Created attachment 6964 [details]
2.6.11 dmesg and config (device is working)

Created attachment 6965 [details]
2.6.15 dmesg and config (device not works)

Created attachment 6980 [details]
2.6.12.5 dmesg and config (device is working)

Created attachment 6981 [details]
2.6.13 demsg an config (device not works)

Te bug seems to be related to the new version of bttv (0.9.15 to 0.9.16) 
The change occur AFAIK from 2.6.12.5 to 2.6.13 
 
First impression: 
Or the new version of bttv dont detect btaudio properly  
Or ALSA don't use well "a posible" API change of btaudio 

btaudio is OSS. What happens with the alsa module (snd-87x)?

Sorry for misunderstood, I don't know if there are btaudio in OSS.
I don't use OSS, The dmesg and config don't show OSS
I use snd-87x (ALSA) but compiled in kernel I pass the following parameters
to boot (like I showed in attach):
root=/dev/sda2 ro bttv.tuner=5 bttv.card=70 bttv.radio=1 bttv.bttv_gpio=1
msp3400.once=1 snd_bt87x.enable=1 snd_intel8x0.index=0 snd_bt87x.index=1
snd_virmidi.index=2 snd_bt87x.load_all=1 snd_bt87x.digital_rate=48000 

I assign the index number 1 to snd_bt87x and snd_bt87x.load_all=1 to try all
models (i wonder if not was docummented last one)...

Now I have and idea in 2.6.12.5, I will try snd_bt87x.load_all=0 
Wait a minute ...

-----
bttv: driver version 0.9.15 loaded
bttv: using 8 buffers with 2080k (520 pages) each for capture
bttv: Bt8xx card found (0).
bttv0: Bt878 (rev 2) at 0000:04:00.0, irq: 16, latency: 64, mmio: 0xd0001000
bttv0: using: Prolink Pixelview PV-BT878P+ (Rev.4C,8E) [card=70,insmod option]
bttv0: gpio: en=00000000, out=00000000 in=00ffffff [pre-init]
bttv0: gpio: en=00000000, out=00000000 in=00ffffff [init]
bttv0: using tuner=5
bttv0: registered device video0
bttv0: registered device vbi0
bttv0: registered device radio0
bttv0: gpio: en=0000000f, out=0000000d in=00fffff0 [audio: off]
bttv0: gpio: en=0000000f, out=0000000d in=00fffff0 [audio: off]
bttv0: PLL: 28636363 => 35468950 .. ok
bttv0: add subdevice "remote0"
tvaudio: TV audio decoder + audio/video mux driver
tvaudio: known chips: tda9840,tda9873h,tda9874h/a,tda9850,tda9855,tea6300,tea632
0,tea6420,tda8425,pic16c54 (PV951),ta8874z
bttv0: gpio: en=0000000f, out=0000000d in=00fffff0 [audio: off]
ir-kbd-gpio: bttv IR (card=70) detected at pci-0000:04:00.0/ir0
tuner 0-0061: chip found @ 0xc2 (bt878 #0 [sw])
tuner 0-0061: type set to 5 (Philips PAL_BG (FI1216 and compatibles))

....
....

intel8x0_measure_ac97_clock: measured 49229 usecs
intel8x0: clocking to 48000
ALSA device list:
  #0: Intel ICH6 with AD1981B at 0xdfebfe00, irq 23


------

As I supossed snd_bt87x is not detected like 2.6.13 and higher.

summup:
if snd_bt87x.load_all=1 my snd_bt87x is detected.
if snd_bt87x.load_all=0 is not detected like 2.6.13 and higher.

maybe my chipset is blacklisted?

Regards

seeing changelog in 2.6.13

-------------
commit 4dcef52400fa6b9eb2de589300ae0151a1c65b3b
Author: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Date:   Thu Aug 4 12:53:30 2005 -0700

    [PATCH] v4l: oopsfix for BTTV on badly behaved PCI chipsets

    no_overlay bttv parameter implemented to fix OOPS on some PCI chipsets
    (like some VIA) with these behaviors:

    1) If pci_quicks does identify the chip as having troubles to
       handle PCI2PCI transfers, no_overlay defaults to 1. The user may force
       it to 0, to reenable (not recommended).

    2) For newer chipsets not blacklisted, no_overlay=1 is provided as a
       workaround until PCI chipset included on /drivers/pci/quirks.c
---------------

and seeing patch 2.6.13


-----------
snd_bt87x

-       supported = pci_match_device(snd_bt87x_ids, pci);
+       supported = pci_match_device(&driver, pci);

-----------

in 2.6.15 sound/pci/bt87x.c 
at static int __devinit snd_bt87x_detect_card(struct pci_dev *pci)
---------
        /* Add to debug */
        printk(KERN_NOTICE "juampe debug card %#04x:%#04x-%#04x:%#04x
\n",pci->vendor, pci->device,pci->subsystem_vendor, pci->subsystem_device);
        if (supported)
                return supported->driver_data;
        printk(KERN_NOTICE "juampe not supported\n");

-----------
Advanced Linux Sound Architecture Driver Version 1.0.10rc3 (Mon Nov 07 13:30:21
2005 UTC).
juampe debug card 0x109e:0x878-0xffff:0xffff
ACPI: PCI Interrupt 0000:00:1e.2[A] -> GSI 23 (level, low) -> IRQ 22
PCI: Setting latency timer of device 0000:00:1e.2 to 64
intel8x0_measure_ac97_clock: measured 58344 usecs
intel8x0: clocking to 48000
ALSA device list:
  #0: Intel ICH6 with AD1981B at 0xdfebfe00, irq 22
  #2: Virtual MIDI Card 1
--------

The card is supported in 2.6.15 but not showed by ALSA device list
I continue following the trhead ...

