Htc Universal

[Sam Liddicott]

Charging time:
I may be posting all my excitement to the wrong place - if so, sorry:

When running of usb, /proc/apm gives screwy values, these were about 1 second apart

Code: [Select]

htcuni:~# cat /proc/apm
1.13 1.2 0x02 0x01 0x03 0x03 77% -1044 min
htcuni:~# cat /proc/apm
1.13 1.2 0x02 0x01 0x03 0x03 77% -498 min
htcuni:~# cat /proc/apm
1.13 1.2 0x02 0x01 0x03 0x03 77% -588 min
htcuni:~# cat /proc/apm
1.13 1.2 0x02 0x01 0x03 0x03 77% -10792 min
htcuni:~# cat /proc/apm
1.13 1.2 0x02 0x01 0x03 0x03 77% -490 min
htcuni:~# cat /proc/apm
1.13 1.2 0x02 0x01 0x03 0x03 77% -2023 min

[Sam Liddicott]

I also have to restart /etc/rcS.d/S99htcuni-keypress after suspend, or backlight button doesn't work.

And a tip for others, to set the timezone:
dpkg-reconfigure tzdata

Sam

[datathief]

I also have to restart /etc/rcS.d/S99htcuni-keypress after suspend, or backlight button doesn't work.

Already fixed.  Add my repository to /etc/apt/sources:

Code: [Select]

deb http://debian.neilandtheresa.co.uk/ titchy/and install titchy-button-daemon (note the name change).  Install titchy-base to get all the updated packages.

[datathief]

I'm loving this, thanks guys; some questions though:

Can I specify a different timeout for htcuni-autosuspend when running on USB?

Yep.  Install titchy-autosuspend-daemon (note the name change) from my repository.

[datathief]

I rang my phone and it went straight on to voicemail...

Maybe the phone hadn't finished registering with the network.

Code: [Select]

htcunid network

And finally, datathief and ShiroiKuma; how did you setup your build system to cross-compile debs?
I'm running ubuntu with some from-source debian packages; I'll generally just get a deb-src, but please could you tell me your spell for cross-building debs?

I haven't cross-compiled any debs.  All my work is in the form of shell scripts and python...

[Sam Liddicott]

I also have to restart /etc/rcS.d/S99htcuni-keypress after suspend, or backlight button doesn't work.

Already fixed.  Add my repository to /etc/apt/sources:

Code: [Select]

deb http://debian.neilandtheresa.co.uk/ titchy/and install titchy-button-daemon (note the name change).  Install titchy-base to get all the updated packages.

Do they appropriately obsolete the older htc* packages or will I have to first remove those manually, do you think?

Sam

[datathief]

Do they appropriately obsolete the older htc* packages or will I have to first remove those manually, do you think?

They do correctly specify the "Replaces:" tag, so the old packages will be uninstalled.  They'll also migrate your configuration files from the old packages.  It's probably worth checking that everything works before you rely on it as your primary phone, but I haven't had any problems so far.

I have found some new bugs unrelated to the repackaging, though:  alarms don't work until the phone has been suspended and resumed at least once, and there's no error message if you try to send SMS with no credit...  I need to build a bug tracker.

[Sam Liddicott]

Do they appropriately obsolete the older htc* packages or will I have to first remove those manually, do you think?

They do correctly specify the "Replaces:" tag, so the old packages will be uninstalled.  They'll also migrate your configuration files from the old packages.  It's probably worth checking that everything works before you rely on it as your primary phone, but I haven't had any problems so far.

I have found some new bugs unrelated to the repackaging, though:  alarms don't work until the phone has been suspended and resumed at least once, and there's no error message if you try to send SMS with no credit...  I need to build a bug tracker.

Aye...

apm (as has been noted) gives bogus values and it's hard to tell if the battery charges via usb. Something screwy is going on because just now it wouldn't even charge via usb when running WM6.1 with my laptop running winXP, until I did a hard-reset-forget-everything on the universal.

