i810_audio broken?

i810_audio broken?

[Pawel Worach]

[-- Attachment #1: Type: text/plain, Size: 1034 bytes --]

God afternoon!

The i810 audio driver is broken on my Fujitsu Lifebook
S-4546. All output is just noise. Here is a snip's from
the kernel log.

Intel 810 + AC97 Audio, version 0.02, 19:41:16 Apr 23 2001
PCI: Found IRQ 9 for device 00:00.1
PCI: The same IRQ used for device 00:00.2
PCI: The same IRQ used for device 00:13.1
PCI: Setting latency timer of device 00:00.1 to 64
i810: Intel 440MX found at IO 0x1cc0 and 0x1000, IRQ 9
ac97_codec: AC97 Audio codec, id: 0x594d:0x4800 (Unknown)
i810_audio: only 48Khz playback available

And the lspci:
[root@whyami src]# lspci -vv -s 00:00.1
00:00.1 Multimedia audio controller: Intel Corporation 82440MX AC'97
Audio Controller
	Subsystem: Citicorp TTI: Unknown device 10d0
	Control: I/O+ Mem- BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B-
	Status: Cap- 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort-
<MAbort- >SERR- <PERR-
	Latency: 0
	Interrupt: pin B routed to IRQ 9
	Region 0: I/O ports at 1000 [size=256]
	Region 1: I/O ports at 1cc0 [size=64]



[-- Attachment #2: Card for <pworach@mysun.com> --]
[-- Type: text/x-vcard, Size: 103 bytes --]

begin:vcard
n:Worach;Pawel
fn:Pawel Worach
version:2.1
email;internet:pawel.worach@mysun.com
end:vcard

Re: i810_audio broken?

[Alan Cox]

> The i810 audio driver is broken on my Fujitsu Lifebook
> S-4546. All output is just noise. Here is a snip's from
> the kernel log.
> 
> Intel 810 + AC97 Audio, version 0.02, 19:41:16 Apr 23 2001
> PCI: Found IRQ 9 for device 00:00.1
> PCI: The same IRQ used for device 00:00.2
> PCI: The same IRQ used for device 00:13.1
> PCI: Setting latency timer of device 00:00.1 to 64
> i810: Intel 440MX found at IO 0x1cc0 and 0x1000, IRQ 9
> ac97_codec: AC97 Audio codec, id: 0x594d:0x4800 (Unknown)
> i810_audio: only 48Khz playback available

The dump looks fine. Its a cheap and cheeerful codec however by the look of it.
Make sure the applications you use properly handle 48Khz only audio. That
may be the problem or maybe not.

Also try the very latest kernels as a fair bit of work has been done on them

Re: i810_audio broken?

[Pawel Worach]

[-- Attachment #1: Type: text/plain, Size: 1483 bytes --]

sorry the kernel version is 2.4.3-ac12, so it's kind of latest...

I was using mpg123 (xmms and c/o does exactly the same)
if I run it like this Moby sounds very stupid... :)
[root@whyami mp3]# mpg123 -r 48000 Moby_01.wav.mp3 
unsupported playback rate: 44100
Audio device open for 44.1Khz, stereo, 16bit failed
Trying 44.1Khz, 8bit stereo.
unsupported sound format: 32
Audio device open for 44.1Khz, stereo, 8bit failed
Trying 48Khz, 16bit stereo.

Cheers
Pawel Worach
PostGirot Bank AB
----- Original Message -----
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Monday, April 23, 2001 9:53 pm
Subject: Re: i810_audio broken?

