Help - Search - Members - Calendar
Full Version: Htc Universal
OESF Forums > Distros, Development, and Model Specific Forums > Distro Support and Discussion > Debian
Pages: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13
ArchiMark
Knock on wood, things are working pretty well for me right now with Titchy/HTC Uni....

However, just a couple minor things I've noticed....

During boot-up, I see in console the following:
CODE
activating swap [20.550000] unable to find swap-space signature
[color="#FF0000"]failed.[/color]
fsck.....


then Uni rebooted on its own...when it restarts it come back up in WM of course, but I find that it gets 'stuck' on the 'Owner Info' screen, such that it doesn't respond to screen taps and thus I need to do a reset..the next time it comes back up, I can get to the haret file and start up Titchy...

Hmm...maybe I spoke too soon....

Meanwhile, a few minutes ago, I was in the middle of doing an 'apt-get install...' and all of a sudden screen went dark and Uni went into suspend...clicked power button to wake it up and I saw console for a few seconds then the ts_calibration screen came on by itself....after a short time, it disappeared by itself (before I could get to it and calibrate display) and then I saw the 'X' startup window and now I'm back at the xdm login screen....

I logged in and then I saw the those 'ts_calibrate' registration marks faintly in the X background.... I clicked on one and then the iceWM window disappeared and now I'm back in the login console....although I can see a bit of the IceWm taskbar on the right where the time is shown...very odd....

Now, I just clicked somewhere else on the taskbar and the terminal window popped back on the display...but there's just a black background behind it...

So, decided to reboot, now I see some strange sights on the display, first a yellowish/white coloration on the display which now gone and now there's a sort of random series of horizontal dark grey and white bands of varying widths on the display....

OK, just did a reset again....and restarted haret and now back to xdm login again.....

rolleyes.gif
datathief
QUOTE(ArchiMark @ May 19 2008, 03:54 AM) *
Knock on wood, things are working pretty well for me right now with Titchy/HTC Uni....

However, just a couple minor things I've noticed....

During boot-up, I see in console the following:
CODE
activating swap [20.550000] unable to find swap-space signature
[color="#FF0000"]failed.[/color]
fsck.....

This, along with the other symptoms, makes it sound like corrupted swap space. Are you using a swap partition on the SD card? (What is the "swap" line in /etc/fstab?)
ArchiMark
QUOTE(ArchiMark @ May 19 2008, 03:54 AM) *
Knock on wood, things are working pretty well for me right now with Titchy/HTC Uni....

However, just a couple minor things I've noticed....

During boot-up, I see in console the following:
CODE
activating swap [20.550000] unable to find swap-space signature
[color="#FF0000"]failed.[/color]
fsck.....


QUOTE(datathief @ May 19 2008, 12:46 AM) *
This, along with the other symptoms, makes it sound like corrupted swap space. Are you using a swap partition on the SD card? (What is the "swap" line in /etc/fstab?)


Here's that line:
CODE
/dev/mmcblk0p3    swap       swap     pri=1       0    0



UPDATE:
Other weirdness now.....

Having the 'wild cursor' and 'non-responsive cursor' problems in IceWM....

Have tried to do 'ts_calibrate' but terminal says 'command not found'....

Got out of IceWM and back into console...

Now notice other strangeness....

I have 2 'ts.conf' files....one in '/etc/ and the other in '/opt/kdrive/', which one do I want to use?

Meanwhile, checked both files and then did 'ts_calibrate' and get message 'segmentation fault'

Any suggestions as to how to clean up this mess?


Thanks!
ShiroiKuma
QUOTE(ArchiMark @ May 19 2008, 03:36 PM) *
Having the 'wild cursor' and 'non-responsive cursor' problems in IceWM....

Have tried to do 'ts_calibrate' but terminal says 'command not found'....

Got out of IceWM and back into console...

Now notice other strangeness....

I have 2 'ts.conf' files....one in '/etc/ and the other in '/opt/kdrive/', which one do I want to use?

Meanwhile, checked both files and then did 'ts_calibrate' and get message 'segmentation fault'

Any suggestions as to how to clean up this mess?

