Broken AC3 on EMU10K1 since July 2-nd

Broken AC3 on EMU10K1 since July 2-nd

[Georgi Georgiev]

Hello,

When trying to play an AC3 sound on my Creative Live I got heavy
skipping. The "Digital" light on my decoder was flashing for less than a
second and then remained turned off for a little longer (but still less
than a second).

I tried different versions of alsa-driver, and versions prior to 1.0.5
were OK. I thought the problem is related to the fact that the
passthrough device changed from subdevice 2 to subdevice 3 in version
1.0.6, but it seems it is unrelated.

I checked out the CVS in my kernel source tree, and after playing with
different dates:

cvs up -D "1  Jul 2004" -- sound plays OK
cvs up -D "2  Jul 2004" -- sound skips heavily

However, there were a lot of changes between these two versions, some of
them pretty extensive, especially in the emu10k1 driver.

I sent the AC3 file with:

$ ac3dec -C city.ac3 
Using PCM device 'plug:iec958:{AES0 0x2 AES1 0x82 AES2 0x0 AES3 0x2}'
AC3 Stream 48.0 KHz 448 kbps

but the effect was identical to when using mplayer with -ac hwac3 on a
DVD for example.

-- 
 >   Georgi Georgiev    > Love thy neighbor as thyself, but choose      >
<     chutz@gg3.net    <  your neighborhood. -- Louise Beal            <
 >  +81(90)6266-1163    >                                               >


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Broken AC3 on EMU10K1 since July 2-nd

[Georgi Georgiev]

maillog: 28/09/2004-01:03:14(+0900): Georgi Georgiev types
> I checked out the CVS in my kernel source tree, and after playing with
> different dates:
> 
> cvs up -D "1  Jul 2004" -- sound plays OK
> cvs up -D "2  Jul 2004" -- sound skips heavily

I did some more checking:

cvs up -dP -D "1 Jul 2004 18:0:0" -- OK
cvs up -dP -D "1 Jul 2004 19:0:0" -- BAD

Still, I cannot understand the changes made to emu10k1, so I don't know
what the problem might be. I hope someone finds the fix.

-- 
|    Georgi Georgiev   |  You will be given a post of trust and        |
|     chutz@gg3.net    |  responsibility.                              |
|   +81(90)6266-1163   |                                               |


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Lee Revell]

On Mon, 2004-09-27 at 12:20, Georgi Georgiev wrote:
> maillog: 28/09/2004-01:03:14(+0900): Georgi Georgiev types
> > I checked out the CVS in my kernel source tree, and after playing with
> > different dates:
> > 
> > cvs up -D "1  Jul 2004" -- sound plays OK
> > cvs up -D "2  Jul 2004" -- sound skips heavily
> 
> I did some more checking:
> 
> cvs up -dP -D "1 Jul 2004 18:0:0" -- OK
> cvs up -dP -D "1 Jul 2004 19:0:0" -- BAD
> 
> Still, I cannot understand the changes made to emu10k1, so I don't know
> what the problem might be. I hope someone finds the fix.

Can you please send me the diff via private mail?  I might be able to
figure it out.

Lee



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Lee Revell]

On Tue, 2004-09-28 at 01:10, Georgi Georgiev wrote:
> maillog: 27/09/2004-14:50:40(-0400): Lee Revell types
> > On Mon, 2004-09-27 at 12:20, Georgi Georgiev wrote:
> > > maillog: 28/09/2004-01:03:14(+0900): Georgi Georgiev types
> > > > I checked out the CVS in my kernel source tree, and after playing with
> > > > different dates:
> > > > 
> > > > cvs up -D "1  Jul 2004" -- sound plays OK
> > > > cvs up -D "2  Jul 2004" -- sound skips heavily
> > > 
> > > I did some more checking:
> > > 
> > > cvs up -dP -D "1 Jul 2004 18:0:0" -- OK
> > > cvs up -dP -D "1 Jul 2004 19:0:0" -- BAD
> > > 
> > > Still, I cannot understand the changes made to emu10k1, so I don't know
> > > what the problem might be. I hope someone finds the fix.
> > 
> > Can you please send me the diff via private mail?  I might be able to
> > figure it out.
> 
> It's pretty big, but there it is:
> 

OK, the problem is almost certainly the introduction of pcm_indirect for
the FX8010 PCM.  Takashi-san, do you agree?