> > The i810 audio driver is broken on my Fujitsu Lifebook
> > S-4546. All output is just noise. Here is a snip's from
> > the kernel log.
> > 
> > Intel 810 + AC97 Audio, version 0.02, 19:41:16 Apr 23 2001
> > PCI: Found IRQ 9 for device 00:00.1
> > PCI: The same IRQ used for device 00:00.2
> > PCI: The same IRQ used for device 00:13.1
> > PCI: Setting latency timer of device 00:00.1 to 64
> > i810: Intel 440MX found at IO 0x1cc0 and 0x1000, IRQ 9
> > ac97_codec: AC97 Audio codec, id: 0x594d:0x4800 (Unknown)
> > i810_audio: only 48Khz playback available
> 
> The dump looks fine. Its a cheap and cheeerful codec however by the 
> look of it.
> Make sure the applications you use properly handle 48Khz only 
> audio. That
> may be the problem or maybe not.
> 
> Also try the very latest kernels as a fair bit of work has been 
> done on them
> 
> 

[-- Attachment #2: Card for <pworach@mysun.com> --]
[-- Type: text/x-vcard, Size: 103 bytes --]

begin:vcard
n:Worach;Pawel
fn:Pawel Worach
version:2.1
email;internet:pawel.worach@mysun.com
end:vcard

Re: i810_audio broken?

[Jeff Garzik]

Pawel Worach wrote:
> sorry the kernel version is 2.4.3-ac12, so it's kind of latest...
> 
> I was using mpg123 (xmms and c/o does exactly the same)
> if I run it like this Moby sounds very stupid... :)

"very stupid" means "broken" obviously, but can you be more specific? 
music is faster? slower?  garbled?

> [root@whyami mp3]# mpg123 -r 48000 Moby_01.wav.mp3
> unsupported playback rate: 44100
> Audio device open for 44.1Khz, stereo, 16bit failed
> Trying 44.1Khz, 8bit stereo.
> unsupported sound format: 32
> Audio device open for 44.1Khz, stereo, 8bit failed
> Trying 48Khz, 16bit stereo.

so, since you provided no more output than this, I assume that
48Khz/16bit succeeded, which appears perfectly normal for a locked-rate
codec.

You may need the 'clocking' module option, not sure...

-- 
Jeff Garzik      | The difference between America and England is that
Building 1024    | the English think 100 miles is a long distance and
MandrakeSoft     | the Americans think 100 years is a long time.
                 |      (random fortune)

Re: i810_audio broken?

[Pawel Worach]

[-- Attachment #1: Type: text/plain, Size: 1698 bytes --]

It sounds like when you play music very loud with bad speakers
and it's kind of slow. It's kind of "clinking", describing sound via
e-mail can be very hard.

what value shall i put for the clocking parameter?
is it trial-and-error or is there some formula?

And no, the cut off output does not mean that it worked. :(
xmms says this:
Audio device open for 44.1Khz, stereo, 8bit failed
Trying 48Khz, 16bit stereo.
But it still sounds strange.


----- Original Message -----
From: Jeff Garzik <jgarzik@mandrakesoft.com>
Date: Tuesday, April 24, 2001 0:07 am
Subject: Re: i810_audio broken?

> Pawel Worach wrote:
> > sorry the kernel version is 2.4.3-ac12, so it's kind of latest...
> > 
> > I was using mpg123 (xmms and c/o does exactly the same)
> > if I run it like this Moby sounds very stupid... :)
> 
> "very stupid" means "broken" obviously, but can you be more 
> specific? 
> music is faster? slower?  garbled?
> 
> > [root@whyami mp3]# mpg123 -r 48000 Moby_01.wav.mp3
> > unsupported playback rate: 44100
> > Audio device open for 44.1Khz, stereo, 16bit failed
> > Trying 44.1Khz, 8bit stereo.
> > unsupported sound format: 32
> > Audio device open for 44.1Khz, stereo, 8bit failed
> > Trying 48Khz, 16bit stereo.
> 
> so, since you provided no more output than this, I assume that
> 48Khz/16bit succeeded, which appears perfectly normal for a locked-
> ratecodec.
> 
> You may need the 'clocking' module option, not sure...
> 
> -- 
> Jeff Garzik      | The difference between America and England is that
> Building 1024    | the English think 100 miles is a long distance and
> MandrakeSoft     | the Americans think 100 years is a long time.
>                 |      (random fortune)
> 

[-- Attachment #2: Card for <pworach@mysun.com> --]
[-- Type: text/x-vcard, Size: 103 bytes --]

begin:vcard
n:Worach;Pawel
fn:Pawel Worach
version:2.1
email;internet:pawel.worach@mysun.com
end:vcard

Re: i810_audio broken?

