Re: QtMoko v22
On Thursday 27 May 2010 11:50:23 Yann SLADEK wrote: Hi Radek, after a test on Gand's FR with v22, I decided to reflash from v21 to v22 But, every times the FR goes into sleep, I got a WSOD This is strange, i havent seen wsod for long time except when working with xrandr or with qvga. I tried to see if I didn't mess with files when flashing but everything was ok Then, I realized that, at start, QI says the kernel is qtmoko-v21 (creation date : 06-04-2010). For the record, I always boot by pressing the AUX+power buttons Is there any problems with qtmoko-v22.bin ? Maybe, it's exactly the same version as qtmoko-v21 ? Sometimes i use the kernel from previous release. I do when there are no kernel improvements/changes and i dont have time or i am not brave enough to recompile just to increase version. I wanted to release v24 week ago. I was hunting for 5 days why GPRS does not work. I have finally figured it out: vodafone changed their APN! So next release is hopefully on the way... Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi Radek, after a test on Gand's FR with v22, I decided to reflash from v21 to v22 But, every times the FR goes into sleep, I got a WSOD I tried to see if I didn't mess with files when flashing but everything was ok Then, I realized that, at start, QI says the kernel is qtmoko-v21 (creation date : 06-04-2010). For the record, I always boot by pressing the AUX+power buttons Is there any problems with qtmoko-v22.bin ? Maybe, it's exactly the same version as qtmoko-v21 ? Regards, Yann Hi, i have uploaded new stable QtMoko debian images [1]. QtMoko is stable distribution based on Qtopia (or QtExtended). For more info check our homepage. This new version is mainly bug fix release. Changes from previous version include: - fixed power management to correctly detect cable - fixed GPS power up script (NeronGPS now works just fine again) - workaround for the GSM (restart not needed now) - added omhacks for controlling Neo hardware from command line - new application Neocontrol for controlling Neo sysfs and other tasks - implemented power management for locked/unlocked screen - build fixes and code improvements (Bartlomiej Zimon) As for the fixes: first two are quite clear. The problem with GSM not registering can be partially solved by putting sleep after we power the modem on. GSM registering now failed for me only first time after flashing. Then i tried many times and haven't hit the problem again so let's hope it's fixed now. Anyway 2.6.32 does not have this problem - so 2.6.32 is long term solution. Omhacks [4] is great for various low level openmoko setting. It's from Debian and credits to it's authors. I will slowly migrate everything to use it's API. Neocontrol is work in progress application. Most of the functionality is not implemented yet. The switching to QVGA is only experimental and it will most probably not work for you. Watching the sysfs is on the other hand working and is useful. I have also implemented power management for locked/unlocked screen. If you set intervals to 0 in When locked mode - it will not be used and Qtmoko will use suspend intervals based on power source. This feature is still quite experiment and i think sometimes it does now properly. As for upgrading to new version: the kernel, rootfs and all except /opt is the same. You can just grab qtmoko-update-v22.tar.gz, stop Qtopia and unpack it. ssh r...@192.168.0.202 cd / wget http://sourceforge.net/projects/qtmoko/files/qtmoko-update - v22.tar.gz/download /etc/init.d/qpe.sh stop tar xzvpf qtmoko-update-v22.tar.gz Btw please check MD5SUMs: 7bd189c4ab152f92517b82d748f22a8a qtmoko-debian-v22.jffs2 79012074ea1c2c6498b7524ab2e70fb7 qtmoko-debian-v22.tar.gz 96e7c199a9001b6b1d0a4f8a45effc47 qtmoko-update-v22.tar.gz 8ea022a3eaaba471ad6809b8923934e0 uImage-v22.bin Plans for next version remain the same except that i have now GTA01 device (thanks to Pavel Machek), so if all goes right i will release also gta01 images. Hope you'll like this new version Cheers Radek [1] http://sourceforge.net/projects/qtmoko/files/ [2] http://qtmoko.org/ [3] http://activationrecord.net/radekp/qtmoko/ [4] http://git.debian.org/?p=pkg-fso/omhacks.git ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
I have the same 'problem' here with the v21 instead of v22 ... but mine is rock stable ^^ -- Gand' On Thu, May 27, 2010 at 11:50 AM, Yann SLADEK yann.sla...@free.fr wrote: Hi Radek, after a test on Gand's FR with v22, I decided to reflash from v21 to v22 But, every times the FR goes into sleep, I got a WSOD I tried to see if I didn't mess with files when flashing but everything was ok Then, I realized that, at start, QI says the kernel is qtmoko-v21 (creation date : 06-04-2010). For the record, I always boot by pressing the AUX+power buttons Is there any problems with qtmoko-v22.bin ? Maybe, it's exactly the same version as qtmoko-v21 ? Regards, Yann Hi, i have uploaded new stable QtMoko debian images [1]. QtMoko is stable distribution based on Qtopia (or QtExtended). For more info check our homepage. This new version is mainly bug fix release. Changes from previous version include: - fixed power management to correctly detect cable - fixed GPS power up script (NeronGPS now works just fine again) - workaround for the GSM (restart not needed now) - added omhacks for controlling Neo hardware from command line - new application Neocontrol for controlling Neo sysfs and other tasks - implemented power management for locked/unlocked screen - build fixes and code improvements (Bartlomiej Zimon) As for the fixes: first two are quite clear. The problem with GSM not registering can be partially solved by putting sleep after we power the modem on. GSM registering now failed for me only first time after flashing. Then i tried many times and haven't hit the problem again so let's hope it's fixed now. Anyway 2.6.32 does not have this problem - so 2.6.32 is long term solution. Omhacks [4] is great for various low level openmoko setting. It's from Debian and credits to it's authors. I will slowly migrate everything to use it's API. Neocontrol is work in progress application. Most of the functionality is not implemented yet. The switching to QVGA is only experimental and it will most probably not work for you. Watching the sysfs is on the other hand working and is useful. I have also implemented power management for locked/unlocked screen. If you set intervals to 0 in When locked mode - it will not be used and Qtmoko will use suspend intervals based on power source. This feature is still quite experiment and i think sometimes it does now properly. As for upgrading to new version: the kernel, rootfs and all except /opt is the same. You can just grab qtmoko-update-v22.tar.gz, stop Qtopia and unpack it. ssh r...