Re: is "dbus-uuidgen --ensure" ok in an install script?

[Brad Midgley]

Faheem

> I don't think this will help your problem described in your reply but
> instead of using dbus-uuidgen, you can use run-standalone.sh  program> instead.
>
> Maybe a su - user -c '' may work.

fwiw, both the dbus-uuidgen and run-standalone.sh approaches result in
an execution where osso_initialize returns null.

so actually "su -" works. It even makes the gps icon appear properly
in the system tray after it's powered up. I'll have to replicate that
programatically since a simple seteuid() obviously doesn't cut it and
I need some root behavior in early execution. Maybe it's something in
the environment.

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

Re: is "dbus-uuidgen --ensure" ok in an install script?

[Brad Midgley]

hmm

fwiw, maybe it's not so simple... turning on the gps from my daemon
doesn't activate the gps status display in the system tray. Is there a
better way to do this stuff from the background?

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

is "dbus-uuidgen --ensure" ok in an install script?

[Brad Midgley]

Hey

I have an issue that goes away after running "dbus-uuidgen --ensure".
Is this safe enough to put in a post-install script?

I'm writing a daemon that needs to be able to keep the tablet awake
(long story).

osso_initialize will complete if run from the terminal as "user" but
will fail and abort if I sudo gainroot first or if it's run as an
init.d service. I do seteuid() so we're running as user before making
the call but it doesn't help. The error message is:

initializing osso
process 1669: D-Bus library appears to be incorrectly set up; failed
to read machine uuid: Failed to open "/var/lib/dbus/machine-id":No
such file or directory
See the manual page for dbus-uuidgen to correct this issue.

Should this be filed as a maemo bug?

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

Re: Some questions about liblocation and gpsd

[Brad Midgley]

David

NAN is used by gpsd for "out of bounds" values. You should test the
float using isnan(value) before you assume it has something valid.

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

Re: javascript interruption

[Brad Midgley]

fwiw,

> I've noticed that when using the browser, if I haven't clicked
> anything on my n810 for a while, the javascript execution in the
> browser is halted.

it's sort of cosmic... the answer was 42.

I had to build and periodically execute "acpi_fakekey 42" to keep the
browser awake. 42 is the value for the left shift key... plus I get to
hear the keyboard tapping away by itself.

I would feel better with a hack-free way to do this. It won't work on
my n800 since acpi_fakekey can't find a keyboard. I'm not sure how to
change it to insert events into a different device or if there even is
a no-op I can insert there that won't get ignored.

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

javascript interruption

[Brad Midgley]

Hey

I've noticed that when using the browser, if I haven't clicked
anything on my n810 for a while, the javascript execution in the
browser is halted. It doesn't matter if the browser is the front
application, it doesn't matter if the screen has blanked, etc. As soon
as I tap something again, unplug the AC, etc then the javascript in
the page runs again.

Is there any way to modify this behavior so the javascript will
continue to run for longer than the ~5 minute timeout? It would be
nice if it was driven by the screen blank state since I can control
that programatically and through preferences.

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

Re: powering on gps, playing nice with others

[Brad Midgley]

Till

> Do you call gpsbt_stop when closing your app?
>
> gpxview also uses gpsbt_start and _stop and i have never encountered
> any problems.

I tried again but couldn't replicate the issue I had using
gpsbt_start/stop. Maybe operator error. :)

I'll stick with using those.

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

powering on gps, playing nice with others

[Brad Midgley]

Hey

I'm having trouble finding the way to power on the gps for my own
program in a way that plays nice with maemo mapper. IIt might be that
maemo mapper is bending some rules too but it's hard to tell.

If I use the call gpsbt_start(NULL, 0, 0, 0, errors, 256, 0, &ctx)
then the gps will start and before long I have coordinates. However,
if I then start up maemo mapper, stop the gps from my app, then stop
the gps from maemo mapper, the gps is left running.

If I use the call gpsmgr_start("/usr/sbin/gpsd", NULL,
"/tmp/.gpsd_ctrl_sock", 0, 0, &ctx) then gpsd apparently starts but is
not connected properly to anything because I never get coordinates.
Also if I then run maemo mapper, it keeps saying it's looking for a
gps, not even to the point where it's trying to get a fix.

The docs are a bit confusing on this and have a mix of old and new
approaches. I'm using diablo, I expect the user to set the gps
location in the gps preferences panel, so I don't expect to have to
specify the gps device. Maemo mapper even seems a little mixed up on
this front since what you do is specify in its prefs that you have a
bt gps but the address you specify there is ignored and the system
preference is used.

thanks for any tips
Brad
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers

Re: USB Host - Class 1 Bluetooth

[Brad Midgley]

Daniel

> I am interested in using class 1 USB bluetooth adapters in Maemo.

it sounds like fun. Any bluetooth usb adapter "should" work but i'd
suggest you get one with a csr bluecore4 so it's more similar to
what's in the n8xx.

the bluetooth-proximity app would be a great way to try it out. You're
more likely to reach distant class 2 devices if your adapter has an
external antenna.

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

Re: DSP SBC debugging

[Brad Midgley]

Simon

> Right, well goodish news, the dsp sbc encoder appears to work. It can be
> used with mplayer to play ~1s of audio, then it all goes quiet.

this is how it was when trying the floating point version on an
emulated fpu. There's no way it could keep up.

> though as I understood that the code could run in real-time on a 200MHz
> StrongARM chip, I'd thought the DSP would be able to also run it without any
> tweaks?

the dsp is not going to perform well on general-purpose code. There
are a few matrix ops that need to be rewritten first probably.

sbc_analyze_{four,eight} are the main culprits. These have been
optimized to combine & remove multiplications whenever unnecessary,
but to preserve symmetry on a dsp you'll undo that to get it optimal.

> Anyway, what I'm saying is that at first glance I can't really see a faster
> way of transferring the data across (though I will do some testing to see
> what the cold hard figures say), but if anyone spots a mistake or two in my
> logic please pipe up.

I'm not sure how this impacts you but the soft codec could benefit
from zero copy. We'd need to change the internal representation to
match host-endian stereo audio and only copy it if the incoming data
is not in that format. There is a complication in that the codec has
some state from previous encode calls, so we'd need to rework things
so a buffer we refer to isn't changed by the client between calls.

