[maemo-developers] VNC from maemo to ubuntu edgy desktop now working

2006-12-08 Thread Laurent GUERBY
Hi,

Just to let you know that the ubuntu people were kind
enough to integrate the vino CoRRE bug fix into their edgy-proposed
repository (Version: 2.16.0-0ubuntu2.3), so now vncviewer 0.4.1
works when connecting to an edgy desktop: I tested from my Nokia 770 to
my 1680x1050 desktop (password protected) without problem so far (not
that reactive - a bit better in 8bits - but it works :).

The story is here:
https://launchpad.net/distros/ubuntu/+source/vino/+bug/70986

VNCviewer 0.4.1:
http://physik.de/770/

In particular when some web pages aren't displaying, I
can use my desktop browser. Now let's try openvpn :).

Laurent
http://guerby.org/blog/


___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers


[maemo-developers] Specification for Epson S1D13742 Mobile Graphics Engine

2006-12-08 Thread Frantisek Dufka

Hello,

I've got good news. Few days ago I tried to contact once again Epson 
Electronics (last time I tried got no reply) and today I got nice 126 
pages long specification PDF with register description and other 
programming information. If someone is interested, let me know. I was 
not told to keep it secret or anything so I guess I can resend it to 
people interested. Not sure if putting it freely to website isn't too much.


Looks like the rotation mode is two lowest bits in one register at 
offset 34h


00 0 degrees(Normal)
01 90 degrees
10 180 degrees
11 270 degrees

and in theory can be simply enabled and all display updates are 
transparently rotated. I'll try it.


It also looks like there is register (offset 58h) which contains display 
(vertical) blank timing information - 2 read only bits (number 7 and 6) 
- bit contains 1 when display is in vertical/horizontal non-display 
period. Also the chip has TE (tearing effect) output pin which can be 
configured to signal this condition but it is hard to say whether it can 
triger interrupt or whether it is connected to anything at all in N770. 
But still it can be possible to check these bits (mainly the vertical 
one) in register and try to synchronize display updates with it to 
minimize tearing.


There is also double buffering functionality but it works only in lower 
resolutions, not 800x480 (as there is no memory for 2 buffers) and I 
guess LCD supports only 800x480. Maybe timing for one line can be 
doubled to have 400x480 and then enable double buffering? Probably not.


Frantisek
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers


RE: [maemo-developers] BadWindow error when calling hildon_program_add_window?

2006-12-08 Thread Braimah, AL (Adam)
 
I hadn't even heard of Matchbox until you mentioned it; I ran it within
Scratchbox ("/usr/bin/matchbox-window-manager &") and then tried my
program but the same error occurred. Did I miss something? I should
mention that I'm using Xephyr as my display too.
Thanks,
Adam

-Original Message-
From: Johan Bilien [mailto:[EMAIL PROTECTED] 
Sent: 08 December 2006 12:26
To: Braimah, AL (Adam)
Cc: maemo-developers@maemo.org
Subject: Re: [maemo-developers] BadWindow error when calling
hildon_program_add_window?

On Fri, Dec 08, 2006, Braimah, AL (Adam) wrote:
> I'm just starting off coding with Maemo (Scirocco, on Debian) and am 
> running into an X error straight away; hopefully a more experienced 
> person can spot if there's an obvious mistake here!
> 
> The code is a basic "Hello, world" as shown at the foot of this
message.
> However, trying to run it causes an X error; 'BadWindow,' (serial 68 
> error_code 3 request_code 20 minor_code 0) The line which appears to 
> cause the problem is a call to hildon_program_add_window(). I've 
> compiled one of the included example programs (which it turns out only

> uses direct GTK calls) with the same includes and libraries and that 
> runs fine, and I can also bring up the emulated 770 desktop so I'm 
> really at a loss!

Can you try running the matchbox window manager before starting the
program?

I filled #904

Thanks,
Johan.

___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers


Re: [maemo-developers] BadWindow error when calling hildon_program_add_window?

2006-12-08 Thread Johan Bilien
On Fri, Dec 08, 2006, Braimah, AL (Adam) wrote:
> I'm just starting off coding with Maemo (Scirocco, on Debian) and am
> running into an X error straight away; hopefully a more experienced
> person can spot if there's an obvious mistake here!
> 
> The code is a basic "Hello, world" as shown at the foot of this message.
> However, trying to run it causes an X error;
> 'BadWindow,' (serial 68 error_code 3 request_code 20 minor_code 0)
> The line which appears to cause the problem is a call to
> hildon_program_add_window(). I've compiled one of the included example
> programs (which it turns out only uses direct GTK calls) with the same
> includes and libraries and that runs fine, and I can also bring up the
> emulated 770 desktop so I'm really at a loss!

Can you try running the matchbox window manager before starting the program?

I filled #904

Thanks,
Johan.
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers


[maemo-developers] BadWindow error when calling hildon_program_add_window?

2006-12-08 Thread Braimah, AL (Adam)
Hi,
I'm just starting off coding with Maemo (Scirocco, on Debian) and am
running into an X error straight away; hopefully a more experienced
person can spot if there's an obvious mistake here!