Lee



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Takashi Iwai]

At Tue, 28 Sep 2004 12:24:01 -0400,
Lee Revell wrote:
> 
> On Tue, 2004-09-28 at 01:10, Georgi Georgiev wrote:
> > maillog: 27/09/2004-14:50:40(-0400): Lee Revell types
> > > On Mon, 2004-09-27 at 12:20, Georgi Georgiev wrote:
> > > > maillog: 28/09/2004-01:03:14(+0900): Georgi Georgiev types
> > > > > I checked out the CVS in my kernel source tree, and after playing with
> > > > > different dates:
> > > > > 
> > > > > cvs up -D "1  Jul 2004" -- sound plays OK
> > > > > cvs up -D "2  Jul 2004" -- sound skips heavily
> > > > 
> > > > I did some more checking:
> > > > 
> > > > cvs up -dP -D "1 Jul 2004 18:0:0" -- OK
> > > > cvs up -dP -D "1 Jul 2004 19:0:0" -- BAD
> > > > 
> > > > Still, I cannot understand the changes made to emu10k1, so I don't know
> > > > what the problem might be. I hope someone finds the fix.
> > > 
> > > Can you please send me the diff via private mail?  I might be able to
> > > figure it out.
> > 
> > It's pretty big, but there it is:
> > 
> 
> OK, the problem is almost certainly the introduction of pcm_indirect for
> the FX8010 PCM.  Takashi-san, do you agree?

Most likely.  I'll take a look.


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Takashi Iwai]

[-- Attachment #1: Type: text/plain, Size: 283 bytes --]

At Tue, 28 Sep 2004 18:28:27 +0200,
I wrote:
> 
> > OK, the problem is almost certainly the introduction of pcm_indirect for
> > the FX8010 PCM.  Takashi-san, do you agree?
> 
> Most likely.  I'll take a look.

The attached oneliner fixes this bug.
Already applied to CVS.


Takashi

[-- Attachment #2: Type: text/plain, Size: 760 bytes --]

Index: alsa-kernel/pci/emu10k1/emupcm.c
===================================================================
RCS file: /suse/tiwai/cvs/alsa/alsa-kernel/pci/emu10k1/emupcm.c,v
retrieving revision 1.35
retrieving revision 1.36
diff -u -r1.35 -r1.36
--- alsa-kernel/pci/emu10k1/emupcm.c	15 Sep 2004 09:43:12 -0000	1.35
+++ alsa-kernel/pci/emu10k1/emupcm.c	30 Sep 2004 10:01:42 -0000	1.36
@@ -1192,7 +1192,7 @@
 	}
 	snd_emu10k1_fx8010_playback_tram_poke1((unsigned short *)emu->fx8010.etram_pages.area + tram_pos,
 					       (unsigned short *)emu->fx8010.etram_pages.area + tram_pos + tram_size / 2,
-					       src, frames, tram_shift++);
+					       src, frames, tram_shift);
 	tram_pos -= frames;
 	pcm->tram_pos = tram_pos;
 	pcm->tram_shift = tram_shift;

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Georgi Georgiev]

maillog: 30/09/2004-12:10:34(+0200): Takashi Iwai types
> At Tue, 28 Sep 2004 18:28:27 +0200,
> I wrote:
> > 
> > > OK, the problem is almost certainly the introduction of pcm_indirect for
> > > the FX8010 PCM.  Takashi-san, do you agree?
> > 
> > Most likely.  I'll take a look.
> 
> The attached oneliner fixes this bug.
> Already applied to CVS.

Not sure what the proper procedure is, so I will just confirm that
latest CVS works now.