@192.168.0.202 cd / wget http://sourceforge.net/projects/qtmoko/files/qtmoko-update- v22.tar.gz/download /etc/init.d/qpe.sh stop tar xzvpf qtmoko-update-v22.tar.gz Btw please check MD5SUMs: 7bd189c4ab152f92517b82d748f22a8a qtmoko-debian-v22.jffs2 79012074ea1c2c6498b7524ab2e70fb7 qtmoko-debian-v22.tar.gz 96e7c199a9001b6b1d0a4f8a45effc47 qtmoko-update-v22.tar.gz 8ea022a3eaaba471ad6809b8923934e0 uImage-v22.bin Plans for next version remain the same except that i have now GTA01 device (thanks to Pavel Machek), so if all goes right i will release also gta01 images. Hope you'll like this new version Cheers Radek [1] http://sourceforge.net/projects/qtmoko/files/ [2] http://qtmoko.org/ [3] http://activationrecord.net/radekp/qtmoko/ [4] http://git.debian.org/?p=pkg-fso/omhacks.git __ _ Openmoko community mailing list commun...@lists.openmoko.orghttp://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi, On Sat, May 22, 2010 at 6:57 PM, Al Johnson openm...@mazikeen.demon.co.ukwrote: On Saturday 22 May 2010, Linus Gasser wrote: Le 14.05.10 23:28, Torfinn Ingolfsen a écrit : Hi, The new version appears to work fine here (just a quick test so far). BTW, my Ubuntu laptop (Xubuntu 9.10) exhibits a strange symptom: each time I connect my FR, the network interface name increases: eg. it started out with 'eth9' (what happened with usb0?), the next time I connected the FR the interface was 'eth10', then 'eth11', 'eth12' and so on. What gives? Same here on MacOSX10.5, if somebody has a solution for that problem, it would be great... A randomly assigned MAC address would probably do that. Check whether it's different each time you get a different ethX. Nice theaory, but the MAC address is consistent all the time: [11317.984463] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [11318.092042] udev: renamed network interface eth1 to eth9 [11328.293019] eth9: no IPv6 routers present [18287.197308] eth9: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device [18444.245380] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [18444.303889] udev: renamed network interface eth1 to eth10 [18454.737021] eth10: no IPv6 routers present [18845.694406] eth10: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device [19086.795400] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [19086.831189] udev: renamed network interface eth1 to eth11 [19097.128157] eth11: no IPv6 routers present [19551.746854] eth11: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device [19688.926889] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [19688.955066] udev: renamed network interface eth1 to eth12 [19698.976064] eth12: no IPv6 routers present [20157.117309] eth12: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device [20461.402924] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [20461.444637] udev: renamed network interface eth1 to eth13 [20471.792055] eth13: no IPv6 routers present [24397.170339] eth13: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device [24417.319627] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [24417.373764] udev: renamed network interface eth1 to eth14 [24425.441322] eth14: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device [24427.792533] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [24427.822051] udev: renamed network interface eth1 to eth15 [24438.289131] eth15: no IPv6 routers present [61417.122323] eth15: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device [69011.133835] eth1: register 'cdc_ether' at usb-:00:1d.1-1, CDC Ethernet Device, 00:1f:11:01:0d:39 [69011.617916] udev: renamed network interface eth1 to eth16 [69022.348130] eth16: no IPv6 routers present [69263.593382] eth16: unregister 'cdc_ether' usb-:00:1d.1-1, CDC Ethernet Device Any other suggestions? FWIW, the last time I used my FR with QtMoko (v14) it didn't behave like this, and I don't see why it should start now. I think its more likely that Ubuntu is the sinner here. -- Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi, On Sat, May 22, 2010 at 9:24 PM, Dr. H. Nikolaus Schaller h...@goldelico.com wrote: It is a probolem of the OM kernel that provides a random Mac address Now, why would the OM kernel do a stupid thing like that? More to the point: why would it change MAC address every time I connect my FR to my machine, even if I have not rebooted the FR in the meantime? As you can see from my other post in this thread, the MAC address is consistent, som something else is going on here. -- Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
if it happens with both an ubuntu box and macos x, it's most likely caused by the fr. nevertheless, on ubuntu you may try to set udevadm control --log-priority=debug and watch the log for more information. another idea that occurs to me: is the interface always correctly shut down? what does ifconfig -a say? is there something in /etc/network/interfaces ? ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
i forgot: what does /etc/udev/rules.d/70-persistent-net.rules look like? ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi, On Sun, May 23, 2010 at 6:51 PM, arne anka openm...@ginguppin.de wrote: if it happens with both an ubuntu box and macos x, it's most likely caused by the fr. nevertheless, on ubuntu you may try to set udevadm control --log-priority=debug and watch the log for more information. I might try this later. another idea that occurs to me: is the interface always correctly shut down? I think so (but I just disconnect the FR when I'm done, like I always have done it). what does ifconfig -a It says that there are network interfaces eth0, wlan0, wmaster0, pan0 and lo0 (wired, wireless and bluetooth, pus local) and now eth18 (usb to my FR). is there something in /etc/network/interfaces I think it is okay: ti...@kg-home:~$ more /etc/network/interfaces auto lo iface lo inet loopback -- Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi,
On Sun, May 23, 2010 at 6:53 PM, arne anka openm...@ginguppin.de wrote:
i forgot:
what does
/etc/udev/rules.d/70-persistent-net.rules
look like?
Now, this is an interesting file. Here are a few lines:
# This file maintains persistent names for network interfaces.
# See udev(7) for syntax.
#
# Entries are automatically added by the 75-persistent-net-generator.rules
# file; however you are also free to add your own entries.
# PCI device 0x14e4:0x169d (tg3)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:16:36:54:a9:ae, ATTR{type}==1, KERN
EL==eth*, NAME=eth0
# PCI device 0x8086:0x4222 (iwl3945)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:13:02:3e:d4:ce, ATTR{type}==1, KERN
EL==wlan*, NAME=wlan0
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth9
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth10
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth11
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth12
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth13
# USB device 0x0525:0xa4a1 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth14
# USB device 0x0525:0xa4a2 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth15
# USB device 0x0525:0xa4a2 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth16
I had a look at /lib/udev/rules.d/75-persistent-net-generator.rules but
didn't find anything funky.
--
Regards,
Torfinn Ingolfsen
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Torfinn Ingolfsen tin...@gmail.com writes:
On Sun, May 23, 2010 at 6:53 PM, arne anka openm...@ginguppin.de wrote:
what does
/etc/udev/rules.d/70-persistent-net.rules
look like?
Now, this is an interesting file. Here are a few lines:
...
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth9
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth10
# USB device 0x1457:0x5122 (usb)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:1f:11:01:0d:39, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth11
...
Apparently your host system fails to recognise the device by its MAC
and keeps assigning new interfaces. Debug/fix your network rules
generator script.
--
Be free, use free (http://www.gnu.org/philosophy/free-sw.html) software!
mailto:fercer...@gmail.com
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Tue, Apr 27, 2010 at 7:33 PM, Radek Polak pson...@seznam.cz wrote: Hi, i have uploaded new stable QtMoko debian images [1]. .. - workaround for the GSM (restart not needed now) .. As for the fixes: first two are quite clear. The problem with GSM not registering can be partially solved by putting sleep after we power the modem on. GSM registering now failed for me only first time after flashing. Then i tried many times and haven't hit the problem again so let's hope it's fixed now. Anyway 2.6.32 does not have this problem - so 2.6.32 is long term solution. I have reflashed to v22 and found that it cycled on searching for network. A qt restart did not help at first but then seemed to register as my carrier was shown. I have since restarted QT and the device a few times and sometimes it cycles searching and other times it just seems to register. Nonetheless, I cannot send sms, simply get given error error sending sms and the sms is placed into drafts. I also cannot establish a call... it goes through the motions after I dial a number, but pretty quickly shows disconnected and quits the dialler. Happy to be guinea pig with versions or any other debugging as I have not been using the fr as primary phone for months due to similar problems. If there are any logs or debugging or versions you would like me to try, please respond with step by step instructions cheers Denis BTW, thanks for all your work on this distro ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Le 14.05.10 23:28, Torfinn Ingolfsen a écrit : Hi, The new version appears to work fine here (just a quick test so far). BTW, my Ubuntu laptop (Xubuntu 9.10) exhibits a strange symptom: each time I connect my FR, the network interface name increases: eg. it started out with 'eth9' (what happened with usb0?), the next time I connected the FR the interface was 'eth10', then 'eth11', 'eth12' and so on. What gives? Same here on MacOSX10.5, if somebody has a solution for that problem, it would be great... Linus ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Saturday 22 May 2010, Linus Gasser wrote: Le 14.05.10 23:28, Torfinn Ingolfsen a écrit : Hi, The new version appears to work fine here (just a quick test so far). BTW, my Ubuntu laptop (Xubuntu 9.10) exhibits a strange symptom: each time I connect my FR, the network interface name increases: eg. it started out with 'eth9' (what happened with usb0?), the next time I connected the FR the interface was 'eth10', then 'eth11', 'eth12' and so on. What gives? Same here on MacOSX10.5, if somebody has a solution for that problem, it would be great... A randomly assigned MAC address would probably do that. Check whether it's different each time you get a different ethX. If it is you can fix it in a number of ways. SHR sets it in /etc/modprobe.d/g_ether.conf but you should also be able to do it in /etc/network/interfaces if g_ether is compiled in. The thread below should help. http://kerneltrap.org/mailarchive/openmoko-community/2008/7/20/2584124 ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Linus Gasser l...@markas-al-nour.org writes: The new version appears to work fine here (just a quick test so far). BTW, my Ubuntu laptop (Xubuntu 9.10) exhibits a strange symptom: each time I connect my FR, the network interface name increases: eg. it started out with 'eth9' (what happened with usb0?), the next time I connected the FR the interface was 'eth10', then 'eth11', 'eth12' and so on. What gives? Same here on MacOSX10.5, if somebody has a solution for that problem, it would be great... Check the mac address of FR. Usually it is supposed that your udev rules will assign ethX device to FR by MAC and every time you plug it in, it will rename the created interface. Take a look at your persistent-net udev rules file. To get persistent MAC on FR, you need to fetch it from the factory ext2 partition and use as a kernel (or g_ether) parameter. Recent Qi does that automatically. -- Be free, use free (http://www.gnu.org/philosophy/free-sw.html) software! mailto:fercer...