Thanks!

Your card is failing man. Forget about "cleaning up this mess". Mount it on a PC, backup your documents, and user settings if you still can.

Then throw it in the garbage, buy a new one, and reinstall debian.

That's the reason why with swap on SD you need to do the installed rootfs backups. I do them now twice a week with rdiff-backup, so in case of failure it's easy to reinstall, without starting from scratch.
datathief
QUOTE(ShiroiKuma @ May 19 2008, 05:05 PM) *
Your card is failing man. Forget about "cleaning up this mess". Mount it on a PC, backup your documents, and user settings if you still can.

Then throw it in the garbage, buy a new one, and reinstall debian.

This seems like a serious problem. I don't use swap - partly because of the "swap will kill flash drives" rumours. Could this perhaps be helped by setting /proc/sys/vm/swappiness to 0?
ArchiMark
QUOTE(ArchiMark @ May 19 2008, 03:36 PM) *
Having the 'wild cursor' and 'non-responsive cursor' problems in IceWM....

Have tried to do 'ts_calibrate' but terminal says 'command not found'....

Got out of IceWM and back into console...

Now notice other strangeness....

I have 2 'ts.conf' files....one in '/etc/ and the other in '/opt/kdrive/', which one do I want to use?

Meanwhile, checked both files and then did 'ts_calibrate' and get message 'segmentation fault'

Any suggestions as to how to clean up this mess?

Thanks!


QUOTE(ShiroiKuma @ May 19 2008, 10:05 AM) *
Your card is failing man. Forget about "cleaning up this mess". Mount it on a PC, backup your documents, and user settings if you still can.

Then throw it in the garbage, buy a new one, and reinstall debian.

That's the reason why with swap on SD you need to do the installed rootfs backups. I do them now twice a week with rdiff-backup, so in case of failure it's easy to reinstall, without starting from scratch.



Thanks for the 'good news', ShiroiKuma..... blink.gif

The card you're suggesting I throw away is the 16GB A-DATA SD card that I bought about a month ago..... sad.gif

Think I should reformat card back to one vfat partition and return it to store I got it from????

wink.gif
ShiroiKuma
QUOTE(ArchiMark @ May 19 2008, 09:54 PM) *
The card you're suggesting I throw away is the 16GB A-DATA SD card that I bought about a month ago..... sad.gif

Think I should reformat card back to one vfat partition and return it to store I got it from????

wink.gif

Hahahah... You should've said so. The 16GB A-Data cards are trash. I've gone through four in the past two weeks, all of them failed after one or two days. And that was not because of swap, it was just when copying massive amount of data to them. They are trash...

I've had no other card fail.

Now I've had a Kingston 16GB card for more than a couple of weeks, no probs...

BTW, I think, based on my experiences so far, that really, if you have a swap partition, it'll not kill the whole card and your data ain't gonna get corrupted. If you use just a swap file, it's gonna cause problems.

But a swap partition on a good SD seems to be no prob at all. But still, backups are in order.

Nevertheless, go throw it in the garbage and get a good one... rolleyes.gif
ShiroiKuma
QUOTE(datathief @ May 19 2008, 12:29 AM) *
QUOTE(ShiroiKuma @ May 18 2008, 09:23 PM) *
QUOTE(datathief @ May 15 2008, 02:00 PM) *
Edit: It looks like the bluetooth module is powered up by default. I've updated titchy-hardware-support to turn it off. I've no idea whether it will make much difference, though.

So how do I bring bluetooth up now?

hciconfig gives me nothing, no change when insmoding the two bluetooth module files in /lib/modules

This should power it up again:
CODE
devmem2 0x10000104 halfword 0xC6A0


Doesn't seem to work. hciconfig still gives me nothing.
ArchiMark
QUOTE(ShiroiKuma @ May 19 2008, 01:27 PM) *
QUOTE(ArchiMark @ May 19 2008, 09:54 PM) *
The card you're suggesting I throw away is the 16GB A-DATA SD card that I bought about a month ago..... sad.gif

Think I should reformat card back to one vfat partition and return it to store I got it from????

wink.gif

Hahahah... You should've said so. The 16GB A-Data cards are trash. I've gone through four in the past two weeks, all of them failed after one or two days. And that was not because of swap, it was just when copying massive amount of data to them. They are trash...