Both devices are back to linux and I'll watch and see if the battery is charging or discharging; apm says:

Code: [Select]

mojo:~# for g in 1 2 3 4 5; do printf "`date` "; apm; sleep 1; done
Thu May  8 12:37:09 BST 2008 On-line, battery status high: 96% (6d+22:29:00)
Thu May  8 12:37:10 BST 2008 On-line, battery status high: 96% (20:26:00)
Thu May  8 12:37:11 BST 2008 On-line, battery status high: 96% (2d+0:33:00)
Thu May  8 12:37:12 BST 2008 On-line, battery status high: 96% (4d+1:07:00)
Thu May  8 12:37:13 BST 2008 On-line, battery status high: 96% (1d+12:59:00)
However:

Code: [Select]

mojo:/# cat /sys/class/power_supply/ds2760-battery.0/status
Charging
While I had time over christmas I was improving lazarus packaging in anticipation of this, so I think my first compile will be the complete lazarus IDE...

Sam

[Sam Liddicott]

I also have to restart /etc/rcS.d/S99htcuni-keypress after suspend, or backlight button doesn't work.

Already fixed.  Add my repository to /etc/apt/sources:

Code: [Select]

deb http://debian.neilandtheresa.co.uk/ titchy/and install titchy-button-daemon (note the name change).  Install titchy-base to get all the updated packages.

titchy-base did not install titchy-phone-daemon

also; dmesg output relating to charging:

Code: [Select]

[  328.720000] generic ttyUSB0: generic converter now disconnected from ttyUSB0
[  328.730000] usbserial_generic 1-1:1.0: device disconnected
[  328.740000] generic ttyUSB1: generic converter now disconnected from ttyUSB1
[  328.740000] usbserial_generic 1-1:1.1: device disconnected
[  328.810000] done.
[  329.050000] usb 1-1: new full speed USB device using pxa27x-ohci and address 4
[  329.230000] usb 1-1: configuration #1 chosen from 1 choice
[  329.230000] usbserial_generic 1-1:1.0: generic converter detected
[  329.230000] usb 1-1: generic converter now attached to ttyUSB0
[  329.230000] usbserial_generic 1-1:1.1: generic converter detected
[  329.230000] usb 1-1: generic converter now attached to ttyUSB1
[  330.260000] phone mic enable called. on=1
[  368.120000] udc: USB reset
[  368.360000] udc: USB reset
[  368.470000] udc: USB reset
[  368.670000] usb0: full speed config #1: 100 mA, Ethernet Gadget, using CDC Ethernet Subset
[  369.100000] ds2760-battery ds2760-battery.0: not enough power to charge
[  566.860000] htcuniversal configure phone: 2
[  566.860000] tried to call phone_off, stack will follow:
[  566.860000] htcuniversal configure phone: 3
[  616.110000] htcuniversal powerdown_lcd: called
[  728.670000] htcuniversal powerup_lcd: called
[  729.190000] ds2760-battery ds2760-battery.0: not enough power to charge
[ 1336.860000] htcuniversal powerdown_lcd: called
[ 1936.970000] htcuniversal powerdown_lcd: called
[ 2285.050000] htcuniversal configure phone: 0
[ 2285.050000] htcuniversal configure phone: 1
[ 2314.870000] htcuniversal powerup_lcd: called
[ 2349.480000] ds2760-battery ds2760-battery.0: not enough power to charge
"not enough power to charge" is strange cos charge has gone from 96% to 99%

I sent an sms to myself from an ssh login, and after acknowledging receipt via whiptail the screen went blank;
chvt 7
does nothing and chvt 1, hangs, eventually my X session died and restarted.

I don't mean these to sound like whining, I'm just reporting what happens.

[datathief]

titchy-base did not install titchy-phone-daemon

True.  Titchy will eventually run on both Zaurus and Universal  -  the phone stuff doesn't make much sense on the Zaurus, so I left it out...