@gmail.com ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Am 22.05.2010 um 18:29 schrieb Linus Gasser: Le 14.05.10 23:28, Torfinn Ingolfsen a écrit : Hi, The new version appears to work fine here (just a quick test so far). BTW, my Ubuntu laptop (Xubuntu 9.10) exhibits a strange symptom: each time I connect my FR, the network interface name increases: eg. it started out with 'eth9' (what happened with usb0?), the next time I connected the FR the interface was 'eth10', then 'eth11', 'eth12' and so on. What gives? Same here on MacOSX10.5, if somebody has a solution for that problem, it would be great... It is a probolem of the OM kernel that provides a random Mac address and OS X takes this to distinguish known from unknown devices. A solution is described on: http://wiki.openmoko.org/wiki/Mac_OS_X#AJZaurusUSB_driver Linus ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
no - andy-tracking does not have kms stuff. All 2.6.32 kernels were until now also without kms - they are built from om-gta02-2.6.32. At that time KMS branch had some problem, which were fixed. Last time i tried KMS branch it worked fine. So maybe next 2.6.32 based release will be with kms. yes, there was this white screen issue. Thomas fixed it and I also tested it, and no problem since then... -- rgrds, mobi phil being mobile, but including technology http://mobiphil.com ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Saturday 15 May 2010 17:47:47 mobi phil wrote: but still do not know if it is based on andy tracking. All 2.6.29 kernels in QtMoko are based on andy-tracking. They are built with nodebug config and wifi patch from here: whatever kernel it is.. does it have the Thomas Whites Kms stuff merged? On Sat, May 15, 2010 at 1:54 PM, Torfinn Ingolfsen tin...@gmail.com wrote: On Sat, May 15, 2010 at 1:31 PM, mobi phil m...@mobiphil.com wrote: the uImage-v22.bin is 2.6.29 isn't it? Yes, QtMoko V22 has this kernel: neo:~# uname -a Linux neo 2.6.29-rc3-v21 #10 Tue Apr 6 22:54:31 CEST 2010 armv4tl GNU/Linux was not following for a while the list... could anybody tell me where I could find a 2.6.32.bin? QtMoko V23 (testing) has this kernel: neo:~# uname -a Linux neo 2.6.32v20 #16 Tue May 4 21:11:16 CEST 2010 armv4tl GNU/Linux HTH -- Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Saturday 15 May 2010 17:47:47 mobi phil wrote: but still do not know if it is based on andy tracking. All 2.6.29 kernels in QtMoko are based on andy-tracking. They are built with nodebug config and wifi patch from here: http://github.com/radekp/qtmoko/tree/master/devices/neo/linux_kernel/ whatever kernel it is.. does it have the Thomas Whites Kms stuff merged? no - andy-tracking does not have kms stuff. All 2.6.32 kernels were until now also without kms - they are built from om-gta02-2.6.32. At that time KMS branch had some problem, which were fixed. Last time i tried KMS branch it worked fine. So maybe next 2.6.32 based release will be with kms. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
la. den 15. 05. 2010 klokka 00.28 (+0200) skreiv Torfinn Ingolfsen: Hi, The new version appears to work fine here (just a quick test so far). BTW, my Ubuntu laptop (Xubuntu 9.10) exhibits a strange symptom: each time I connect my FR, the network interface name increases: eg. it started out with 'eth9' (what happened with usb0?), the next time I connected the FR the interface was 'eth10', then 'eth11', 'eth12' and so on. What gives? This is happening to me too. Ubuntu 10.04 and 9.10 Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
the uImage-v22.bin is 2.6.29 isn't it? was not following for a while the list... could anybody tell me where I could find a 2.6.32.bin? On wiki it is stated that andy tracking would be used for 2.6.32, but not sure if that is from the 2.6.29 times.. rgrds, mobi phil being mobile, but including technology http://mobiphil.com ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Sat, May 15, 2010 at 1:31 PM, mobi phil m...@mobiphil.com wrote: the uImage-v22.bin is 2.6.29 isn't it? Yes, QtMoko V22 has this kernel: neo:~# uname -a Linux neo 2.6.29-rc3-v21 #10 Tue Apr 6 22:54:31 CEST 2010 armv4tl GNU/Linux was not following for a while the list... could anybody tell me where I could find a 2.6.32.bin? QtMoko V23 (testing) has this kernel: neo:~# uname -a Linux neo 2.6.32v20 #16 Tue May 4 21:11:16 CEST 2010 armv4tl GNU/Linux HTH -- Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
thanks but still do not know if it is based on andy tracking. whatever kernel it is.. does it have the Thomas Whites Kms stuff merged? On Sat, May 15, 2010 at 1:54 PM, Torfinn Ingolfsen tin...@gmail.com wrote: On Sat, May 15, 2010 at 1:31 PM, mobi phil m...@mobiphil.com wrote: the uImage-v22.bin is 2.6.29 isn't it? Yes, QtMoko V22 has this kernel: neo:~# uname -a Linux neo 2.6.29-rc3-v21 #10 Tue Apr 6 22:54:31 CEST 2010 armv4tl GNU/Linux was not following for a while the list... could anybody tell me where I could find a 2.6.32.bin? QtMoko V23 (testing) has this kernel: neo:~# uname -a Linux neo 2.6.32v20 #16 Tue May 4 21:11:16 CEST 2010 armv4tl GNU/Linux HTH -- Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community -- rgrds, mobi phil being mobile, but including technology http://mobiphil.com ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Friday 14 May 2010 00:33:48 Gand' wrote: Do you mean it's significantly more unstable than the v20 ? 