I've had no other card fail.

Now I've had a Kingston 16GB card for more than a couple of weeks, no probs...

BTW, I think, based on my experiences so far, that really, if you have a swap partition, it'll not kill the whole card and your data ain't gonna get corrupted. If you use just a swap file, it's gonna cause problems.

But a swap partition on a good SD seems to be no prob at all. But still, backups are in order.

Nevertheless, go throw it in the garbage and get a good one... rolleyes.gif



Thanks again for the input....

Meanwhile, for fun, I deleted partitions on A-DATA SD card and created one new FAT partition and made a vfat filesystem on it, then mounted it on my Zaurus, appears to be OK....

So, wonder if card has really failed or not???

Thanks.
ShiroiKuma
QUOTE(ArchiMark @ May 19 2008, 11:58 PM) *
Meanwhile, for fun, I deleted partitions on A-DATA SD card and created one new FAT partition and made a vfat filesystem on it, then mounted it on my Zaurus, appears to be OK....

So, wonder if card has really failed or not???

Thanks.

Well, I couldn't believe it that it'd develop problems just copying data etc. consistently, thought it was a bad specimen, but have since then determined, that these cards are really very bad. Low price - low quality.
ShiroiKuma
QUOTE(ShiroiKuma @ May 19 2008, 11:45 PM) *
QUOTE(datathief @ May 19 2008, 12:29 AM) *
QUOTE(ShiroiKuma @ May 18 2008, 09:23 PM) *
QUOTE(datathief @ May 15 2008, 02:00 PM) *
Edit: It looks like the bluetooth module is powered up by default. I've updated titchy-hardware-support to turn it off. I've no idea whether it will make much difference, though.

So how do I bring bluetooth up now?

hciconfig gives me nothing, no change when insmoding the two bluetooth module files in /lib/modules

This should power it up again:
CODE
devmem2 0x10000104 halfword 0xC6A0


Doesn't seem to work. hciconfig still gives me nothing.

I've tried many ways now, even rebooted, can't get it up. hconfig empty, ifconfig hci0 up - hci0: ERROR while getting interface flags: No such device, what's wrong?
Sam Liddicott
I've noticed twice now that if I send a text to a non-usb'd suspended universal and do not acknowledge the text, that the battery depletion rate increases.

Test:

fully charge phone., apm shows 99%
suspend for 10 hours
apm shows battery about 90%

fully charge phone
suspend
send text (don't acknowledge)
wait for screen to go black
wait 10 hours
apm shows battery about 81%

I also get
1. backlight button mostly doesn't work after resume
2. sometimes mouse pointer doesn't work after resume
3. sometimes device crashes on reboot
4. sometimes after reboot to wm6.1 (ranju's latest rom) mouse pointer or keyboard don't work until reset.

Sam
Sam Liddicott
I also find that the keyboard is not sensitive enough, I generally don't seem to hold the key down for long enough to register, I guess I need to reduce the debounce settings.

Sam
Sam Liddicott
128MB upgrade

I'm just picking who to do my 128MB upgrade, does anybody here have the upgrade knows someone who can make a recommendation?
(I'm based in the UK)

Sam
datathief
QUOTE(Sam Liddicott @ May 20 2008, 07:58 AM) *
I've noticed twice now that if I send a text to a non-usb'd suspended universal and do not acknowledge the text, that the battery depletion rate increases.

Test:

fully charge phone., apm shows 99%
suspend for 10 hours
apm shows battery about 90%

fully charge phone
suspend
send text (don't acknowledge)
wait for screen to go black
wait 10 hours
apm shows battery about 81%

I also get
1. backlight button mostly doesn't work after resume
2. sometimes mouse pointer doesn't work after resume
3. sometimes device crashes on reboot
4. sometimes after reboot to wm6.1 (ranju's latest rom) mouse pointer or keyboard don't work until reset.