The code is a basic "Hello, world" as shown at the foot of this message.
However, trying to run it causes an X error;
'BadWindow,' (serial 68 error_code 3 request_code 20 minor_code 0)
The line which appears to cause the problem is a call to
hildon_program_add_window(). I've compiled one of the included example
programs (which it turns out only uses direct GTK calls) with the same
includes and libraries and that runs fine, and I can also bring up the
emulated 770 desktop so I'm really at a loss!

Thanks,
Adam

#include 
#include 
#include 

int main (int argc, char* argv[])
{
HildonProgram* program;
HildonWindow* window;

gtk_init(&argc, &argv);

program=HILDON_PROGRAM(hildon_program_get_instance());
g_set_application_name("Just testing!");

window= HILDON_WINDOW (hildon_window_new());

hildon_program_add_window(program, window); //@@

gtk_container_add(GTK_CONTAINER(window),
GTK_WIDGET(gtk_label_new("This is a GTK label. ")));

gtk_widget_show_all(GTK_WIDGET(window));

 
g_signal_connect(G_OBJECT(window),"delete_event",G_CALLBACK(gtk_main_qui
t),NULL); 

gtk_main();

return 0;
}
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers


Re: [maemo-developers] Debugging maemo_af_desktop crashes

2006-12-08 Thread Johan Bilien
On Thu, Dec 07, 2006, Marius Gedminas wrote:
> I believe I've mentioned before that my 770 likes to reboot spontaneously.
> /proc/bootreason is sw_rst, /var/lib/dsme/stats/lifeguard_resets usually
> indicates /usr/bin/maemo_af_desktop (46 times already, out of 54
> resets).  I mkdir'ed /media/mmc1/core-dumps and now have a core file
> from the last crash.  Question: what can I do with it?
> 
> gdb on the 770 doesn't appear to be very useful:
> 
> /media/mmc1/core-dumps $ gdb maemo_af_desktop core.959
> ...
> warning: core file may not match specified executable file.
> Core was generated by `/usr/bin/maemo_af_desktop'.
> Program terminated with signal 6, Aborted.
> Reading symbols from /lib/libc.so.6...(no debugging symbols found)...done.
> Loaded symbols for /lib/libc.so.6
> Reading symbols from /lib/ld-linux.so.3...(no debugging symbols found)...done.
> Loaded symbols for /lib/ld-linux.so.3
> #0  0x40047c14 in kill () from /lib/libc.so.6
> (gdb) bt
> #0  0x40047c14 in kill () from /lib/libc.so.6
> #1  0x40047a2c in raise () from /lib/libc.so.6
> #2  0x8f20 in ?? ()
> 
> 
> I'm still running mistral.  I've two status bar applets installed
> (panelclock 0.5 and load-applet-run 0.4.3-2), but previously I've seen
> these spontaneous reboots with only osso-statusbar-cpu installed instead
> of these two.
> 
> (I also happen to have four other core files from previous crashes:
> browser, maemo-launcher, dsme, osso_hss.)

A few things:

. maemo_af_desktop uses maemo-launcher. This means that the
maemo-launcher process is the one actually running the maemo_af_desktop
code. When a crash happens you should see 2 core files (with consecutive
pids):
  .  a small one from maemo_af_desktop, which is just coming from the
 maemo-invoker process which tells maemo-launcher to load
 maemo_af_desktop.launch
  . a big one (several MB) which is maemo-launcher running
maemo_af_desktop.launch. That's the interesting one.

so to get something useful you need to run gdb /usr/bin/maemo-launcher
thebigcorefile

. you will need all the debug symbols. Unfortunately it seems the debug
packages (-dbg) are all missing from mistral and scirocco. Ferenc, Tran,
do you know why? Now the only way would be to apt-get source libc6, gtk,
glib, and maemo-af-destkop and rebuild them :(

You can also send me the core file I can take a look.

Regards,
-- 
Johan Bilien
<[EMAIL PROTECTED]>
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers


[maemo-developers] Debugging maemo_af_desktop crashes

2006-12-08 Thread Eero Tamminen
Hi,

> Maybe we should have a place where users could upload their cores  ;) 

770 core dumps from asserts cannot be debugged because C-library abort()
function in 770 has "noreturn" attribute and it's compiled without
frame-pointer. Same thing with the Glib asserts (which call abort()).


> I mkdir'ed /media/mmc1/core-dumps and now have a core file
> from the last crash.  Question: what can I do with it?
> 
> gdb on the 770 doesn't appear to be very useful:
> 
> /media/mmc1/core-dumps $ gdb maemo_af_desktop core.959
> ...
> warning: core file may not match specified executable file.
> Core was generated by /usr/bin/maemo_af_desktop'.
> Program terminated with signal 6, Aborted.
> Reading symbols from /lib/libc.so.6...(no debugging symbols found)...done.

You need to:
- re-compile C-library with additional -g -fno-omit-frame-pointer
  compiler options and install it on to the device unstripped (you can
  strip other debug sections than .debug_frame if the library is too large)
- do same for Glib 
- reproduce the crash

Then Gdb works a bit better.


- Eero
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers