linuxconsole-dev

AW: Kernel Version

[Unger R. <ric...@te...>]

Hi!

Thanks for the tip, I should obviously have checked for tags...

Is the linuxconsole stuff still under active development?=20
Will the redesigned console layer make it into the mainline kernel?

To me it seems like an obvious idea that if a multiuser system has =
multiple screens and keyboards, multiple (independant) terminals should =
be possible. In fact, to me, any other situation indicates a little bit =
of a design problem.

Currently I'm playing with the 2.6.9-ruby kernel I compiled. Don't have =
much time as I am moving these days, but it looks like the kernel boots =
fine and the /dev/bus/console devices are now available...

I do have a few questions, documentation on this is so hard to find...

1) Is it possible to have more than one VGA console, or is the VGA =
console always only bound to the primary graphics device?
2) What's the status on framebuffer support? All the docs indicate =
framebuffer console isn't ready - is it worth trying out multiple =
fbcons?

Thanks!

Richard

=20

> -----Urspr=FCngliche Nachricht-----
> Von: lin...@li...=20
> [mailto:lin...@li...] Im=20
> Auftrag von Aivils Stoss
> Gesendet: Mittwoch, 25. J=E4nner 2006 14:56
> An: Unger Richard
> Cc: lin...@li...
> Betreff: Re: Kernel Version
>=20
> On Tre=F0diena, 25. Janv=E2ris 2006 15:26, Unger Richard wrote:
> > Hi!
> >
> > Which kernel version should I apply the current CVS version=20
> of ruby to?
>=20
> Tagged version should work:
> cvs -z3=20
> -d:pserver:ano...@cv...:/cvsroot/linuxconsole co -r
> v2-6-9 ruby
>=20
> Same version:
> http://www.ltn.lv/~aivils/files/ruby-2.6.9-20041021.diff.bz2
> Nowadays for extreme enthusiasts.
>=20
> Aivils
> > I've tried 2.6.9 and 2.6.15.1 without success.
> >
> > Also, is the following the correct installation procedure=20
> for kernel=20
> > 2.6?
> >
> > 1) Get project from CVS
> > 2) cd /usr/src/linux
> > 3) cp -av /path/to/my/ruby/ruby-2.6/* .
> > 4) /path/to/my/ruby/utils/remove_obsolete_files
> > 5) make xconfig and build kernel, modules
> >
> > Thanks!
> >
> > Richard Unger
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep=20
> through log files for problems?  Stop!  Download the new AJAX=20
> search engine that makes searching your log files as easy as=20
> surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dk&kid=103432&bid#0486&dat=121642
> _______________________________________________
> Linuxconsole-dev mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linuxconsole-dev
>=20

Re: X11R7.0

[Zephaniah E. H. <wa...@ae...>]

On Fri, Jan 27, 2006 at 09:32:20AM +0200, Aivils Stoss wrote:
> On Ceturtdiena, 26. Janv=C4=81ris 2006 23:09, Peter Schrammel wrote:
> > Yop, that seems to work. I'm useing
> > setxkbmap de -keycodes 'xfree86+aliases(qwertz)+aliases(evdev)' -option
> >
> > but my up/down keys are a mess. Pressing them sometimes causes a 100%
> > CPU load (Xorg and kwin). But this happens only on console :0 with a
> > ps/2 keyboard and only with the extra cursor keys (not those on the
> > numpad).
> >
> > On console :2 (usb) it runs smoothly, everything's fine.
> >
> > Tell me if I can provide you with some debug info.
> >
> > Sorry, avilis. I don't have a X build system installed here and (if
> > debian helps me) never will so I can't test the patches. But thanks any=
way.
>=20
> If You are not fanatic compile-from-scratch sympatizer, then You can try =
out
> this one:
> http://www.ltn.lv/~aivils/files/evdev-zeh-060127.tar.bz2
>=20
> Must have X server sources. Edit SERVER_SRC line in the Makefile
>=20
> Aivils

Thanks a ton for putting that together.

However there's been a slight bug fix patch update which helps
compiling with some compilers.

If you'd like I can generate a diff between that version and current?

Zephaniah E. Hull.

--=20
	  1024D/E65A7801 Zephaniah E. Hull <wa...@ae...>
	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
	    CCs of replies from mailing lists are requested.

Here's your cable.  We made it fifty feet long, just in case.  In case
what, in case tectonic movement makes the serial ports farther apart?
  -- Carl Jacobs on ASR.