'Cause i've been using it for a while, and there was nothing critical in it ... It should be better then v20, but both have the problem that it eats battery in suspend - much more then stable ones. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Oh ok, so i'll give it a try :) Improved reactivity worth the shorter battery :D -- Gand' On Fri, May 14, 2010 at 10:24 AM, Radek Polak pson...@seznam.cz wrote: On Friday 14 May 2010 00:33:48 Gand' wrote: Do you mean it's significantly more unstable than the v20 ? 'Cause i've been using it for a while, and there was nothing critical in it ... It should be better then v20, but both have the problem that it eats battery in suspend - much more then stable ones. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Friday 14 May 2010 11:52:09 Gand' wrote: Oh ok, so i'll give it a try :) Improved reactivity worth the shorter battery :D I think v22 is most reactive - after debug stuff was removed i subjectively think it's the fastest what we had. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi, The new version appears to work fine here (just a quick test so far). BTW, my Ubuntu laptop (Xubuntu 9.10) exhibits a strange symptom: each time I connect my FR, the network interface name increases: eg. it started out with 'eth9' (what happened with usb0?), the next time I connected the FR the interface was 'eth10', then 'eth11', 'eth12' and so on. What gives? -- Regards, Torfinn Ingolfsen ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Is this possible to update from v20 to v23 without flashing ? -- Gand' On Wed, May 12, 2010 at 7:43 AM, Radek Polak pson...@seznam.cz wrote: On Wednesday 12 May 2010 02:13:13 Denis Johnson wrote: I re-flashed with latest v23 files Just a warning note: v23 is still not good enough for daily use. It's based on 2.6.32 and it has problems with power management. For daily use i'd recommend v22. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Gand' wrote: Is this possible to update from v20 to v23 without flashing ? You can always update manually if you replace /opt and check the differences in rootfs. But v23 is experimental based on 2.6.32. I'd recommend it only for testing not for serious using now. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Do you mean it's significantly more unstable than the v20 ? 'Cause i've been using it for a while, and there was nothing critical in it ... -- Gand' On Thu, May 13, 2010 at 10:01 PM, Radek Polak pson...@seznam.cz wrote: Gand' wrote: Is this possible to update from v20 to v23 without flashing ? You can always update manually if you replace /opt and check the differences in rootfs. But v23 is experimental based on 2.6.32. I'd recommend it only for testing not for serious using now. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi all, I started using v22 as my default phone a couple days ago, and for the most part it has been very stable and worked great. However, I've noticed that the contact name resolution (for incoming calls) doesn't work. If I initiate the call it works fine, but whenever I have something incoming, I have no idea who it is from. Strangely enough, it works fine on incoming sms though. Is there something similar to SHR's phoneutils.conf file that I should be editing? Thanks, -Dan Staley On Wed, May 12, 2010 at 1:43 AM, Radek Polak pson...@seznam.cz wrote: On Wednesday 12 May 2010 02:13:13 Denis Johnson wrote: I re-flashed with latest v23 files Just a warning note: v23 is still not good enough for daily use. It's based on 2.6.32 and it has problems with power management. For daily use i'd recommend v22. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Thanks for all your hard work and efforts. I re-flashed with latest v23 files and my first call, although I could hear the other party pretty well (although I would like to increase the volume), the other party complained of bad distortion so I guess it is time to find the correct mix of state files. Mine is a GTA02 A5 buzz fixed. So if someone has good working state files then please share. cheers Denis On Tue, Apr 27, 2010 at 7:33 PM, Radek Polak pson...@seznam.cz wrote: Hi, i have uploaded new stable QtMoko debian images [1]. ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Wednesday 12 May 2010 02:13:13 Denis Johnson wrote: I re-flashed with latest v23 files Just a warning note: v23 is still not good enough for daily use. It's based on 2.6.32 and it has problems with power management. For daily use i'd recommend v22. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Feedback for sms. There are no problems in receving messages (short sms or long sms). The problem in sending long sms still remains. If the sms is 3 pages long then an error appears and the recipient receives only 2 parts of 3 of the sms. If the sms is 2 pages long there is no error and the recipient receives the sms correctly. But in both cases, after sending the long sms, it seems that qtmoko has a problem in receving messages (like the missed messages old problem) and it is necessary to restart qtmoko. Also another problem that persists from the old versions is that while I'm writing a message, if I click in a certain point of the text to edit a part of it, it often happens that the text from that point until the end is highlighted and to make the text not highlighted I have to move through it using the arrows of the docked keyboard. -- View this message in context: http://openmoko-public-mailinglists.1958.n2.nabble.com/QtMoko-v22-tp4967705p4990027.html Sent from the Openmoko Community mailing list archive at Nabble.com. ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi Please take a piggy bank: - googlecontactsync still misses phones, but now found only if they are the same type (2 mobile, 2 workers, etc.) Mail in this case does not pass. - Thank you for your attention. -- View this message in context: http://openmoko-public-mailinglists.1958.n2.nabble.com/QtMoko-v22-tp4967705p4981214.html Sent from the Openmoko Community mailing list archive at Nabble.com. ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