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

Re: Replacing the standard maemo bluetooth pairing dialog.

[Brad Midgley]

Fheem

> This program works on the principle of default security pins and the program
> comes with a script to do just this.

one application registers itself as the passkey agent using d-bus. A
second agent trying to register will fail.

maybe there's a way to intercept at a different layer.

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

Re: DSP SBC encoder task

[Brad Midgley]

Simon

I only say stepping around gstreamer since the gst way would be to
make one gst plugin for soft and one for dsp codec and let gst choose
one when building a pipeline. ie, they're prepared for multiple
implementations of a given codec.

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

Re: DSP SBC encoder task

[Brad Midgley]

Simon

I have been thinking more about this and I think another approach
could be considered.

It would be easier to plug your work into everything else if you wrote
it up as a patch to the regular sbc.c so it transparently chooses the
soft or dsp codec at runtime. It would work with the alsa plugin, gst,
and eventually pulse without extra work.

Marcel will have to weigh in if it's to be accepted upstream.

In any case, we'd need an override. It could be done with an
environment variable like "SBC_CODEC" with values eg "soft", "dsp",
"auto" with auto the default if it's not set. This does step around
gstreamer a little, but anyway, alsa and pulse don't have the greatest
gstreamer integration to start with...

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

Re: DSP SBC encoder task

[Brad Midgley]

Simon

> Atm we have to use ALSA to produce A2DP output, is the data sent
> thought here simply piped straight into sbcenc running standalone?

the alsa plugin with most of the bluez logic is a shared object, so
the alsa audio client is effectively transmitting directly to the
headset.

> I'd like to make sure that the people who currently use it (e.g. from
> mplayer) are also able to continue doing so using the DSP.
>
> Certainly a GStreamer wrapper is next on the list of ToDos after I
> stop the DSP from crashing ;).

unlike the alsa plugin, the gstreamer stuff breaks up the problem.
There is one gstreamer component for encoding and a separate one for
transmission. You wouldn't have to change the transmission part.

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

Re: DSP SBC encoder task

[Brad Midgley]

Simon

> Just a quick note to say that the DSP task encodes the test .au file
> correctly. Still a minor ;) issue that the DSP crashes

It sounds like an awesome achievement. I'm looking over things now.

If it can be coded generically, it would be nice if the standard bluez
gstreamer plugin could choose to use the dsp task if it's available.

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

Re: How can I test a GPS application without using the device?

[Brad Midgley]

Andrea

I haven't tried with n810 yet, but making gumstix or neo1973 act as a
bluetooth gps is fairly straightforward as long as gpspipe is
installed.

sdptool add SP
rfcomm -r watch 0 1 sh -c "/usr/bin/gpspipe -r >/dev/rfcomm0" &

Brad

On Feb 7, 2008 9:42 AM, Andrea Grandi <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I'd like to write a Python application that interface itself to GPS
> (using gpsd). I'd like to be able to develop it on my PC as well,
> without using the real hardware. Not because I don't want to use my
> N810, but because I should use it outside to test it!
>
> I've discovered this application: http://gpsfeed.sourceforge.net/
>
> "gpsfeed+ is a utility that feeds the PC with continuous GPS data as
> if a GPS were moving in a car or airplane. NMEA 0183 GPS sentences are
> sent to the application through a socket (TCP/IP) connection or a UDP
> multicast message.!"
>
> but I still cannot understand if it works also with gpsd.
>
> I'd like to test it on my Ubuntu machine, but while Maemo uses this
> library for python-gps:
> http://pymaemo.garage.maemo.org/documentation/python_gps_examples/index.html
>
> Linux seems to have a totally different library. I did "apt-get
> install python-gps" the unique python gps library available in
> repository and it comes without any documentation.
>
> So I've to develop in Scratchbox at least but I would need a GPS
> "simulator", a very simple one... so I could test a GPS application
> even in Scratchbox environment, without using a real GPS device.
>
> Any idea? What are, all those people that wrote a gps-application,
> using to test it? In particular maemo-mapper, gps-camera ecc.
>
> Thanks for your support!
>
> --
> Andrea Grandi
> email: a.grandi [AT] gmail [DOT] com
> website: http://www.andreagrandi.it
> PGP Key: http://www.ptlug.org/andreagrandi.asc
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>



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

gpsd data quality

[Brad Midgley]

Hey

I haven't seen this myself, but does it qualify as a bug? Is there an
intentional change to the data gpsd produces? I couldn't find any docs
or existing bugs talking about it.

http://www.kismetwireless.net/Forum/General/Messages/1201749708.949826

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

Re: USB host mode on N800

[Brad Midgley]

Sebastian

Is it any harder or easier to get the host mode cable sorted out with
the n810 and its new connector?

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

gps listener

[Brad Midgley]

Hey

Is there a way with the gps api to watch the gpsd/gps status without
affecting it? I'd like to leave an app running which doesn't force
gpsd startup & gps power-on but would use it whenever another app has
started it... hopefully such an app would not force the gps to stay
on...

fwiw, I couldn't find gpsbt.h/gpsmgr.h in the chinook beta (I filed a
bug). I'm using the docs I could find in & linked from
http://maemo.org/development/documentation/how-tos/4-x/maemo_connectivity_guide.html#GPS

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

Re: Developing for maemo...

[Brad Midgley]

Tim

It seems the ruby bindings to maemo-specific stuff aren't as complete (sadly)

Brad

On 10/10/07, Tim <[EMAIL PROTECTED]> wrote:
>
>
> Hey everyone,
>
> I'm going to have some time on my hands pretty soon and I thought I'd try to
> start learning a programming language... And, I want it to be something that
> will easily apply to maemo.
>
> My first inclination was to learn Python/Gtk+, as it seems to be the norm
> for most people who program for Gtk/maemo...
>
> Although, today I started researching Ruby/Gtk. Since I have some experience
> with Ruby already, as well as a large local group of friends who are pretty
> apt in Ruby/RonR, I was thinking that might be the route to go for me...
> But, my question is this:
>
>
> Would it be as possible (i.e., as _easy_) to port a Ruby/Gtk app to maemo as
> it would a Python/Gtk app?
>
> Or, in other words, is it Python that counts or is it Gtk+?
>
> I apologize if these are entirely ignorant questions, but as most of you
> know, I'm a designer, not a coder... But, I'd like that to change. And, I'd
> like some advice as to the proper route towards programming specifically for
> maemo. Also, I'm thinking that my first phase of education will be a
> pre-Scratchbox phase, as I am going to be learning from (nearly) scratch (no
> pun intended). Or, if I'm wrong in that area, let me know that too.
>
> Thanks!
>  Tim
>
>
>  ---
>  Weblog ~ http://tim.samoff.com
>  Baby Blog ~ http://kc.samoff.com
>  Photography ~ http://www.flickr.com/photos/timsamoff
>  Film ~ http://www.youtube.com/timsamoff
>  Music ~ http://www.adkoc.com
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>
>
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers

Re: Anyone know where to get the VMWare application for maemo development?

