linux-apus-devel

Re: high quality sound

[Glenn H. <gh...@c2...>]

Hello, 

On 06-Oct-2000, Geert wrote:

> New try. Now the code is in AmiInit().

> For now I'd like to know whether the custom chip programming below works,
> if amifb is not compiled in or disabled (video=amifb:off).


I tried this with a 2.4.0-test8 kernel without amifb support compiled in,
and it seems to work fine (I haven't done a lot of testing, but everything I
tried worked).
Finally we can get high quality sound without the need to switch consoles
:-)


- glenn

Buddha Interface and Linux(PPC)

[Giorgio T. <de...@ip...>]

Hi !,
Last year at Pianeta Amiga i have buyed from Vesalia a 32k Buddha Flash.

All is gone right from the Amiga side, but in the Linux(PPC) side i have
found a bug:
When i try to reboot the machine Buddha locks the process.

You said something about this problem in your docs, so with these hints
i began to test the 2.2 kernel Linux's driver software and have found 
the bug :
In first i have disabled the writing of offset $fc0 but it do not was
the (only) problem.

The BUDDHA_CONTROL(ALTSTATUS) was the guilty! 

I tried to disable the use of this register offset and all works well
now !!!
Now the problem is that 2.2 kernel's ideprobe identify function must
use STATUS reg instead of ALTSTATUS, complaining for it.
So, I think that my solution is not the best way out.

Please can you give me some suggestion about this and, if possible,
the offset of CONTROL(ALTSTATUS) register because i do not know if $11a
is right.

I am attaching the source code of linux's interface with my comments
(search for -gterzi- )

Thanking you in advance

Regards
				Giorgio Terzi

Re: high quality sound (was: Re: XFree86 4.01. Q&A)

[Geert U. <ge...@li...>]

On Fri, 6 Oct 2000, Michel [iso-8859-1] Dänzer wrote:
> Frank Petzold wrote:
> > On Thu, Oct 05, 2000 at 05:27:22PM +0200, Geert Uytterhoeven wrote:
> > > There's one disadvantage of this method: in the rare occasion you have a
> > > modular amifb and rmmod amifb, you cannot get the high quality sound back.
> > > With the original idea, you could (rmmod and) insmod dmasound_paula.
> 
> Why even the need to reinitialize the sound module? Isn't it possible to set
> the mode each time the sound device is opened or similar?

Good idea!

> > > What do you prefer?
> > 
> > You should not rely on the order of initialization of things anywhere.
> 
> True.

Which is difficult if different functionalities depend on common hardware
settings.


New try. Now the code is in AmiInit().

Perhaps we should replace the {request,release}_mem_region() stuff with some
simpler access method, but that requires changes to amifb as well.

For now I'd like to know whether the custom chip programming below works, if
amifb is not compiled in or disabled (video=amifb:off).

Thanks for testing!

--- linux-2.4.0-test9/drivers/sound/dmasound/dmasound_paula.c	Tue Jul 18 13:54:40 2000
+++ geert-31kHz-2.4.0-test9/drivers/sound/dmasound/dmasound_paula.c	Fri Oct  6 17:12:01 2000
@@ -27,8 +27,6 @@
    /*
     *	The minimum period for audio depends on htotal (for OCS/ECS/AGA)
     *	(Imported from arch/m68k/amiga/amisound.c)
-    *
-    *	FIXME: if amifb is not used, there should be a method to change htotal
     */
 
 extern volatile u_short amiga_audio_min_period;
@@ -366,6 +364,21 @@
 	int period, i;
 
 	AmiSilence();
+
+	if ((amiga_chipset == CS_ECS || amiga_chipset == CS_AGA) &&
+	    amiga_audio_min_period > 57 &&
+	    request_mem_region(CUSTOM_PHYSADDR+0xe0, 0x120, "31 kHz hack")) {
+	    /* program Denise/Lisa for a higher maximum play rate */
+	    /* (may be overridden by amifb) */
+	    custom.htotal = 113;	/* 31 kHz */
+	    custom.vtotal = 223;	/* 70 Hz */
+	    custom.beamcon0 = 0x4390;	/* HARDDIS, VAR{BEAM,VSY,HSY,CSY}EN */
+	    /* suspend the monitor */
+	    custom.hsstrt = custom.hsstop = 116;
+	    custom.vsstrt = custom.vsstop = 226;
+	    amiga_audio_min_period = 57;
+	    release_mem_region(CUSTOM_PHYSADDR+0xe0, 0x120);
+	}
 
 	if (dmasound.soft.speed)
 		period = amiga_colorclock/dmasound.soft.speed-1;