My bad.. I follow instructions on french wiki, it tells to modify DeepSleep in Active=always (as shr in fact) I modified it in the right way (Active=yes) and I got Apr 28 09:10:37 neo Qtopia: AtChat : T : AT%SLEEP=4 So it sounds good, let's wait for an experiment in the real life now :) Regards, Yann On Tuesday 27 April 2010 17:00:52 Yann SLADEK wrote: The only weird thing I got is a poor battery life (24h max) althought I have a buzzfixed GTA02 (the battery goes up to 80h with SHR for example). I did the modification of the right file but I have the same behaviour than with Android who didn't implement the software fix. I have tweaked my config: neo:~# cat /opt/qtmoko/etc/default/Trolltech/Modem.conf [DeepSleep] Active=yes [Multiplexing] Active=yes Now i run NeoControl and suspend my phone. After waking up the battery current in NeoContol displays: 11437 which is ~12mA. Since the battery has capacity 1200mAh it should last 1200/12=100 hours in this state. I had full battery yesterday afternoon. At home i used wifi for like 40minutes otherwise it was suspended. Now after 12 hours the battery indicator in Qtopia displays 4/5 full. So i think deep sleep is working. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Wednesday 28 April 2010 09:16:09 Yann SLADEK wrote: My bad.. I follow instructions on french wiki, it tells to modify DeepSleep in Active=always (as shr in fact) I modified it in the right way (Active=yes) and I got Apr 28 09:10:37 neo Qtopia: AtChat : T : AT%SLEEP=4 So it sounds good, let's wait for an experiment in the real life now :) It's the same if you have always or yes. The code checks for never and activates deep sleep otherwise. http://github.com/radekp/qtmoko/blob/master/devices/neo/src/plugins/phonevendors/neo/vendor_neo.cpp if (deepsleep == never) chat(AT%SLEEP=2); else chat(AT%SLEEP=4); Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
QtMoko v22
Hi, i have uploaded new stable QtMoko debian images [1]. QtMoko is stable distribution based on Qtopia (or QtExtended). For more info check our homepage. This new version is mainly bug fix release. Changes from previous version include: - fixed power management to correctly detect cable - fixed GPS power up script (NeronGPS now works just fine again) - workaround for the GSM (restart not needed now) - added omhacks for controlling Neo hardware from command line - new application Neocontrol for controlling Neo sysfs and other tasks - implemented power management for locked/unlocked screen - build fixes and code improvements (Bartlomiej Zimon) As for the fixes: first two are quite clear. The problem with GSM not registering can be partially solved by putting sleep after we power the modem on. GSM registering now failed for me only first time after flashing. Then i tried many times and haven't hit the problem again so let's hope it's fixed now. Anyway 2.6.32 does not have this problem - so 2.6.32 is long term solution. Omhacks [4] is great for various low level openmoko setting. It's from Debian and credits to it's authors. I will slowly migrate everything to use it's API. Neocontrol is work in progress application. Most of the functionality is not implemented yet. The switching to QVGA is only experimental and it will most probably not work for you. Watching the sysfs is on the other hand working and is useful. I have also implemented power management for locked/unlocked screen. If you set intervals to 0 in When locked mode - it will not be used and Qtmoko will use suspend intervals based on power source. This feature is still quite experiment and i think sometimes it does now properly. As for upgrading to new version: the kernel, rootfs and all except /opt is the same. You can just grab qtmoko-update-v22.tar.gz, stop Qtopia and unpack it. ssh r...@192.168.0.202 cd / wget http://sourceforge.net/projects/qtmoko/files/qtmoko-update- v22.tar.gz/download /etc/init.d/qpe.sh stop tar xzvpf qtmoko-update-v22.tar.gz Btw please check MD5SUMs: 7bd189c4ab152f92517b82d748f22a8a qtmoko-debian-v22.jffs2 79012074ea1c2c6498b7524ab2e70fb7 qtmoko-debian-v22.tar.gz 96e7c199a9001b6b1d0a4f8a45effc47 qtmoko-update-v22.tar.gz 8ea022a3eaaba471ad6809b8923934e0 uImage-v22.bin Plans for next version remain the same except that i have now GTA01 device (thanks to Pavel Machek), so if all goes right i will release also gta01 images. Hope you'll like this new version Cheers Radek [1] http://sourceforge.net/projects/qtmoko/files/ [2] http://qtmoko.org/ [3] http://activationrecord.net/radekp/qtmoko/ [4] http://git.debian.org/?p=pkg-fso/omhacks.git ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Hi Radek, Thanks a lot for this version Have you planned testing release with 2.6.32 kernel ? Regards, Yann - Mail Original - De: Radek Polak pson...