Hi,
 after some effort, the solution..

Given that I posed this question a while ago, and now have the answer, 
I felt it could help some to have the answer.

Gave up on OSS.. This has it's origins from pre 1.0 kernel days. The API 
designer/coders etc had no concept of someone wanting 6 sound cards in a 
machine.

In the end, I saw no option but to use Alsa.
Alsa did not work immediately. There were buglets in the sample program.
The alsa website is a mass(mess) of information, which requires some 
effort to follow.

My conclusion out of the alsa code, 
   Alsa programming is not for the fainthearted. QT was easier. Much easier.

Derek.
===========================================================================

On Fri, 8 Oct 2004, Synco Reynders wrote:

> Hi,
>   Something that's tripped me up once when scaling up a PCI system like what 
> you're doing, yet completely unrelated to sound cards is this weird option in 
> the BIOS. Sorry can't be anymore specific, but check your PCI bios 
> settings... (and hopefully it'll jump out at you)
> /s
> 
> On Fri, 08 Oct 2004 15:11, Derek Smithies wrote:
> > Hi,
> >  I have installed 6 sb live cards on a redhat 9.0 box.
> >    - the box has had the kernel upgraded to 2.4.26
> >
> > redhat-config-soundcard runs fine, and reports all 6 cards.
> >
> > redhat-config-soundcard has an option to play a test sound out the sound
> > card. This only worked for the first card..
> >
> >
> > I would appreciate some suggestions on gettting it to go.
> > Below is a list of my thoughts and meanderings. Apologies for the length.
> >
> >
> > ==================================================================
> >
> > Yet, there are no extra devices added to the devices list
> >   - thus, it failed to create /dev/dsp2, /dev/dsp3, dev/dsp4
> >                               /dev/sound2  etc..
> >
> > Ok, examine the python sources of redhat-config-soundcard
> >   no instances of mknod or proc were found
> >
> > Google. Many helpful suggestions, such as switch to alsa.
> >      Not an option.
> > Google did suggest to create the requisite devices.
> >   fair enough, redhat-config-soundcard does not appear to have
> >    (in the source code) the ability to create the requisite devices
> >
> > Manually create (with mknod) /dev/dsp2, /dev/dsp3
> >                             /dev/mixer2, /dev/mixer3  etc....
> > ls -l /dev/  | grep \ 14,
> > crw-------    1 root     root      14,   4 Jan 30  2003 audio
> > crw-------    1 root     root      14,  20 Jan 30  2003 audio1
> > crw-------    1 root     root      14,   7 Jan 30  2003 audioctl
> > crw-------    1 root     root      14,   3 Jan 30  2003 dsp
> > crw-------    1 root     root      14,  19 Jan 30  2003 dsp1
> > crw-------    1 root     root      14,  35 Oct  6 04:41 dsp2
> > crw-------    1 root     root      14,  51 Oct  6 04:41 dsp3
> > crw-------    1 root     root      14,  67 Oct  6 04:42 dsp4
> > crw-------    1 root     root      14,  83 Oct  6 04:42 dsp5
> > crw-------    1 root     root      14,  97 Oct  6 23:18 dsp6
> > crw-------    1 root     root      14,   2 Jan 30  2003 midi00
> > crw-------    1 root     root      14,  18 Jan 30  2003 midi01
> > crw-------    1 root     root      14,  34 Jan 30  2003 midi02
> > crw-------    1 root     root      14,  50 Jan 30  2003 midi03
> > crw-------    1 root     root      14,   0 Jan 30  2003 mixer
> > crw-------    1 root     root      14,  16 Jan 30  2003 mixer1
> > crw-------    1 root     root      14,  32 Oct  6 17:59 mixer2
> > crw-------    1 root     root      14,  48 Oct  6 17:59 mixer3
> > crw-------    1 root     root      14,  64 Oct  6 18:00 mixer4
> > crw-------    1 root     root      14,  80 Oct  6 23:18 mixer5
> > crw-------    1 root     root      14,  96 Oct  6 23:18 mixer6
> > crw-------    1 root     root      14,  17 Jan 30  2003 patmgr0