[Brad Midgley]

ok

I've been using the bora vm. Setting up chinook was much easier than
previous versions so I'm running it natively.

you may still need the dns fix. I didn't see a ticket for it.

Brad

On 10/10/07, Acadia Secure Networks <[EMAIL PROTECTED]> wrote:
> Brad,
>
> I am talking about the version of the maemo SDK developed as a VM by
> Marcelo Lira and available at the maemo garage www page whose url is:
>
> https://garage.maemo.org/projects/maemovmware
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers

Re: Anyone know where to get the VMWare application for maemo development?

[Brad Midgley]

Hey

You're talking bora right?

When you do get it, the first thing you need to do so dns works in sb is

 ln -sf /etc/resolv.conf /scratchbox/etc/resolv.conf

I can't host this but I could mail it to you on a dvd. Might not get
there faster than 1kbps but certainly faster than 0kbps.

Brad

On 10/10/07, Acadia Secure Networks <[EMAIL PROTECTED]> wrote:
> George,
>
> you had better success than I did. I was getting 0 kbps from the torrent
> using the Auzreus torrent client!
>
> It was the first time that I used that particular client so it occurs to
> me that I may be somehow using it incorrectly.
>
> A while ago I downloaded, using the Bittorrent client, an earlier
> version of the SDK that is still referenced on the VMware www site but
> that version, of course, is now obsolete.
>
> It surprises me that Nokia does not sponsor the bandwidth/storage
> necessary to support this kind of thing since it will only help the
> process of software development for the Internet Tablets.
>
>
> Best Regards,
>
>
>
> John Holmblad
>
>
>
> Acadia Secure Networks, LLC
>
> * *
>
> *Serving the SmartDigital^TM home, entrepreneurial enterprise, and
> emerging network service provider markets*
>
> * *
>
>
>
> Best Regards,
>
>
>
> John Holmblad
>
>
>
> Televerage International
>
> GSEC Gold,   GCWN Gold,   GAWN,  GGSC-0100,   NSA-IAM,  NSA-IEM
>
> Information security, telecommunications, and information technology
> consulting
>
>
>
> (M) 703 407 2278
>
> (F)  703 620 5388
>
> primary email address:  [EMAIL PROTECTED]
>
> backup email address:  [EMAIL PROTECTED]
>
>
>
>
>
> George Nachman wrote:
> > I'm trying to get a dev environment set up for Maemo. I spent a few
> > hours wrestling with the install scripts and have given up for now. I
> > did find references to a VMWare appliance called
> > maemo-sdk-vmware_0.2.tar.bz2, but I can't find a working link for it.
> > The bittorrent is coming down at about 1kb/minute :). Does anyone have
> > a copy squirrelled away that I might be able to get?
> >
> > Thanks,
> > George
> > 
> >
> > ___
> > maemo-developers mailing list
> > maemo-developers@maemo.org
> > https://lists.maemo.org/mailman/listinfo/maemo-developers
> >
>
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-developers
>
___
maemo-developers mailing list
maemo-developers@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-developers

[patch] maemo-mapper: load tracks from logging gps

[Brad Midgley]

Hey

This patch allows maemo-mapper to load gpx tracks from a wintec gps.
You have to install gpsbabel to /usr/bin or the menu items won't
activate. The gps needs to be connected to load the log. For a recent
(but not packaged) gpsbabel: http://www.daria.co.uk/gps/

It's a quick and dirty thing and my first real work in scratchbox
(just noticed vim wasn't expanding tabs to spaces, blah).

One reason I wanted to get this out there is so it would be clear
there are downsides to running gpsd. We'd have to stop gpsd during the
track transfer if it was in the loop.

Working with a logging gps can be useful if you want to save the
n800's batteries. I sometimes leave the n800 off and let the gps log
things by itself. The next time I'm hiking I do not want to have to
drop to a shell to pull this off :)

Brad
Index: src/maemo-mapper.c
===
--- src/maemo-mapper.c	(revision 110)
+++ src/maemo-mapper.c	(working copy)
@@ -1061,6 +1061,8 @@
 static GtkWidget *_menu_track_distlast_item = NULL;
 static GtkWidget *_menu_track_distfirst_item = NULL;
 static GtkWidget *_menu_track_clear_item = NULL;
+static GtkWidget *_menu_track_load_gps_item = NULL;
+static GtkWidget *_menu_track_clear_gps_item = NULL;
 
 /* Menu items for the "Maps" submenu. */
 static GtkWidget *_menu_maps_submenu = NULL;
@@ -1215,6 +1217,9 @@
 static gchar XML_TZONE[7];
 static gint _gmtoffset = 0;
 
+static gboolean _gpsbabel_present = FALSE;
+static gchar *_rfcomm_device = NULL;
+
 /
  * ABOVE: DATA **
  /
@@ -1290,6 +1295,10 @@
 menu_cb_track_distfirst(GtkAction *action);
 static gboolean
 menu_cb_track_clear(GtkAction *action);
+static gboolean
+menu_cb_track_load_gps(GtkAction *action);
+static gboolean
+menu_cb_track_clear_gps(GtkAction *action);
 
