new interface driver usbfs
  [    0.050000] usbcore: registered new interface driver hub
  [    0.050000] usbcore: registered new device driver usb
  [    0.050000] Linux video capture interface: v2.00
  [    0.050000] Advanced Linux Sound Architecture Driver Version 1.0.25.
  [    0.050000] Init eGon pin module V2.0
  [    0.050000] Switching to clocksource aw 64bits couter
  [    0.050000] FS-Cache: Loaded
  [    0.050000] CacheFiles: Loaded
  [    0.050000] [usb_manager]: CONFIG_USB_SW_SUN5I_USB0_HOST_ONLY
  [    0.050000] [sw_hcd0]: usb host driver initialize........
  [    0.050000] [sw_hcd0]: [sw_hcd_host0]: open_usb_clock
  [    0.050000] [hcd0]: open, 0x60(0xc141), 0xcc(0x143)
  [    0.050000] [sw_hcd0]: host_init_state = 0
  [    0.050000] [sw_hcd0]: platform is usb host
  [    0.050000] [sw_hcd0]: sw_hcd_host0: ConfigData=0xde (UTMI-8, dyn FIFOs, bulk combine, bulk split, HB-ISO Rx (X), HB-ISO Tx (X), SoftConn)
  [    0.050000] [sw_hcd0]: sw_hcd_init_controller: sw_hcd_host0: USB Host mode controller at f1c13000 using PIO, IRQ 38
  [    0.050000] sw_hcd_host0 sw_hcd_host0: sw_hcd host driver
  [    0.050000] sw_hcd_host0 sw_hcd_host0: new USB bus registered, assigned bus number 1
  [    0.050000] hub 1-0:1.0: USB hub found
  [    0.050000] hub 1-0:1.0: 1 port detected
  [    0.050000] [sw_hcd_host0]: Set USB Power ON
  [    0.050000] axp driver uning configuration failed(562)
  [    0.050000] axp driver uning configuration failed(574)
  [    0.050000] NET: Registered protocol family 2
  [    0.050000] IP route cache hash table entries: 4096 (order: 2, 16384 bytes)
  [    0.050000] TCP established hash table entries: 16384 (order: 5, 131072 bytes)
  [    0.050000] TCP bind hash table entries: 16384 (order: 4, 65536 bytes)
  [    0.050000] TCP: Hash tables configured (established 16384 bind 16384)
  [    0.050000] TCP: reno registered
  [    0.050000] UDP hash table entries: 256 (order: 0, 4096 bytes)
  [    0.050000] UDP-Lite hash table entries: 256 (order: 0, 4096 bytes)
  [    0.050000] NET: Registered protocol family 1
  [    0.060000] cfg80211: Calling CRDA to update world regulatory domain
  [    0.060000] RPC: Registered named UNIX socket transport module.
  [    0.060000] RPC: Registered udp transport module.
  [    0.060000] RPC: Registered tcp transport module.
  [    0.060000] RPC: Registered tcp NFSv4.1 backchannel transport module.
  [    0.060000] [pm]aw_pm_init!
  [    0.060000] audit: initializing netlink socket (disabled)
  [    0.060000] type=2000 audit(0.050:1): initialized
  [    0.060000] VFS: Disk quotas dquot_6.5.2
  [    0.060000] Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
  [    0.060000] NTFS driver 2.1.30 [Flags: R/W].
  [    0.060000] fuse init (API version 7.18)
  [    0.070000] msgmni has been set to 774
  [    0.070000] alg: No test for stdrng (krng)
  [    0.070000] Block layer SCSI generic (bsg) driver version 0.4 loaded (major 253)
  [    0.070000] io scheduler noop registered
  [    0.070000] io scheduler deadline registered
  [    0.070000] io scheduler cfq registered (default)
  [    0.070000] sunxi disp driver loaded (/dev/disp api 1.0)
  [    0.070000] Serial: 8250/16550 driver, 4 ports, IRQ sharing disabled
  [    0.070000] [uart]: failed to get uart2's used information
  [    0.070000] [uart]: failed to get uart3's used information
  [    0.070000] [uart]: used uart info.: 0x02
  [    0.070000] [uart]: serial probe 1 irq 2 mapbase 0x01c28400
  [    0.080000] sunxi-uart.1: ttyS0 at MMIO 0x1c28400 (irq = 2) is a U6_16550A
  [    0.090000] console [ttyS0] enabled
  [    0.100000] brd: module loaded
  [    0.100000] loop: module loaded
  [    0.100000] 'Low Performance USB Block' driver is deprecated. Please switch to usb-storage
  [    0.110000] usbcore: registered new interface driver ub
  [    0.120000] [NAND]nand driver, init.
  [    0.120000] cmu_clk is 408
  [    0.120000] nand clk init end
  [    0.130000] offset 0xc:  0xe141
  [    0.130000] offset 0x14:  0x8200000a
  [    0.130000] [NAND] nand gpio_request
  [    0.140000] [NAND] nand driver version: 0x2 0x9
  [    0.140000] nand interrupte register ok
  [    0.150000] ret of NFC_ChangMode is 0
  [    0.150000] dma_hdle  is 0
  [    0.150000] dma_hdle  is 10000008
  [    0.160000] [SCAN_ERR] search nand physical architecture parameter failed!
  [    0.160000] [NAND]init_blklayer fail
  [    0.170000] sun4i gpio driver init
  [    0.170000] [spi]: sw spi init !!
  [    0.170000] [spi]: sw spi init fetch spi0 uning configuration failed
  [    0.180000] [spi]: Get spi devices number failed
  [    0.190000] [spi]: register spi devices board info failed
  [    0.190000] spi: cannot find any using configuration for                     all 4 spi controllers, return directly!
  [    0.200000] usbcore: registered new interface driver rtl8192cu
  [    0.210000] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
  [    0.210000] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver
  [    0.220000] [sw-ehci1]: open clock
  [    0.240000] [sw-ehci1]: Set USB Power ON
  [    0.250000] sw-ehci sw-ehci.1: SW USB2.0 'Enhanced' Host Controller (EHCI) Driver
  [    0.250000] sw-ehci sw-ehci.1: new USB bus registered, assigned bus number 2
  [    0.260000] sw-ehci sw-ehci.1: irq 39, io mem 0xf1c14000
  [    0.290000] sw-ehci sw-ehci.1: USB 0.0 started, EHCI 1.00
  [    0.290000] ehci_irq: port change detect
  [    0.290000] hub 2-0:1.0: USB hub found
  [    0.300000] hub 2-0:1.0: 1 port detected
  [    0.300000] [sw-ohci1]: open clock
  [    0.330000] sw-ohci sw-ohci.1: SW USB2.0 'Open' Host Controller (OHCI) Driver
  [    0.330000] sw-ohci sw-ohci.1: new USB bus registered, assigned bus number 3
  [    0.340000] sw-ohci sw-ohci.1: irq 64, io mem 0xf1c14400
  [    0.410000] hub 3-0:1.0: USB hub found
  [    0.410000] hub 3-0:1.0: 1 port detected
  [    0.420000] Initializing USB Mass Storage driver...
  [    0.420000] usbcore: registered new interface driver usb-storage
  [    0.430000] USB Mass Storage support registered.
  [    0.430000] usbcore: registered new interface driver ums-alauda
  [    0.440000] usbcore: registered new interface driver ums-cypress
  [    0.440000] usbcore: registered new interface driver ums-datafab
  [    0.450000] usbcore: registered new interface driver ums_eneub6250
  [    0.460000] usbcore: registered new interface driver ums-freecom
  [    0.460000] usbcore: registered new interface driver ums-isd200
  [    0.470000] usbcore: registered new interface driver ums-jumpshot
  [    0.480000] usbcore: registered new interface driver ums-karma
  [    0.480000] usbcore: registered new interface driver ums-onetouch
  [    0.490000] usbcore: registered new interface driver ums-realtek
  [    0.490000] usbcore: registered new interface driver ums-sddr09
  [    0.500000] usbcore: registered new interface driver ums-sddr55
  [    0.510000] usbcore: registered new interface driver ums-usbat
  [    0.510000] mousedev: PS/2 mouse device common for all mice
  [    0.520000] ========RTC Inital ===================
  [    0.520000] rtc_fetch_sysconfig_para: after: rtc_twi_addr is 0x51, dirty_addr_buf: 0x51. dirty_addr_buf[1]: 0xfffe
  [    0.530000] rtc_fetch_sysconfig_para: rtc_twi_id is 1.
  [    0.540000] rtc_fetch_sysconfig_para:ok
  [    0.540000] pcf8563_init: after fetch_sysconfig_para:  normal_i2c: 0x51. normal_i2c[1]: 0xfffe
  [    0.550000] i2c /dev entries driver
  [    0.550000] !!! base_Addr = 0xf1c2ac00
  [    0.560000] config i2c gpio with gpio_config api
  [    0.560000] twi0, apb clock = 24000000
  [    0.570000] axp_mfd 0-0034: AXP (CHIP ID: 0x21) detected
  [    0.570000] [AXP]axp driver uning configuration failed(323)
  [    0.580000] [AXP]power_start = 0
  [    0.580000] rtc_detect,line:157,twi_id:1,adapter->nr:0
  [    0.590000] rtc_detect,line:166
  [    0.590000] I2C: i2c-0: AW16XX I2C adapter
  [    0.600000] **********start************
  [    0.600000] 0x40
  [    0.600000] 0xf8
  [    0.600000] 0x59
  [    0.610000] 0x0
  [    0.610000] 0x0
  [    0.610000] **********end************
  [    0.610000] !!! base_Addr = 0xf1c2b000
  [    0.620000] usb 2-1: new high-speed USB device number 2 using sw-ehci
  [    0.620000] twi1, apb clock = 24000000
  [    0.630000] rtc_detect,line:157,twi_id:1,adapter->nr:1
  [    0.630000] rtc_detect: Detected chip pcf8563 at adapter 1, address 0x51
  [    0.640000] rtc_detect,line:162
  [    0.640000] pcf8563_probe,line:657
  [    0.650000] pcf8563 1-0051: chip found, driver version 0.4.3
  [    0.650000] pcf8563 1-0051: low voltage detected, date/time is not reliable.
  [    0.660000] pcf8563_get_datetime,raw data is st1=08, st2=c0, sec=c8, min=80, hr=c0, mday=c5, wday=c8, mon=40, year=00
  [    0.670000] pcf8563_get_datetime: tm is secs=48, mins=0, hours=0,mday=5, mon=-1, year=110, wday=0
  [    0.680000] pcf8563 1-0051: retrieved date/time is not valid.
  [    0.690000] pcf8563 1-0051: rtc core: registered pcf8563 as rtc0
  [    0.690000] I2C: i2c-1: AW16XX I2C adapter
  [    0.700000] **********start************
  [    0.700000] 0x40
  [    0.700000] 0xf8
  [    0.700000] 0x59
  [    0.710000] 0x0
  [    0.710000] 0x0
  [    0.710000] **********end************
  [    0.710000] !!! base_Addr = 0xf1c2b400
  [    0.720000] twi2, apb clock = 24000000
  [    0.720000] rtc_detect,line:157,twi_id:1,adapter->nr:2
  [    0.730000] rtc_detect,line:166
  [    0.730000] I2C: i2c-2: AW16XX I2C adapter
  [    0.730000] **********start************
  [    0.740000] 0x40
  [    0.740000] 0xf8
  [    0.740000] 0x59
  [    0.740000] 0x0
  [    0.740000] 0x0
  [    0.750000] **********end************
  [    0.750000] [cedar dev]: install start!!!
  [    0.750000] [cedar dev]: install end!!!
  [    0.760000] [ace_drv] start!!!
  [    0.760000] [ace_drv] init end!!!
  [    0.760000] axp20_ldo1: 1300 mV
  [    0.770000] axp20_ldo2: 1800 <--> 3300 mV at 3000 mV
  [    0.780000] hub 2-1:1.0: USB hub found
  [    0.780000] hub 2-1:1.0: 4 ports detected
  [    0.790000] axp20_ldo3: 700 <--> 3500 mV at 3300 mV
  [    0.790000] axp20_ldo4: 1250 <--> 3300 mV at 3300 mV
  [    0.800000] axp20_buck2: 700 <--> 2275 mV at 1400 mV
  [    0.810000] axp20_buck3: 700 <--> 3500 mV at 1200 mV
  [    0.810000] axp20_ldoio0: 1800 <--> 3300 mV at 2800 mV
  [    0.820000] input: axp20-supplyer as /devices/platform/sun5i-i2c.0/i2c-0/0-0034/axp20-supplyer.28/input/input0
  [    0.860000] md: linear personality registered for level -1
  [    0.870000] md: multipath personality registered for level -4
  [    0.870000] md: faulty personality registered for level -5
  [    0.880000] device-mapper: ioctl: 4.22.0-ioctl (2011-10-19) initialised: dm-devel@redhat.com
  [    0.890000] device-mapper: multipath: version 1.3.0 loaded
  [    0.890000] device-mapper: multipath round-robin: version 1.0.0 loaded
  [    0.900000] device-mapper: multipath queue-length: version 0.1.0 loaded
  [    0.910000] device-mapper: multipath service-time: version 0.2.0 loaded
  [    0.910000] [mmc_pm]: no sdio card used in configuration
  [    0.920000] [mmc]: sunximmc_init
  [    0.920000] [mmc]: sunxi mmc controller using config : 0x1
  [    0.930000] [mmc]: sunxi-mmc.0: pdev->name: sunxi-mmc, pdev->id: 0
  [    0.930000] [mmc]: mmc 0 power off !!
  [    0.940000] [mmc]: mmc0 Probe: base:0xf1c0f000 irq:32 dma:0 pdes:0xdf9dc000, ret 0.
  [    0.950000] usbcore: registered new interface driver usbhid
  [    0.950000] usbhid: USB HID core driver
  [    0.960000] enter sun5i Audio codec!!!
  [    0.960000] sun5i audio support initialized
  [    0.970000] sun5i Audio codec successfully loaded..
  [    0.970000] IPv4 over IPv4 tunneling driver
  [    0.980000] TCP: cubic registered
  [    0.980000] Initializing XFRM netlink socket
  [    0.980000] NET: Registered protocol family 17
  [    0.990000] NET: Registered protocol family 15
  [    0.990000] 8021q: 802.1Q VLAN Support v1.8
  [    1.000000] VFP support v0.3: implementor 41 architecture 3 part 30 variant c rev 3
  [    1.000000] Registering SWP/SWPB emulation handler
  [    1.010000] registered taskstats version 1
  [    1.940000] [mmc]: mmc 0 detect change, present 1
  [    1.940000] Console: switching to colour frame buffer device 100x37
  [    1.960000] axp20_buck3: incomplete constraints, leaving on
  [    1.960000] axp20_buck2: incomplete constraints, leaving on
  [    1.970000] axp20_ldo4: incomplete constraints, leaving on
  [    1.970000] axp20_ldo3: incomplete constraints, leaving on
  [    1.980000] axp20_ldo2: incomplete constraints, leaving on
  [    1.990000] axp20_ldo1: incomplete constraints, leaving on
  [    1.990000] pcf8563 1-0051: low voltage detected, date/time is not reliable.
  [    2.000000] pcf8563_get_datetime,raw data is st1=08, st2=c0, sec=c9, min=80, hr=c0, mday=c5, wday=c8, mon=40, year=00
  [    2.010000] pcf8563_get_datetime: tm is secs=49, mins=0, hours=0,mday=5, mon=-1, year=110, wday=0
  [    2.020000] pcf8563 1-0051: retrieved date/time is not valid.
  [    2.030000] pcf8563 1-0051: hctosys: invalid date/time
  [    2.030000] ALSA device list:
  [    2.030000]   #0: sun5i-CODEC  Audio Codec
  [    2.040000] md: Waiting for all devices to be available before autodetect
  [    2.040000] md: If you don't use raid, use raid=noautodetect
  [    2.050000] md: Autodetecting RAID arrays.
  [    2.060000] md: Scanned 0 and added 0 devices.
  [    2.060000] md: autorun ...
  [    2.060000] md: ... autorun DONE.
  [    2.070000] Waiting for root device /dev/mmcblk0p2...
  [    2.240000] [mmc]: mmc 0 power on !!
  [    2.280000] [mmc]: sdxc_request_done(L1113): smc 0 err, cmd 52,  RTO !!
  [    2.280000] [mmc]: sdxc_request_done(L1113): smc 0 err, cmd 52,  RTO !!
  [    2.290000] [mmc]: sdxc_request_done(L1113): smc 0 err, cmd 5,  RTO !!
  [    2.300000] [mmc]: sdxc_request_done(L1113): smc 0 err, cmd 5,  RTO !!
  [    2.310000] [mmc]: sdxc_request_done(L1113): smc 0 err, cmd 5,  RTO !!
  [    2.320000] [mmc]: sdxc_request_done(L1113): smc 0 err, cmd 5,  RTO !!
  [    2.360000] mmc0: new high speed SDHC card at address 1234
  [    2.360000] mmcblk0: mmc0:1234 SA04G 3.63 GiB
  [    2.370000]  mmcblk0: p1 p2
  [    2.370000] mmcblk0: p2 size 7709696 extends beyond EOD, truncated
  [    2.470000] List of all partitions:
  [    2.470000] b300         3813376 mmcblk0  driver: mmcblk
  [    2.470000]   b301           16384 mmcblk0p1 00000000-0000-0000-0000-000000000000
  [    2.480000]   b302         3795968 mmcblk0p2 00000000-0000-0000-0000-000000000000
  [    2.490000] No filesystem could mount root, tried:  ext3 ext2 ext4 vfat msdos ntfs fuseblk
  [    2.500000] Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(179,2)
  [    2.510000] Rebooting in 10 seconds..[/b]
  