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: high quality sound (was: Re: XFree86 4.01. Q&A)

[Roman Z. <zi...@fh...>]

Hi,

On Thu, 5 Oct 2000, Geert Uytterhoeven wrote:

> There's one disadvantage of this method: in the rare occasion you have a
> modular amifb and rmmod amifb, you cannot get the high quality sound back.
> With the original idea, you could (rmmod and) insmod dmasound_paula.

We can also make it a inline function and call it from cleanup_module(),
too.

> Oh, and anybody who'd like to test this patch (no testing done, not even a
> compile test)? Preferably someone who has a monitor (with insurance :-) on the
> Amiga graphics output as well...

We should make it depending on wether amifb is compiled in or not,
otherwise during every boot you restart, you turn off the monitor for a
short period. IMO this should only be done for amifb:off.

bye, Roman

Re: high quality sound (was: Re: XFree86 4.01. Q&A)

[Michel <dae...@st...>]

Frank Petzold wrote:
> 
> On Thu, Oct 05, 2000 at 05:27:22PM +0200, Geert Uytterhoeven wrote:
> > There's one disadvantage of this method: in the rare occasion you have a
> > modular amifb and rmmod amifb, you cannot get the high quality sound back.
> > With the original idea, you could (rmmod and) insmod dmasound_paula.

Why even the need to reinitialize the sound module? Isn't it possible to set
the mode each time the sound device is opened or similar?


> > What do you prefer?
> 
> You should not rely on the order of initialization of things anywhere.

True.


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and The DRI Project

Re: high quality sound (was: Re: XFree86 4.01. Q&A)

[<fp...@zu...>]

On Thu, Oct 05, 2000 at 05:27:22PM +0200, Geert Uytterhoeven wrote:
> There's one disadvantage of this method: in the rare occasion you have a
> modular amifb and rmmod amifb, you cannot get the high quality sound back.
> With the original idea, you could (rmmod and) insmod dmasound_paula.
> 
> What do you prefer?

You should not rely on the order of initialization of things anywhere.
-- 
Frank Petzold, IBM Zurich Research Laboratory, Säumerstrasse 4,
CH-8803 Rüschlikon/Switzerland, Tel.  +41-1-724-84-42  Fax.   +41-1-724-89-56
 Business email: fp...@zu...          Private email: pe...@he...
   The opinions expressed here are mine and not necessarily those of IBM.

high quality sound (was: Re: XFree86 4.01. Q&A)

[Geert U. <ge...@li...>]

On Tue, 26 Sep 2000, Geert Uytterhoeven wrote:
> On Mon, 25 Sep 2000, Alan Buxey wrote:
> > PS on the sound subject, I'm locating the info for how to get the sound to
> > operate at 44KHz instead of the standard 22 of AGA  (bootstrap line
> > required) but cant locate it
> 
> You need to program htotal for a 31 kHz mode.
> 
> Once I was working on a patch for that, but I haven't finished it yet. I'll
> dig for it when I find some time...

My original idea was to request_mem_region() the Denise/Lisa registers in
dmasound_paula_init() and program a 31 kHz mode when amifb is not active.

However, then I came up with this simpler solution: just setup a 31 kHz mode
in amiga_init_sound(). If amifb is used, it will be initialized later and
override the mode. If it's not used, you keep on having 56 kHz sound (and no
video output on the Amiga graphics output).

There's one disadvantage of this method: in the rare occasion you have a
modular amifb and rmmod amifb, you cannot get the high quality sound back.
With the original idea, you could (rmmod and) insmod dmasound_paula.

What do you prefer?


Oh, and anybody who'd like to test this patch (no testing done, not even a
compile test)? Preferably someone who has a monitor (with insurance :-) on the
Amiga graphics output as well...

Good luck!