 /* Callbacks for the "Maps" submenu. */
 static gboolean
@@ -3866,7 +3875,10 @@
 _connect_sid = g_io_add_watch_full(_channel, G_PRIORITY_HIGH_IDLE,
 G_IO_OUT, channel_cb_connect, NULL, NULL);
 }
-g_free(fdpath);
+	if(_rfcomm_device) {
+		g_free(_rfcomm_device);
+	}
+_rfcomm_device = fdpath;
 
 vprintf("%s(): return\n", __PRETTY_FUNCTION__);
 }
@@ -5287,6 +5299,10 @@
 FALSE);
 gtk_widget_set_sensitive(GTK_WIDGET(_menu_gps_reset_item),
 FALSE);
+gtk_widget_set_sensitive(GTK_WIDGET(_menu_track_load_gps_item),
+FALSE);
+gtk_widget_set_sensitive(GTK_WIDGET(_menu_track_clear_gps_item),
+FALSE);
 }
 }
 else if(!_rcvr_mac || strcmp(_rcvr_mac,
@@ -6157,6 +6173,12 @@
 = gtk_menu_item_new_with_label(_("Show Distance from Beginning")));
 gtk_menu_append(submenu, _menu_track_clear_item
 = gtk_menu_item_new_with_label(_("Clear")));
+gtk_menu_append(submenu, _menu_track_load_gps_item
+= gtk_menu_item_new_with_label(_("Load track from logging gps")));
+gtk_widget_set_sensitive(GTK_WIDGET(_menu_track_load_gps_item), _enable_gps && _gpsbabel_present);
+gtk_menu_append(submenu, _menu_track_clear_gps_item
+= gtk_menu_item_new_with_label(_("Clear tracks in logging gps")));
+	gtk_widget_set_sensitive(GTK_WIDGET(_menu_track_clear_gps_item), _enable_gps && _gpsbabel_present);
 
 /* The "Maps" submenu. */
 gtk_menu_append(menu, menu_item
@@ -6330,7 +6352,12 @@
   G_CALLBACK(menu_cb_track_distfirst), NULL);
 g_signal_connect(G_OBJECT(_menu_track_clear_item), "activate",
   G_CALLBACK(menu_cb_track_clear), NULL);
+g_signal_connect(G_OBJECT(_menu_track_load_gps_item), "activate",
+  G_CALLBACK(menu_cb_track_load_gps), NULL);
+g_signal_connect(G_OBJECT(_menu_track_clear_gps_item), "activate",
+  G_CALLBACK(menu_cb_track_clear_gps), NULL);
 
+
 /* Connect the "Maps" signals. */
 g_signal_connect(G_OBJECT(_menu_maps_repoman_item), "activate",
   G_CALLBACK(menu_cb_maps_repoman), NULL);
@@ -8105,6 +8132,8 @@
 
 config_init();
 
+	_gpsbabel_present = g_file_test("/usr/bin/gpsbabel", G_FILE_TEST_EXISTS);
+
 /* Initialize _program. */
 _program = HILDON_PROGRAM(hildon_program_get_instance());
 g_set_application_name("Maemo Mapper");
@@ -10635,6 +10664,67 @@
 }
 
 static gboolean
+menu_cb_track_load_gps(GtkAction *action)
+{
+gchar *buffer;
+gint size;
+	gchar command[256];
+	char *tmpfile = "/tmp/tracklog.gpx";
+	char *file_uri_str;
+	pid_t pid;
+	int status;
+
+printf("%s()\n", __PRETTY_FUNCTION__);
+	
+	// would work better with a popen if the gpx parser could use a stream
+	
+	if(!(pid = fork())) {
+		unlink(tmpfile);
+		execl("/usr/bin/gpsbabe

Re: Computerworld article

[Brad Midgley]

Guys

More than any nationalist agenda, I think the guy is just trying to
posit an unlikely outcome--a single playing sewing up what has been a
fragmented market. He then tossed together some weak subjective
evidence he thought would support it.

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

beta 4 sdk vs. maemo-mapper

[Brad Midgley]

Hey

Do regular apps need a lot of tweaks to be built in chinook?
maemo-mapper trunk fails the configure stage:

./autogen.sh ; ./configure --prefix=/usr
 ...
checking for OSSO... configure: error: Package requirements (libosso
>= 1 libossohelp osso-ic) were not met:

No package 'libossohelp' found
No package 'osso-ic' found

the required packages seem to be there... and I can't find any
packages to add that would help...

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

Re: a2dp bluez 3.11 and libhal-dev

[Brad Midgley]

Jesse

It might only be the "sniff" change that made it work for you. Could you
tinker a bit and eliminate the unnecessary changes?

Brad

On 6/9/07, Jesse Guardiani <[EMAIL PROTECTED]> wrote:


That worked perfectly. It now plays back without constant skipping. I
think there are still some quirks with a2dpd. It's CPU usage fluctuates
a bit depending on how many times I run mplayer... but this is very
promising.




On Sat, 9 Jun 2007 07:04:48 -0600
"Brad Midgley" <[EMAIL PROTECTED]> wrote:

> try putting "master" in hcid.conf's lm setting and take "rswitch" and
> "sniff" out of the lp setting (then reboot or restart hcid)

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

Re: a2dp bluez 3.11 and libhal-dev

[Brad Midgley]

try putting "master" in hcid.conf's lm setting and take "rswitch" and
"sniff" out of the lp setting (then reboot or restart hcid)
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: a2dp bluez 3.11 and libhal-dev

[Brad Midgley]

Jesse

If you're using plugz it seems that you would be linking the external
libsbc, not the version inside bluez. As of about two months ago, the
standalone sbc project had our best version of the (mostly) 32-bit
fixed-point codec. fwiw, the sbc inside bluez was only put there once we had
it nailed down in the external project.

the a2dpd2 alsa plugin was introduced by Frederic Dalleau to try to observe
best practices for an alsa plugin (things in a2dpd plugin like usleep()
calls were avoided) so you could try that next.

there's some room for improvement in the sbc codec (see the todo list in
sbc.c) and we may even see some of the Brazilian dsp masters try their hand
at moving it to the dsp.

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

Re: maemo-mapper crashes on os2007 hacker edition

[Brad Midgley]

hey

as a data point, mm was crashing for me on loading topo maps because there
was a corrupt map in the cache. Cleaning the cache fixed the crashing.