Sam

Do you get messages in syslog like 'Proxy exited with status 1'? That would mean some apmd scripts are not being run, which would cause some of those symptoms.

If so, try running:
CODE
/etc/apm/apmd-proxy resume suspend
and see if you get any error messages that might indicate which apm script is failing. If one script fails, the rest don't get run at all (which is a bug in apmd in my opinion).
ShiroiKuma
QUOTE(Sam Liddicott @ May 20 2008, 09:58 AM) *
3. sometimes device crashes on reboot
4. sometimes after reboot to wm6.1 (ranju's latest rom) mouse pointer or keyboard don't work until reset.

Don't use reset. It always messes up. When you want to reset to Windows, use halt, then soft reset with the stylus.
datathief
QUOTE(ShiroiKuma @ May 19 2008, 10:42 PM) *
QUOTE(ShiroiKuma @ May 19 2008, 11:45 PM) *
QUOTE(datathief @ May 19 2008, 12:29 AM) *
QUOTE(ShiroiKuma @ May 18 2008, 09:23 PM) *
QUOTE(datathief @ May 15 2008, 02:00 PM) *
Edit: It looks like the bluetooth module is powered up by default. I've updated titchy-hardware-support to turn it off. I've no idea whether it will make much difference, though.

So how do I bring bluetooth up now?

hciconfig gives me nothing, no change when insmoding the two bluetooth module files in /lib/modules

This should power it up again:
CODE
devmem2 0x10000104 halfword 0xC6A0


Doesn't seem to work. hciconfig still gives me nothing.

I've tried many ways now, even rebooted, can't get it up. hconfig empty, ifconfig hci0 up - hci0: ERROR while getting interface flags: No such device, what's wrong?

There's a bluetooth reset line (I'm getting my information from http://wiki.xda-developers.com/index.php?p...=UniversalASIC3 if you're wondering). Try:
CODE
devmem2 0x10000104 halfword 0xC620 # bt power on, reset low
devmem2 0x10000104 halfword 0xC6A0 # bt power on, reset high


Edit: You don't need to do the above smile.gif

CODE
echo ttyS1 swave >/etc/bluetooth/uart
/etc/init.d/bluetooth restart
datathief
(Whilst trying to get bluetooth to work, I rebooted. Out of curiosity I've just checked my syslog, and as far as I can tell, the last time I rebooted was May 7. With the amount of surfing, e-mailing, chatting and SSHing I do, I'm shocked. I think it's the most reliable and most useful phone i've ever owned. Still feel a bit silly talking into it, though. smile.gif )
ShiroiKuma
QUOTE(datathief @ May 20 2008, 11:57 AM) *
Edit: You don't need to do the above smile.gif
CODE
echo ttyS1 swave >/etc/bluetooth/uart
/etc/init.d/bluetooth restart

This doesn't seem to be working for me, after restarting bluetooth I get:
CODE
# /etc/init.d/bluetooth restart
Restarting bluetooth: hcid.

then hciconfig still gives empty output.
datathief
QUOTE(ShiroiKuma @ May 20 2008, 06:40 PM) *
QUOTE(datathief @ May 20 2008, 11:57 AM) *
Edit: You don't need to do the above smile.gif
CODE
echo ttyS1 swave >/etc/bluetooth/uart
/etc/init.d/bluetooth restart

This doesn't seem to be working for me, after restarting bluetooth I get:
CODE
# /etc/init.d/bluetooth restart
Restarting bluetooth: hcid.

then hciconfig still gives empty output.

Does the following do anything?
CODE
hciattach /dev/ttyS1 swave

Does the blue LED start flashing?
ShiroiKuma
The hciattach worked, the bluetooth device is up now and the led flashing, I'll experiment with it now. You said the above devmems are unnecessary. What's the best way to bring it down now, battery-wise?
datathief
QUOTE(ShiroiKuma @ May 21 2008, 05:43 AM) *
What's the best way to bring it down now, battery-wise?