@seznam.cz À: List for Openmoko community discussion community@lists.openmoko.org Envoyé: Mardi 27 Avril 2010 11:33:37 GMT +01:00 Amsterdam / Berlin / Berne / Rome / Stockholm / Vienne Objet: QtMoko v22 Hi, i have uploaded new stable QtMoko debian images [1]. QtMoko is stable distribution based on Qtopia (or QtExtended). For more info check our homepage. This new version is mainly bug fix release. Changes from previous version include: - fixed power management to correctly detect cable - fixed GPS power up script (NeronGPS now works just fine again) - workaround for the GSM (restart not needed now) - added omhacks for controlling Neo hardware from command line - new application Neocontrol for controlling Neo sysfs and other tasks - implemented power management for locked/unlocked screen - build fixes and code improvements (Bartlomiej Zimon) As for the fixes: first two are quite clear. The problem with GSM not registering can be partially solved by putting sleep after we power the modem on. GSM registering now failed for me only first time after flashing. Then i tried many times and haven't hit the problem again so let's hope it's fixed now. Anyway 2.6.32 does not have this problem - so 2.6.32 is long term solution. Omhacks [4] is great for various low level openmoko setting. It's from Debian and credits to it's authors. I will slowly migrate everything to use it's API. Neocontrol is work in progress application. Most of the functionality is not implemented yet. The switching to QVGA is only experimental and it will most probably not work for you. Watching the sysfs is on the other hand working and is useful. I have also implemented power management for locked/unlocked screen. If you set intervals to 0 in When locked mode - it will not be used and Qtmoko will use suspend intervals based on power source. This feature is still quite experiment and i think sometimes it does now properly. As for upgrading to new version: the kernel, rootfs and all except /opt is the same. You can just grab qtmoko-update-v22.tar.gz, stop Qtopia and unpack it. ssh r...@192.168.0.202 cd / wget http://sourceforge.net/projects/qtmoko/files/qtmoko-update- v22.tar.gz/download /etc/init.d/qpe.sh stop tar xzvpf qtmoko-update-v22.tar.gz Btw please check MD5SUMs: 7bd189c4ab152f92517b82d748f22a8a qtmoko-debian-v22.jffs2 79012074ea1c2c6498b7524ab2e70fb7 qtmoko-debian-v22.tar.gz 96e7c199a9001b6b1d0a4f8a45effc47 qtmoko-update-v22.tar.gz 8ea022a3eaaba471ad6809b8923934e0 uImage-v22.bin Plans for next version remain the same except that i have now GTA01 device (thanks to Pavel Machek), so if all goes right i will release also gta01 images. Hope you'll like this new version Cheers Radek [1] http://sourceforge.net/projects/qtmoko/files/ [2] http://qtmoko.org/ [3] http://activationrecord.net/radekp/qtmoko/ [4] http://git.debian.org/?p=pkg-fso/omhacks.git ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Tuesday 27 April 2010 16:19:20 Yann SLADEK wrote: Thanks a lot for this version Have you planned testing release with 2.6.32 kernel ? I have not worked on 2.6.32 since last release. I want some stable version which has no annoying bugs (hopefully it's v22?). But then i would like to finish 2.6.32 up and make it stable. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
What are the annoying bugs you're talking about ? I'm on v21 (or v20 I don't remember but 2.6.32 kernel) for a month now and I didn't see any regressions (touchscreen is smoother, speed has been improved, etc..) The only weird thing I got is a poor battery life (24h max) althought I have a buzzfixed GTA02 (the battery goes up to 80h with SHR for example). I did the modification of the right file but I have the same behaviour than with Android who didn't implement the software fix. But it's not kernel related Regards, Yann - Mail Original - De: Radek Polak pson...@seznam.cz À: List for Openmoko community discussion community@lists.openmoko.org Envoyé: Mardi 27 Avril 2010 16:44:39 GMT +01:00 Amsterdam / Berlin / Berne / Rome / Stockholm / Vienne Objet: Re: QtMoko v22 On Tuesday 27 April 2010 16:19:20 Yann SLADEK wrote: Thanks a lot for this version Have you planned testing release with 2.6.32 kernel ? I have not worked on 2.6.32 since last release. I want some stable version which has no annoying bugs (hopefully it's v22?). But then i would like to finish 2.6.32 up and make it stable. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Tuesday 27 April 2010 17:00:52 Yann SLADEK wrote: What are the annoying bugs you're talking about ? I'm on v21 (or v20 I don't remember but 2.6.32 kernel) for a month now and I didn't see any regressions (touchscreen is smoother, speed has been improved, etc..) Oki, nice to hear it. The only weird thing I got is a poor battery life (24h max) althought I have a buzzfixed GTA02 (the battery goes up to 80h with SHR for example). I did the modification of the right file but I have the same behaviour than with Android who didn't implement the software fix. I have #1024 fix since Saturday, but havent tried it yet. But in the Modem AT logs it seems ok. This is my log line after modifying DeepSleep to yes: Apr 27 18:08:53 neo Qtopia: AtChat : T : AT%SLEEP=4 if it was not active you should get AT%SLEEP=2. But maybe the command has to be issued also after resume. I'll have to check if it's correctly implemented... Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
I'm also using kernel 2.6.32, and there is no problem with it (if you except the poor battery life ...) -- Gand' On Tue, Apr 27, 2010 at 6:15 PM, Radek Polak pson...@seznam.cz wrote: On Tuesday 27 April 2010 17:00:52 Yann SLADEK wrote: What are the annoying bugs you're talking about ? I'm on v21 (or v20 I don't remember but 2.6.32 kernel) for a month now and I didn't see any regressions (touchscreen is smoother, speed has been improved, etc..) Oki, nice to hear it. The only weird thing I got is a poor battery life (24h max) althought I have a buzzfixed GTA02 (the battery goes up to 80h with SHR for example). I did the modification of the right file but I have the same behaviour than with Android who didn't implement the software fix. I have #1024 fix since Saturday, but havent tried it yet. But in the Modem AT logs it seems ok. This is my log line after modifying DeepSleep to yes: Apr 27 18:08:53 neo Qtopia: AtChat : T : AT%SLEEP=4 if it was not active you should get AT%SLEEP=2. But maybe the command has to be issued also after resume. I'll have to check if it's correctly implemented... Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