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

Re: [maemo-developers] N800 GUI Improvements from a Newton Developer's Perspective

[Brad Midgley]

Sean

> I did indeed, and a semi-snark-filled article deserves semi-snark-filled
> feedback.  It's reasonable.  Several of the criticisms were valid, and
> I'll tweak the article a bit appropriately.  And respond later on
> perhaps to some of the criticisms that I think were off the mark; at the
> very least the discussion might help me get a clue.

I'm not sure we'll ever get rotation. If we ever do, it would be nice if
the icons represent the new orientation. I had to roll this myself for
my tablet:
 http://www.xmission.com/~bmidgley/p1510/

I had the feeling the obscuring-things-popup was displaying broadcast
system messages, but messages like 'deleting' shouldn't have to be
broadcast.

Stronger app integration through dragging/routing would be nice.

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

Re: [maemo-developers] N800 GUI Improvements from a Newton Developer's Perspective

[Brad Midgley]

Levi

> Overall, you made some good points, but it felt to me like they were
> semi-submerged within the general current of, "I'm angry because the
> N800 isn't a Newton with updated hardware and a stripped-down OSX."

I enjoyed reading Sean's criticism. He states at the start of the
document that he likes the n800 and I believe him.

One thing I really like about maemo is there are things improving in
every rev. Has anyone else been disgusted about the same gui problems in
release after release of pocketpc? Things that look like hyperlinks are
really buttons with side effects... that awful dialog when you start to
edit a recurring event... these rankle me every time. MS seems
determined to never improve these but at least with maemo there's a real
chance to see fundamental improvements.

No one should get offended by any facet of this discussion. I won't get
religious about an object store and if there is someone out there who is
passionate about having extra gutters, you should be able to voice it
without recriminations. :)

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

Re: [maemo-developers] Bluetooth Headset Support - Status update?

[Brad Midgley]

Martin

>> * resulting audio is not very loud and occasionally makes a "pop" sound
>> on high-volume input. 64-bit version runs without these problems.
> 
> This are the problems I also noticed with libsbc from sf.net. Where
> can I find the 64-bit Version?

I have actually been tinkering again with the 32-bit version--I think
dynamic compression could mostly work around the problems.

In any case I will create a separate sbc64 cvs module and announce it on
the lists.

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

Re: [maemo-developers] Bluetooth Headset Support - Status update?

[Brad Midgley]

Gopi

> I'm pretty sure that you could write an A2DP music player that worked on
> the 770 without any help from Nokia. It wouldn't be nicely integrated,
> but it should work. I tested an app awhile ago, and got music coming out
> of my headset, it was just not real time - a second of music, 5 seconds
> silence, etc.
> 
> The reason for this is that it needed to transcode the music into a
> different CODEC, for which there was no integer version available.

you must have tried our sbc encoder before any optimization. we have a
version that runs ok* in 32 bits that should execute in realtime without
burning up too much battery life.

> you were to, say, just store all your music in that format, transcoding
> as you uploaded to the 770, then no problem.

sbc typically takes twice the space of mp3 and then you'd need to build
an sbc decoder into gstreamer to play your files through wired audio.

fyi, there are headsets out there that have an mp3 decoder. our a2play
test app at least recognizes when the mp3 decoder is present even though
it doesn't try to use it.

> Thus, there's often more config and poking needed to get the chipset
> into the right mode.
> 
> Does anybody know if the N800 uses the same Bluetooth chipset as the 770?

it's a bluecore4. you could use bccmd to route its audio to hci. if you
do mess around with bccmd, i'd recommend you make all changes to ram and
just don't even bother with the permanent settings. it would be so sad
to brick the bluetooth adapter in a new n800.

brad

* resulting audio is not very loud and occasionally makes a "pop" sound
on high-volume input. 64-bit version runs without these problems.
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: [maemo-developers] Using the N770/N800 as a remote control.

[Brad Midgley]

hey

> There is another problem: The N770 has a USB-Host mode (which is 
> necessary if you want to run such IrDA adapters) but it does not power 
> the USB port. I have made an adapter cable to power 5 Volts to an 
> external HDD drive via USB and it worked very well. But I think for a IR 
> transmitter this solution would be very unhandy.

by the time you do all this you'll wonder why you're not using wifi or
bluetooth to send commands. It would be easier to set up a pc that will
relay commands over IR or a media pc that will just act on them.

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

Re: [maemo-developers] ALSA

[Brad Midgley]

Devesh

> https://garage.maemo.org/plugins/scmsvn/viewcvs.php/alsa-plugin/?root=dsp-alsa
> 
> If I understood this correct, this provides the ALSA interface through
> ALSA-DSP PCM plugin.

would this help with integrating userspace output plugins? I have been
working on alsa plugins for a2dp audio but I haven't spent the time on
gst plugins.

I really need to get going with gstreamer. A gst plugin could pass mp3
files as-is to a headset like my r35 with its own mp3 decoder. but I'm
not totally sure the maemo apps can be told to use a different gst output...

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

Re: [maemo-developers] Bluetooth headset status on OS2006

[Brad Midgley]

Pierre

> gizmo developers claims on:
> http://support.gizmoproject.com/FAQs/nokia770.php
> 
> That you can use a bluetooth headset on the 770, but various other
> sources tells that it cant:
> http://www.gossamer-threads.com/lists/maemo/developers/3167?search_string=bluetooth%20headset;#3167
> 
> http://linuxdevices.com/news/NS9255811301.html
> http://www.maemo.org/maemowiki/BluetoothHeadset
> 
> So are the last links outdated or are gizmo guys doiing disinformation?

gizmo seems to be assuming it'll be in there just like it is for every
other nokia device.

I haven't heard from Marcel on this but you could pose the question on
bluez-devel to see if anyone got him a dev kit for the TI adapter.

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

[maemo-developers] Re: [maemo-announce] Maemo 2.0 beta published

[Brad Midgley]

erkko

> Today we published the maemo 2.0 beta on _http://www.maemo.org_. This
> release includes both the maemo 2.0 beta development environment and the
> Internet Tablet 2006 OS Beta image. For more details about the release
> you can check out _http://maemo.org/downloads/releases.html_

Should I load this new image via:

./flasher-2.0 -r BETA.SU-18_2006SE_0.2006.22-21_PR_F5_MR0_ARM.bin -f -R

I'm hesitant since it has a .bin instead of .jffs2 extension

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

Re: [maemo-developers] New Nokia 770 software image available

[Brad Midgley]

hey

>>Isn't SD a closed format requiring non-disclosure agreements (and it does 
>>not mix well with Linux for this reason) ???
> 
> 
> Plenty of linux devices with SD exist. (The entire zaurus line, for 
> example.)
> 
> Even skipping the whole SD thing, how about a full-size mmc? 