> > crw-------    1 root     root      14,  33 Jan 30  2003 patmgr1
> > crw-------    1 root     root      14,   1 Jan 30  2003 sequencer
> > crw-------    1 root     root      14,   8 Jan 30  2003 sequencer2
> > crw-------    1 root     root      14,   6 Jan 30  2003 sndstat
> >
> >
> > however, my code to examine all the available sound devices does not work.
> > It found just 5 devices, not 6.
> > well, examine source for the driver, es1371.c
> >   #define NR_DEVICES 5
> >
> >  Change this, recompile and install kernel..
> > Nope - did not help
> > kernel messages that may be of interest::
> >
> > Tests on sound working are a little hard. That box is in America.
> > Although, I have a report that redhat-config-soundcard play test sound
> > does play sound fine for the first card, but not the others...
> >
> > Kernel log messages::
> > PCI: Found IRQ 9 for device 01:05.0
> > PCI: Sharing IRQ 9 with 00:1d.2
> > PCI: Sharing IRQ 9 with 00:1f.1
> > PCI: Sharing IRQ 9 with 01:02.0
> > es1371: found chip, vendor id 0x1274 device id 0x5880 revision 0x04
> > es1371: found es1371 rev 4 at io 0xd080 irq 9
> > es1371: features: joystick 0x0
> > ac97_codec: AC97 Audio codec, id: 0x8384:0x7608 (SigmaTel STAC9708)
> > PCI: Found IRQ 3 for device 01:04.0
> > PCI: Sharing IRQ 3 with 00:1f.3
> > es1371: found chip, vendor id 0x1274 device id 0x5880 revision 0x04
> > es1371: found es1371 rev 4 at io 0xd400 irq 3
> > es1371: features: joystick 0x0
> > ac97_codec: AC97 Audio codec, id: 0x8384:0x7608 (SigmaTel STAC9708)
> > PCI: Found IRQ 5 for device 01:03.0
> > PCI: Sharing IRQ 5 with 00:1d.1
> > es1371: found chip, vendor id 0x1274 device id 0x5880 revision 0x04
> > es1371: found es1371 rev 4 at io 0xd480 irq 5
> > es1371: features: joystick 0x0
> > ac97_codec: AC97 Audio codec, id: 0x8384:0x7608 (SigmaTel STAC9708)
> > PCI: Found IRQ 9 for device 01:02.0
> > PCI: Sharing IRQ 9 with 00:1d.2
> > PCI: Sharing IRQ 9 with 00:1f.1
> > PCI: Sharing IRQ 9 with 01:05.0
> > es1371: found chip, vendor id 0x1274 device id 0x5880 revision 0x04
> > es1371: found es1371 rev 4 at io 0xd800 irq 9
> > es1371: features: joystick 0x0
> > ac97_codec: AC97 Audio codec, id: 0x8384:0x7608 (SigmaTel STAC9708)
> > PCI: Found IRQ 11 for device 01:01.0
> > es1371: found chip, vendor id 0x1274 device id 0x5880 revision 0x04
> > es1371: found es1371 rev 4 at io 0xd880 irq 11
> > es1371: features: joystick 0x0
> > es1371: cannot register misc device
> > PCI: Found IRQ 11 for device 01:00.0
> > es1371: found chip, vendor id 0x1274 device id 0x5880 revision 0x04
> > es1371: found es1371 rev 4 at io 0xdc00 irq 11
> > es1371: features: joystick 0x0
> > es1371: cannot register misc device
> >
> > See - the last two card did not go in properly. And, they were
> > all created as character devices. (See the sound dev listing)
> >
> > Try another box.
> > just put 2 SB16 cards into a redhat 9 box with 2.4.26 kernel.
> >
> > Create a couple of extra /dev entries for the new cards.
> > Many mesages about
> > Can't locate module sound-service-0-0
> > Can't locate module sound-service-1-0
> >
> > lspci -n - have a look at the device id numbers.
> >   Ahh, 1274 - ensoniq.
> >    5880 is handled by the es1371 driver - there is a comment to
> > explicitely turn on the AC97 if we are a 5880.
> >
> >
> >
> > Gack - anyone with suggestions on where to look next ?
> >
> >
> > Derek.
> 
> 
> 

-- 
Derek Smithies Ph.D.                           This PC runs pine on linux for 
email
IndraNet Technologies Ltd.                     If you find a virus apparently 
from me, it has
Email: [EMAIL PROTECTED]                    forged  the e-mail headers on 
someone else's machine
ph +64 3 365 6485                              Please do not notify me when 
(apparently) receiving a
Web: http://www.indranet-technologies.com/     windows virus from me......

Reply via email to