I agree with John in that the problem may be the insufficient amount in the uSD (3.9 GB), as reported by the dd utility.
I will try to get a bigger one and flash it. Then I will report here if it worked out correctly.
BTW, thanks to jwischka for building these images  8).

Best,

agcr

The boot log shows that the problem is with the mmc being too small.

As I posted already, if the dd fails then there is no point booting.

Cards are so cheap now, just get one big enough!

(As an alternative: mount via loopback, mount a smaller card and copy files.  This is hard for an inexperienced person to get right and very painful due to the slow speed.  I'd just get a bigger card, which is what I did.)

John

This (http://www.amazon.com/SanDisk-microSDHC-Memory-Adapter-SDSDQY-004G-AFFP-A/dp/B007L4RTOS/ref=sr_1_13?ie=UTF8&qid=1361213072&sr=8-13&keywords=sandisk+4gb+micro+sd+ultra), incidentally, is the card I am using. Though I'd recommend getting an 8gb (or 32, as I use for my personal purposes).

I bought a 8GB one, but I'm having trouble flashing it with dd.
Can you please tell me the steps you followed for flashing the image in your uSD?
Thanks in advance.

QuoteI bought a 8GB one, but I'm having trouble flashing it with dd.
Can you please tell me the steps you followed for flashing the image in your uSD?
Thanks in advance.

Hey agcr ,
try reading this thread https://www.olimex.com/forum/index.php?topic=888.0 (https://www.olimex.com/forum/index.php?topic=888.0)
Regards :D

Hi ellecieffe,

Thanks for the help  8)! I was doing what you say in the thread, but still I'm getting a corrupted image.
I use GParted to create a 4GB partition (/dev/sdb1) and then use dd to flash it there.
I've tried 2 differente card readers, but still get the same result.
I am using an 8GB uSD card, and there's left 3.9GB as free space.
I think I'm missing something though, any thoughts?

Best,
agcr

hi agcr, as said by jwischka,

Quoteyou don't need to partition the card - DD will write all the partitioning information for you.

In realty,I  always use gparted to "reset" the card (to be sure, it can be said  ::))
However,you also can try with win32 disk imager in windows OS. I did so with this the very first time  ;D
Finally, remember to  update your drivers

My 2 cents
Luigi

Do NOT run gparted!

Just dd the IMG file straight to /dev/sdb - NOT to sdb1

John

Thanks guys, it did work  :)

Hello jwischka,

  is framebuffer support available in R15/R17 ? I see no log info in dmesg about it

  BTW: I'm trying to run qt-embedded on my olinuxino

       thanks

              Giampaolo

Quote from: iw2lsi on February 20, 2013, 08:40:39 PM
Hello jwischka,

  is framebuffer support available in R15/R17 ? I see no log info in dmesg about it

  BTW: I'm trying to run qt-embedded on my olinuxino

       thanks

              Giampaolo

Should be, but I'll have to check and make sure. If not, I can compile it in.

Hello jwischka,
 
  first of all, thanks a lot for your time and for the great job you are doing...
  I'm totally new on A13 so your images are very useful for me

  BTW: are the .config for the various releases saved somewhere ?
 
     thanks again

          Giampaolo

Quote from: iw2lsi on February 20, 2013, 10:07:34 PM
Hello jwischka,
 
  first of all, thanks a lot for your time and for the great job you are doing...
  I'm totally new on A13 so your images are very useful for me

  BTW: are the .config for the various releases saved somewhere ?
 
     thanks again

          Giampaolo

Only since R10 or so.

I tryed the R17 twice but I get this error:


>unzip olinuxino_xfce-r17.zip
Archive:  olinuxino_xfce-r17.zip
  inflating: olinuxino_xfce-r17.img   bad CRC a9ad228b  (should be 7c897ce4)


Which Linux & kernel version are you using?  Some have a broken (something, probably a .so) that causes that.

Try a 2.x kernel-based system if you can.  7zip seems OK, too

John

Thanks for the reply (I've been up north in the land of no internet)... I'll check at my local hobbyist electronics shop to see if I can get one of those, and figure out how to use it.

Quote from: jwischka on February 17, 2013, 02:49:48 PM
It sounds like there's some sort of kernel panic happening, perhaps. Is there any chance you have a usb-serial adapter and could get a log of the messages? It's going to be really hard to debug what's going on without that (unless I actually figure out a way to redirect it to the VGA, which is on the eventual to-do list).

Quote from: tehowe on February 17, 2013, 08:52:48 AM
Hello;

By way of report, I haven't had any luck yet with R17, R15, R12 on my board.

After numerous iterations, the most promising outcome has been that my OLinuXino enters a sort of boot-loop, where the screen fades in grey then goes black again every 15s, sometimes with a fine line about 1/3 of the way down the LCD.

The board's a rev.c WIFI with the 7" LCD and touchscreen, as well as the Olimex-supplied AC adapter. It's been flashed with the LCD image to internal NVRAM.

I've tried a limited variety of sizes and brands of uSD card (Adata 16GB, Kingston 4GB, Phillips 4GB). DD actually returns an out-of-room error when I try flashing the images to either 4GB card. I've tried both 'sudo dd if=yourfile.img of=/dev/sdX' as well as 'sudo dd bs=4M oflag=sync if=yourfile.img of=/dev/sdX' (where X is the letter representing the flash card in question). I 'Eject This Card Safely' from Nautilus each time as well postflash.

The board boots Android ok. I also have tried an Olimex-supplied Debian uSD card. That just boots into Android with a message about a damaged SD card. So I'm out of ideas.

Any tips? Sorry to bug jwischka and other contributors - without you, we neophytes wouldn't have an in, so thanks.

Quote from: JohnS on February 21, 2013, 12:46:35 PM
Which Linux & kernel version are you using?  Some have a broken (something, probably a .so) that causes that.

Try a 2.x kernel-based system if you can.  7zip seems OK, too

John

I use Ubuntu 12.04.2.
I'll try 7z, thanks.

and kernel ver?

I`m new to all this. So here`s my problem, I managed to get the R17 to run, but I can`t run the video... I`m using olinuxino wifi and when i run /home/olinuxino/go-video it says: cannot run kernels/video/Uimage.
Here is the log from the serial :


* Stopping Network connection manager wicd                                             [ OK ]
                                                                         [ OK ] * Asking all remaining processes to terminate...
                                                                         [ OK ] * All processes ended within 4 seconds....
                                                                         [ OK ] * Deactivating swap...
                                                                               umount: /run/lock: not mounted
                                                                                                             umount: /run/shm: not mounted
                                                                                                                                           * Will now restart
                                                                                                                                                             [  211.680000] Restarting system.

U-Boot SPL 2012.10-04268-gf925eea-dirty (Jan 20 2013 - 15:46:58)
DRAM: 512MB
SUNXI SD/MMC: 0


Thanks in advance
Hristiyan

Quote from: ozko on February 23, 2013, 12:53:19 PM
I`m new to all this. So here`s my problem, I managed to get the R17 to run, but I can`t run the video... I`m using olinuxino wifi and when i run /home/olinuxino/go-video it says: cannot run kernels/video/Uimage.
Here is the log from the serial :

Code Select Expand


* Stopping Network connection manager wicd                                             [ OK ]
                                                                         [ OK ] * Asking all remaining processes to terminate...
                                                                         [ OK ] * All processes ended within 4 seconds....
                                                                         [ OK ] * Deactivating swap...
                                                                               umount: /run/lock: not mounted
                                                                                                             umount: /run/shm: not mounted
                                                                                                                                           * Will now restart
                                                                                                                                                             [  211.680000] Restarting system.

U-Boot SPL 2012.10-04268-gf925eea-dirty (Jan 20 2013 - 15:46:58)
DRAM: 512MB
SUNXI SD/MMC: 0


Thanks in advance
Hristiyan

Is the video not working? Can you post the whole log? Are you using sudo (e.g. "sudo ./go-video")?

Hi Jwishchka,
great help so far, having an issue im hoping you cna help with.

I am trying to use a USB wifi adapter TL-WN721N which is supported by ATH9K_HTC driver.

In the headless kernel appears to have this module however the video version does not.

Is there a reason for the video version not to have this module.
if not is it possible to include this in the next release.
Thanks

Probably an oversight. I'll try to get a new kernel out tomorrow, and make sure the wireless adapter support is common across the two.

Quote from: saand on February 26, 2013, 03:32:31 AM
Hi Jwishchka,
great help so far, having an issue im hoping you cna help with.

I am trying to use a USB wifi adapter TL-WN721N which is supported by ATH9K_HTC driver.

In the headless kernel appears to have this module however the video version does not.

Is there a reason for the video version not to have this module.
if not is it possible to include this in the next release.
Thanks

I've been testing your image and I've found some things that can be improved.
I think that you could better use the Linaro 12.11 http://releases.linaro.org/12.11/ubuntu/leb-panda, that comes with Ubuntu 12.04.2 LTS whitch is more polished and stable. I have found some programs that don't work in the last Linaro release.