The driver seems to power it down properly. Just use:
CODE
/etc/init.d/bluetooth stop

Or if all else fails, kill the 'hciattach' process.
Sam Liddicott
QUOTE(datathief @ May 20 2008, 10:48 AM) *
QUOTE(Sam Liddicott @ May 20 2008, 07:58 AM) *
I've noticed twice now that if I send a text to a non-usb'd suspended universal and do not acknowledge the text, that the battery depletion rate increases.

Test:

fully charge phone., apm shows 99%
suspend for 10 hours
apm shows battery about 90%

fully charge phone
suspend
send text (don't acknowledge)
wait for screen to go black
wait 10 hours
apm shows battery about 81%

I also get
1. backlight button mostly doesn't work after resume
2. sometimes mouse pointer doesn't work after resume
3. sometimes device crashes on reboot
4. sometimes after reboot to wm6.1 (ranju's latest rom) mouse pointer or keyboard don't work until reset.

Sam

Do you get messages in syslog like 'Proxy exited with status 1'? That would mean some apmd scripts are not being run, which would cause some of those symptoms.

If so, try running:
CODE
/etc/apm/apmd-proxy resume suspend
and see if you get any error messages that might indicate which apm script is failing. If one script fails, the rest don't get run at all (which is a bug in apmd in my opinion).


It seems like it is nothing to do with sms or autosuspend, I'm just back to losing 2% an hour again instead of 1% an hour which I was getting after you fixed the bluetooth suspend.

Sam
datathief
QUOTE(Sam Liddicott @ May 22 2008, 06:43 AM) *
It seems like it is nothing to do with sms or autosuspend, I'm just back to losing 2% an hour again instead of 1% an hour which I was getting after you fixed the bluetooth suspend.

Sam

It's difficult for me to test battery life - my phone is in use most of the day, and on charge overnight. I'll try to find an opportunity.

In the meantime:
  • Do you get good battery life on your first suspend after a reboot, and bad battery life on every subsequent suspend? If so, the kernel must be doing something upon resume.
  • When you're getting bad battery life, run:
    CODE
    devmem2 0x10000105 byte
    If the result (the last number on the last line of output) is 0x86, bluetooth is powered down. If the result is 0xC6, it's not, and that could be the culprit.
  • Do you have the htcuniversal_acx and/or acx modules loaded?
Sam Liddicott
QUOTE(datathief @ May 22 2008, 09:52 AM) *
QUOTE(Sam Liddicott @ May 22 2008, 06:43 AM) *
It seems like it is nothing to do with sms or autosuspend, I'm just back to losing 2% an hour again instead of 1% an hour which I was getting after you fixed the bluetooth suspend.

Sam

It's difficult for me to test battery life - my phone is in use most of the day, and on charge overnight. I'll try to find an opportunity.

In the meantime:
  • Do you get good battery life on your first suspend after a reboot, and bad battery life on every subsequent suspend? If so, the kernel must be doing something upon resume.
  • When you're getting bad battery life, run:
    CODE
    devmem2 0x10000105 byte
    If the result (the last number on the last line of output) is 0x86, bluetooth is powered down. If the result is 0xC6, it's not, and that could be the culprit.
  • Do you have the htcuniversal_acx and/or acx modules loaded?


bluetooth is off according to that and no acx modules are loaded.

I'll do more experiments tonight, and still losing 2% per hour.

Sam
datathief
QUOTE(Sam Liddicott @ May 22 2008, 03:05 PM) *
I'll do more experiments tonight, and still losing 2% per hour.

Sam

I've rebooted and tested the battery for a few hours, and haven't found anything that makes any significant difference (except 'modprobe acx', which does terrible, terrible things to the battery life).

I'm currently losing almost exactly 0.5% per hour on standby, with the 3800mAh battery. I guess that would equate to about 1.2% per hour with the standard 1620mAh battery. I'll re-test after a day of use, and see if anything has changed.
webdork
Is it possible to run a usb wifi card via an externally powered USB hub with the Universal?
datathief
QUOTE(webdork @ May 23 2008, 02:17 AM) *
Is it possible to run a usb wifi card via an externally powered USB hub with the Universal?

