Summary: | Toshiba P100, Nosound from speakers after upgrading to gentoo-sources-2.6.23 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Luís Costa <luisfcosta> |
Component: | [OLD] Core system | Assignee: | Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel> |
Status: | RESOLVED FIXED | ||
Severity: | major | CC: | endgame.dos |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | x86 | ||
OS: | Linux | ||
Whiteboard: | linux-2.6.24 | ||
Package list: | Runtime testing required: | --- | |
Attachments: |
Kernel 2.6.22 info
Kernel 2.6.23 info |
Description
Luís Costa
2007-12-04 21:02:23 UTC
Your USB headphones and bluetooth devices do not use the sound card. It sounds like the sound card is not working under these kernels. Please post output of "cat /proc/asound/cards" and attach dmesg from a working 2.6.22 kernel and a broken 2.6.23 kernel. Created attachment 137743 [details]
Kernel 2.6.22 info
Created attachment 137745 [details]
Kernel 2.6.23 info
Thank you for the quick response. I have attached the info you requested. I had never used bugzilla's attachments before. Should I use something else? What arguments to you pass to the kernel in grub.conf? For 2.6.22 kernels I use acpi_osi=!Linux rather than editing the DSDT but this also stops working on 2.6.23-gentoo{,-r1,-r2,-r3}. Jack, I don't pass any arguments at all with 2.6.22. It just worked. However, I have tried adding "acpi_osi=!Linux" to the 2.6.23kernel (I read it somewhere, but it still didn't work. Something must have changed in 2.6.23. Just in case it helps complete the picture: My Toshiba Satellite P100-j01 would have no sound on some 2.6 kernels unless I replaced the DSDT, which I did until 2.6.22 when I discovered acpi_osi=!Linux. If I did neither the sound wouldn't work at all. I've only tried acpi_osi=!Linux on 2.6.23 (I can no longer find the patch for the 3.30 DSDT), but no luck. I have very limited understanding of what it needs to be done in order to correctly patch a DSDT. That said, I tried fixing mine, at least the parts that were supposedly connected to the sound card, but the probem still persists. It is highly likely that I did something wrong, but the iasl compiler didn't complain about any of the things I changed- just a few warnings. The thing that intrigues me here, is the fact that I didn't have to add anything like "acpi_osi=!Linux" during the 2.6.22* series, which means that somehow the kernel was able to bypass the bugged DSDT, or at least ignore it. I'm keeping 2.6.22 for the time being, as sound is a major feature for me. I have the same problem running 2.6.23-tuxonice-r4 on a Toshiba Satego P100-10F (other name for Satellite). Interestingly, I already fixed my DSDT so that it builds without errors _or_ warnings, and it still doesn't work. acpi=off "fixes" it, but I kinda need my acpi :) Still no progress regarding this matter? I am looking forward to going from 2.6.22 to 2.6.23. boroshan on f.g.o has claimed that this is fixed on 2.6.24 (http://forums.gentoo.org/viewtopic.php?p=4795322). Can someone confirm? If this is true, is it worth closing this bug? Yes, I can confirm the bug has been fixed in kernel 2.6.24. I have the alsa-drivers built in and I didn't need any DSDT patch. Great great work :D! Thanks for your testing efforts, everyone. I'm closing this as we expect to be stabilizing the 2.6.24 version of gentoo-sources sometime this week. |