On Mon, 2008-06-16 at 18:00 +0200, Takashi Iwai wrote:

At Tue, 17 Jun 2008 01:52:08 +1000, Travis Place wrote:

...

On Mon, 2008-06-16 at 12:38 +0200, Takashi Iwai wrote:

...

At Sat, 14 Jun 2008 03:18:33 +1000, Travis Place wrote:

...

Just been speaking with a guy in #alsa about the problems with his "MSI K8N Diamond MB" which has an onboard "SB Live 24bit" or "Audigy".

While looking through ca0106_main.c in GIT, i saw the following section:

/* MSI K8N Diamond Motherboard with onboard SB Live 24bit without AC97 */ /* SB0438 * CTRL:CA0106-DAT * ADC: WM8775SEDS * DAC: CS4382-KQZ */ { .serial = 0x10091462, .name = "MSI K8N Diamond MB [SB0438]", .gpio_type = 2, .i2c_adc = 1 } ,

For this guys sound to work, we had to change the serial line to:

{ .serial = 0x10091102,

Does it mean that this number 1009:1102 is missing, or a wrong 1009:1462 entry matches with his hardware?

Well, it appears to me (due to another report filed previously) that the motherboard is a 'newer revision' and has a different ID for the soundcard.

Hm, so you didn't receive alsa-info.sh or "lspci -nv" output?

Of course I got alsa-info.sh output :) (asks himself why he didnt include it though)

http://pastebin.ca/1048644

...

...

In the former case, it happens often. We just need to add a new entry for the newer model. In the latter case, it's odd.

I think that adding the new ID to the driver wont cause any immediate problems. The old value (1009:1462) should remain, until we can confirm it against another K8N Diamond motherboard (although im not sure how common these boards are).

Yeah, I guess so, too. But just to be sure...

Well, we'll just have to see what happens, I guess.

For the record, in both older ALSA versions, and 1.0.16, changing only the serial seems to be sufficient. So adding another serial should be an easy fix.

Thanks, Travis

Takashi