--- linux-2.4.0-test9/arch/m68k/amiga/amisound.c	Fri Jul 28 21:19:00 2000
+++ geert-31kHz-2.4.0-test9/arch/m68k/amiga/amisound.c	Thu Oct  5 17:18:26 2000
@@ -51,6 +51,18 @@
 
 	/* setup divisor */
 	clock_constant = (amiga_colorclock+DATA_SIZE/2)/DATA_SIZE;
+
+	if (amiga_chipset == CS_ECS || amiga_chipset == CS_AGA) {
+	    /* program Denise/Lisa for a higher maximum play rate */
+	    /* (may be overridden by amifb later) */
+	    custom.htotal = 113;	/* 31 kHz */
+	    custom.vtotal = 223;	/* 70 Hz */
+	    custom.beamcon0 = 0x4390;	/* HARDDIS, VAR{BEAM,VSY,HSY,CSY}EN */
+	    /* suspend the monitor */
+	    custom.hsstrt = custom.hsstop = 116;
+	    custom.vsstrt = custom.vsstop = 226;
+	    amiga_audio_min_period = 57;
+	}
 }
 
 static void nosound( unsigned long ignored );

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Mol-diffs

[MICHAEL H. <101...@ge...>]

Hello User

On 02-Okt-00, wrote Samuel Rydh (Developer from MOL) this Mail on me:

Hi,
 
I'm sorry you haven't heard from me (at the moment I've simply
too many things to do, but I'll try to be of some assistance
to MOL on APUS later).

fg

 /Samuel
 
> ----------------------------------------------------------
> E-mail <sa...@ib...>  WWW: <http://www.ibrium.se>
>  Phone/fax: (home) +46 8 4418431, (work) +46 8 7908470   
> ----------------------------------------------------------

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

> Yep, Ken fixed it AFAIK.

I see two changes to fs/affs/file.c, but these are also in 2.4.

bye, Roman

CVS Update: CVSROOT

[Roman Z. <zi...@us...>]

CVSROOT:	/cvsroot/linux-apus
Module name:	CVSROOT
Repository:	./
Changes by:	zi...@sl....	00/10/03 10:11:57

Modified files:
	./: cvsignore 

Log message:
	adding kernel specific temp files

Re: 2.4 test 8

[Michel <da...@re...>]

Roman Zippel wrote:

> > The affs subdir has changed quite a bit since 2.2.  I had a look and did
> > some copying tests. The corruption was similar to what was happening in
> > 2.2.
> 
> Was it fixed there?

Yep, Ken fixed it AFAIK.


> I can't remember hearing of (serious) problems with 2.2.

I have experienced the same problems with 2.2 that Ken is describing with 2.4.


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and the DRI project

Re: 2.4 test 8

[Roman Z. <zi...@fh...>]

Hi,

On Thu, 28 Sep 2000, Ken Tyler wrote:

> Are you getting anywhere with it ?

Slowly. :)
I want to clean up a few things and that will take some time...

> The affs subdir has changed quite a bit since 2.2.  I had a look and did
> some copying tests. The corruption was similar to what was happening in
> 2.2.

Was it fixed there? I can't remember hearing of (serious) problems with
2.2.

bye, Roman

Re: Norwegian keymap

[Michel <dae...@st...>]

Robert Ramiega wrote:
> 
> On Sat, Sep 30, 2000 at 10:17:21AM +0200, Michel Dänzer wrote:
> >
> > Not if you're using X 4.x . It doesn't work better yet there because the
> > Xkb extension doesn't work right. One way to solve it would be to get
> > support for the new input layer, still no takers?
>  I'll give it a go (input layer that is). I somewhat sorted my problems with
> 2.4.0. Test5 boots ok and seems to work ok (i'm using it for last couple of
> days). I see no apparent diffs in pm2fb between test5 and test7 (and the
> latter doesn;t boot). I can only hope that input layer didn't change that
> much between test5 and higher.

AFAIK the kernel version shouldn't matter too much, the code you'd deal with
is in a separate CVS repository at sourceforge.