I believe it's not supported by the hardware (an unconnected pin or something). And even if it was supported, the g_ether driver (which does networking over a USB cable) can't be unloaded to load the USB host drivers, as it's not a module. And no-one here has managed to build a fully working kernel from the source yet.

In other words, no. sad.gif
ShiroiKuma
datathief, regarding APM. Does the Uni turn suspend when a critically low battery level is experienced? I can't check now, since I just ran out of battery juice, so plugged a PC cable in, and unwitfully didn't wait a couple of minutes to test if it'd come back on, just soft-reset it. So am leaving it in Windows Mobile for a while now to charge a little, as the PC cable gives low charge...

Or is the low-bat suspend not taken care of currently?
datathief
QUOTE(ShiroiKuma @ May 23 2008, 10:44 AM) *
datathief, regarding APM. Does the Uni turn suspend when a critically low battery level is experienced? I can't check now, since I just ran out of battery juice, so plugged a PC cable in, and unwitfully didn't wait a couple of minutes to test if it'd come back on, just soft-reset it. So am leaving it in Windows Mobile for a while now to charge a little, as the PC cable gives low charge...

Or is the low-bat suspend not taken care of currently?

I honestly don't know. I haven't written anything to do that, but I don't know if the kernel handles it somehow (like the Zaurus kernel does, IIRC).
ShiroiKuma
Using bluetooth headpones via A2DP
Anyone using bluetooth to listen to songs or movies on the Uni? I've googled and tried various guides like http://gentoo-wiki.com/HOWTO_use_a_bluetooth_headset and http://wiki.debian.org/BluetoothAlsa etc. but I can't get it working.

Does anyone have a working setup?

Got it going using http://fosswire.com/2008/01/11/a2dp-stereo-linux/

However, the audio is really choppy and lags, I've noticed in syslog that hcid sets the bitpool at 32 which would be too low, so probably some tweaking needed. But at least a working start.

If anyone has it working, please share.
datathief
QUOTE(ShiroiKuma @ May 24 2008, 08:18 PM) *
If anyone has it working, please share.

I don't know if this will make any difference... I've found out that the bluetooth chipset is made by Texas Instruments, so the appropriate hciattach command is actually:
CODE
hciattach /dev/ttyS1 texas
datathief
I've uploaded a new UI package, titchy-phoneui-shiny. It's still at the prototype stage - it can only do basic call and SMS notifications so far, and there's no touchscreen or portrait-mode support yet - but I think there's a lot of potential to develop it further.

It works nicely in console mode - I haven't tested it under X yet...

Edit: Seems to work fine under X.
ShiroiKuma
I'll check it out soon, prelim questions before I get to it, what's the relationship to your whiptail frontend, is it on top of that, does it supersede it or replace it, or work alongside?
datathief
QUOTE(ShiroiKuma @ May 26 2008, 05:38 PM) *
I'll check it out soon, prelim questions before I get to it, what's the relationship to your whiptail frontend, is it on top of that, does it supersede it or replace it, or work alongside?

They should coexist fine. It's not anywhere near a finished product yet, so it won't replace the whiptail scripts. Keep your expectations low. smile.gif

Screenshots are up on the wiki now. Also, I've just uploaded a new version to fix a bug (new SMS would cause it to crash).
ShiroiKuma
It seems UTF-8 is not implemented in it. I've replaced the Vera.ttf link to an extended TTF font, containing Japanese characters, however they are not displayed properly, though the font contains them, so a coding issue maybe?
datathief
QUOTE(ShiroiKuma @ May 26 2008, 08:34 PM) *
It seems UTF-8 is not implemented in it. I've replaced the Vera.ttf link to an extended TTF font, containing Japanese characters, however they are not displayed properly, though the font contains them, so a coding issue maybe?

Should be fixed now (you'll still need to replace the font though).
webdork
Where can I find the kernel headers for Shiro's Debian Sid FS? (hh20) I'm trying to build the mac80211 subsystem for the Uni and need to find them!

Thanx in advance!
ShiroiKuma
QUOTE(datathief @ May 25 2008, 12:15 AM) *
I don't know if this will make any difference... I've found out that the bluetooth chipset is made by Texas Instruments, so the appropriate hciattach command is actually:
CODE
hciattach /dev/ttyS1 texas

Oh no, it's a nightmare...

The lag is not caused by swave vs texas, as I've tried on a desktop and once bluetooth audio was setup using the same python script, it had the same lag when playing from mplayer, while totem has none, apparently there is super high cpu usage using mplayer and a2dp, so some bug...

However, not keeping in line with the "if it's not messed up, don't fix it" golden rule, I aptiture removed bluetooth-alsa, as it shouldn't be needed and should be antiquated, rebooted. Now the bluetooth does start up, but the audio will not run.

Now, no matter which combo I try with having the bluetooth-alsa installed etc. it doesn't work, I've noticed:
Can't load plugin: /lib/bluetooth/plugins/libaudio.so in the syslog, the indicated solution to this on the web is recompile bluez-utils.

But it was working, before "I fixed it" so how can this be?

Ahhh....

EDIT 1: Hmmm, some progress, have replaced texas back with swave in /etc/bluetooth/uart, after restart, after stopping bluetooth-alsa, and starting bluetooth, this is fine and audio is initialized... Now just can't get passkey-agent to register the pin to the headphones, which went fine yesterday... Some further twiddlin' probably necessary.
ShiroiKuma
QUOTE(webdork @ May 27 2008, 09:41 AM) *
Where can I find the kernel headers for Shiro's Debian Sid FS? (hh20) I'm trying to build the mac80211 subsystem for the Uni and need to find them!

http://www.oesf.org/forum/index.php?showto...st&p=176793
ShiroiKuma
QUOTE(datathief @ May 27 2008, 01:42 AM) *
Should be fixed now (you'll still need to replace the font though).

It works, displays fine now.

It seems, however, that the mute via loudness button while a call is ringing is not working with it.

EDIT 1: My bad, I guess it's working, let me test again...
ShiroiKuma
Your titchy-phoneui-shiny in the current update is really an elegant interface, very nice and impressive.

There is a nasty bug in the scripts, which just inflicted me, so will describe the situation how it happened to me.

I was making a test call to the Uni from another phone, right around the time when I accepted it by pressing the accept button, don't know if it was a fraction of a second before accept or after, I received an unexpected SMS. This resulted in the phone vibrating, without stopping. I could hit the OK button to dismiss the SMS info, but then the Accept Call? info screen was on top and couldn't be dismissed, the vibration couldn't be killed, I couldn't switch into any other VT, and couldn't kill it with Ctrl-C, had to reboot.
datathief
QUOTE(ShiroiKuma @ May 27 2008, 07:59 AM) *
QUOTE(datathief @ May 27 2008, 01:42 AM) *
Should be fixed now (you'll still need to replace the font though).

It works, displays fine now.

It seems, however, that the mute via loudness button while a call is ringing is not working with it.

EDIT 1: My bad, I guess it's working, let me test again...

This is due to some sort of bug in titchy-button-daemon, where the front panel buttons (backlight, volume, etc.) sometimes stop working after resume. I think I've fixed it (it works for me) in the newest version.
datathief
QUOTE(ShiroiKuma @ May 28 2008, 01:16 PM) *
Your titchy-phoneui-shiny in the current update is really an elegant interface, very nice and impressive.

There is a nasty bug in the scripts, which just inflicted me, so will describe the situation how it happened to me.

I was making a test call to the Uni from another phone, right around the time when I accepted it by pressing the accept button, don't know if it was a fraction of a second before accept or after, I received an unexpected SMS. This resulted in the phone vibrating, without stopping. I could hit the OK button to dismiss the SMS info, but then the Accept Call? info screen was on top and couldn't be dismissed, the vibration couldn't be killed, I couldn't switch into any other VT, and couldn't kill it with Ctrl-C, had to reboot.

I'm not sure how this could happen. I'll look into it.