Regards.

Now i try with probe method....
--------------
static int __devinit snd_bt87x_probe(struct pci_dev *pci,
                                     const struct pci_device_id *pci_id)
.....
.....
        rate = pci_id->driver_data;
        if (! rate)
                if ((rate = snd_bt87x_detect_card(pci)) <= 0){
                        printk(KERN_NOTICE "juampe tring to get rate %d\n",rate);
                        return -ENODEV;
                }
----------------



And i got
--------

Advanced Linux Sound Architecture Driver Version 1.0.10rc3 (Mon Nov 07 13:30:21
2005 UTC).
juampe tring to get rate 0
ACPI: PCI Interrupt 0000:00:1e.2[A] -> GSI 23 (level, low) -> IRQ 22
PCI: Setting latency timer of device 0000:00:1e.2 to 64
intel8x0_measure_ac97_clock: measured 58331 usecs
intel8x0: clocking to 48000
ALSA device list:
  #0: Intel ICH6 with AD1981B at 0xdfebfe00, irq 22
  #2: Virtual MIDI Card 1

---------

rate is 0?
goto again to snd_bt87x_detect_card

supported = pci_match_device(&driver, pci);
        if (supported)
                return supported->driver_data;

and the struct
static struct pci_driver driver;

now i will follow this struct...

Regards

OK, after several probes I get the device working like is expected.  
The problem is at ALSA code in bt87x.c  
I have to force default rate to 48000 couse the detection does not works  
snd_bt87x_detect_card(pci) returns always 0  
 
  
My work is the following-----  
--- linux-2.6.15/sound/pci/bt87x.c      2006-01-03 04:21:10.000000000 +0100  
+++ linux-2.6.15/sound/pci/bt87x.c.new  2006-01-12 08:32:20.000000000 +0100  
@@ -831,7 +831,7 @@  
        if (! rate)  
                if ((rate = snd_bt87x_detect_card(pci)) <= 0)  
                        return -ENODEV;  
-  
+       printk(KERN_NOTICE "juampe rate after force to %d",rate);  
        if (dev >= SNDRV_CARDS)  
                return -ENODEV;  
        if (!enable[dev]) {  
@@ -881,6 +881,7 @@  
  
        pci_set_drvdata(pci, card);  
        ++dev;  
+       printk(KERN_NOTICE "juampe Probe is OK");  
        return 0;  
  
 _error:  
@@ -897,8 +898,8 @@  
 /* default entries for all Bt87x cards - it's not exported */  
 /* driver_data is set to 0 to call detection */  
 static struct pci_device_id snd_bt87x_default_ids[] = {  
-       BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_878, PCI_ANY_ID, PCI_ANY_ID, 0),  
-       BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_879, PCI_ANY_ID, PCI_ANY_ID, 0),  
+       BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_878, PCI_ANY_ID, PCI_ANY_ID, 48000),  
+       BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_879, PCI_ANY_ID, PCI_ANY_ID, 48000),  
        { }  
 };  
  
-----  
result  
  
  
-----  
Advanced Linux Sound Architecture Driver Version 1.0.10rc3 (Mon Nov 07 13:30:21  
2005 UTC).  
juampe rate after force to 48000<6>ACPI: PCI Interrupt 0000:04:00.1[A] -> GSI  
16  
 (level, low) -> IRQ 16  
juampe Probe is OK<6>ACPI: PCI Interrupt 0000:00:1e.2[A] -> GSI 23 (level, low)  
-> IRQ 22  
PCI: Setting latency timer of device 0000:00:1e.2 to 64  
intel8x0_measure_ac97_clock: measured 58333 usecs  
intel8x0: clocking to 48000  
ALSA device list:  
  #0: Intel ICH6 with AD1981B at 0xdfebfe00, irq 22  
  #1: Brooktree Bt878 at 0xd0000000, irq 16  
  #2: Virtual MIDI Card 1  
  
  
;) the device is now working,  
but the probe code remain failing and return 0.  
  
Regards.  

Created attachment 6994 [details]
Patch to test

Could you test this patch? It should fix your problem.

OK, your patch is more ellegant an efficient :P 
It works great, good fix! 
 
---- 
Advanced Linux Sound Architecture Driver Version 1.0.10rc3 (Mon Nov 07 13:30:21 
2005 UTC). 
unknown card 0xffff:0xffff, using default rate 32000 
please mail id, board name, and, if it works, the correct digital_rate option 
to <alsa-devel@lists.sf.net> 
ACPI: PCI Interrupt 0000:04:00.1[A] -> GSI 16 (level, low) -> IRQ 16 
ACPI: PCI Interrupt 0000:00:1e.2[A] -> GSI 23 (level, low) -> IRQ 22 
PCI: Setting latency timer of device 0000:00:1e.2 to 64 
intel8x0_measure_ac97_clock: measured 58342 usecs 
intel8x0: clocking to 48000 
ALSA device list: 
  #0: Intel ICH6 with AD1981B at 0xdfebfe00, irq 22 
  #1: Brooktree Bt878 at 0xd0000000, irq 16 
  #2: Virtual MIDI Card 1 
---- 
 
Regards 

Ok, I'm closing this bug. The patch is in ALSA CVS thus it will be exported to
mainstream (2.6.16).