"not enough power to charge" is strange cos charge has gone from 96% to 99%

I think it's safe to ignore this if it does actually charge.

I sent an sms to myself from an ssh login, and after acknowledging receipt via whiptail the screen went blank;
chvt 7
does nothing and chvt 1, hangs, eventually my X session died and restarted.

This could be a problem with X.  Can you try (with X running):

Code: [Select]

openvt -w -s -- sleep 2X should come back after 2 seconds.

[Sam Liddicott]

titchy-base did not install titchy-phone-daemon

True.  Titchy will eventually run on both Zaurus and Universal  -  the phone stuff doesn't make much sense on the Zaurus, so I left it out...

"not enough power to charge" is strange cos charge has gone from 96% to 99%

I think it's safe to ignore this if it does actually charge.

Aye, although I had to do the hard-reset in order to make it actually charge...

I sent an sms to myself from an ssh login, and after acknowledging receipt via whiptail the screen went blank;
chvt 7
does nothing and chvt 1, hangs, eventually my X session died and restarted.

This could be a problem with X.  Can you try (with X running):

Code: [Select]

openvt -w -s -- sleep 2X should come back after 2 seconds.

It works repeatedly, although it often doesn't switch back from whiptail...

I also note they key-repeat has stopped working.

Am I daft to think I need to install wpasupplicant?

Sam

[datathief]

It works repeatedly, although it often doesn't switch back from whiptail...

Are there any other scripts in /etc/titchy-phone-daemon/event.d that could be doing something to affect it?  Does X come back after:

Code: [Select]

openvt -w -s -- whiptail --msgbox "test" 20 50

I also note they key-repeat has stopped working.

Are KEYBOARD_RATE and KEYBOARD_DELAY set properly in /etc/console-tools/config?  You can use this command to set the key repeat rate and delay back to the defaults:

Code: [Select]

kbdrate -r 30 -d 250

Am I daft to think I need to install wpasupplicant?

I've no idea whether WPA works with this wifi driver.

[Sam Liddicott]

It works repeatedly, although it often doesn't switch back from whiptail...

Are there any other scripts in /etc/titchy-phone-daemon/event.d that could be doing something to affect it?  Does X come back after:

Code: [Select]

openvt -w -s -- whiptail --msgbox "test" 20 50

That killed the x session again the first time I did it, but worked afterwards.

I noticed that "ps -fade" showed a couple of whiptail notifications that were still running, which I also killed.
I'll see what data I can gather if it goes wrong again.

I also note they key-repeat has stopped working.

Are KEYBOARD_RATE and KEYBOARD_DELAY set properly in /etc/console-tools/config?  You can use this command to set the key repeat rate and delay back to the defaults:

Code: [Select]

kbdrate -r 30 -d 250

Code: [Select]

# kbdrate -r 30 -d 250
Cannot open /dev/port: No such file or directory
I don't have a /dev/port; the times were commented out in the config file, I have un-commented and we will see what happens -
however, I think unrelated to that config, the keyboard repeat has come back when the X session died. Weird.

Am I daft to think I need to install wpasupplicant?

I've no idea whether WPA works with this wifi driver.

Thanks for your help on this.

[p3t3r]

wpa_supplicant does not work with 2.6.21-hh20s' acx driver - i tried it, but as it isn't supported officially, it does not work. There is a mac80211 acx driver, which supports WPA, but it isn't available in a kernel for HTC Universal, yet.  

Am I daft to think I need to install wpasupplicant?

Sam

[Sam Liddicott]

It works repeatedly, although it often doesn't switch back from whiptail...

Am I daft to think I need to install wpasupplicant?

I've no idea whether WPA works with this wifi driver.

I will try the max80211 + acx as explained here:
http://acx100.sourceforge.net/wiki/WPA

building my kernel according to:
http://handhelds.org/moin/moin.cgi/UniversalKernel

using the config of my current sumuo supplied kernel