Quote from: iso9660 on February 27, 2013, 03:44:12 PM
I've been testing your image and I've found some things that can be improved.
I think that you could better use the Linaro 12.11 http://releases.linaro.org/12.11/ubuntu/leb-panda, that comes with Ubuntu 12.04.2 LTS whitch is more polished and stable. I have found some programs that don't work in the last Linaro release.

Thanks for your suggestion. I prefer to be more on the cutting edge of things, and given that 12.04 LTS is about to be two major versions behind (with the release of Raring in two months), I'd prefer not to go back to it.

Which programs are you finding that don't work? Admittedly I don't do a lot of UI type stuff, mostly dev work, but I haven't found anything that's given me any problems. In general, I've found more support in Quantal for packages than in Precise, though not by a huge margin. I suspect that as Raring is going to be released specifically with ARM support in mind, the direction to go is forward, not back.

All of that said, you're certainly welcome to build your own Precise image for distribution. This is just a side product of a base image we are exploring for use at work, that I thought people might find useful.

Quote from: jwischka on February 27, 2013, 04:21:49 PM
Which programs are you finding that don't work?

Codelite IDE that I use to develop directly on board.

Quote from: jwischka on February 27, 2013, 04:21:49 PM
All of that said, you're certainly welcome to build your own Precise image for distribution.

I first tested with your image (changing the uboot and kernel) but as I said, some programs don't work. Then I moved to the stable version (Linaro 12.11). Remember the version your image use is a developer (Linaro developer) one.

Quote from: iso9660 on February 28, 2013, 01:49:59 PM

Quote from: jwischka on February 27, 2013, 04:21:49 PM
Which programs are you finding that don't work?

Codelite IDE that I use to develop directly on board.

Quote from: jwischka on February 27, 2013, 04:21:49 PM
All of that said, you're certainly welcome to build your own Precise image for distribution.

I first tested with your image (changing the uboot and kernel) but as I said, some programs don't work. Then I moved to the stable version (Linaro 12.11). Remember the version your image use is a developer (Linaro developer) one.

FWIW, I just installed Codelite and it seems to work fine, both on my image and on my Cubieboard image (also Quantal).

The Linaro developer version is not "dev" because it is "unstable", but because it has development tools (g++, etc) preinstalled. Quantal is the current stable version of Ubuntu, and the Linaro dev image is a Quantal image, not a Precise image. One huge reason not to use the LTS release, if for no other reason, is that it can only be upgraded in place to another LTS release - which won't happen for another year or two. With Raring expected to give significant improvement to ARM support, I don't want an image that I'll have to reformat (something I potentially can't do when they're sitting in a box in the field) in order to upgrade.

Look - again, I'm happy you're on the desktop release and that it's working out for you. But there are, in fact, actual engineering reasons I've chosen to use the Quantal release (and why I'll bump to Raring when it is released). I'm happy to make changes to the image for people if it's something that is broadly helpful and doesn't conflict with my own needs for the board. Moving everything back to Precise (and maintaining two sets of images), though, is a fairly major change, not to mention a lot more work for me, and something that probably isn't going to be in the cards.

As I said, if you think it would be a huge benefit to the community to have a flashable Linaro Desktop release, I'd be more than happy for you to support it. I'll even host the files for you. But at the moment I've got my hands full supporting one release, and it feels like I'm behind on it most of the time. If I add support for something else, it would be the A13-Micro, or the A10s-Micro, when it comes out.

First off, let me start by saying that I love the linux image. Once if figured out that root wasn't a member of the sound group it was trivial to install Squeezeslave on it so now it's plugging away as my squeezebox replacement in the kitchen.

On thing though, it does consume quite a bit of energy even in it's headless state. Is there a way to make it use less energy? I was thinking about disabling the XServer alltogether but I can't figure out how to do that (without breaking anything)

Any thoughts on both?

Cheers,

Leon

Quote from: ontsurt on March 05, 2013, 11:27:11 AM
First off, let me start by saying that I love the linux image. Once if figured out that root wasn't a member of the sound group it was trivial to install Squeezeslave on it so now it's plugging away as my squeezebox replacement in the kitchen.

On thing though, it does consume quite a bit of energy even in it's headless state. Is there a way to make it use less energy? I was thinking about disabling the XServer alltogether but I can't figure out how to do that (without breaking anything)

Any thoughts on both?

Cheers,

Leon

I have cpufreq disabled, so the board is running @ 1GHz the entire time. I can add it in a kernel update, if people are interested.

The way I look at it: at 50 watt continuously the board uses about 95 Euro's worth of energy per year. That just feels so wasteful for a radio/squeezebox that I only listen too 2 or 3 hours per day when having breakfast or dinner.

Then again I don't know it there are any drawbacks from using cpu freq and allowing the board to go into near complete idle mode. If that makes it completely unsuitable for other purposes for instance then obviously it makes sense not to implement it.

I've tried to do some reading on it but this and came across some references to using cpufreq in combination with sched_mc_power_savings which apparently gives the best results or something but honestly this is beyond my current understanding of Linux.

If you do it you make me a happy man is about all I can 'meaningfully' say.

Kind regards,

Leon

Just turn it off when not in use.

It's hardly its main market!

John

Hi John,

Sure I can do that and I will if there is no other way to resolve this.
However there are two thoughts I have around that.

• If there are no serious ill effects of enabling the dynamic CPU frequency then enabling it will at least make all usages a little more efficient. Nothing wrong with that as far as I can see
• having to turn the device on and off reduces the amount of locations one can usefully deploy it. I was thinking of mounting one in the garden so we can listen to music there. Having to turn it on or off means that it's more difficult to do that than when I can just lock it in a waterproofed box and only control it through Wifi

I realize these are all not significant items in the grand vast emptiness of space but if there are no ill effects why not?

Quote from: ontsurt on March 06, 2013, 12:10:45 AM
The way I look at it: at 50 watt continuously the board uses about 95 Euro's worth of energy per year. That just feels so wasteful for a radio/squeezebox that I only listen too 2 or 3 hours per day when having breakfast or dinner.

Then again I don't know it there are any drawbacks from using cpu freq and allowing the board to go into near complete idle mode. If that makes it completely unsuitable for other purposes for instance then obviously it makes sense not to implement it.

I've tried to do some reading on it but this and came across some references to using cpufreq in combination with sched_mc_power_savings which apparently gives the best results or something but honestly this is beyond my current understanding of Linux.

If you do it you make me a happy man is about all I can 'meaningfully' say.

Kind regards,

Leon

I'm sorry, but there is absolutely no way the board is using 50 watts of energy. If it were, you wouldn't be able to hold it. Where are you getting that number? I suspect the board is using less than 5 watts at full power, if that. Do you actually have something that is measuring the power, or are you just looking at the rating of your supply? Put in perspective, many modern day laptops don't consume more than 40 watts - and even some desktop are down in the 45-50 watt range.

I'm in the process of doing some kernel updating, and I'll compile cpufreq support in, but I don't think you have to worry about the board using that much energy.

Kernel updates later today, perhaps.

Thanks for the pointer. I have to admit that I found it odd myself but it was technically possible as the PSU I am using delivers 12 volt at a maximum of 4 Amps. So I was just assuming it was maxing out the PSU. The PSU is regulated so it shouldn't consume this much power if it's not used.

However if you suspect it's only drawing about 5 watts of power it could well be that the sensor I am using is simply not sensitive enough and somehow bottoming out at 50 Watt. I'll borrow a more accurate sensor and see what that reads.

Cheers

Quote from: ontsurt on March 08, 2013, 05:23:56 PM
Thanks for the pointer. I have to admit that I found it odd myself but it was technically possible as the PSU I am using delivers 12 volt at a maximum of 4 Amps. So I was just assuming it was maxing out the PSU. The PSU is regulated so it shouldn't consume this much power if it's not used.

However if you suspect it's only drawing about 5 watts of power it could well be that the sensor I am using is simply not sensitive enough and somehow bottoming out at 50 Watt. I'll borrow a more accurate sensor and see what that reads.

Cheers

Remember that in electronic terms, power always goes to heat. If the board is not hot, then you're not drawing that much power. If your PSU is hot, then the PSU may be consuming that much power, but it's not the board's fault.

R18 is posted.

Added FTDI support, and fixed some wireless driver issues in the video kernel. Preliminary support for cpufreqd, but it's not working yet. And I'm not going to spend a ton of time working on it at the moment. Maybe in future releases.

Kernels:
R15 Video Kernel (http://files.wisch.org/3.4-video-R18.tar.gz)
R15 Headless Kernel (http://files.wisch.org/3.4-headless-R18.tar.gz)



Image:
Get current version (R18) here. (http://files.wisch.org/olinuxino_xfce-r18.zip)

Hi!

These Ubuntu images seem very nice. I would like to use them but I want to be
able to compile the kernel myself. To get the exactly same kernel it is however not
enough to use the kernel sources that can be downloaded according to the
general instructions about building the kernel. The vital part is the .config file.
Would it be possible to obtain the .config file for the kernels in these
images. I would like to have it at least for R17 and R18.

Thank you in advance!

Already in the tar files.

John

Quote from: alp on March 11, 2013, 10:50:27 AM
Hi!

These Ubuntu images seem very nice. I would like to use them but I want to be
able to compile the kernel myself. To get the exactly same kernel it is however not
enough to use the kernel sources that can be downloaded according to the
general instructions about building the kernel. The vital part is the .config file.
Would it be possible to obtain the .config file for the kernels in these
images. I would like to have it at least for R17 and R18.

Thank you in advance!

As John said, they are included in the tars, and in the image itself.

Thank you, now I have found the config files. But then I had problems using
them. I followed the instructions in

https://www.olimex.com/wiki/Build_Bootable_SD_Card_with_Debian

under "Building the Kernel". But apparently this caused the download of the version
3.0.62 of kernel. According to the jwischka changelog I see I should have
version 3.4.29. I am confused now... Where exactly should I look to get this version of the
kernel source tree which includes all of the necessary support for the
Olimex A13-WIFI board? Can I simply use the general version 3.4.29 on www.kernel.org?
Do I need to apply some patches to it for the Olimex A13-WIFI board support?

Once again... thanks for the help!

Quote from: alp on March 11, 2013, 07:44:49 PM
Thank you, now I have found the config files. But then I had problems using
them. I followed the instructions in

https://www.olimex.com/wiki/Build_Bootable_SD_Card_with_Debian

under "Building the Kernel". But apparently this caused the download of the version
3.0.62 of kernel. According to the jwischka changelog I see I should have
version 3.4.29. I am confused now... Where exactly should I look to get this version of the
kernel source tree which includes all of the necessary support for the
Olimex A13-WIFI board? Can I simply use the general version 3.4.29 on www.kernel.org?
Do I need to apply some patches to it for the Olimex A13-WIFI board support?

Once again... thanks for the help!

git checkout sunxi-3.4

Are you wanting to compile this for an academic exercise, or are you wanting to actually add/remove something from the kernel?

Thank you very much, I will try that.

At the moment I do not have exact plan of
what I would do with the kernel. I want to use the GPIOs
and it is always good to have possibility to tweak the kernel.
I also can not get the touchscreen to work correctly and maybe
I will put some logs into the kernel driver to see what's
happening.

I'm running R18 now. It's good to see some of the problems from old versions are gone :) thanks for that


Still two grudges:

- cpufreqd is not working correctly -maybe it's time to upgrade to something else?


- when i check who's on my board I always get a root user that doesnt go away :)

root     tty1                      Wed21    2days  0.09s  0.03s -bash

even when I reboot my board, he stays there. in this case I have rebooted my card last time 2 days ago lol


thanks for all the work you've place into this jwischka

I think this is my first post here, and I love my olinuxino wifi which was running the standard image from Olimex.

Today I switched to the R18 version of this image and the audio was dead, much head breaking and comparison between the other image and this one and reading from this thread itself, here is what fixed the issue...

olinuxino@ubuntu:~/drive/Songs$ grep audio /etc/group
audio:x:29:linaro,olinuxino

If the audio line does not have olinuxino as the user defined there, aplay -l will say something like this "aplay: device_list:207: no soundcards found..."

Happy to have the audio going but for some reason mplayer is hitting 100% load at times, and the playback is choppy, which doesn't happen on the other image, not sure what could be causing this but am investigating. This is with an mp3 not video.

Also I have the webcam that is integrated in a DELL SP2309W throwing video out using motion, it was a cakewalk there. :)

Installing anything using apt-get always throws this error...

Errors were encountered while processing:
cpufreqd
E: Sub-process /usr/bin/dpkg returned an error code (1)

What could be causing that?

cpufreqd support is just broken. I shouldn't have included it. I'd recommend uninstalling for the moment. I may try to get it working in the future, but honestly it's pretty low on the priority list.

Quote from: jwischka on March 17, 2013, 01:55:27 PM
cpufreqd support is just broken. I shouldn't have included it. I'd recommend uninstalling for the moment. I may try to get it working in the future, but honestly it's pretty low on the priority list.

There's no point in trying to fix it, just don't include it in future images.

If you really want there are other more recent options.

Do you know which headers to use to compile the source against the kernel?

I am talking about "olinuxino_xfce-r15.zip".

"uname -a" is giving:
"3.4.29-jwischka-3.4-video-20130209-R15+ #20 PREEMPT"

Can I get the "github" link for the kernel source, you have used?

Thanks.