I have also implemented power management for locked/unlocked screen. If you set intervals to 0 in When locked mode - it will not be used and Qtmoko will use suspend intervals based on power source. This feature is still quite experiment and i think sometimes it does now properly. I upgraded via the method you suggested last night. I'm happy to say it does seem less buggy initially, I'll have to play with it a bit more to really test out the new changes. There's one minor thing I noticed with the new power management function. When on battery, the power manager defaults to the 'When locked' option. When I then plug in the phone it switches to 'On Battery' instead of 'Plugged In'. Not sure if we still have to restart QTMoko after plugging in to switch to the 'Plugged in' mode by default, or if this is a bug with the new setup. Again, this is very preliminary usage, so I'd still like to try it out for a while with different circumstances and see how things go. -- Steve Woodruff swo...@ubuntu.com https://wiki.ubuntu.com/swoody ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Looks like need to upgrade to v22 - i'm on shr now because of no rcv smses on QtMoko ;/ Thx Radek! Dnia 27 kwietnia 2010 19:52 Gand' lt;gand...@viroenforce.comgt; napisał(a): I'm also using kernel 2.6.32, and there is no problem with it (if you except the poor battery life ...) -- Gand' On Tue, Apr 27, 2010 at 6:15 PM, Radek Polak lt;pson...@seznam.czgt; wrote: On Tuesday 27 April 2010 17:00:52 Yann SLADEK wrote: gt; What are the annoying bugs you're talking about ? gt; I'm on v21 (or v20 I don't remember but 2.6.32 kernel) for a month now and gt; I didn't see any regressions (touchscreen is smoother, speed has been gt; improved, etc..) Oki, nice to hear it. gt; The only weird thing I got is a poor battery life (24h gt; max) althought I have a buzzfixed GTA02 (the battery goes up to 80h with gt; SHR for example). I did the modification of the right file but I have the gt; same behaviour than with Android who didn't implement the software fix. I have #1024 fix since Saturday, but havent tried it yet. But in the Modem AT logs it seems ok. This is my log line after modifying DeepSleep to yes: Apr 27 18:08:53 neo Qtopia: AtChat : T : AT%SLEEP=4 if it was not active you should get AT%SLEEP=2. But maybe the command has to be issued also after resume. I'll have to check if it's correctly implemented... Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community___Openmoko community mailing listcommun...@lists.openmoko.orghttp://lists.openmoko.org/mailman/listinfo/community___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
Ok one bug - have Message Icon on Home screen with info SMS Full - looks like smses are not deleted from sim. System Info shows 25 smses used/ 0 free. How to easy delete all of them just from sim? When locked kicks ass. Best regards.Bartlomiej Zimon.ps. i have still problems with qt/qtmoko compilation on x86_64 and need it to develop on qtmoko :( Dnia 27 kwietnia 2010 21:23 Bartłomiej Zimoń lt;uz...@o2.plgt; napisał(a): Looks like need to upgrade to v22 - i'm on shr now because of no rcv smses on QtMoko ;/ Thx Radek! Dnia 27 kwietnia 2010 19:52 Gand' lt;gand...@viroenforce.comgt; napisał(a): I'm also using kernel 2.6.32, and there is no problem with it (if you except the poor battery life ...) -- Gand' On Tue, Apr 27, 2010 at 6:15 PM, Radek Polak lt;pson...@seznam.czgt; wrote: On Tuesday 27 April 2010 17:00:52 Yann SLADEK wrote: gt; What are the annoying bugs you're talking about ? gt; I'm on v21 (or v20 I don't remember but 2.6.32 kernel) for a month now and gt; I didn't see any regressions (touchscreen is smoother, speed has been gt; improved, etc..) Oki, nice to hear it. gt; The only weird thing I got is a poor battery life (24h gt; max) althought I have a buzzfixed GTA02 (the battery goes up to 80h with gt; SHR for example). I did the modification of the right file but I have the gt; same behaviour than with Android who didn't implement the software fix. I have #1024 fix since Saturday, but havent tried it yet. But in the Modem AT logs it seems ok. This is my log line after modifying DeepSleep to yes: Apr 27 18:08:53 neo Qtopia: AtChat : T : AT%SLEEP=4 if it was not active you should get AT%SLEEP=2. But maybe the command has to be issued also after resume. I'll have to check if it's correctly implemented... Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community___Openmoko community mailing listcommun...@lists.openmoko.orghttp://lists.openmoko.org/mailman/listinfo/community___Openmoko community mailing listcommun...@lists.openmoko.orghttp://lists.openmoko.org/mailman/listinfo/community___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
Re: QtMoko v22
On Tuesday 27 April 2010 17:00:52 Yann SLADEK wrote: The only weird thing I got is a poor battery life (24h max) althought I have a buzzfixed GTA02 (the battery goes up to 80h with SHR for example). I did the modification of the right file but I have the same behaviour than with Android who didn't implement the software fix. I have tweaked my config: neo:~# cat /opt/qtmoko/etc/default/Trolltech/Modem.conf [DeepSleep] Active=yes [Multiplexing] Active=yes Now i run NeoControl and suspend my phone. After waking up the battery current in NeoContol displays: 11437 which is ~12mA. Since the battery has capacity 1200mAh it should last 1200/12=100 hours in this state. I had full battery yesterday afternoon. At home i used wifi for like 40minutes otherwise it was suspended. Now after 12 hours the battery indicator in Qtopia displays 4/5 full. So i think deep sleep is working. Regards Radek ___ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community