Re: X11R7.0

[Aivils S. <ai...@un...>]

On Ceturtdiena, 26. Janv=E2ris 2006 23:09, Peter Schrammel wrote:
> Yop, that seems to work. I'm useing
> setxkbmap de -keycodes 'xfree86+aliases(qwertz)+aliases(evdev)' -option
>
> but my up/down keys are a mess. Pressing them sometimes causes a 100%
> CPU load (Xorg and kwin). But this happens only on console :0 with a
> ps/2 keyboard and only with the extra cursor keys (not those on the
> numpad).
>
> On console :2 (usb) it runs smoothly, everything's fine.
>
> Tell me if I can provide you with some debug info.
>
> Sorry, avilis. I don't have a X build system installed here and (if
> debian helps me) never will so I can't test the patches. But thanks anywa=
y.

If You are not fanatic compile-from-scratch sympatizer, then You can try out
this one:
http://www.ltn.lv/~aivils/files/evdev-zeh-060127.tar.bz2

Must have X server sources. Edit SERVER_SRC line in the Makefile

Aivils

Re: X11R7.0

[Mikhail K. <kl...@la...>]

Hi Peter,

> I've installed the new Debian Xorg 6.9 packages. They dropped the 
> proprietary support for evdev and implemented a clean evdev module (which is
> a good news). But actually I can't get my keyboard working any more with this
> new driver. It seems that the evdev driver doesn't support the XkbLayout
> option of Xorg.
> 
> Any ideas?

you'll need to patch your evdev module.
see https://bugs.freedesktop.org/show_bug.cgi?id=3912 for details

cheers,
Mikhail Kshevetskiy

Re: X11R7.0

[Peter S. <pet...@gm...>]

Yop, that seems to work. I'm useing
setxkbmap de -keycodes 'xfree86+aliases(qwertz)+aliases(evdev)' -option

but my up/down keys are a mess. Pressing them sometimes causes a 100% 
CPU load (Xorg and kwin). But this happens only on console :0 with a 
ps/2 keyboard and only with the extra cursor keys (not those on the numpad).

On console :2 (usb) it runs smoothly, everything's fine.

Tell me if I can provide you with some debug info.

Sorry, avilis. I don't have a X build system installed here and (if 
debian helps me) never will so I can't test the patches. But thanks anyway.

Peter

Ruben Faelens schrieb:

> About your evdev problem:
>
> I had the same problem. The large problem is the evdev keyboard driver 
> not outputting the same keycodes as the kbd driver. As such, some 
> keycodes which are assigned do not exist, whereas some are wrongly 
> assigned. For now, this workaround helps:
> https://bugs.freedesktop.org/show_bug.cgi?id=5559
>
> The bug has been assigned to the input/evdev developers, but I haven't 
> heard from it ever since, so I guess it's low priority for them. Also, 
> they probably should clean op Xkb-support for evdev first anyway, the 
> current implementation is a hack IIRC.
>
> Good luck using this workaround. It's strange, without Xkb everything 
> works correctly (with a qwerty layout).
>
> Could you tell me if it worked or add a comment on the bug page? Thanks.
>
>
>

Re: X11R7.0

[Aivils S. <ai...@un...>]

On Ceturtdiena, 26. Janv=E2ris 2006 14:49, Zephaniah E. Hull wrote:
> On Thu, Jan 26, 2006 at 11:56:47AM +0100, Ruben Faelens wrote:
> > About your evdev problem:
> >
> > I had the same problem. The large problem is the evdev keyboard driver
> > not outputting the same keycodes as the kbd driver. As such, some
> > keycodes which are assigned do not exist, whereas some are wrongly
> > assigned. For now, this workaround helps:
> > https://bugs.freedesktop.org/show_bug.cgi?id=3D5559
> >
> > The bug has been assigned to the input/evdev developers, but I haven't
> > heard from it ever since, so I guess it's low priority for them. Also,
> > they probably should clean op Xkb-support for evdev first anyway, the
> > current implementation is a hack IIRC.
> >
> > Good luck using this workaround. It's strange, without Xkb everything
> > works correctly (with a qwerty layout).
> >
> > Could you tell me if it worked or add a comment on the bug page? Thanks.
>
> Could you take a look at https://bugs.freedesktop.org/show_bug.cgi?id=3D5=
696?
>
> In theory the enhanced xkb support should work quite a bit better.
> (It also needs additional testing.)