Quote from: nvd on March 19, 2013, 04:08:50 PM
Do you know which headers to use to compile the source against the kernel?

I am talking about "olinuxino_xfce-r15.zip".

"uname -a" is giving:
"3.4.29-jwischka-3.4-video-20130209-R15+ #20 PREEMPT"

Can I get the "github" link for the kernel source, you have used?

Thanks.

It's just the standard linux-sunxi github, with the modifications made for the i2c bus on the Olinuxino-A13.

OK.
I was looking for sublevel 29 on the "sunxi-3.4" branch.
As it seems, only one sublevel exists; 29.

Hi Jwischka thanks for R18, the included driver works well.

I am noticing the startup of this version and a few before it takes significant time before anything is visible on the screen.

Note sure if others have seen this issue, didn't see any mention in the last few pages of this thread.

this long delay was seen from right after the fresh reimage of the SD card.

running dmesg in the terminal shows there is a long pause or apparent pause between "[15.85] init: plymouth-splash main process (255) terminated with status 2"
then the next line "[141.28] init: plymouth-stop pre-start process (740) terminated with status 1"

just wondering if this is normal or an issue
thanks

Hello saand

   check your network connection editing /etc/interfaces

   if I'm not wrong, R18 try to setup wifi...

   BTW: what's plymouth for ? some time ago there was a splash manager with this name...
   but I see no splash during boot...

          giampaolo
   

thanks, removed a double entry in my /etc/network/interfaces file now booting is down to 28.2 sec

a bit longer than id like and i can see the main booting has really finished at 12.8 but its a lot better anyway.

for reference the last lines of dmesg are
[12.84] wlan1: associated
[19.74] init: failsafe main process (292) killed by TERM signal
[28.16] init: plymouth-stop pre-start process (788) terminated with status 1
[28.20] init tty1 main process (789) killed by TERM signal

if anyone knows what any of this means or why its chewing up 16 seconds of boot would be nice to know. either way im happy with it at the moment

not sure what plymouth is either

Quote from: saand on March 22, 2013, 04:38:45 PM
thanks, removed a double entry in my /etc/network/interfaces file now booting is down to 28.2 sec

a bit longer than id like and i can see the main booting has really finished at 12.8 but its a lot better anyway.

for reference the last lines of dmesg are
[12.84] wlan1: associated
[19.74] init: failsafe main process (292) killed by TERM signal
[28.16] init: plymouth-stop pre-start process (788) terminated with status 1
[28.20] init tty1 main process (789) killed by TERM signal

if anyone knows what any of this means or why its chewing up 16 seconds of boot would be nice to know. either way im happy with it at the moment

not sure what plymouth is either

It's a splash screen, basically.

Have you installed any additional packages?

Hi jwischka

if plymouth is, as supposed, a splash screen... where is is ? I see no splash during boot... at least
not on LCD... can I safetly remove it ?

for saand: my R18 boot in less than 15 seconds... so maybe there is something wrong somewhere on your setup...

   best regards

         Giampaolo

The additional packages are i think gtk.

i only reimaged the SD card with R18 today. plugged it in, it took a long time to show screen to start up but there was a network interface defined by default so that would explain it being slow initially.

i got network going then tried to start installing GTK so i could start developing.

thats all i have done with it so far.

I am happy with the speed of it currently just thought id mention in the event it could be speed up for others

thanks iw2lsi.

good to know others are getting fast booting.

I also do not see a splash screen. maybe its just not configured and therefore black, i do see the screen start up and turn on the backlight as soon as the system is powered.

I suspect based on the error msgs i listed that some settings are not working well for me as you suggest. Wonder if anyone has any ideas how to debug this. dmesg doesn't give me too much info

Hi all,

I am trying to recompile kernel to enable LCD display rotation (we need "portrait" mode). I've read this thread and all available tutorials, but I cannot reproduce kernel on my computer what works with LCD. I use (linux mint 14) arm-linux-gnueabihf  version 4.7, script.bin for LCD 800x480 (which perfectly works with r18 kernel for example), .config form this distro and sources from both linux-sunxi branch 3.4 and jwischka git repo. Only difference I could point out is i2c patch that I'm not able to locate yet.

I'm aware that most of you do not own LCD but any hint is highly appreciated.

BR, Cedomir

I have successfully booted R18
I am unable to get the wifi working.
I got it working under Debian SD.
using the same steps I can not get an IP.
I am using WEP 128.
I do a iwlist wlan1 scan and see my AP
I have modified the interfaces file and
did the manual steps.

any Ideas?

First idea would be to make sure you have your key entered correctly (I know you've probably already checked that). Specifically, I'd make sure you don't have a space after that.

Next, I'd try removing encryption briefly and see if that fixes the problem.

Finally, while I don't tend to do this sort of thing, I'd recommend going to WPA, unless you have a legacy device that will only do WEP. Given the well known and documented vulnerabilities in WEP, it's a good idea not to use it.

Thank you jwischka, this is fantastic!!!

sorry I forgot to update.
wicd finally listed my aP so was able to configure for Wpa.
now it boots with the wifi on

Anybody have any idea how can I access the GPIOs ? (on the Pi it's under /sys/class/gpio/)

Quote from: aquarat on May 08, 2013, 12:23:40 AM
Anybody have any idea how can I access the GPIOs ? (on the Pi it's under /sys/class/gpio/)

Easiest way is probably to install the python GPIO library - if that fits within the scheme of what you're trying to do.

Hi jwischka,

Thank you for your work wich is very helpful to me.

I'm actually working on a project on A13-OlinuXino-Wifi board. I compiled a Linaro Developper OS for the board and a QT GUI which automatically run at the boot.

In order to optimize the boot time I'm trying to build Mali driver in the kernel and not to load it as a module.

A saw that you did it since the R15 version and tried to use it with the R18 version.
I installed the client driver using sunxi-mali with the config

EGL_TYPE=framebuffer VERSION=r3p0

when I run the "test" example I got the next error

UMP:ump_arch_open() failed to open UMP device driver
Could not initialize egl display

EGL error
Aborded

It is as if the kernel driver were not loaded. Do you have an idea of where I did a mistake ?
I just precise that everything is working when I use mali as a module.

Thank you very much in advance

Martin

Did your compile work correctly with MALI built into the kernel? (In other words, are you sure that your kernel does contain MALI support, not as a module?)

Quote from: milimilimoy on May 22, 2013, 05:43:50 PM
Hi jwischka,

Thank you for your work wich is very helpful to me.

I'm actually working on a project on A13-OlinuXino-Wifi board. I compiled a Linaro Developper OS for the board and a QT GUI which automatically run at the boot.

In order to optimize the boot time I'm trying to build Mali driver in the kernel and not to load it as a module.

A saw that you did it since the R15 version and tried to use it with the R18 version.
I installed the client driver using sunxi-mali with the config

Code Select Expand

EGL_TYPE=framebuffer VERSION=r3p0

when I run the "test" example I got the next error

Code Select Expand

UMP:ump_arch_open() failed to open UMP device driver
Could not initialize egl display

EGL error
Aborded

It is as if the kernel driver were not loaded. Do you have an idea of where I did a mistake ?
I just precise that everything is working when I use mali as a module.

Thank you very much in advance

Martin

Using your R18 image and it looks quite nice.
But, I want to use a Davicom DM9601 usb-ethernet adapter.
There is a driver for this in my laptop Ubuntu machine, but not your
distribution. Can you compile a driver for this device?

Here is a link to the manufacturer's web site with source code:
http://www.davicom.com.tw/page1.aspx?no=209814

Quote from: bobmelville on May 27, 2013, 04:36:50 AM
Using your R18 image and it looks quite nice.
But, I want to use a Davicom DM9601 usb-ethernet adapter.
There is a driver for this in my laptop Ubuntu machine, but not your
distribution. Can you compile a driver for this device?

Here is a link to the manufacturer's web site with source code:
http://www.davicom.com.tw/page1.aspx?no=209814

Bob -

I'll have to look into it, and how much (if any) will have to be changed to make it work on ARM. I'm out this week, but can you PM me next Monday to remind me?

Much appreciated!
I am already finding this image to be quite useful, but
need to have internet access to install additional packages,
like a Fortran compiler.

long time no activity here - am I still at the right place?  Love the R18 build!  Works like a charm - running apache and all for me...

@bobmelville: I know, this is a late response, but did you put the lines

auto eth0
iface eth0 inet dhcp

into your /etc/network/interfaces file?  I guess they have been eliminated in favor of the WiFi customization, which is what most people seem to be using.  In your case you can comment out all the WLAN stuff to speed up boot time.

@jwischka: Could you tell us how to disable the "auto login" feature?  When I start up the board with a video-enabled kernel and a console attached, I get logged into the olinuxino account with full superuser privileges.  For initial configuration this may be what you want, but in the long run it's a bit too much power without password for my taste. I was looking for the /etc/lightdm/lightdm.conf file to edit to disable it, but there is no such file, not even such a directory.  The older /etc/gdm directory with the gdm.conf file does not exist either.  So I am running out of ideas.

So I'm clear, you're saying that you are auto-logged in when you have the video kernel enabled and have a keyboard, monitor, etc attached? That isn't behavior in the normal release, as far as I am aware. Have you changed something?

My suggestion would be to look into getty and see if you have autologin enabled there.

So everyone knows, I'm wanting to release a new kernel and image based on Raring, but frankly the current sunxi 3.4 kernel is, well, broken for the A13. That and I'm trying to track down a couple of annoying bugs in my spare time before releasing it. Hopefully soon, when a few more changes get committed in the sunxi branch.

Quote from: Ekkehard on June 26, 2013, 01:17:07 AM
long time no activity here - am I still at the right place?  Love the R18 build!  Works like a charm - running apache and all for me...

@bobmelville: I know, this is a late response, but did you put the lines

Code Select Expand


auto eth0
iface eth0 inet dhcp

into your /etc/network/interfaces file?  I guess they have been eliminated in favor of the WiFi customization, which is what most people seem to be using.  In your case you can comment out all the WLAN stuff to speed up boot time.

@jwischka: Could you tell us how to disable the "auto login" feature?  When I start up the board with a video-enabled kernel and a console attached, I get logged into the olinuxino account with full superuser privileges.  For initial configuration this may be what you want, but in the long run it's a bit too much power without password for my taste. I was looking for the /etc/lightdm/lightdm.conf file to edit to disable it, but there is no such file, not even such a directory.  The older /etc/gdm directory with the gdm.conf file does not exist either.  So I am running out of ideas.

Yup - that is exactly what I am seeing.  I did install a bunch of stuff (gawk, realpath, apache2, php5, php5-cgi, calibre), but I am not aware that I changed anything other than /etc/hostname and /etc/hosts to change the host name and of course /etc/network/interfaces to get my USB Ethernet adaptor to work.  If you are not seeing this - I'll start over with a fresh install directly from your image.  Maybe I was doing something else without noticing.....

Thanks for the very prompt response!!!

Quote from: Ekkehard on June 26, 2013, 01:33:47 AM
Yup - that is exactly what I am seeing.  I did install a bunch of stuff (gawk, realpath, apache2, php5, php5-cgi, calibre), but I am not aware that I changed anything other than /etc/hostname and /etc/hosts to change the host name and of course /etc/network/interfaces to get my USB Ethernet adaptor to work.  If you are not seeing this - I'll start over with a fresh install directly from your image.  Maybe I was doing something else without noticing.....

Thanks for the very prompt response!!!

Let me know if you still see the same behavior - like I said, I don't think it should be doing that, but if it continues with a fresh install, let me know.

I started fresh with an image I downloaded today.  Unzipped it and did

mkdosfs -F 32 -v -I /dev/sdd
dd bs=4M oflag=sync if=olinuxino_xfce-r18.img of=/dev/sdd

then put it into the board, hooked up monitor, keyboard and mouse (via KVM switch), and powered it up.
After the boot time I saw:
ubuntu login:
but after a second or so that disappeared and I was logged in and getting the prompt
root@ubuntu:~#
with full superuser privileges.  The working directory was /root.  No extra package was installed, no files were changed.  I didn't have that behavior with previous images.  If you don't see that, that's even a lot weirder...

That is definitely strange. I certainly don't see that on my updated builds, but I'd have to test on R18. Does there look to be anything in the getty settings for your build that might cause it? Really, really weird.

not sure I found all getty settings.  There is no file /etc/inittab, which is where I looked first.  But I then looked in /etc/init/tty{x}.conf and there was nothing out of the ordinary in there.  What I also found in /etc/init was an auto-serial-console.conf, which is something I don't have on my host (also running Ubuntu 13.04).  I also checked out a file autogetty in /etc/defaults (which again does not exist on my host).  The nasty AUTOGETTY_ARGS="-n -1 auto-root-login" line which I would have blamed for this weird behavior is commented out and the AUTOGETTY_ARGS is set to the empty string instead.  I even set AUTOGETTY_ARGS explicitly to /bin/login but it didn't run /bin/login at the beginning and still logged me in.

I also took both files (auto-serial-console.conf and autogetty) away - still the same behavior.

Not sure what else I could try.

You have no /etc/inittab?  I'd add one.  Maybe init can't decide what runlevel to enter.

Try an earlier build such as R17 and see what it does & what inittab it has.

John

Hm... strange. I'd suggest what John said as the next step as well.

Normally the console root autologin feature is in inittab:

T0:23:respawn:/sbin/getty -L -a root ttyS0 115200 vt102

Just remove the "-a root"

Linaro file-system has this feature enabled by default.

