Thank you, alsa developers, for the prompt resolution of my previous sound issue with the Thinkpad X61. Unfortunately, I've now turned up an apparently different problem on very similar hardware.
Sound doesn't work on my DQ965GF motherboard under Ubuntu 7.10. Here's what I've found out about it. I'm writing in hope that someone on the ALSA devel list can suggest diagnostics to pin down the problem exactly.
The device has PCI ID [8086:284b] (rev 02). It is an Intel 82801H.
This appears to be an Ubuntu-specific error. At http://www.linuxquestions.org/questions/linux-hardware-18/intel-dq965gf-mobo... I found a post implying that at least the rear jacks on the mobo are active under Fedora Core 6. I have therefore filed an Ubuntu bug at https://bugs.launchpad.net/ubuntu/+bug/162410 but seen no response.
This also appears to be a very hardware-configuration-specific error. I have an Intel HDA audio device on my X61 laptop with PCI ID [8086:284b] (rev 03), another 82801H, that works just fine under Ubuntu 7.10 (the alsa version is nominally 1.0.14 but some 1.10.15 fixes, including the right one, were backported). Note the 03 as opposed to 02 subrev.
aplay -l, which is supposed to list sound playback devices, returns no listing on the DQ965GF. As a check, on the X61 where it works it returns this:
card 0: Intel [HDA Intel], device 0: AD198x Analog [AD198x Analog] Subdevices: 1/1 Subdevice #0: subdevice #0
lsmod also returns suggestively different results. On the laptop, its output includes this line.
snd_hda_intel 337192 1
On the DQ965GF, here are the sound device lines:
snd_hda_intel 337192 0 snd_usb_audio 96640 1 snd_pcm_oss 50048 0 snd_mixer_oss 20096 1 snd_pcm_oss snd_pcm 94344 3 snd_hda_intel,snd_usb_audio,snd_pcm_oss snd_usb_lib 20352 1 snd_usb_audio snd_seq_dummy 5380 0 snd_seq_oss 36864 0 snd_seq_midi 11008 0 snd_seq_midi_event 9984 2 snd_seq_oss,snd_seq_midi snd_seq 62496 6 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq_midi_event snd_timer 27272 2 snd_pcm,snd_seq snd_rawmidi 29824 2 snd_usb_lib,snd_seq_midi snd_seq_device 10260 5 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq,snd_rawmidi snd_hwdep 12168 1 snd_usb_audio snd 69288 13 snd_hda_intel,snd_usb_audio,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_seq_oss,snd_seq,snd_timer,snd_rawmidi,snd_seq_device,snd_hwdep soundcore 10272 1 snd snd_page_alloc 12560 2 snd_hda_intel,snd_pcm usbcore 161584 8 usbhid,snd_usb_audio,snd_usb_lib,xpad,gspca,ehci_hcd,uhci_hcd
Note that snd_hda_intel has no users and that snd_usb_audio has 1.
On the DQ965GF, the Sound Preferences tab on my Gnome toolbar thinks the sound device is "USB Device 0x46d:0x8ad (ALsa mixer)". I don't know what tool GNOME is running to determine this.
This arch-linux thread gave me some insight: http://bugs.archlinux.org/task/6617
It appears the snd_usb_audio driver is a red herring, or at least a separate bug. I checked, and my dmesg log does indeed say "hda-intel: no codecs found!" on boot, which suggests that snd_hda_intel is loading and then failing to initialize properly, accounting for the usage count of 0.
I tried reverting the kernel patch mentioned in the arch thread (e.g. recompiling the snd_hda_intel driver with #define AZX_MAX_CODECS 4) but this did *not* solve the problem.
Any suggestions would be appreciated. I need my music back!