More consrevative C compilator support patch:
=2D-- evdev_rel.c.orig	2006-01-26 15:08:57.000000000 +0200
+++ evdev_rel.c	2006-01-26 16:07:33.000000000 +0200
@@ -68,24 +68,24 @@
 static struct {
     const char	*name;
     const int	i;
=2D} axis_names =3D {
=2D    "X",	REL_X,
=2D    "Y",	REL_Y,
=2D    "Z",	REL_Z,
=2D    "RX",	REL_RX,
=2D    "RY",	REL_RY,
=2D    "RZ",	REL_RZ,
=2D    "HWHEEL",	REL_HWHEEL,
=2D    "DIAL",	REL_DIAL,
=2D    "WHEEL",	REL_WHEEL,
=2D    "MISC",	REL_MISC,
=2D    "10",	10,
=2D    "11",	11,
=2D    "12",	12,
=2D    "13",	13,
=2D    "14",	14,
=2D    "15",	15,
=2D    NULL,	0
+} axis_names[] =3D {
+    {"X",	REL_X},
+    {"Y",	REL_Y},
+    {"Z",	REL_Z},
+    {"RX",	REL_RX},
+    {"RY",	REL_RY},
+    {"RZ",	REL_RZ},
+    {"HWHEEL",	REL_HWHEEL},
+    {"DIAL",	REL_DIAL},
+    {"WHEEL",	REL_WHEEL},
+    {"MISC",	REL_MISC},
+    {"10",	10},
+    {"11",	11},
+    {"12",	12},
+    {"13",	13},
+    {"14",	14},
+    {"15",	15},
+    {NULL,	0}
 };
=20
 static void

Re: X11R7.0

[Zephaniah E. H. <wa...@ae...>]

On Thu, Jan 26, 2006 at 11:56:47AM +0100, Ruben Faelens wrote:
> About your evdev problem:
> 
> I had the same problem. The large problem is the evdev keyboard driver 
> not outputting the same keycodes as the kbd driver. As such, some 
> keycodes which are assigned do not exist, whereas some are wrongly 
> assigned. For now, this workaround helps:
> https://bugs.freedesktop.org/show_bug.cgi?id=5559
> 
> The bug has been assigned to the input/evdev developers, but I haven't 
> heard from it ever since, so I guess it's low priority for them. Also, 
> they probably should clean op Xkb-support for evdev first anyway, the 
> current implementation is a hack IIRC.
> 
> Good luck using this workaround. It's strange, without Xkb everything 
> works correctly (with a qwerty layout).
> 
> Could you tell me if it worked or add a comment on the bug page? Thanks.

Could you take a look at https://bugs.freedesktop.org/show_bug.cgi?id=5696?

In theory the enhanced xkb support should work quite a bit better.
(It also needs additional testing.)

-- 
	  1024D/E65A7801 Zephaniah E. Hull <wa...@ae...>
	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
	    CCs of replies from mailing lists are requested.

   "<green>From</yellow>"
   "Wow. The green word From is no longer yellow. That's deep, man."
 -- Marcus Meissner & Lars Balker Rasmussen in the Scary Devil Monastery

Re: X11R7.0

[Ruben F. <par...@gm...>]

About your evdev problem:

I had the same problem. The large problem is the evdev keyboard driver 
not outputting the same keycodes as the kbd driver. As such, some 
keycodes which are assigned do not exist, whereas some are wrongly 
assigned. For now, this workaround helps:
https://bugs.freedesktop.org/show_bug.cgi?id=5559

The bug has been assigned to the input/evdev developers, but I haven't 
heard from it ever since, so I guess it's low priority for them. Also, 
they probably should clean op Xkb-support for evdev first anyway, the 
current implementation is a hack IIRC.

Good luck using this workaround. It's strange, without Xkb everything 
works correctly (with a qwerty layout).

Could you tell me if it worked or add a comment on the bug page? Thanks.

Re: X11R7.0

[Helge H. <hel...@ai...>]

Peter Schrammel wrote:

> Hi,
>
> I've installed the new Debian Xorg 6.9 packages. They dropped the 
> proprietary
> support for evdev and implemented a clean evdev module (which is a 
> good news).
> But actually I can't get my keyboard working any more with this new 
> driver. It seems
> that the evdev driver doesn't support the XkbLayout option of Xorg.
>
> Any ideas?
>
> Here some pices of my config:
>
> vanilla linux 2.6.13
> debian sid
> xorg 6.9.0
>
> no kernel pathces needed
>
> Thanks Peter
>
>
>
> Section "Module"
>        Load    "glx"
>        Load    "bitmap"
>        Load    "dbe"
>        Load    "ddc"
>        Load    "extmod"
>        Load    "freetype"
>        Load    "record"
>        Load    "type1"
>        Load    "evdev"
> EndSection
> Section "InputDevice"
> Identifier "logitech0"
> Driver     "evdev"
> #Option     "Protocol" "evdev"
> #Option     "Dev Phys" "usb-0000:00:10.1-1/input0"
> Option     "Device" "/dev/input/event2"
> Option     "Buttons" "3"
> Option     "CorePointer"
> Option     "ZAxisMapping" "4 5"
> EndSection
> #-------------------------------------Tatstatur
> Section "InputDevice"
> Identifier "keyboard0"
> Driver     "evdev"
> #Option     "Protocol" "evdev"
> Option     "Device" "/dev/input/event0"
> #Option     "Dev Phys" "isa0060/serio0/input0"
> Option     "XkbModel" "pc104"
> Option     "XkbLayout" "de"
> Option     "CoreKeyboard"
> EndSection