Now I am really confused.  @JohnS, @oldpinguin, are you all running R18, or at least some (maybe newer) Ubuntu version?  I was initially looking for /etc/inittab out of memory back from the days when I thought I knew all that stuff.  But Ubuntu does definitely not have an /etc/inittab anymore.  Therefore, R18 should not have one either - so I don't know where you guys are getting yours from.

Also, I'm afraid I need Ubuntu, as in Debian versions I couldn't get apache with php5 to work right, so going back to a Debian version won't cut it for me, as figuring out how to run apache/php5 under Debian still seems to be more work than fixing the autologin problem in Ubuntu (yes, I also read that "it was decided" that Linaro is having this "feature" enabled by default).

Here's the inittab deal: From 6.10 on, Ubuntu uses upstart to handle starting of tasks and services during boot, stopping them during shutdown and supervising them while the system is running.  So there is no more /etc/inittab by default.  However, I read that upstart would still honor an /etc/inittab file if it was there.  So, following all three suggestions, I added the line that @oldpenguin suggested (i.e. without the -a root) as the only line in my /etc/inittab file, but it didn't change anything.  This is not horribly surprising, since I want to remove an autologin command from wherever it is lingering, not add anything.

I checked the runlevel, and I am running in runlevel 2, as I should.  So I also checked /etc/rc2.d, as upstart likes to put stuff in /etc/rc{n}.d, where n is the runlevel - nothing in there either that would autologin as root (technically, it is olinuxino with superuser privileges).

Is anybody out there who's currently running R18 and doesn't have the autologin problem, or maybe has even solved it? Or somebody having an idea I haven't tried yet?  I seem to be running out of them soon  :(

Let me put a beta image together that is a stock Ubuntu Raring (not Linaro) and PM it to you to see if it works for you - and to see if you're seeing a bug that I'm experiencing in my testing. I'm about to be on a marathon conference call, so it may be a bit before I get it uploaded.

Quote from: Ekkehard on June 26, 2013, 08:41:50 PM
Now I am really confused.  @JohnS, @oldpinguin, are you all running R18, or at least some (maybe newer) Ubuntu version?  I was initially looking for /etc/inittab out of memory back from the days when I thought I knew all that stuff.  But Ubuntu does definitely not have an /etc/inittab anymore.  Therefore, R18 should not have one either - so I don't know where you guys are getting yours from.

Also, I'm afraid I need Ubuntu, as in Debian versions I couldn't get apache with php5 to work right, so going back to a Debian version won't cut it for me, as figuring out how to run apache/php5 under Debian still seems to be more work than fixing the autologin problem in Ubuntu (yes, I also read that "it was decided" that Linaro is having this "feature" enabled by default).

Here's the inittab deal: From 6.10 on, Ubuntu uses upstart to handle starting of tasks and services during boot, stopping them during shutdown and supervising them while the system is running.  So there is no more /etc/inittab by default.  However, I read that upstart would still honor an /etc/inittab file if it was there.  So, following all three suggestions, I added the line that @oldpenguin suggested (i.e. without the -a root) as the only line in my /etc/inittab file, but it didn't change anything.  This is not horribly surprising, since I want to remove an autologin command from wherever it is lingering, not add anything.

I checked the runlevel, and I am running in runlevel 2, as I should.  So I also checked /etc/rc2.d, as upstart likes to put stuff in /etc/rc{n}.d, where n is the runlevel - nothing in there either that would autologin as root.

Is anybody out there who's currently running R18 and doesn't have the autologin problem, or maybe has even solved it? Or somebody having an idea I haven't tried yet?  I seem to be running out of them soon  :(

I'm not using R18

John

Thanks $JohnS - now at least I am less confused  :)

Not sure if this is useful but (thanks to oldpenguin) I've got this image working on the A13's nand... I'm hoping it'll be possible to flash the image to the nand on other devices using dd.

@aquarat - "this image" meaning R18?  If so, did you solve the auto-login problem, or not run into it at all?  If you run headless, do you see, in addition to yourself, root logged in when you do a "who"?

@jwischka - I'd be more than happy do help you with any debugging to the full extent of my abilities.  You've done so much for this community, it would be great to give you at least a little bit in return.

Hi Ekkehard

I realise this is not good security protocol but I usually log into systems like the Pi/A13/etc. as root. I don't bother with sudo on those devices.

I noticed on the LiveSuit image oldpenguin prepp'ed for me that the auto-login feature on ttyS0 was present, but I erased the rootfs in his image and replaced it with a copy of this R18 image (which was running on an SD card at the time).

I'll check now and see if it auto-logs in via ttyS0... I usually disable getty on ttyS0 and use the machine via ssh on wifi.

-update-
Login via ssh isn't automatic.
Login via ttyS0 isn't automatic, but it takes approximately a minute for the login prompt on ttyS0 to appear (the device can be logged into via ssh on wifi long before the ttyS0 login prompt presents).
One user, on tty1 shows as logged in, the user is root. This happens using the headless kernel.

Thanks @aquarat!  When hooking up a monitor and keyboard I come in over tty1,  You come in via a serial port - am I right (hence ttyS0)?  Correct me if I'm wrong.  That would be a little different, but still you don't have root logged in via tty1 if I understand you right.

Disabling getty on the console seems a bit drastic to me, since I need to use it occasionally to do maintenance (for instance on calibre, where I don't think there is a CLI).

Yes, you are right!  It isn't good security protocol to always log in as root on any networked machine  :) :) :)  I have it easy, as I am not tempted to do that.  I have so many scripts that require extra manual interaction if they detect that they are run with special privileges.  Maybe I'm a little paranoid, but just because I'm paranoid doesn't mean that they are not after me  ;)

haha @ Ekkehard / they're after me :P .

My machine does have root logged in automatically on tty1 (the first on screen terminal)... but I've disabled that getty. When the getty on tty1 is enabled root is automatically logged in after boot.

I use the serial port for debugging but I usually disable the getty on that when everything is working (I sometimes use the serial port for other things).

If this machine was a "server", ie. sharing files or running specialised scripts in an environment that was directly connected to the internet... I wouldn't be doing stuff as root... but because the machine only has one user - me - and it's running behind a firewall and it exists to run an application that doesn't require network connectivity I figure root is acceptable :P . I run an ODroid machine as a small server in an office (not my office) and that has properly set up user accounts. I also have a Pi in PCExtreme's data center and that also has a properly set up user account which uses key-based auth via ssh.

I also usually move the port for ssh to something higher up for inet-linked machines... like 30022.

Also, my thanks to jwischka for this image!

Aha!!!

QuoteWhen the getty on tty1 is enabled root is automatically logged in after boot.

That is what I was wondering. 

So root being logged in after boot appears to be a systemic issue.  Of course, disabling getty on tty1 will fix the symptom, but it will not solve root cause.  And since I am not only paranoid but also pedantic (I guess), I'm still after finding and solving root cause :)  At any rate, your help with identifying this is much appreciated!!  It also tells me that I didn't do something really stupid to cause root to be logged in after boot, which still makes me feel good, in spite of all the paranoia and pedantry ;)

There's a script : /bin/auto-root-login
which logs root in.

It is referred to in /etc/init/openvt.conf , the last line is :
exec /bin/openvt -e -c 1 -f -- /bin/auto-root-login

I'm fairly sure that if you remove that script root will not longer automatically log on on tty1.

Thanks @aquarat,

I was aware of that.  I had already commented out that one line in /etc/init/openvt.conf, but since it didn't change anything, I re-enabled it.

Next thing on the list is to take the x-flag away from /bin/auto-root-login.  If that makes the problem go away, I should be able to find who calls the darn thing - since it ain't /etc/init/openvt.conf.

I'll let you know how it goes.

Weird - openvt reports "permission denied" when I do that.  Don't know why commenting out that line didn't fix it then.  Now, of course, the main console doesn't log in at all anymore but hangs - but I am still logged in via ssh (actually from two different machines - did I mention that I'm paranoid ;)?) - so I gave auto-root-login its x-flag back.  Next thing to try is change auto-root-login to simply /bin/login in openvt.conf - maybe that'll solve it.

Success!!

Changing /bin/auto-root-login into /bin/login in the file /etc/init/openvt.conf fixed the issue.  Thank you so much for bringing it up @aquarat - I had already put it on the pile of "things that didn't work" - too early, as it turns out.  Now I can leave getty enabled on tty1 to be able to log in and use X for maintenance on calibre.

@jwischka, of course I will still check out your native Raring beta!!

I'll send you a PM when I get it ripped. Sorry... the past couple of days have been filled with 5+ hour conference calls for a new project I've been assigned to... which has cut down my ability to play on fun projects I'd rather be working on :)

Hopefully this weekend!

Quote from: Ekkehard on June 28, 2013, 02:08:05 AM
Success!!

Changing /bin/auto-root-login into /bin/login in the file /etc/init/openvt.conf fixed the issue.  Thank you so much for bringing it up @aquarat - I had already put it on the pile of "things that didn't work" - too early, as it turns out.  Now I can leave getty enabled on tty1 to be able to log in and use X for maintenance on calibre.

@jwischka, of course I will still check out your native Raring beta!!

No rush @jwischka - as you know, my burning issue is solved.  I also just got a couple of PCBs back from the fab - so I am off to soldering for a while - not so much fun when the parts are getting smaller at the same rate as your eyes are getting weaker...

hi, hows things? i have been away from the forum a few months. i am still using 3.4.29-jwischka-3.4-video-20130209-R15+ and its been great so far. i did just get a Realtek RTL2832U based dongle and dmesg didnt pick it up other than a usb device. is there a shortcut way to get this working without a recompile?

Quote from: drspastic on June 29, 2013, 07:48:54 PM
hi, hows things? i have been away from the forum a few months. i am still using 3.4.29-jwischka-3.4-video-20130209-R15+ and its been great so far. i did just get a Realtek RTL2832U based dongle and dmesg didnt pick it up other than a usb device. is there a shortcut way to get this working without a recompile?

Probably need a new kernel. I'll see if I can get the new 3.4 branch to be a bit more stable and look into a release...

i still cant believe how stable this old version has been. maybe one crash in months of uptime and that was my fault!

Ok... update time.

You will note this has a big fat BETA tag on it. I think there may be a couple of regression issues. Or it could just be my board. If that scares you, don't use it.

First, there should be broader driver support. We're also on 3.4.43 now, for whatever that's worth. Someone decided to break SPI support in the kernel, so there is no SPI support, allowing things to compile. There may or may not be an issue with rebooting more than one time. There is for me. If people could test and see if their board hangs after a second reboot, I'd be grateful.

You will also note that there is a different user name. If you try to login with olinuxino as your user, you will not be able to. New U/P is ubuntu/olinuxino. The big news in this update is that we are on an official Raring filesystem, which should make everyone happier. At least, it would make me happier if I could nail down this stupid reboot issue.

Download: Get R19-BETA image here. (http://files.wisch.org/olinuxino-r19-beta.zip)

Good job, jwischka, but for me is still the bug that don't allows doing a reboot the second time.... :S

What's the last version of your distribution without this bug? Is possible to compile it (i'm noob with kernels...) with ftdi support? At this moment, your distribution is the only with FTDI, but still need to reboot remotely....

Thanks for your effort!

Quote from: jpacha on July 08, 2013, 02:56:11 PM
Good job, jwischka, but for me is still the bug that don't allows doing a reboot the second time.... :S

What's the last version of your distribution without this bug? Is possible to compile it (i'm noob with kernels...) with ftdi support? At this moment, your distribution is the only with FTDI, but still need to reboot remotely....

Thanks for your effort!

R18 should work for you.

Quote from: jwischka on July 08, 2013, 03:02:00 PM

Quote from: jpacha on July 08, 2013, 02:56:11 PM
Good job, jwischka, but for me is still the bug that don't allows doing a reboot the second time.... :S

What's the last version of your distribution without this bug? Is possible to compile it (i'm noob with kernels...) with ftdi support? At this moment, your distribution is the only with FTDI, but still need to reboot remotely....

Thanks for your effort!

R18 should work for you.

Wow! Instant reply!
:( No, R18 has the same problem with rebooting... the second reboot hangs

Quote from: jpacha on July 08, 2013, 03:09:29 PM

Quote from: jwischka on July 08, 2013, 03:02:00 PM

Quote from: jpacha on July 08, 2013, 02:56:11 PM
Good job, jwischka, but for me is still the bug that don't allows doing a reboot the second time.... :S

What's the last version of your distribution without this bug? Is possible to compile it (i'm noob with kernels...) with ftdi support? At this moment, your distribution is the only with FTDI, but still need to reboot remotely....

Thanks for your effort!

Did you flash the whole R18 image, or just replace the kernel? R18 reboots fine for me.

The reboot problem seems to be a kernel issue, and one that is not so easy to track down. If I get a chance, along with the million other things I have to do this week, I may try to see if I can debug it some more. Until then, hope someone on linux-sunxi fixes what they broke :).

R18 should work for you.

Wow! Instant reply!
:( No, R18 has the same problem with rebooting... the second reboot hangs

Quote from: jwischka on July 08, 2013, 03:12:40 PM

Quote from: jpacha on July 08, 2013, 03:09:29 PM

Quote from: jwischka on July 08, 2013, 03:02:00 PM

Quote from: jpacha on July 08, 2013, 02:56:11 PM
Good job, jwischka, but for me is still the bug that don't allows doing a reboot the second time.... :S

What's the last version of your distribution without this bug? Is possible to compile it (i'm noob with kernels...) with ftdi support? At this moment, your distribution is the only with FTDI, but still need to reboot remotely....