I noticed the rs-mmc has more connectors than a plain mmc card. In
theory, requiring rs-mmc might make it so the nokia doesn't have to run
the card in a slower compatibility mode.

I couldn't find any decent info on this so it's all a guess.

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

Re: [maemo-developers] sound architecture question

[Brad Midgley]

Dale

Is there any work toward making zinf work more directly as a gstreamer
client? It wouldn't have to go through emulation layers that way.

It looks like rhythmbox is sometimes grouped together with zinf but it
has gstreamer support now

 http://software.newsforge.com/print.pl?sid=04/04/30/1416253

Brad

> Hello.  I'm new to this arena but have searched hi and low and haven't
> been able to clearly understand the sound architecture.
> 
> I'm attempting to port the package 'zinf' to the N770 and am having
> difficulty getting the audio out. 
> I've read this:
> http://maemo.org/platform/docs/multimedia/multimedia_architecture.html
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

[maemo-developers] Re: Bluetooth headset news?

[Brad Midgley]

Laurent

> Any news on this? I've been unable to locate any information using google.
> 
> I've seen some nice Nokia Bluetooth Headsets while shopping this
> afternoon, would be great if Nokia Headsets worked with Nokia 770 :).

I'm curious as well as to where Nokia is with voice headsets. There's a
big piece missing right now in the bluetooth driver for our TI adapter.
No one has been discussing it over on the bluez lists so I don't
perceive any progress.

Myself... I have not been working on voice headsets at all. I've done
some work on stereo sets like finishing the 32-bit fixed-point sbc codec
and starting on gstreamer plugins but nothing to show yet.

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

Re: [maemo-developers] Re: backlight

[Brad Midgley]

Frantisek

Nice work.

Can you put a unified diff on your page or post it here (I realize it's
a work in progress, but I'd like to see where it all applies)

What is the process for patching and building the kernel .deb?

Brad

>>  It means that either UI dynamically adjusts itself to the value of
>> sysfs backlight_max or it uses directly tahvo.
> 
> 
> Yes, it uses sysfs. I implemented translation table and also hardcoded
> values for level 1 and 2 to values 1 and 2 (instead of 8 and 16) and it
> works like expected. Display is really dark on minimum UI level. Now I
> need to implement some way of setting this table from userspace. Then
> each UI backlight level can be customized.
> 
> Frantisek
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers

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

[maemo-developers] bluetooth development

[Brad Midgley]

Hi

I am working with the arm1.1 scratchbox root and the laika 1.1 plugin
for using eclipse with maemo. I'm trying to get A2DP working. My main
question:

Is there any way for the emulator to see a real usb bluetooth adapter on
my host machine? I could stop hcid on my host if I needed to. I can't
match the TI chip but fwiw, I know this usb dongle I have uses a CSR chip.

And a couple of annoyances:

When I run xterm inside the emulator, I get the dreaded
enter-hides-and-shows-soft-keyboard problem. The workaround of switching
to another app and back works on the real nokia but not in the emulator.
I have to try to remember to use control-m instead.

qemu-arm keeps running after I stop the emulator (nicely) from eclipse.
qemu-arm seems to be running some kind of busy loop because my laptop
gets really hot when it's left running.

Brad

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

Re: [maemo-developers] Re: backlight

[Brad Midgley]

Frantisek

I think the cleanest way to deal with this is to keep 15(?) levels so
the gui doesn't change but change the actual curve to be nonlinear
inside the sysfs interface.

Does Nokia provide everything we need to install a modified kernel?

brad

> I hope the darkest level won't be too bright either. I see darker level
> when writing 1 via sysfs then it is possible via gui. When writing 2 it
> is same level as possible with GUI. But when writing 2 via sysfs it
> immediatelly returns 1 on read (and the brightness is same). I'll try
> the ioctl too. Thank you.
>
> Frantisek

>> when you write 1, you'll write 1*0x7f/0x0f=0x08


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

Re: [maemo-developers] Re: backlight

[Brad Midgley]

Arnaud

The strange thing is that when you're on the lowest level the gui will
allow (1), the display goes a little dimmer than that when idle and the
sys file actually reports the value 0. But if you *write* a 0 to the sys
file, the backlight turns off altogether.

This leaves two interpretations for a 0 value.

Brad

>>Looks like it can be controlled via ioctl on the framebuffer device.
>>http://maemo.org/lxr/source/osso-af-utils/src/omapfb.h
>>
>>struct lcd_panel has pointer
>>int (*set_bklight_level)(struct lcd_panel *panel,unsigned int level);
>>
>>I'll try to figure out how it can be called. If anyone knows, don't
>>hesitate to answer :)
> 
> 
> well, you can call it by echoing to
> /sys/bus/platform/devices/omapfb/panel/backlight_level
> 
> In reality, the function called is lph8923_panel_set_bklight_level() in
> lcd_lph8923.c (which in turn calls tahvo_set_backlight_level). The code
> is saying that 1 is the lowest value (except 0 :P) and 15 the highest.
> 
> So, you can't go lower unless there's an other way to play with the
> level.
> 
> 
> Arnaud
> 
> 
> 
>>Frantisek
>>
>>Brad Midgley wrote:
>>
>>>Frantisek
>>>I noticed this too, but we know it can go lower... the backlight goes
>>>just a bit lower when you're idle and before it turns off altogether.
>>>Brad
>>>
>>
>>___
>>maemo-developers mailing list
>>maemo-developers@maemo.org
>>https://maemo.org/mailman/listinfo/maemo-developers
> 
> 

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

Re: [maemo-developers] Bluetooth keyboard update

[Brad Midgley]

Tomas

This is working great for me using a think outside/stowaway keyboard. I
did get one reset while trying to re-do the pairing but my nokia is
behaving again. Excellent work!

Now your app is a complete model for the ui for other bluetooth stuff
like bluetooth stereo. (I need to get busy :) Nokia may change things
with a system-wide bluetooth manager but we'll see.