This worked for me:
Section "InputDevice"
        Identifier      "Generic Keyboard"
#        Driver          "kbd"
        Driver          "evdev"
#        Option          "Protocol"      "evdev"
       Option          "Device"        "/dev/input/event2"
#        Option          "Dev Phys"      "isa0060/serio0/input0"
        Option          "CoreKeyboard"
        Option          "XkbRules"      "xfree86"
        Option          "XkbModel"      "pc102"
        Option          "XkbLayout"     "no"
EndSection

You don't have the "XkbRules", perhaps that is the problem?
Also check that you have the correct event device.
(Try cat /dev/input/event0 at the console, then type a bit to
verify that the event device is working.)
If none of this helps, try using the same order of statements.
Finally, X.org 6.9 is in unstable for a reason, there may be some bugs.
Perhaps "pc104" or "de" is broken.  If so, use "reportbug xserver-xorg"

Helge Hafting

Re: Kernel Version

[samar j. s. <sa...@ef...>]

Dear Aivils

Hope 2006 is treating you well. I can see you are keeping as busy as usual=
=20
from the list.

I had a question that I wonder if you could help with.

The reason for my asking this question is that I would like to be able to s=
how=20
my work (lesson)  to a class   on the projector screen and to be able to se=
e=20
the files that your software produces of their usb keyboard inputs on my=20
laptop screen.

My first question is that -  is this possible and  if it is, whether you ca=
n=20
advise me on how to do it.

My second question is that can I link each of these screens to a different=
=20
desktop in linux's multiple desktop system. That would make the whole proce=
ss=20
of using it very simple.

I am hopeful 2006 will see us implementing this. I have had very many=20
opportunities recently where it would be have been very exciting but I want=
=20
this to be the special characteristic of the school. The more time I spend =
on=20
this the more exciting opportunities I can see for small group learning.

I just want to say a very big thank you for making it possible. Let us hope=
 we=20
get the school started this year and really  get it going.

best regards

samar

I think most laptops have two display adaptors, so theoretically it should =
be=20
possible.

The next issue is whether we can connect one screen to the desktop windows =
in=20
linux and the other screen to a different desktop.

On Wednesday 25 January 2006 19:25, Aivils Stoss wrote:
> On Tre=F0diena, 25. Janv=E2ris 2006 15:26, Unger Richard wrote:
> > Hi!
> >
> > Which kernel version should I apply the current CVS version of ruby to?
>
> Tagged version should work:
> cvs -z3 -d:pserver:ano...@cv...:/cvsroot/linuxconsole co
> -r v2-6-9 ruby
>
> Same version:
> http://www.ltn.lv/~aivils/files/ruby-2.6.9-20041021.diff.bz2
> Nowadays for extreme enthusiasts.
>
> Aivils
>
> > I've tried 2.6.9 and 2.6.15.1 without success.
> >
> > Also, is the following the correct installation procedure for kernel
> > 2.6?
> >
> > 1) Get project from CVS
> > 2) cd /usr/src/linux
> > 3) cp -av /path/to/my/ruby/ruby-2.6/* .
> > 4) /path/to/my/ruby/utils/remove_obsolete_files
> > 5) make xconfig and build kernel, modules
> >
> > Thanks!
> >
> > Richard Unger
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> files for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=103432&bid#0486&dat=121642
> _______________________________________________
> Linuxconsole-dev mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linuxconsole-dev

=2D-=20
samar j. singh
bangalore

Re: X11R7.0

[Peter S. <pet...@gm...>]

Hi,

I've installed the new Debian Xorg 6.9 packages. They dropped the 
proprietary
support for evdev and implemented a clean evdev module (which is a good 
news).
But actually I can't get my keyboard working any more with this new 
driver. It seems
that the evdev driver doesn't support the XkbLayout option of Xorg.

Any ideas?