Thanks for your effort!

Did you flash the whole R18 image, or just replace the kernel? R18 reboots fine for me.

The reboot problem seems to be a kernel issue, and one that is not so easy to track down. If I get a chance, along with the million other things I have to do this week, I may try to see if I can debug it some more. Until then, hope someone on linux-sunxi fixes what they broke :).

R18 should work for you.

Wow! Instant reply!
:( No, R18 has the same problem with rebooting... the second reboot hangs

Tested with R18, blank installation, full image, ... can't reboot twice: Second hangs.

I'll try to put together a new kernel again this week and see if the problem is fixed.

Hello,
it's possible to have your kernel source, in addition to bin?

I'm trying to compile module rtl8188eu, but I need original kernel source.
If I get kernel from linux-sunxi (git checkout v3.4.29-r2), using your .config file, and compiling module, I cannot insmod module.

Thank you

Quote from: pongo on July 09, 2013, 04:40:00 PM
Hello,
it's possible to have your kernel source, in addition to bin?

I'm trying to compile module rtl8188eu, but I need original kernel source.
If I get kernel from linux-sunxi (git checkout v3.4.29-r2), using your .config file, and compiling module, I cannot insmod module.

Thank you

Do a git checkout sunxi-3.4 instead. My source is the source from linux-sunxi, with a couple of changes made to the i2c files to make them work on the Olinuxino.

Hello,
alredy done, but last version is 3.4.43, your is 3.4.29.

I would not recompile the whole kernel.
I tried, but it failed. I'm cross compiling from Ubuntu 12.04 (gnueabihf)

Thank you

You'll probably need to recompile the whole kernel. You can also grab my latest (beta) image, which has a 3.4.43 kernel (with your driver built in.)

I already tried to recompile the whole kernel, but it failed.
I will try your new image.

Thanks

Another thing: I do not understand why my compilation failed, but you succeded.
These are my steps:

~$ sudo apt-get install build-essential git u-boot-tools ccache zlib1g-dev gawk bison flex gettext uuid-dev libusb-1.0-0-dev uboot-mkimage pkg-config libncurses5-dev
~$ sudo apt-get install gcc-arm-linux-gnueabihf
~$ git clone https://github.com/linux-sunxi/linux-sunxi.git
~$ cd linux-sunxi
~/linux-sunxi$ git checkout sunxi-3.4
~/linux-sunxi$ cp ../config-video-r18 .config
~/linux-sunxi$ make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- menuconfig
~/linux-sunxi$ make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- uImage

Finally I got


  ...
  CC      drivers/spi/spi_sunxi.o
In file included from arch/arm/mach-sun5i/include/mach/gpio.h:22:0,
                 from drivers/spi/spi_sunxi.c:27:
include/asm-generic/gpio.h: In function 'gpio_get_value_cansleep':
include/asm-generic/gpio.h:218:2: error: implicit declaration of function '__gpio_get_value' [-Werror=implicit-function-declaration]
include/asm-generic/gpio.h: In function 'gpio_set_value_cansleep':
include/asm-generic/gpio.h:224:2: error: implicit declaration of function '__gpio_set_value' [-Werror=implicit-function-declaration]
drivers/spi/spi_sunxi.c: In function 'spi_sunxi_register_spidev':
drivers/spi/spi_sunxi.c:1836:2: error: implicit declaration of function 'gpio_request' [-Werror=implicit-function-declaration]
drivers/spi/spi_sunxi.c:1837:3: error: implicit declaration of function '__gpio_to_irq' [-Werror=implicit-function-declaration]
cc1: some warnings being treated as errors
make[2]: *** [drivers/spi/spi_sunxi.o] Error 1
make[1]: *** [drivers/spi] Error 2
make: *** [drivers] Error 2

PS: I think that the driver is in branch wip/sunxi-3.4/rtl8188eu

Disable SPI. It's been broken for a month or so.

Disabling SPI I compiled successfully kernel using your .config R19-beta.
Thanks

Amazing! This is the only one image which works. Thank you.

I'm having a question. How I can configure the WiFi?

Quote from: SlyCom on July 11, 2013, 11:51:27 AM
Amazing! This is the only one image which works. Thank you.

I'm having a question. How I can configure the WiFi?

Edit the /etc/network/interfaces file.

Thanks a lot!

Is it possible to install java runtime?

Network

root@ubuntu:~# wpa_passphrase SSID password >> /etc/wpa_supplicant/wpa_supplicant.conf
olinuxino@ubuntu:~$ cat /etc/network/interfaces
# interfaces(5) file used by ifup(8) and ifdown(8)
auto lo
iface lo inet loopback

auto wlan0
iface wlan0 inet dhcp
        wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf

auto wlan1
iface wlan1 inet dhcp
        wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf

auto wlan2
iface wlan2 inet dhcp
        wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf

auto wlan3
iface wlan3 inet dhcp
        wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf


Java
http://linux-sunxi.org/Oracle_Java

Quote from: SlyCom on July 11, 2013, 03:32:25 PM
Thanks a lot!

Is it possible to install java runtime?

You can install Java by:

sudo apt-get update
sudo apt-get install default-jre

Thanks!

I wrote small java application on my Win machine with RXTX library for (serial ports). Now I'd like to move the generated JAR file to my Olinuxino. How I can do this???

P.S.
I'm not sure but I've recently installed 'librxtx-java' maybe it would help?

Quote from: SlyCom on July 12, 2013, 07:10:31 PM
Thanks!

I wrote small java application on my Win machine with RXTX library for (serial ports). Now I'd like to move the generated JAR file to my Olinuxino. How I can do this???

P.S.
I'm not sure but I've recently installed 'librxtx-java' maybe it would help?

Use filezilla to sftp it.

It works once again. Thanks!

Sorry but I'm having another question. I have USB=>serial. lsusb command shows it as CP210x UART bridge but it doesn't appears if I run dmesg command. How I can configure the device to work properly?

Not sure. Are you seeing an error message, or does it appear in /dev?

No error messages. I've got "usb 2-1.2.2: new full-speed USB device number 7 using sw-ehci" and no ttUSB available in /dev :(

Quote from: SlyCom on July 13, 2013, 10:50:27 AM
No error messages. I've got "usb 2-1.2.2: new full-speed USB device number 7 using sw-ehci" and no ttUSB available in /dev :(

Hmm... not sure what to tell you. Which version are you using?

I'm using the latest one.
The USB -> Serial FTDI converter is recognised correctly but not CP210x drivers.

I'll check and see if it's enabled in the kernel.

Thank you. If its disabled, do I need to recompile the Kernel??

Try this image. R19 Beta 2 (http://files.wisch.org/olinuxino-r19-beta2.zip)

It stops:
[    2.630000] Kernel panic - not syncing: Attempted to kill init!
[    2.630000] [<c003e204>] (unwind_backtrace+0x0/0x138) from [<c05a5c2c>] (panic+0x6c/0x18c)
[    2.640000] [<c05a5c2c>] (panic+0x6c/0x18c) from [<c0067f14>] (do_exit+0x6ec/0x720)
[    2.650000] [<c0067f14>] (do_exit+0x6ec/0x720) from [<c003c1a4>] (die+0x260/0x2b8)
[    2.660000] [<c003c1a4>] (die+0x260/0x2b8) from [<c0042758>] (__do_kernel_fault+0x74/0x84)
[    2.670000] [<c0042758>] (__do_kernel_fault+0x74/0x84) from [<c00428f4>] (do_page_fault+0x18c/0x2ec)
[    2.680000] [<c00428f4>] (do_page_fault+0x18c/0x2ec) from [<c003121c>] (do_DataAbort+0x34/0x9c)
[    2.680000] [<c003121c>] (do_DataAbort+0x34/0x9c) from [<c0037950>] (__dabt_svc+0x70/0xa0)
[    2.690000] Exception stack(0xdf83bf50 to 0xdf83bf98)
[    2.700000] bf40:                                     ffffffed 00000001 00000000 00000000
[    2.710000] bf60: ffffffed c002bc08 df83a000 00000000 c000e8c4 00000000 00000000 00000000
[    2.710000] bf80: c0842354 df83bf98 c000e98c c02e9a00 a0000013 ffffffff
[    2.720000] [<c0037950>] (__dabt_svc+0x70/0xa0) from [<c02e9a00>] (regulator_get_voltage+0x8/0x34)
[    2.730000] [<c02e9a00>] (regulator_get_voltage+0x8/0x34) from [<c000e98c>] (sun4i_cpufreq_initcall+0xc8/0x124)
[    2.740000] [<c000e98c>] (sun4i_cpufreq_initcall+0xc8/0x124) from [<c0031464>] (do_one_initcall+0x34/0x17c)
[    2.750000] [<c0031464>] (do_one_initcall+0x34/0x17c) from [<c0008a18>] (kernel_init+0xc8/0x158)
[    2.760000] [<c0008a18>] (kernel_init+0xc8/0x158) from [<c0038edc>] (kernel_thread_exit+0x0/0x8)
[    2.780000] [mmc]: mmc 0 detect change, present 1

This is going to sound silly, but pop the card, out and back in, and power cycle the board.

Quote from: SlyCom on July 23, 2013, 11:41:05 AM
It stops:
[    2.630000] Kernel panic - not syncing: Attempted to kill init!
[    2.630000] [<c003e204>] (unwind_backtrace+0x0/0x138) from [<c05a5c2c>] (panic+0x6c/0x18c)
[    2.640000] [<c05a5c2c>] (panic+0x6c/0x18c) from [<c0067f14>] (do_exit+0x6ec/0x720)
[    2.650000] [<c0067f14>] (do_exit+0x6ec/0x720) from [<c003c1a4>] (die+0x260/0x2b8)
[    2.660000] [<c003c1a4>] (die+0x260/0x2b8) from [<c0042758>] (__do_kernel_fault+0x74/0x84)
[    2.670000] [<c0042758>] (__do_kernel_fault+0x74/0x84) from [<c00428f4>] (do_page_fault+0x18c/0x2ec)
[    2.680000] [<c00428f4>] (do_page_fault+0x18c/0x2ec) from [<c003121c>] (do_DataAbort+0x34/0x9c)
[    2.680000] [<c003121c>] (do_DataAbort+0x34/0x9c) from [<c0037950>] (__dabt_svc+0x70/0xa0)
[    2.690000] Exception stack(0xdf83bf50 to 0xdf83bf98)
[    2.700000] bf40:                                     ffffffed 00000001 00000000 00000000
[    2.710000] bf60: ffffffed c002bc08 df83a000 00000000 c000e8c4 00000000 00000000 00000000
[    2.710000] bf80: c0842354 df83bf98 c000e98c c02e9a00 a0000013 ffffffff
[    2.720000] [<c0037950>] (__dabt_svc+0x70/0xa0) from [<c02e9a00>] (regulator_get_voltage+0x8/0x34)
[    2.730000] [<c02e9a00>] (regulator_get_voltage+0x8/0x34) from [<c000e98c>] (sun4i_cpufreq_initcall+0xc8/0x124)
[    2.740000] [<c000e98c>] (sun4i_cpufreq_initcall+0xc8/0x124) from [<c0031464>] (do_one_initcall+0x34/0x17c)
[    2.750000] [<c0031464>] (do_one_initcall+0x34/0x17c) from [<c0008a18>] (kernel_init+0xc8/0x158)
[    2.760000] [<c0008a18>] (kernel_init+0xc8/0x158) from [<c0038edc>] (kernel_thread_exit+0x0/0x8)
[    2.780000] [mmc]: mmc 0 detect change, present 1

This little thing is quite amazing - running apache, php, and quite a bit of my code connecting it to other machines via sockets and all (running R18).  So I'm almost done  ;)

But here is something that puzzles me: Just looked at my syslog to find out why cron isn't working, and I found it is getting filled with lines of the form

Jul 25 06:27:44 olinuxino kernel: [2205272.270000] init: openvt main process ended, respawning
Jul 25 06:28:44 olinuxino kernel: [2205332.990000] init: openvt main process ended, respawning
Jul 25 06:29:45 olinuxino kernel: [2205393.700000] init: openvt main process ended, respawning

So every minute openvt fails.  Not only does that clutter syslog quite a bit, something ain't quite right. Anybody having the same issue?  Anybody knows what causes it?  Anybody knows how to fix it?

There is some chatter on the net regarding Linaro and this symptom, but all connected to some upstart jobs - not something that happens every minute.  It being there every minute could mean it is cron-related (which runs every minute and is also in upstart).

Now I know why cron wasn't working - not sure that this is related to the other problem though.

cron is working - kind of - but it is using GMT instead of the local time.  That is definitely a bug.  What goes into crontab files should be local time, not GMT.  Other than using the wrong time, cron is working very reliably as usual.

Anybody knows how to fix this, i.e. where to start looking? @jwischka any ideas?

BTW, the "date" command shows the local time as expected - not GMT.

Quote from: Ekkehard on July 27, 2013, 12:10:41 AM
Now I know why cron wasn't working - not sure that this is related to the other problem though.

cron is working - kind of - but it is using GMT instead of the local time.  That is definitely a bug.  What goes into crontab files should be local time, not GMT.  Other than using the wrong time, cron is working very reliably as usual.