Brad

> A new version of the Bluetooth keyboard plugin is available from
> http://770.fs-security.com/keyboards.html
> 
> New stuff:
> - You can search for keyboards and choose a specific one to connect to.
> - You can pair the keyboard to the 770. Once paired, powering on the
>   keyboard will associate it with the 770 and you can simply start
>   typing away. Convenient!
> - You can choose a keyboard layout, currently I have a Scandinavian
>   and a German layout bundled. Send me your layout if you want
>   it included (see the hacking section on the homepage).
> - Matthew's screen blanking fix is integrated (thanks!), but it doesn't
>   seem to work all the time, needs some looking into.
> 
> Regards,
> Tomas
> 
> 
> 
> 
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers

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

Re: backlight , Re: [maemo-developers] How to prevent display from shutting down after idle timeout?

[Brad Midgley]

Frantisek

I noticed this too, but we know it can go lower... the backlight goes
just a bit lower when you're idle and before it turns off altogether.

Brad

> As for the backlight - the default minimum is too bright at night. Is
> this hardware or software limitation? I cannot find API for this and for
> volume control in libOSSO. Is it elsewhere?
> 
> Frantisek
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers

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

Re: [maemo-developers] swapfile is a huge improvement!

[Brad Midgley]

Andy

> What do you mean "30mb wouldn't work"?  I put a 32meg swapFILE on the
> 64meg mmc I received and it worked fine (/proc/meminfo and /proc/swaps
> showed swap usage, and I could get numerous things running with out
> problems.  Now I'm trying to partition a 512meg MMC into two partitions,
> and neither 128meg nor 64meg of swap as a dedicated partition
> (/dev/mmcblk0p2) works.  As soon as I issue swap on, the "low on memory,
> close some applications" window pops up.

I was always trying a swap partition. I had the same issues you're
reporting with constant out-of-memory alerts coming up when my partition
was 30mb but they went away and things started performing when it was
down at 25mb.

I wanted to have a swap partition rather than a swapfile inside the fat
partition because I don't want things to blow up when I plug to in to a
host via usb.

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

Re: [maemo-developers] swapfile is a huge improvement!

[Brad Midgley]

Urho

> If user has swapon. User opens MMC door. System pops up a large RED GUI
> that states: You have swap active. Please turn off swap before removing
> MMC from the slot. GUI would have one large button (turn off swap).
> After swapoff, GUI would turn green and say, it's safe to detach MMC now.

if a user opens the door, you must assume they are going to eject the
card. this is what nokia does with the fat partition. you should pop up
your gui and make a beep and run swapoff. the message should say 'do not
eject the card yet'

i believe it's possible for swapoff to fail if memory is overcommitted,
so you have another case to worry about. the device should tell you
you'll have to close apps or turn off the nokia first and possibly give
you buttons to do that.

> On USB cable insertion, the same GUI would popup and tell user that MMC
> cannot be mounted on PC until swap is off. Again, nice large button to
> turn swap off.

why bother the user? we should be using a swap partition--not a swapfile
on the fat fs--right? every usb-host OS will leave the swap partition
alone by default.

> When MMC inserted back from either USB connection or physically, system
> would check if previous state was swapon and popup a gui asking if swap
> should be resumed.

i would make the swapon automatic whenever a swap partition is found on
rsmmc, but then it could be a 'remember this decision' checkbox on your gui.

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

Re: [maemo-developers] swapfile is a huge improvement!

[Brad Midgley]

Vladislav

> with swap on, what happens to the system when you manually remove the
> mmc card? Or when you connect the USB cable to the unit?

I am pretty sure the nokia will just crash if I eject it when swap is
active. We need to latch into the cardslot-opening and -closing events
if we want to handle this right (and display a busy... dialog while the
swapoff is working and a dialog if swapoff fails...)

Plugging it into a host is no trouble. The data partition is unmounted
on the 770 side and I can use it from the host. Fortunately, the os is
not trying to outsmart me by turning off swap as well when it unmounts
the partition. That would not be good.

Of course the host pc sees the whole card and can see there's a swap
partition but I didn't try accessing it from the host. In fact, when I
was starting from a card with just one fat partition I just plugged the
nokia (with card inserted) into my linux machine and did the
repartitioning and mkdosfs/mkswap from the host pc.

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

[maemo-developers] swapfile is a huge improvement!

[Brad Midgley]

hey

I decided to try the swap partition on my flash card and it made a HUGE
difference. Even with only 25mb swap (30mb wouldn't work for me), now
memory is not nearly the concern it was. I can run multiple apps, look
at complex pdf documents, etc. Just like I expected my 770 to do in the
first place!

I think the vm may need to be tuned if nokia plans to keep shipping
these things with 64mb and no swap. Swap changes the character of the
vm, so even adding ram may not whip things into shape in the same way.

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

[maemo-developers] keyboard notes

[Brad Midgley]

Hey

I do like the wireless keyboard overall (stowaway model). The caveats
are probably well known...

 - activate the keyboard with something other than the xterm in the
foreground or switch between apps once (enter key won't work)
 - backlight is dimmed and then turned off in spite of typing activity
 - not a lot of global keyboard shortcuts? only alt-tab as far as i can tell
 - the icon image should be a keyboard in combo with the bluetooth icon
so it's clear what thing is connected over bluetooth

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

Re: [maemo-developers] Bluetooth headset - not really working?

[Brad Midgley]

Johannes

>>The two types of bluetooth audio are very different. You linked to a
>>discussion about the other type.

> And the answer to that in
> http://article.gmane.org/gmane.linux.bluez.devel/6748 by Ville Tervo
> from Nokia:
> 
> "I'll try to arrange something."

Ok. I see it. It does sound encouraging (either Nokia will do it
themselves or they will also involve Marcel). BTW, Marcel is brilliant
with this stuff. They'd be doing themselves a real favor to get him on
board.

[a2dp]
> So that is used for bluetooth stereo earphones to listen to e.g. CD
> music?

yes. That's what I've been tinkering with mostly. I like the gstreamer
architecture... I want to put our audio encoder that runs on the cpu in
a gstreamer plugin and Nokia could snap in a replacement that runs on
the dsp.

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

Re: [maemo-developers] Bluetooth headset - not really working?

