Distribution: Debian sid, but using vanilla kernel Hardware Environment: ECS k7s5a motherboard with onboard sound, Athlon 1800+, 512MB mem Problem Description: modprobing the snd-intel8x0 crashes my computer There is nothing useful in the logs and no screen output I'll attach a lspci dump; the relevant pci device is 00:02.7
Created attachment 2925 [details] lspci output Relevant pci device is 00:02.7
I did some more debugging: I have been able to trace to lockup to the snd_ac97_read(ac97, AC97_VENDOR_ID1) call in ac97_reset_wait() in ac97_codec.c. The call trace is: snd_intel8x0_probe() --> snd_intel8x0_mixer() --> snd_ac97_mixer() --> ac97_reset_wait() Hope this helps. Please let me know if some more info is needed.
In mail exchange Bas tells me this was resolved in 2.6.7, closing.
I think you misread my message. I meant to say that iirc, the bug is still there in 2.6.7
You are correct, sorry about that. Upgrading bug to 2.6.7 and giving back to correct owner. Jaroslav, do you know if this has been resolved?
Could you add some additional printk lines to snd_intel8x0_codec_semaphore() and snd_intel8x0_codec_read() routines in sound/pci/intel8x0.c? I need to know the exact lockup location. Also, you may try to compile ALSA 1.0.7 driver package against the 2.6 tree.
I'll try, but it could take me a while as I recently upgraded my motherboard. The one triggering the bug is now disconnecteed and cpuless.
I bought this motherboard from Bas. It seems to have some weird IRQ routing problems. Loading the de4x5 NIC driver caused a hard lockup. The last messages on the console were IRQ related. Disabling ACPI didn't help. Putting the card in another PCI slot also don't help. In the end I swapped the NIC with a rtl8139 card with 8139too driver. This one seems to work. I'll try to load the sound driver later on.
Erik, what is the status of this issue?
The motherboard broke down, so I can't test anymore :/