Anybody knows how to fix this, i.e. where to start looking? @jwischka any ideas?

BTW, the "date" command shows the local time as expected - not GMT.

Which image are you on?

R18

Quote from: Ekkehard on July 27, 2013, 12:21:06 AM
R18

Can you try with R19 beta 2 and see if it is still there?

will do - but it will take me a little while before I have my stuff set up again

Quote from: Ekkehard on July 27, 2013, 01:49:17 AM
will do - but it will take me a little while before I have my stuff set up again

Ok - The main reason I ask is because the entire rootfs has changed between the two versions, meaning that the bug may have already been fixed. Let me know if not and I'll look into it.

OK - tried it out - same result using R19 beta...

However, I found the problem - it was mine!!  I had used tzselect to select the time zone, which works fine but not system wide.  So cron does not know about my local time zone changes, even though I put the setting of the TZ variable into /etc/profile.  I should have used

sudo dpkg-reconfigure tzdata

to select the correct time zone instead.  Only this program modifies /etc/localtime correctly to reflect the proper local time zone.  With that mod, cron works perfectly in R19.  I'm going back now to R18 and my full configuration - I'm sure it will work fine there too.  I will switch to R19 once I got everything working.

Thanks for your help anyways!

Quote from: Ekkehard on July 27, 2013, 05:33:24 AM
OK - tried it out - same result using R19 beta...

However, I found the problem - it was mine!!  I had used tzselect to select the time zone, which works fine but not system wide.  So cron does not know about my local time zone changes, even though I put the setting of the TZ variable into /etc/profile.  I should have used

Code Select Expand


sudo dpkg-reconfigure tzdata

to select the correct time zone instead.  Only this program modifies /etc/localtime correctly to reflect the proper local time zone.  With that mod, cron works perfectly in R19.  I'm going back now to R18 and my full configuration - I'm sure it will work fine there too.  I will switch to R19 once I got everything working.

Thanks for your help anyways!

Glad you got it working

Hi,

I'm still using R18 version. How I can use I2C in JAVA with RXTX library or other?
The i2c tool is working fine but when I use rxtx in Java I have followinf available ports:
ttyUSB0 (USB to serial converter FTDI)
ttS0 (I guess its console).

Quote from: SlyCom on July 27, 2013, 12:24:50 PM
Hi,

I'm still using R18 version. How I can use I2C in JAVA with RXTX library or other?
The i2c tool is working fine but when I use rxtx in Java I have followinf available ports:
ttyUSB0 (USB to serial converter FTDI)
ttS0 (I guess its console).

I've only tried to do this in python, and never on the A13. I might start a separate thread and see if anyone has suggestions.

I'm trying to get my stuff working under R19 beta to get rid of the error messages in syslog that I get under R18, but I keep getting this error message:

perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
        LANGUAGE = (unset),
        LC_ALL = (unset),
        LANG = "en_US.UTF-8"
    are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory

Any idea what causes it?  It happens in a lot of system programs run under sudo.  Looks like some Perl directory isn't there?  Anybody else having that problem?  Should I try re-installing Perl?

One more thing:

build-essential is not installed in R19 beta by default.  If it is intentional, that's OK, of course.  Just didn't want it to get dropped by accident.  I think it was in R18.

Quote from: Ekkehard on July 31, 2013, 08:08:31 PM
One more thing:

build-essential is not installed in R19 beta by default.  If it is intentional, that's OK, of course.  Just didn't want it to get dropped by accident.  I think it was in R18.

It was. I'll add that to the next release.

Quote from: Ekkehard on July 31, 2013, 07:37:58 PM
I'm trying to get my stuff working under R19 beta to get rid of the error messages in syslog that I get under R18, but I keep getting this error message:

Code Select Expand


perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
        LANGUAGE = (unset),
        LC_ALL = (unset),
        LANG = "en_US.UTF-8"
    are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory

Any idea what causes it?  It happens in a lot of system programs run under sudo.  Looks like some Perl directory isn't there?  Anybody else having that problem?  Should I try re-installing Perl?

The problem, I believe, is that the location settings aren't set. I honestly just haven't had time to look into it, but I don't believe it actually causes any problems.

Oh no - you are right - it does not cause any harm - not to my knowledge anyways.  It's just something that ain't quite right and that catches my attention when I try to figure out what went wrong with my stuff.  When I have it all running, I'll check some more into it.  Will let you know what I find.

yup - you were right @jwischka.

sudo locale-gen en

fixes the problem. Of course "en" is for the English language only, "de" for German, and so on...

R19 looks great - it also fixed my problem with the syslog entries that happened every minute.  While I am not paranoid about the write cycles to the SD card, writing to it every minute of every hour of every day may get to it eventually.  Haven't had this problem in R19 ever!  Thanks for R19 @jwischka!!!

One more question: Are the audio drivers in there?  mplayer doesn't create any audio output for me.  Don't need them right now - just curious...

Can I connect usb camera to the R18 or R19?

Quote from: Ekkehard on August 01, 2013, 12:57:12 AM
R19 looks great - it also fixed my problem with the syslog entries that happened every minute.  While I am not paranoid about the write cycles to the SD card, writing to it every minute of every hour of every day may get to it eventually.  Haven't had this problem in R19 ever!  Thanks for R19 @jwischka!!!

One more question: Are the audio drivers in there?  mplayer doesn't create any audio output for me.  Don't need them right now - just curious...

I think so, but I could have accidentally dropped them. There wasn't a lot of regression testing on this one. I'll check when I get back to my build machine.

Quote from: SlyCom on August 02, 2013, 06:03:05 PM
Can I connect usb camera to the R18 or R19?

Yes.

dmesg just returns ' usb 2-1.2: new full-speed USB device number 5 using sw-ehci' when I attach USB camera or for example USB-bluetooth dongle.

The lsusb command:
Bus 002 Device 005: ID 046d:092f Logitech, Inc. QuickCam Express Plus
Bus 002 Device 006: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode)

Quote from: SlyCom on August 05, 2013, 04:01:49 PM
dmesg just returns ' usb 2-1.2: new full-speed USB device number 5 using sw-ehci' when I attach USB camera or for example USB-bluetooth dongle.

The lsusb command:
Bus 002 Device 005: ID 046d:092f Logitech, Inc. QuickCam Express Plus
Bus 002 Device 006: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode)

Is the camera UVC compliant?

I don't know if the camera is UVC but what about the bluetooth usb dongle?

Quote from: jwischka on June 29, 2013, 09:44:40 PM

Quote from: drspastic on June 29, 2013, 07:48:54 PM
hi, hows things? i have been away from the forum a few months. i am still using 3.4.29-jwischka-3.4-video-20130209-R15+ and its been great so far. i did just get a Realtek RTL2832U based dongle and dmesg didnt pick it up other than a usb device. is there a shortcut way to get this working without a recompile?

Probably need a new kernel. I'll see if I can get the new 3.4 branch to be a bit more stable and look into a release...

hm... dmesg does see it now  ::)
[ 6243.420000] usb 2-1.3: new high-speed USB device number 7 using sw-ehci

olinuxino@debian:~$ lsusb
Bus 002 Device 002: ID 05e3:0608 Genesys Logic, Inc. USB-2.0 4-Port HUB
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 003: ID 04d9:1603 Holtek Semiconductor, Inc. Keyboard
Bus 002 Device 006: ID 046d:c001 Logitech, Inc. N48/M-BB48 [FirstMouse Plus]
Bus 002 Device 007: ID 0bda:2838 Realtek Semiconductor Corp. RTL2838 DVB-T
Bus 002 Device 005: ID 13d3:3273 IMC Networks 802.11 n/g/b Wireless LAN USB Mini-Card
olinuxino@debian:~$ lsmod
Module                  Size  Used by
rt2800usb              12206  0
rt2800lib              41704  1 rt2800usb
crc_ccitt               1054  1 rt2800lib
rt2x00usb               7495  1 rt2800usb
rt2x00lib              31405  3 rt2x00usb,rt2800lib,rt2800usb
led_class               2407  1 rt2x00lib

and the led comes on so its not so bad after all.
just need the module.

by the way, these dongles have 2 uses, requiring different modules that i hear can clash:
first is the dvb-t / dab standard function, second is rtl-sdr taking raw sample stream for
software defined radio like gnuradio etc. 24-1850 mhz can be had from the rtl2832/r820t dongles.
i read that the r820t tuner chip can also be software disabled somehow to allow raw samples from dc-24 mhz to be decoded too.

anyway, im rambling..... just thought it might be interesting as it opens up a big chunk of cheap connectivity

Why sometimes the board boot correct but sometimes it try to boot from nand? Is it a SD card issue?

Quote from: SlyCom on August 19, 2013, 11:41:45 PM
Why sometimes the board boot correct but sometimes it try to boot from nand? Is it a SD card issue?

I'm not sure to be honest, though I can tell you I experience the same behavior.

So is there any solution for this issue? My device works remotely and in case off power cut it sometimes not starting.
Is there any solution to write the image to NAND?

P.S.
What is the user and password for version R19 ?? I can not log in using olinuxino/olinuxino.

I'm the only one who can run this image of Debian un the A13MICRO board ?!

Quote from: SlyCom on August 20, 2013, 10:29:42 AM
So is there any solution for this issue? My device works remotely and in case off power cut it sometimes not starting.
Is there any solution to write the image to NAND?

P.S.
What is the user and password for version R19 ?? I can not log in using olinuxino/olinuxino.

ubuntu/olinuxino

I haven't spent a lot of time looking into the boot issues - writing to NAND would likely be a workable solution, though I haven't looked into doing that either. I know that others have looked at booting from NAND, but I haven't kept up with that line of posts recently.

Hello

I'm wondering why I can't get the image to fit on any 4gb card, with Linux or Image Writer or Roadkils image writer on windows

for example:

jonne@Ibuntu:~/Downloads$ sudo dd if=~/Downloads/olinuxino_xfce-r18.img of=/dev/sdb
dd: writing to `/dev/sdb': No space left on device
7626753+0 records in
7626752+0 records out
3904897024 bytes (3.9 GB) copied, 1975.27 s, 2.0 MB/s

... running out of space with something like 1 or 2 sectors to go

Obviously, the solution is to buy a 8 or 16 gb card to write the 4gb image. i just dont understand why the image is 3.7 GB when all my 4GB cards seem to be 3.68 in capacity.

i wonder if this is related to:

https://bugs.launchpad.net/win32-image-writer/+bug/912259 (https://bugs.launchpad.net/win32-image-writer/+bug/912259)

What brand of card are you using? The image is made from a legit, 4GB Sandisk card. Cards with bad sectors (or fewer sectors) will obviously experience problems. I've used WinDiskImager and dd to successfully write the images to other (Sandisk) cards, however.

I suppose I could try to pull an image with DD, if you would like me to try, but I suspect it's a physical card problem, not a software issue.

Quote from: johnjonh on September 02, 2013, 07:57:00 PM
Hello

I'm wondering why I can't get the image to fit on any 4gb card, with Linux or Image Writer or Roadkils image writer on windows

for example:

jonne@Ibuntu:~/Downloads$ sudo dd if=~/Downloads/olinuxino_xfce-r18.img of=/dev/sdb
dd: writing to `/dev/sdb': No space left on device
7626753+0 records in
7626752+0 records out
3904897024 bytes (3.9 GB) copied, 1975.27 s, 2.0 MB/s

... running out of space with something like 1 or 2 sectors to go

Obviously, the solution is to buy a 8 or 16 gb card to write the 4gb image. i just dont understand why the image is 3.7 GB when all my 4GB cards seem to be 3.68 in capacity.

i wonder if this is related to:

https://bugs.launchpad.net/win32-image-writer/+bug/912259 (https://bugs.launchpad.net/win32-image-writer/+bug/912259)

wow. Fast reply. Thanks.

No need to bother with pulling a different image with dd. All my cards are Kingston... I've suspected them for a while. Your 4 gb sd images are not the only ones I cant get to fit.

Quote from: jwischka on September 02, 2013, 08:00:49 PM
What brand of card are you using? The image is made from a legit, 4GB Sandisk card. Cards with bad sectors (or fewer sectors) will obviously experience problems. I've used WinDiskImager and dd to successfully write the images to other (Sandisk) cards, however.

I suppose I could try to pull an image with DD, if you would like me to try, but I suspect it's a physical card problem, not a software issue.

That's good information to have. I may try to grab a Kingston card in the future, then, for making images. Obviously you can put a smaller image on a larger card, so having the "smallest" 4GB card possible would be good - from the perspective of the image maker.

Quote from: johnjonh on September 02, 2013, 09:02:53 PM
wow. Fast reply. Thanks.

No need to bother with pulling a different image with dd. All my cards are Kingston... I've suspected them for a while. Your 4 gb sd images are not the only ones I cant get to fit.

Quote from: jwischka on September 02, 2013, 08:00:49 PM
What brand of card are you using? The image is made from a legit, 4GB Sandisk card. Cards with bad sectors (or fewer sectors) will obviously experience problems. I've used WinDiskImager and dd to successfully write the images to other (Sandisk) cards, however.

I suppose I could try to pull an image with DD, if you would like me to try, but I suspect it's a physical card problem, not a software issue.