[Pawel Worach]

[-- Attachment #1: Type: text/plain, Size: 822 bytes --]

Ok building mpg123 without eSound worked for me too,
so guess this is not a Linux kernel issue, sorry for this.

I tried the fstodell hack but it seems to be obsoluted.
Now it works without any tweaks.

eSound sux?

Thanks guys!
Back to work (with music :)

----- Original Message -----
From: Chmouel Boudjnah <chmouel@mandrakesoft.com>
Date: Tuesday, April 24, 2001 1:34 am
Subject: Re: i810_audio broken?

> "Pawel Worach" <pworach@mysun.com> writes:
> 
> > I was using mpg123 (xmms and c/o does exactly the same)
> > if I run it like this Moby sounds very stupid... :)
> 
> i got the same problem when using mpg123 compiled with esd on my dell
> workstation (which has a need to have set explictely to a clocking of
> 41194 via ftsodell option), compiling without esd seems fix the prob
> for me.
> 
> -- 
> Chmouel
> 

[-- Attachment #2: Card for <pworach@mysun.com> --]
[-- Type: text/x-vcard, Size: 103 bytes --]

begin:vcard
n:Worach;Pawel
fn:Pawel Worach
version:2.1
email;internet:pawel.worach@mysun.com
end:vcard

Re: i810_audio broken?

[Pawel Worach]

[-- Attachment #1: Type: text/plain, Size: 1093 bytes --]

I tried that one to, got even worse,
with the -tcp -p 16001 and conigured xmms to use that one.

The funny thing is that when use start the esd server
it plays a fanfare and it sounds ok (maybe it's to
short to notice the buggy sound)

----- Original Message -----
From: Jeff Garzik <jgarzik@mandrakesoft.com>
Date: Tuesday, April 24, 2001 0:59 am
Subject: Re: i810_audio broken?

> Pawel Worach wrote:
> > 
> > Ok building mpg123 without eSound worked for me too,
> > so guess this is not a Linux kernel issue, sorry for this.
> > 
> > I tried the fstodell hack but it seems to be obsoluted.
> > Now it works without any tweaks.
> > 
> > eSound sux?
> 
> Are you guys running esd with any special arguments?
> 
> esd needs a special argument, -r RATE [iirc], in order to tell esd 
> thatit is dealing with a locked rate codec.
> 
> -- 
> Jeff Garzik      | The difference between America and England is that
> Building 1024    | the English think 100 miles is a long distance and
> MandrakeSoft     | the Americans think 100 years is a long time.
>                 |      (random fortune)
> 

[-- Attachment #2: Card for <pworach@mysun.com> --]
[-- Type: text/x-vcard, Size: 103 bytes --]

begin:vcard
n:Worach;Pawel
fn:Pawel Worach
version:2.1
email;internet:pawel.worach@mysun.com
end:vcard

Re: i810_audio broken?

[Jeff Garzik]

Pawel Worach wrote:
> 
> Ok building mpg123 without eSound worked for me too,
> so guess this is not a Linux kernel issue, sorry for this.
> 
> I tried the fstodell hack but it seems to be obsoluted.
> Now it works without any tweaks.
> 
> eSound sux?

Are you guys running esd with any special arguments?

esd needs a special argument, -r RATE [iirc], in order to tell esd that
it is dealing with a locked rate codec.

-- 
Jeff Garzik      | The difference between America and England is that
Building 1024    | the English think 100 miles is a long distance and
MandrakeSoft     | the Americans think 100 years is a long time.
                 |      (random fortune)

Re: i810_audio broken?