[Brad Midgley]

Razvan

> [4.046356] Bluetooth: SCO (Voice Link) ver 0.4
> [4.051147] Bluetooth: SCO socket layer initialized

this will be printed both for adapters with and without sco support. The
driver does not detect and disable the interface on unsupported
hardware. (Although it should--we get questions all the time about
everything working right up to the point that they get silence on the
connection instead of audio)

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

Re: [maemo-developers] Bluetooth headset - not really working?

[Brad Midgley]

Ralph

> Is there a bandwidth problem, or just people being silly with the 
> design?

There is a bandwidth problem. High-quality audio is compressed on the
fly before transmission. We will probably be able to build in the delay
to the video player so it slips the video stream a bit. gstreamer may
even have a way to specify in the codec plugin how much latency it
introduces.

Some of these problems could be fixed if a new audio spec was written
that required the extended data rate of bluetooth 2.0 and left out
compression.

The a2dp spec says nothing about duplex high-quality audio, but there's
no reason a headset couldn't try to connect back to the source node on
an independent connection when it starts receiving audio. That's
basically how the AVRCP connection works.

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

Re: [maemo-developers] Bluetooth headset - not really working?

[Brad Midgley]

Ralph

> Hmm. Linux support aside, does this mean I'm never going to find a bt
> headset with both a mic for voice and high quality stereo playback for 
> music in the same device? More or less what all the game consoles use,
> for example?

I assumed you're aware that a lot of the stereo sets will switch into
low-quality 2-way mode. It's the simultaneous stereo/mic that no one can
do now.

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

Re: [maemo-developers] Bluetooth headset - not really working?

[Brad Midgley]

Ralph

The latency for every stereo set out there is so bad (>500ms!) that
you'd lose every game or just throw yourself out the nearest window.

In theory the latency could be tightened up and in theory you could get
two independent a2dp connections, one in each direction. I doubt anyone
will make this headset for you... building it into your 770 is probably
the only way you'll get it.

Brad

>>low-latency voice-quality mono audio uses a special transport (SCO) and
>>is delivered from the bluetooth chip either over the regular data
>>connection to the cpu (HCI) or it is connected to dedicated audio/analog
>>hardware (PCM). The HCI formatting and negotiation differs from chip to
>>chip and the linux driver is only complete for CSR chipsets.
> 
> 
> Hmm. Linux support aside, does this mean I'm never going to find a bt
> headset with both a mic for voice and high quality stereo playback for 
> music in the same device? More or less what all the game consoles use,
> for example?
> 
>  -r

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

Re: [maemo-developers] Bluetooth headset - not really working?

[Brad Midgley]

Johannes

The two types of bluetooth audio are very different. You linked to a
discussion about the other type.

low-latency voice-quality mono audio uses a special transport (SCO) and
is delivered from the bluetooth chip either over the regular data
connection to the cpu (HCI) or it is connected to dedicated audio/analog
hardware (PCM). The HCI formatting and negotiation differs from chip to
chip and the linux driver is only complete for CSR chipsets.

a2dp specifies one-way high-quality, traditionally stereo, audio. It is
delivered over an L2CAP socket so linux can send and receive this audio
with any bluetooth chip that has even the most basic support.

We know what has to be done for a2dp. The SCO stuff is what I would like
to see help from Nokia on. I would hope they can help us with the HCI
interaction with the SCO transport since that gives us the most generic
control over the audio stream.

Brad

> On Tue, 2005-12-20 at 10:16 -0700, Brad Midgley wrote:
> 
>>I'm overseeing the bluetooth audio on linux project...
>>
>>Linux only supports SCO transfers for CSR chips. To use the current
>>btsco stuff, Nokia is going to have to contribute a SCO driver to the
>>kernel for whatever bluetooth chip they put in the 770.
> 
> 
> I found some further info at:
> http://thread.gmane.org/gmane.linux.bluez.devel/6713
> 
> Can anyone of the Nokia developers give us some more hints on the head
> set topic please? Can we only expect that to work when the 2006 software
> edition will be released or will the head set support be included in an
> earlier firmware update?
> 
> This missing feature is very much appreciated!
> 
> Greets, Jonek.
> 
> 
> ___
> maemo-developers mailing list
> maemo-developers@maemo.org
> https://maemo.org/mailman/listinfo/maemo-developers

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

Re: [maemo-developers] Bluetooth headset - not really working?

[Brad Midgley]

Razvan

> This seems to have been reported before (not for the 770):
> http://sourceforge.net/tracker/index.php?func=detail&aid=1273754&group_id=116589&atid=678258
> 
>  .
> However, the solution "pskey mapsco 0" only seems to work with CSR
> Bluetooth chips, and the 770 has a TI chip.

I'm overseeing the bluetooth audio on linux project...

Linux only supports SCO transfers for CSR chips. To use the current
btsco stuff, Nokia is going to have to contribute a SCO driver to the
kernel for whatever bluetooth chip they put in the 770.

It's possible that they will route SCO to PCM hardware (direct to audio
hardware or even the dsp) rather than HCI (to the cpu). If this is the
case, btsco etc will need a little hacking to work that way. (Or even
better, to be rewritten)

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

[maemo-developers] bluetooth stereo devel

[Brad Midgley]

Hi

I put a placeholder for this project on
http://maemo.org/maemowiki/ApplicationCatalogWip

Some questions:

Is there work being done to unify all the bluetooth
scan/connect/disconnect operations together? Audio and keyboard should
probably be siblings in the bluetooth menu button that you get when you
install the bluetooth keyboard plugin.

What about AVRCP events? Is there a clean path through DBUS for sending
next/prev/play/pause events back to whatever apps are generating audio
and also getting volume up/down events from the audio clients?

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

[maemo-developers] A2DP/bluetooth stereo

[Brad Midgley]

Hi

I am working on the bluetooth-audio-linux project
(http://bluetooth-alsa.sf.net) and I'd like to get bluetooth stereo
working on maemo.

Is this something Nokia is working on?

What about the codec? We can run what we have now in integer math on the
arm side, but the SBC codec would be a great fit for a dsp task if Nokia
could help out there.

Is there a nice place for all this to fit in with maemo 2? (eg browsing
for bluetooth devices, audio plugins, etc)

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