Hi jwischka,

Your ubuntu distribution is brilliant. I wonder if the destribution will work with Olinuxino iMX233-Olinuxino board?

regards

Quote from: SlyCom on September 13, 2013, 01:27:02 PM
Hi jwischka,

Your ubuntu distribution is brilliant. I wonder if the destribution will work with Olinuxino iMX233-Olinuxino board?

regards

With the right kernel, probably, though it would be quite slow, I suspect.

Hi,

I'm going to use devices connected to Olinuxino A13 via UART (Ubuntu r18). I'm using USB to UART dongle. The question is: Is there any option to use internal uart port? I read somewhere that there is UART but it's connected to sd card.

If you need actual RS232 signal levels then an easy solution is to use an EXT module MOD-RS232. At €5.95 it's not worth getting the soldering irion out.

Alternatively, if TTL is ok then 3.3V serial is available for UART1 and UART0 on the appropriately named connectors.

David Goadby

Hi jwischka,

will it be a image-version with TFT and Touchsrcreen (Olimex-TS7-TFT) support in future?

Best regards
Fluxxy

Hi,

I'm still trying to work with uart ports. David above mentioned about UART0 and UART1. I'm having a problem with UART1. How I can start using this port as a standard UART? If it's possible I'd like to avoid usb-uart conventers. The UART1 what is the baud rate: 115200?

Assuming you have a REV-B board then UART1 is NOT RS232 levels but 3.3v with TX data on pin 2 and RX data on pin 3. If you use an FTDI serial/USB adaptor then make sure it is 3.3v version and not 5V.

Something you also need to check is /etc/inittab settings. As standard I think that UART1 is /dev/ttyS0. The default baud rate is usually 115200.

David Goadby

Hi,

I checked the version and there is "rev. E".
I simply connect the pins TX,RX to my FTDI converter and it works ok (It's ttyS0 I guess, 115200). You mentioned that UART1 is ttyS0. I use it when booting the board. My questions is about UART0. Is it possible use other serial ports on board (not UART1!)? If yes, then where are the pins?

Assuming the UART0 you are using is the one provided by the group of pins next to the GPIO-1 connector then UART1 is on the opposite side of the PCB next to the UEXT connector. Also, the UART1 connections are available on the UEXT connector on pins 3 (TX) and 4 (RX) with Pin 2 as 0V return.

UART3 is also available on the chip but it uses multi-function pins and I think that they are configured for USB use. Alternate pin configurations are possible but you really need to know what you are doing before making changes.

David Goadby

Is it possible to run Java on this.
What do i download.
http://www.oracle.com/technetwork/java/embedded/downloads/javase/index.html (http://www.oracle.com/technetwork/java/embedded/downloads/javase/index.html)

Also I am having problems with UK keyboard.
lots of wrong chars.

Hi,

I use R19 image and noticed that in high usage network communication very often the board stops working. It's not rebooting just stopped (network restart???).

For example I use wget and trying download the file approx. 3MB during the wget is working the network restarts.

Any one have similar problem?

I checked the /var/log/syslog file. There is an information:

Quote
Oct 23 15:40:46 ubuntu wpa_supplicant[400]: wlan3: CTRL-EVENT-DISCONNECTED bssid=00:8e:f2:b9:87:2e reason=0
Oct 23 15:40:53 ubuntu wpa_supplicant[400]: wlan3: Trying to associate with 00:8e:f2:b9:87:2e (SSID='slycommedia' freq=2437 MHz)
Oct 23 15:40:53 ubuntu wpa_supplicant[400]: wlan3: Association request to the driver failed
Oct 23 15:40:53 ubuntu wpa_supplicant[400]: wlan3: Associated with 00:8e:f2:b9:87:2e
Oct 23 15:40:54 ubuntu wpa_supplicant[400]: wlan3: WPA: Key negotiation completed with 00:8e:f2:b9:87:2e [PTK=CCMP GTK=CCMP]
Oct 23 15:40:54 ubuntu wpa_supplicant[400]: wlan3: CTRL-EVENT-CONNECTED - Connection to 00:8e:f2:b9:87:2e completed (reauth) [id=0 id_str=]

I use two boards A13 with WiFi. The first one has installed R18 and second one R19. There are similar issue when using large amount of data thru network.

[EDIT]
the network is disconnecting without reason. It not depend on the traffic.

[EDIT]
Connected WiFi dongle via USB and the same situation after several minutes.

[UPDATE]
P.S. My board is used to send and receive rest full services every second.

I think that I need implement watchdog as a separate electronic component :(

jwischka,
Is the ipv6 support enabled/disabled in R18 or in R19?

Quote from: SlyCom on October 25, 2013, 01:48:03 AM
jwischka,
Is the ipv6 support enabled/disabled in R18 or in R19?

Sorry for the late reply. I believe it should be enabled. I don't run IPv6 on my local router anymore, so I'm not entirely sure. Do you need me to check?

Yes, please. Is it possible to disable the ipv6 support? or it must be done in kernel?

Quote from: SlyCom on October 25, 2013, 05:40:37 PM
Yes, please. Is it possible to disable the ipv6 support? or it must be done in kernel?

No - you don't have to pull an IPv6 address, even if it is enabled in the kernel. You can configure things to be IPv4, even if IPv6 is supported in kernel.

Why is it important that it be disabled?

If your read my previous post. I have a problem with drop connection, slow connection in R18 and R19 I thought that it can be problem with ipv6. I still don't know if this is a problem but I'm trying to find it out.

Another question. Is it possible to enable board watchdog? Sometimes my board just hangs and I can reset it manually. To enable the watchdog do I need recompile the kernel or maybe there is simplest solution?

Quote from: SlyCom on October 25, 2013, 06:09:55 PM
If your read my previous post. I have a problem with drop connection, slow connection in R18 and R19 I thought that it can be problem with ipv6. I still don't know if this is a problem but I'm trying to find it out.

Another question. Is it possible to enable board watchdog? Sometimes my board just hangs and I can reset it manually. To enable the watchdog do I need recompile the kernel or maybe there is simplest solution?

I suspect your connection issues are related to hardware, not software. In every screenshot I've seen on an A13 wifi (including the ones Olimex have posted), there are a lot of dropped packets. In general, the board has a slow wifi connection, and can experience drops. I've disabled powersave in the kernel, which seems to help, and my board maintains a fairly solid connection. At any rate, I don't think IPv6 is going to be the issue.

There is a hardware watchdog that can be enabled in the kernel, but any time I've tried to enable it, it basically results in the board restarting every 5 minutes.

Are you trying to run a process when your board is hanging? If so, it sounds like you might have some sort of problem with the applications you are running on the board. What happens if you have a "clean" image, that isn't running anything?

I installed fresh R19 beta image set the wlan, tried use wget method and the speed is decreasing and stops after few seconds. Very often the board hangs (or drop the connection). Currently I uploaded fresh R18 image and my java application works (over 3 hours now). I have no idea where is the problem.
In my application I use connection to get information from server every 2 seconds but it's not huge amount of data (approx. 50 bytes). I don't think that it's related to my application as the fresh image also drop the connection (sometimes). I have two boards running close to each other maybe there is some kind of interference? I need more investigation but good information is that only I have similar problem.
So, the power safe mode is off in image R18?

P.S.
Some time ago you released image R19 beta. There is support for ttl converters based on Cx2103. How I can add support for Cx2103 in R18?

Quote from: SlyCom on October 25, 2013, 06:41:44 PM
I installed fresh R19 beta image set the wlan, tried use wget method and the speed is decreasing and stops after few seconds. Very often the board hangs (or drop the connection). Currently I uploaded fresh R18 image and my java application works (over 3 hours now). I have no idea where is the problem.
In my application I use connection to get information from server every 2 seconds but it's not huge amount of data (approx. 50 bytes). I don't think that it's related to my application as the fresh image also drop the connection (sometimes). I have two boards running close to each other maybe there is some kind of interference? I need more investigation but good information is that only I have similar problem.
So, the power safe mode is off in image R18?

P.S.
Some time ago you released image R19 beta. There is support for ttl converters based on Cx2103. How I can add support for Cx2103 in R18?

You'd need to recompile the kernel.

The main issue between R18 and R19 is that the sunxi repository "updated" a lot of code, which simultaneously broke a lot of things - I probably need to try to make a brand new image from scratch based on the most up-to-date sources, but that's going to take more time than I have at the moment. Also I'm trying to do something similar for the A20, though we'll see how well that works :)

Can you confirm that you don't have any memory or handle leaks in your application? Does the board lose connectivity if absolutely nothing is running? (just ping google.com)

Could you provide me some information how I can recompile the kernel for your image? And where I can get the kernel source?

For example in R19 I just uploaded new image, configured the wlan and no other software was running but the connection is dropped every few seconds when I try to use for example wget method. yestreday I had the same issue with R18 but today I uploaded the new image and my application is working without any issues but I need more tests.

Quote from: SlyCom on October 25, 2013, 06:57:40 PM
Could you provide me some information how I can recompile the kernel for your image? And where I can get the kernel source?

For example in R19 I just uploaded new image, configured the wlan and no other software was running but the connection is dropped every few seconds when I try to use for example wget method. yestreday I had the same issue with R18 but today I uploaded the new image and my application is working without any issues but I need more tests.

There are several tutorials in the forum about compiling a kernel, which include instructions on how to get the source.

What happens when you load R19 and simply try to ping google?

There is no results if using ping. But when using the wget it starts download and after few seconds slows down and stops. Often it then drops the connection.

Quote from: SlyCom on October 25, 2013, 07:04:16 PM
There is no results if using ping. But when using the wget it starts download and after few seconds slows down and stops. Often it then drops the connection.

If there is no result using ping, then you don't have a connection at all, and wget should not work.

Have you tried connecting to a different router?

Not yet but today I'm going to do this.

I installed the watchdog using (apt-get install watchdog). Do you think that it will work in case of hangs the board?

P.S.
Can I use your configuration of R18 to recompile the kernel?

Quote from: SlyCom on October 25, 2013, 07:14:32 PM
Not yet but today I'm going to do this.

I installed the watchdog using (apt-get install watchdog). Do you think that it will work in case of hangs the board?

Depends on why the board is hanging. I would work on trying to figure that out. A watchdog should be a last resort, not a planned operation.

Yes I know don't plan to use watchdog but the board will work far away.

P.S.
Can I use your configuration of R18 to recompile the kernel?

Quote from: SlyCom on October 25, 2013, 07:19:32 PM
Yes I know don't plan to use watchdog but the board will work far away.

P.S.
Can I use your configuration of R18 to recompile the kernel?

Yes, though there may still be things broken in the most up-to-date source.

I will play little a bit with the configuration. I will let you know if I found something. The configuration file can be found in kernel folder. Is that right?

Quote from: SlyCom on October 25, 2013, 07:22:53 PM
I will play little a bit with the configuration. I will let you know if I found something. The configuration file can be found in kernel folder. Is that right?

Yes.

Thank you.

Hi,
Touchscreen working in R19 image?

Quote from: Fluxxy on October 25, 2013, 08:13:03 PM
Hi,
Touchscreen working in R19 image?

I'm not sure - I don't have one to test out.

If you can, please try and report back.

Hi,

Strange behavior but the board is working now except sometimes hangs. I couldn't find a source of the problem. It looks like the board and wifi module is very sensitive (I used another device close to the board) maybe this is an issue.

Jwischka,
I know this is not a thread about compilation but I appreciate if you can help me. I'm trying to compile the uboot, kernel and use Linaro ubuntu rootfs. I did similar steps as in
https://www.olimex.com/wiki/Build_Bootable_SD_Card_with_Debian#Write_Uboot_and_Kernel_you_build (https://www.olimex.com/wiki/Build_Bootable_SD_Card_with_Debian#Write_Uboot_and_Kernel_you_build)
but with no success.

Inserted ther prepared card and have following message:

U-Boot SPL 2013.10-rc2-08400-g8a4621c (Oct 28 2013 - 05:14:01)
Board: A13-OLinuXino
DRAM: 512 MiB
CPU: 1008000000Hz, AXI/AHB/APB: 3/2/2
spl: not an uImage at 1600
spl: not an uImage at 80
### ERROR ### Please RESET the board ###

I know this is going to sound basic, but have you checked your SD card to make sure that the uImage file is on the first partition of the file? It looks like (as a first step) the board isn't seeing it.

Quote from: SlyCom on October 29, 2013, 12:22:20 PM
Hi,

Strange behavior but the board is working now except sometimes hangs. I couldn't find a source of the problem. It looks like the board and wifi module is very sensitive (I used another device close to the board) maybe this is an issue.

Jwischka,
I know this is not a thread about compilation but I appreciate if you can help me. I'm trying to compile the uboot, kernel and use Linaro ubuntu rootfs. I did similar steps as in
https://www.olimex.com/wiki/Build_Bootable_SD_Card_with_Debian#Write_Uboot_and_Kernel_you_build (https://www.olimex.com/wiki/Build_Bootable_SD_Card_with_Debian#Write_Uboot_and_Kernel_you_build)
but with no success.

Inserted ther prepared card and have following message:

U-Boot SPL 2013.10-rc2-08400-g8a4621c (Oct 28 2013 - 05:14:01)
Board: A13-OLinuXino
DRAM: 512 MiB
CPU: 1008000000Hz, AXI/AHB/APB: 3/2/2
spl: not an uImage at 1600
spl: not an uImage at 80
### ERROR ### Please RESET the board ###

Yes, the uImage is on the first partition. Now is better (issue with sunxi-spl.bin) but still need write the rootfs to the card.