Here some pices of my config:

vanilla linux 2.6.13
debian sid
xorg 6.9.0

no kernel pathces needed

Thanks Peter



Section "Module"
        Load    "glx"
        Load    "bitmap"
        Load    "dbe"
        Load    "ddc"
        Load    "extmod"
        Load    "freetype"
        Load    "record"
        Load    "type1"
        Load    "evdev"
EndSection
Section "InputDevice"
Identifier "logitech0"
Driver     "evdev"
#Option     "Protocol" "evdev"
#Option     "Dev Phys" "usb-0000:00:10.1-1/input0"
Option     "Device" "/dev/input/event2"
Option     "Buttons" "3"
Option     "CorePointer"
Option     "ZAxisMapping" "4 5"
EndSection
#-------------------------------------Tatstatur
Section "InputDevice"
Identifier "keyboard0"
Driver     "evdev"
#Option     "Protocol" "evdev"
Option     "Device" "/dev/input/event0"
#Option     "Dev Phys" "isa0060/serio0/input0"
Option     "XkbModel" "pc104"
Option     "XkbLayout" "de"
Option     "CoreKeyboard"
EndSection

Re: Kernel Version

[Aivils S. <ai...@un...>]

On Tre=F0diena, 25. Janv=E2ris 2006 15:26, Unger Richard wrote:
> Hi!
>
> Which kernel version should I apply the current CVS version of ruby to?

Tagged version should work:
cvs -z3 -d:pserver:ano...@cv...:/cvsroot/linuxconsole co -=
r=20
v2-6-9 ruby

Same version:
http://www.ltn.lv/~aivils/files/ruby-2.6.9-20041021.diff.bz2
Nowadays for extreme enthusiasts.

Aivils
> I've tried 2.6.9 and 2.6.15.1 without success.
>
> Also, is the following the correct installation procedure for kernel
> 2.6?
>
> 1) Get project from CVS
> 2) cd /usr/src/linux
> 3) cp -av /path/to/my/ruby/ruby-2.6/* .
> 4) /path/to/my/ruby/utils/remove_obsolete_files
> 5) make xconfig and build kernel, modules
>
> Thanks!
>
> Richard Unger

Kernel Version

[Unger R. <ric...@te...>]

Hi!

Which kernel version should I apply the current CVS version of ruby to?

I've tried 2.6.9 and 2.6.15.1 without success.

Also, is the following the correct installation procedure for kernel
2.6?

1) Get project from CVS
2) cd /usr/src/linux
3) cp -av /path/to/my/ruby/ruby-2.6/* .
4) /path/to/my/ruby/utils/remove_obsolete_files
5) make xconfig and build kernel, modules

Thanks!

Richard Unger

Re: multi-seat sound [again]: anybody actually doing it?

[Helge H. <hel...@ai...>]

Hugo Vanwoerkom wrote:

>Hi,
>
>I just found out that my SB/Live 24 bit Audigy works
>good with kernel 2.6.15-ck1 (did not 6 months ago with
>2.6.12.11)
>
>So that opens up the possibility of 2 audio cards for
>2 simultaneous users.
>
>This means alsa however.
>
>Anybody run simultaneous alsa devices for different
>users?
>  
>
I have been using the same device for different users.
(The trident driver, with alsa and OSS emulation.)
This device is supposed to do hardware mixing, so
several processes/user may use the same sound device
at the same time. Provided, of course, that they are next
to each other.

This usually works fine, but in some cases one user gets
strange echoey or chopped-up sound.  A driver bug/limitation
I guess.

The card supports front and rear speakers, perhaps that can
be used to provide independent stereo without having to
use more cards.

Helge Hafting

Phr maceuticeal G et

[Lyric C. <ly...@fu...>]

=20
C
V
V

l
A
l

A
L
A

L
I
G

I
U
R

S
M
A

 from
 from
 from

 $
 $
 $

3
1
3

,
,
,

7
2
3

5
1
3

=20
And many other http://www.tourisoin.com <http://www.tourisoin.com>=20

Re: Inactive device under X

[Aivils S. <ai...@un...>]

Hi, observer!

	Your patch is just evolution of isolateDevice.
However "Inacitve" device is revolution! I post it here, because
of your or someone request. Inactive device bringed into code
by "authentic" developers of multiheaded XFree. In abstract it
should be beter. In the reality it is quite forgotten.