-- 
*>   Georgi Georgiev   *> You will be successful in love.              *>
<*    chutz@gg3.net    <*                                              <*
*>  +81(90)6266-1163   *>                                              *>


-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Takashi Iwai]

At Tue, 28 Sep 2004 01:20:24 +0900,
Georgi Georgiev wrote:
> 
> maillog: 28/09/2004-01:03:14(+0900): Georgi Georgiev types
> > I checked out the CVS in my kernel source tree, and after playing with
> > different dates:
> > 
> > cvs up -D "1  Jul 2004" -- sound plays OK
> > cvs up -D "2  Jul 2004" -- sound skips heavily
> 
> I did some more checking:
> 
> cvs up -dP -D "1 Jul 2004 18:0:0" -- OK
> cvs up -dP -D "1 Jul 2004 19:0:0" -- BAD
> 
> Still, I cannot understand the changes made to emu10k1, so I don't know
> what the problem might be. I hope someone finds the fix.

Are you using the latest alsa-lib, too?


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Georgi Georgiev]

maillog: 27/09/2004-21:56:50(+0200): Takashi Iwai types
> At Tue, 28 Sep 2004 01:20:24 +0900,
> Georgi Georgiev wrote:
> > 
> > maillog: 28/09/2004-01:03:14(+0900): Georgi Georgiev types
> > > I checked out the CVS in my kernel source tree, and after playing with
> > > different dates:
> > > 
> > > cvs up -D "1  Jul 2004" -- sound plays OK
> > > cvs up -D "2  Jul 2004" -- sound skips heavily
> > 
> > I did some more checking:
> > 
> > cvs up -dP -D "1 Jul 2004 18:0:0" -- OK
> > cvs up -dP -D "1 Jul 2004 19:0:0" -- BAD
> > 
> > Still, I cannot understand the changes made to emu10k1, so I don't know
> > what the problem might be. I hope someone finds the fix.
> 
> Are you using the latest alsa-lib, too?

Well, it is the latest *stable* version -- 1.0.6.

-- 
*)   Georgi Georgiev   *)  Amy: Aw, he looks like a little insane      *)
(*    chutz@gg3.net    (* drunken angel.                               (*
*)  +81(90)6266-1163   *)                                              *)


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Takashi Iwai]

At Tue, 28 Sep 2004 14:12:51 +0900,
Georgi Georgiev wrote:
> 
> maillog: 27/09/2004-21:56:50(+0200): Takashi Iwai types
> > At Tue, 28 Sep 2004 01:20:24 +0900,
> > Georgi Georgiev wrote:
> > > 
> > > maillog: 28/09/2004-01:03:14(+0900): Georgi Georgiev types
> > > > I checked out the CVS in my kernel source tree, and after playing with
> > > > different dates:
> > > > 
> > > > cvs up -D "1  Jul 2004" -- sound plays OK
> > > > cvs up -D "2  Jul 2004" -- sound skips heavily
> > > 
> > > I did some more checking:
> > > 
> > > cvs up -dP -D "1 Jul 2004 18:0:0" -- OK
> > > cvs up -dP -D "1 Jul 2004 19:0:0" -- BAD
> > > 
> > > Still, I cannot understand the changes made to emu10k1, so I don't know
> > > what the problem might be. I hope someone finds the fix.
> > 
> > Are you using the latest alsa-lib, too?
> 
> Well, it is the latest *stable* version -- 1.0.6.

If you use the CVS version of driver, use the CVS version of
alsa-lib, too.

Anyway, this problem must be a bug I introduced at that time...


Takashi


-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

Re: Re: Broken AC3 on EMU10K1 since July 2-nd

[Peter Zubaj]

This change was made on 1 jul 2004. I don't think it depends on alsa-lib:

http://cvs.sourceforge.net/viewcvs.py/alsa/alsa-kernel/pci/emu10k1/emupcm.c?r1=1.28&r2=1.29

Peter Zubaj

========================== REKLAMA ==============================
*BUENA VISTA SOCIAL CLUB*          Info: +421 904 666 363
*Ibrahim Ferrer*                   http://www.xl.sk
*3.oktobra 2004 o 20,00 hod*       email: mailto:buena@xl.sk
*Bratislava, INCHEBA - hala C*



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php