I will add the ability to close a screen with Ctrl-C, though, because that's caught me a couple of times while developing it. sad.gif
hakkinen
How can I install Debian on my Uni without a Linux PC??? Anyone can tell me??? I created 3 partitions on my SD card: 1st FAT, 2nd Ext2 and 3rd Swamp space. I unpacked the tar file (Debian) to the 2nd partition by using RAMDISK.cab on my Uni (I saw 4 file in directory dev/mmc/blk0: part1, part2, part5, part6, and I mounted vfat with part1, ext2 with part5). But when I used haret.exe, it warned Failed to load file \zImage. How can I boot in Linux??? I have to edit default.txt and zImage.txt???
Thanks in advance!!!
datathief
QUOTE(hakkinen @ May 29 2008, 05:39 PM) *
How can I install Debian on my Uni without a Linux PC??? Anyone can tell me??? I created 3 partitions on my SD card: 1st FAT, 2nd Ext2 and 3rd Swamp space. I unpacked the tar file (Debian) to the 2nd partition by using RAMDISK.cab on my Uni (I saw 4 file in directory dev/mmc/blk0: part1, part2, part5, part6, and I mounted vfat with part1, ext2 with part5). But when I used haret.exe, it warned Failed to load file \zImage. How can I boot in Linux??? I have to edit default.txt and zImage.txt???
Thanks in advance!!!

The file called zImage is the Linux kernel. If yours is named zImage.txt and you can't rename it to zImage, then try editing default.txt - change the first line to:
CODE
set kernel "zImage.txt"
hakkinen
Thanks very much. Now I can boot in Linux, but it warned Cannot open root device "mmc/blk0/part5" or unknown-block(0,0)...
Please append a correct "root=" boot option... Unable to mount root fs on unknown-block(0,0). How can I boot in Linux? I mounted ext2 with part5 and unpacked the tar file here
datathief
QUOTE(hakkinen @ May 29 2008, 07:12 PM) *
Thanks very much. Now I can boot in Linux, but it warned Cannot open root device "mmc/blk0/part5" or unknown-block(0,0)...
Please append a correct "root=" boot option... Unable to mount root fs on unknown-block(0,0). How can I boot in Linux? I mounted ext2 with part5 and unpacked the tar file here

This is probably due to the kernel using a different naming convention for the device - /dev/mmcblk0p* instead of /dev/mmc/blk0/part*. Try changing the "root=..." option in default.txt to root=/dev/mmcblk0p5
Tom_A
QUOTE(datathief @ Apr 4 2008, 01:03 AM) *
I've been using an HTC Universal for a month now, and I've all but forgotten about my poor Zaurus. ohmy.gif



Wow, I _love_ it. I think a distro for the Universal would be great. I just need to get WLAN working ( with WPA ) to be able to use it. But installing was really easy. Very well done. Could you point me in the direction of some howtos on how to get the UI, the phoneUI and WLAN working?

Thanks a lot!
aku-aku
QUOTE(Tom_A @ May 30 2008, 10:36 AM) *
Wow, I _love_ it. I think a distro for the Universal would be great. I just need to get WLAN working ( with WPA ) to be able to use it. But installing was really easy. Very well done. Could you point me in the direction of some howtos on how to get the UI, the phoneUI and WLAN working?


I've just started using Titchy Mobile too. There are some packages listed on the Titchy Mobile homepage that aren't installed by default. These can be installed by running "apt-get install package-name" (having first set up the USB network per the instructions on the Titchy Mobile homepage).

To get WPA working, you'll need to apt-get install wpasupplicant, then follow one of the many howtos (I used this one as a rough guide, though it's not targeted at Debian). I've got problems with mine at the moment, which look like they're due to the fact that I hide my SSID - it seems the scan setting in the config file may need to be commented out or removed. You'll also find (I think) that you need to set -Dmadwifi rather than -Dndiswrapper in the wpa_supplicant command line.

I was testing out my wpa_supplicant config by running "wpa_supplicant -dd -Dmadwifi -iwlan0 -c/etc/wpa_supplicant.conf" - that'll spew debug to the console and allow you to see what's going on.

If you get your WPA connection working before I fix mine, do let me know how!
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2014 Invision Power Services, Inc.