[Alan Cox]

> I was using mpg123 (xmms and c/o does exactly the same)
> if I run it like this Moby sounds very stupid... :)
> [root@whyami mp3]# mpg123 -r 48000 Moby_01.wav.mp3 
> unsupported playback rate: 44100
> Audio device open for 44.1Khz, stereo, 16bit failed
> Trying 44.1Khz, 8bit stereo.
> unsupported sound format: 32
> Audio device open for 44.1Khz, stereo, 8bit failed
> Trying 48Khz, 16bit stereo.

Ok so its trying to do the right thing. Can you describe what it sounds like
better ?

Re: i810_audio broken?

[Pawel Worach]

[-- Attachment #1: Type: text/plain, Size: 959 bytes --]

Ok i'll try :)

It's a little bit slow and the high tones get kind of very high
The first track of Moby Play sounds like you mixed Moby with
Donald Duck (but it not fast), like I said it's hard to describe sound
in text. :)

I can record it on another box and give You the result if You like? :)

----- Original Message -----
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tuesday, April 24, 2001 1:02 am
Subject: Re: i810_audio broken?

> > I was using mpg123 (xmms and c/o does exactly the same)
> > if I run it like this Moby sounds very stupid... :)
> > [root@whyami mp3]# mpg123 -r 48000 Moby_01.wav.mp3 
> > unsupported playback rate: 44100
> > Audio device open for 44.1Khz, stereo, 16bit failed
> > Trying 44.1Khz, 8bit stereo.
> > unsupported sound format: 32
> > Audio device open for 44.1Khz, stereo, 8bit failed
> > Trying 48Khz, 16bit stereo.
> 
> Ok so its trying to do the right thing. Can you describe what it 
> sounds like
> better ?
> 
> 

[-- Attachment #2: Card for <pworach@mysun.com> --]
[-- Type: text/x-vcard, Size: 103 bytes --]

begin:vcard
n:Worach;Pawel
fn:Pawel Worach
version:2.1
email;internet:pawel.worach@mysun.com
end:vcard

Re: i810_audio broken?

[David Woodhouse]

jgarzik@mandrakesoft.com said:
>  esd needs a special argument, -r RATE [iirc], in order to tell esd
> that it is dealing with a locked rate codec.

Isn't there an ioctl for that?

--
dwmw2

Re: i810_audio broken?

[Alan Cox]

> Ok building mpg123 without eSound worked for me too,
> so guess this is not a Linux kernel issue, sorry for this.

Excellent.

> eSound sux?

esound has very broken rate conversion support (it converts the audio but rather
damages it on the way). Gnome is moving towards using the KDE arts daemon which
currently seems to get it right.

Re: i810_audio broken?

[Chmouel Boudjnah]

"Pawel Worach" <pworach@mysun.com> writes:

> I was using mpg123 (xmms and c/o does exactly the same)
> if I run it like this Moby sounds very stupid... :)

i got the same problem when using mpg123 compiled with esd on my dell
workstation (which has a need to have set explictely to a clocking of
41194 via ftsodell option), compiling without esd seems fix the prob
for me.

-- 
Chmouel

Re: i810_audio broken?

[Doug Ledford]

Chmouel Boudjnah wrote:
> 
> "Pawel Worach" <pworach@mysun.com> writes:
> 
> > I was using mpg123 (xmms and c/o does exactly the same)
> > if I run it like this Moby sounds very stupid... :)
> 
> i got the same problem when using mpg123 compiled with esd on my dell
> workstation (which has a need to have set explictely to a clocking of
> 41194 via ftsodell option), compiling without esd seems fix the prob
> for me.

The latest i810 driver does away with the ftsodell option entirely and should
work on your laptop without having to do anything special.

-- 

 Doug Ledford <dledford@redhat.com>  http://people.redhat.com/dledford
      Please check my web site for aic7xxx updates/answers before
                      e-mailing me about problems