Aivils
On Piektdiena, 13. Janv=C4=81ris 2006 23:33, observer wrote:
> A few months ago I wrote a patch that allows the same functionality
> using isolate device, the file is attached, its working very fine for
> me, but before using the patch read it, Im a a very bad programmer,
> and the patch is basically fine but it breaks the config options for
> isolate device, you must use it like this:
>
> Section "ServerLayout"
> 	Identifier     "Multihead layout"
> 	Screen      0  "Screen0" 0 0
> 	Screen	    1  "Screen1" RightOf "Screen0"
> 	InputDevice    "Mouse0" "CorePointer"
> 	InputDevice    "Keyboard0" "CoreKeyboard"
> 	Option	    "Xinerama" "on"
> #	Option	    "Clone" "on"
> #	Option	SingleCard
> 	Option "IsolateDevice" "PCI:1:0:0-PCI:1:5:0"
> EndSection
>
> Im not sure what options dont work any more at least you cant use
> SingleCard any more you must use
>
> 	Option	"IsolateDevice" "PCI:0:2:0-PCI:0:2:0"
>
> also im not sure if works with more than two devices
>
> Maybe someone can fix it or take the idea, basically it just takes pci
> ids as parameters and makes the xserver to =C2=BFisolate? only those
> devices.
>
> > Hi, All!
> >
> > 	After few months of reading X code i discovered
> > old option. That option added into CVS tree 7 years ago
> > between 3.99 in 4.0.0 versions of XFree.
> > Syntaxis looks like this:
> > Section "ServerLayout"
> >     Identifier "layout1"
> >     InputDevice "Keyboard1" "CoreKeyboard"
> >     InputDevice "Mouse1" "CorePointer"
> >     Screen "screen1"
> >     Inactive "DeviceName-1"
> >     Inactive "DeviceName-2"
> > EndSection
> > I'm not sure, is syntax right.
> >
> > Read about:
> > xc/programs/Xserver/hw/xfree86/RAC.Notes
> >
> > Unfortunately that option lead segfault with X-6.8.2
> > of Mandriva. In the working case "Inactive" device beat
> > isolateDevice, because sysop can set up weird systems,
> > where 1st X have 3 heads, 2nd X - 5 heads ...
> >
> > Aivils

Re: multi-seat sound [again]: anybody actually doing it?

[Aivils S. <ai...@un...>]

> Hugo Vanwoerkom wrote:
> >--- ludovic pollet <pl...@nn...> wrote:
> >
> > =A0
> >
> >>Hello,
> >>
> >>For most alsa application, correctly setting the
> >>following environment
> >>variable will do the trick :
> >>
> >>ALSA_CARD=3D1
> >>ALSA_CTL_CARD=3D1
> >>ALSA_PCM_CARD=3D1
> >>
> >>You can force SDL games to use alsa with :
> >>
> >>SDL_AUDIODRIVER=3Dalsa
> >>
> >>
> >>But applications which use the OSS API will cause
> >>problems (mozilla plugins,
> >>openoffice, old games, ...)
> >>
> >>For those, you will need to make the device entries
> >>point to different files
> >>depending on the DISPLAY : I do that at home. Have a
> >>look at
> >>http://www.nnx.com/~pludov/sessiond/

SDL driver is one , which have hardcoded
/dev/sound/dsp
Please use Ludovic recomendations. Create a user login
script:
SDL_AUDIODRIVER=3D"oss"
AUDIODEVICE=3D"/dev/dsp"
export $SDL_AUDIODRIVER $AUDIODEVICE

I am not udev expert, but that /dev/sound is artifical,
created by udev config scripts - so is _not_ forced
to create /dev/sound at all. Edit udev scripts and You
have dspXX where You want. Under Mandriva edit
/etc/udev/rules.d/50-mdk.rules

=46or appliaction , which does suspicious things
# strace -oapp.log application-bin
Analyze app.log or simpler method
# strings application-bin | grep dsp
or
# strings lib-plugin.so | grep dsp

Aivils

Re: multi-seat sound [again]: anybody actually doing it?

[vinu <vi...@hc...>]

Hugo Vanwoerkom wrote:

>--- ludovic pollet <pl...@nn...> wrote:
>
>  
>
>>Hello,
>>
>>For most alsa application, correctly setting the
>>following environment 
>>variable will do the trick :
>>
>>ALSA_CARD=1
>>ALSA_CTL_CARD=1
>>ALSA_PCM_CARD=1
>>
>>You can force SDL games to use alsa with :
>>
>>SDL_AUDIODRIVER=alsa
>>
>>
>>But applications which use the OSS API will cause
>>problems (mozilla plugins, 
>>openoffice, old games, ...) 
>>
>>For those, you will need to make the device entries
>>point to different files 
>>depending on the DISPLAY : I do that at home. Have a
>>look at 
>>http://www.nnx.com/~pludov/sessiond/
>>
>>    
>>
>
>Indeed, I saw your page, good job!
>
>Did you write sessiond?
>
>Thanks!
>
>Hugo
>
>
>
>
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around 
>http://mail.yahoo.com 
>
>
>-------------------------------------------------------
>This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
>for problems?  Stop!  Download the new AJAX search engine that makes
>searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
>http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
>_______________________________________________
>Linuxconsole-dev mailing list
>Lin...@li...
>https://lists.sourceforge.net/lists/listinfo/linuxconsole-dev
>
>  
>
A few months ago i had made a similar setup, but with shell script only, 
but i used an alsa imulation of oss, and was working fine. after some 
times i tried with the flash plugin for firefox, but wat happened is 
only if v play some sound apps, using firefox, only one user can use it. 
the other users will not b able to use the sound device, whatever be the 
mapping. So we just tracked the working of the plugin and foundout that 
it's creating a lock on the /tmp device which v cannot remove. So what v 
had done is on each session creation the /tmp will be mounted locally 
and do a chroot, so the flash plugin will think that this is another tmp 
and and works happily. But this method is having many security and 
relaibilty issues,and also this setup higly affects the speed. Actually 
v were working on a 6 terminal environment, so the insteability 
increases as each user logged in. Is there some way to fix this?
regards
    vineesh

Re: multi-seat sound [again]: anybody actually doing it?

[Hugo V. <hvw...@ya...>]

--- ludovic pollet <pl...@nn...> wrote:

> Hello,
> 
> For most alsa application, correctly setting the
> following environment 
> variable will do the trick :
> 
> ALSA_CARD=1
> ALSA_CTL_CARD=1
> ALSA_PCM_CARD=1
> 
> You can force SDL games to use alsa with :
> 
> SDL_AUDIODRIVER=alsa
> 
> 
> But applications which use the OSS API will cause
> problems (mozilla plugins, 
> openoffice, old games, ...) 
> 
> For those, you will need to make the device entries
> point to different files 
> depending on the DISPLAY : I do that at home. Have a
> look at 
> http://www.nnx.com/~pludov/sessiond/
> 

Indeed, I saw your page, good job!

Did you write sessiond?

Thanks!

Hugo





__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

Re: multi-seat sound [again]: anybody actually doing it?

[ludovic p. <pl...@nn...>]

Hello,

For most alsa application, correctly setting the following environment=20
variable will do the trick :

ALSA_CARD=3D1
ALSA_CTL_CARD=3D1
ALSA_PCM_CARD=3D1

You can force SDL games to use alsa with :

SDL_AUDIODRIVER=3Dalsa


But applications which use the OSS API will cause problems (mozilla plugi=
ns,=20
openoffice, old games, ...)=20

For those, you will need to make the device entries point to different fi=
les=20
depending on the DISPLAY : I do that at home. Have a look at=20
http://www.nnx.com/~pludov/sessiond/

Regards,
Ludovic P

Le Samedi 14 Janvier 2006 22:17, Hugo Vanwoerkom a =E9crit :
> Hi,
>
> I just found out that my SB/Live 24 bit Audigy works
> good with kernel 2.6.15-ck1 (did not 6 months ago with
> 2.6.12.11)
>
> So that opens up the possibility of 2 audio cards for
> 2 simultaneous users.
>
> This means alsa however.
>
> Anybody run simultaneous alsa devices for different
> users?
>
> Thanks!
>
> Hugo
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> files for problems?  Stop!  Download the new AJAX search engine that ma=
kes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://ads.osdn.com/?ad_id=3D7637&alloc_id=3D16865&op=3Dclick
> _______________________________________________
> Linuxconsole-dev mailing list
> Lin...@li...
> https://lists.sourceforge.net/lists/listinfo/linuxconsole-dev

--=20
If it's worth hacking on well, it's worth hacking on for money.

multi-seat sound [again]: anybody actually doing it?

[Hugo V. <hvw...@ya...>]

Hi,

I just found out that my SB/Live 24 bit Audigy works
good with kernel 2.6.15-ck1 (did not 6 months ago with
2.6.12.11)

So that opens up the possibility of 2 audio cards for
2 simultaneous users.

This means alsa however.

Anybody run simultaneous alsa devices for different
users?

Thanks!

Hugo


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

Ph rmaceutwical Super

[Edorta G. <gui...@co...>]

=20
http://www.soitesily.com <http://www.soitesily.com>=20
=20
So
Xan
ClA
VA
Am
VlA

ma=20
ax
LIS
LIU
bie
GR




M
n
A



from=20




$1.
$1.
$3.
$1.
$2.
$3.

12
42
75
21
89
75

=20

Re: Faketty + gpm weirdness

[vinu <vi...@hc...>]

Hugo Vanwoerkom wrote:

>
>--- vinu <vi...@hc...> wrote:
>
>  
>
>>Hugo Vanwoerkom wrote:
>>
>>    
>>
>>>Hi,
>>>
>>>I realize this is not a hot priority, but I have
>>>repeatedly had the following:
>>>
>>>In a recent kernel (2.6.14/15) gpm (General Purpose
>>>Mouse Interface) will act weird: the mouse cursor
>>>      
>>>
>>does
>>    
>>
>>>not show in a text console (not X), but a line
>>>      
>>>
>>copied
>>    
>>
>>>with gpm copy/paste will appear, *and* in mc lines
>>>      
>>>
>>are
>>    
>>
>>>selected wherever the mouse cursor moves.
>>>
>>>      
>>>
>>there's a possibility that the mouse configuration
>>may go wrong. In the
>>case of some 3 buttn wheel mouse this problem is
>>observerd. the simplest
>>to disable this is reconfigure gpm, as 3 button
>>mouse only. Or 2 btn
>>mouse + emulate 3 buttons.
>>regards
>>vineesh
>>    
>>
>
>Vineesh,
>
>gpm.conf now has:
>
>device=/dev/input/mouse1br
>responsiveness=15
>repeat_type=
>#type=imps2
>type=ps2
>append="-S \"/sbin/poweroff\""
>
>What would have to be changed?
>
>H
>
>
>
>
>
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam?  Yahoo! Mail has the best spam protection around
>http://mail.yahoo.com
>
>DISCLAIMER:
>--------------------------------------------------------------------------------------------------------------
>
>This e-mail contains confidential and/or privileged information. If you are not the intended recipient
>(or have received this e-mail in error)please notify the sender immediately and destroy this e-mail.
>Any unauthorized copying, disclosure, use or distribution of the material in this e-mail is strictly forbidden.
>---------------------------------------------------------------------------------------------------------------
>
>  
>
Actually I don't know on which file the changes are happening
but how I made this working is :
 #mouseconfig
       and then from the following menu select 3 button mouse(not 3 btn 
wheel mouse).or select 2 button mouse and emulate 3 buttons. This 
configuration helped me. I think this will help u.
regards
     vineeh

Re: Inactive device under X

[observer <obs...@bl...>]

A few months ago I wrote a patch that allows the same functionality
using isolate device, the file is attached, its working very fine for
me, but before using the patch read it, Im a a very bad programmer,
and the patch is basically fine but it breaks the config options for
isolate device, you must use it like this:

Section "ServerLayout"
	Identifier     "Multihead layout"
	Screen      0  "Screen0" 0 0
	Screen	    1  "Screen1" RightOf "Screen0"
	InputDevice    "Mouse0" "CorePointer"
	InputDevice    "Keyboard0" "CoreKeyboard"
	Option	    "Xinerama" "on"
#	Option	    "Clone" "on"
#	Option	SingleCard
	Option "IsolateDevice" "PCI:1:0:0-PCI:1:5:0"
EndSection

Im not sure what options dont work any more at least you cant use
SingleCard any more you must use

	Option	"IsolateDevice" "PCI:0:2:0-PCI:0:2:0"

also im not sure if works with more than two devices

Maybe someone can fix it or take the idea, basically it just takes pci
ids as parameters and makes the xserver to =BFisolate? only those
devices.

> Hi, All!
>=20
> 	After few months of reading X code i discovered
> old option. That option added into CVS tree 7 years ago
> between 3.99 in 4.0.0 versions of XFree.
> Syntaxis looks like this:
> Section "ServerLayout"
>     Identifier "layout1"
>     InputDevice "Keyboard1" "CoreKeyboard"
>     InputDevice "Mouse1" "CorePointer"
>     Screen "screen1"
>     Inactive "DeviceName-1"
>     Inactive "DeviceName-2"
> EndSection
> I'm not sure, is syntax right.
>=20
> Read about:
> xc/programs/Xserver/hw/xfree86/RAC.Notes
>=20
> Unfortunately that option lead segfault with X-6.8.2
> of Mandriva. In the working case "Inactive" device beat
> isolateDevice, because sysop can set up weird systems,
> where 1st X have 3 heads, 2nd X - 5 heads ...
>=20
> Aivils

Re: Faketty + gpm weirdness

[Hugo V. <hvw...@ya...>]

--- vinu <vi...@hc...> wrote:

> Hugo Vanwoerkom wrote:
> 
> >Hi,
> >
> >I realize this is not a hot priority, but I have
> >repeatedly had the following:
> >
> >In a recent kernel (2.6.14/15) gpm (General Purpose
> >Mouse Interface) will act weird: the mouse cursor
> does
> >not show in a text console (not X), but a line
> copied
> >with gpm copy/paste will appear, *and* in mc lines
> are
> >selected wherever the mouse cursor moves.
> >
> there's a possibility that the mouse configuration
> may go wrong. In the 
> case of some 3 buttn wheel mouse this problem is
> observerd. the simplest 
> to disable this is reconfigure gpm, as 3 button
> mouse only. Or 2 btn 
> mouse + emulate 3 buttons.
> regards
> vineesh

Vineesh,

gpm.conf now has:

device=/dev/input/mouse1br
responsiveness=15
repeat_type=
#type=imps2
type=ps2
append="-S \"/sbin/poweroff\""

What would have to be changed?

H






__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com