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and The DRI Project

Re: XFree86 CVS sources

[Michel <dae...@st...>]

Sven LUTHER wrote:
> 
> On Sun, Oct 01, 2000 at 08:23:09PM -0400, fh...@at... wrote:
> > I grabbed the XFree86 CVS sources and compiled them on
> > my APUS system.  The server evidenty did not compile.
> > The make complained about some tseng chip or something.
> >
> > Is the APUS compatible server available for download somewhere? I suppose
> > thats all I really need.  Everything else seemed to compile OK.
> 
> Did you try it with the released version ?
> 
> Also to upcome the above mentioned problem, just disable the compilation of
> the driver not really needed. Edit the
> xc/programs/Xserver/hw/xfree86/drivers Imakefile and remove any driver that
> you think will not compile.

The better way is to copy xc/config/cf/xf86site.def to xc/config/cf/host.def
and edit at least the #define XF86CardDrivers line, it doesn't hurt to edit a
few other things there as well.


> (well i think only pm2 and fbdev is needed on apus, don't know about other
> graphic chips)

Right, only fbdev and glint are currently useful on APUS.


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and The DRI Project

Re: XFree86 CVS sources

[Sven L. <lu...@dp...>]

On Sun, Oct 01, 2000 at 08:23:09PM -0400, fh...@at... wrote:
> I grabbed the XFree86 CVS sources and compiled them on
> my APUS system.  The server evidenty did not compile.
> The make complained about some tseng chip or something.
> 
> Is the APUS compatible server available for download somewhere? I suppose
> thats all I really need.  Everything else seemed to compile OK.

Did you try it with the released version ?

Also to upcome the above mentioned problem, just disable the compilation of
the driver not really needed. Edit the xc/programs/Xserver/hw/xfree86/drivers
Imakefile and remove any driver that you think will not compile. (well i think
only pm2 and fbdev is needed on apus, don't know about other graphic chips)

Friendly,

Sven LUTHER

XFree86 CVS sources

[<fh...@at...>]

I grabbed the XFree86 CVS sources and compiled them on
my APUS system.  The server evidenty did not compile.
The make complained about some tseng chip or something.

Is the APUS compatible server available for download somewhere? I suppose
thats all I really need.  Everything else seemed to compile OK.

Fred

Re: Integrating

[Geert U. <ge...@li...>]

On Fri, 29 Sep 2000, Michel [iso-8859-1] Dänzer wrote:
> Geert Uytterhoeven wrote:
> > On Fri, 29 Sep 2000, Michel [iso-8859-1] Dänzer wrote:
> > > Geert Uytterhoeven wrote:
> > > > Is anyone working on integrating the changes in the APUS tree into
> > > > Linus' tree?
> > >
> > > Don't think so. The best way might be via bitkeeper, but noone here has
> > > write permission there.
> > 
> > Then please let someone ask Cort...
> > 
> > (I have access, but don't have APUS nor serious development machines in my
> >  neighborhood ATM)
> 
> And who would that someone be? You? :)

No.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Tools to dump device memory

[Geert U. <ge...@li...>]

On Sat, 30 Sep 2000 fh...@at... wrote:
> Someone mentioned some tools to examine device memory
> and expansion cards.  I guess these run under AmigaOS?
> I want to try them on some expansion cards I have.  Does
> anyone have their names?  Pointers to any other tools 
> would be fine (Linux based too.)

Every tool for debugging is OK. I was used to PowerVisor, which is very
powerful and very complex. Unfortunately it's that complex that the only way to
master PowerVisor is if you know the author personally (he used to live next
door in my student dorm :-)

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Tools to dump device memory

[<fh...@at...>]

Someone mentioned some tools to examine device memory
and expansion cards.  I guess these run under AmigaOS?
I want to try them on some expansion cards I have.  Does
anyone have their names?  Pointers to any other tools 
would be fine (Linux based too.)

Fred

Re: Norwegian keymap

[Michel <dae...@st...>]

Henrik Nordhus wrote:
> 
> hen...@c2...> Henrik Nordhus wrote:
> > > > Although the Swedish one is supposed to accomplish £ and @(+ probably
> > > some others) using RAlt, the buttons for these seem to be placed
> > > elsewhere on the Swedish amiga keyboards than on the Norwegian ones.
> > > (I currently keep @ in my clipboard a lot!)
> >
> > Is this in X or also console?
> 
> That's what I just discovered. It has nothing to do with the signs
> being different places on Swedish and Norwegian keyboards as I
> thought. (It was a bit of a rush-job, so I didn't check the file
> properly). I just never tried using @ in the shell before, but in the
> shell it all works. It's just in X. Perhaps It's my X configuration
> that's wrong?

Not if you're using X 4.x . It doesn't work better yet there because the Xkb
extension doesn't work right. One way to solve it would be to get support for
the new input layer, still no takers?


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and The DRI Project

Re: Norwegian keymap

[Michel <dae...@st...>]

Henrik Nordhus wrote:
> 
> I've attached my Norwegian keymap. Please upload to the ftp site!

Done, thanks.


> Although the Swedish one is supposed to accomplish £ and @(+ probably
> some others) using RAlt, the buttons for these seem to be placed
> elsewhere on the Swedish amiga keyboards than on the Norwegian ones.
> (I currently keep @ in my clipboard a lot!)

Is this in X or also console?


> As I didn't have the reference material, or previous knowledge to
> rectify this, I hope someone else can improve on it.


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and The DRI Project

Norwegian keymap

[Henrik N. <hen...@c2...>]

I've attached my Norwegian keymap. Please upload to the ftp site!

It is basically just the Swedish one by Tommi Leinonen with the
Swedish äÄ and öÖ changed to Norwegian æÆ and øØ.

Other problems with the Swedish one are still there.
Although the Swedish one is supposed to accomplish £ and @(+ probably
some others) using RAlt, the buttons for these seem to be placed
elsewhere on the Swedish amiga keyboards than on the Norwegian ones.
(I currently keep @ in my clipboard a lot!)
As I didn't have the reference material, or previous knowledge to
rectify this, I hope someone else can improve on it. 

All problems aside though, this really has improved the usefulness of
Linux for this particualar Norwegian amigan, and I hope it can for
others.




-- 
***********************************
Henrik Nordhus\\\\\\\\\\\\\\\\\\\\\\\\\\\\
////////////PPC603e+/210 BVision 96MB
4.3GB HD OS3.5\\\\\\\\\\\\\\\\\\\\\\\\\\\\
***********************************

Re: Integrating

[Michel <dae...@st...>]

Geert Uytterhoeven wrote:
> 
> On Fri, 29 Sep 2000, Michel [iso-8859-1] Dänzer wrote:
> > Geert Uytterhoeven wrote:
> > > Is anyone working on integrating the changes in the APUS tree into
> > > Linus' tree?
> >
> > Don't think so. The best way might be via bitkeeper, but noone here has
> > write permission there.
> 
> Then please let someone ask Cort...
> 
> (I have access, but don't have APUS nor serious development machines in my
>  neighborhood ATM)

And who would that someone be? You? :)


Michel


-- 
Earthling Michel Dänzer (MrCooper)  \  CS student and free software enthusiast
Debian GNU/Linux (powerpc,i386) user \   member of XFree86 and The DRI Project

Re: Integrating

[Geert U. <ge...@li...>]

On Fri, 29 Sep 2000, Michel [iso-8859-1] Dänzer wrote:
> Geert Uytterhoeven wrote:
> > Is anyone working on integrating the changes in the APUS tree into Linus'
> > tree?
> 
> Don't think so. The best way might be via bitkeeper, but noone here has write
> permission there.

Then please let someone ask Cort...

(I have access, but don't have APUS nor serious development machines in my
 neighborhood ATM)

> > For my personal interest, I'm mainly thinking about the Amiga specific
> > parts.
> 
> Do you want to get your stuff in here as m68k is more or less dead? ;^)

Exactly my point :-) [thinking about Zorro and Amiga resource stuff]

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

zorroutils-0.04

[Geert U. <ge...@li...>]

I released zorroutils version 0.04. It can be obtained using CVS from
cvs.linux-m68k.org, or from my web page:

    http://home.tvd.be/cr26864/Linux/zorroutils/

Changes (only zorro.ids was really changed):
  - Make sure all entries are sorted
  - Fix a tab in the syntax explanation
  - Better documentation of `stolen' manufacturer IDs

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds