[arch-general] [signoff] kernel26 2.6.28.1-1
Hi guys, new kernel adresses the following things: - bump to latest version - fixed bugs: http://bugs.archlinux.org/task/12821 http://bugs.archlinux.org/task/12816 http://bugs.archlinux.org/task/12820 please signoff for both arches greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa@archlinux.org
On Sun, Jan 18, 2009 at 3:25 PM, Tobias Powalowski <t.powa@gmx.de> wrote:
Hi guys, new kernel adresses the following things: - bump to latest version - fixed bugs: http://bugs.archlinux.org/task/12821 http://bugs.archlinux.org/task/12816 http://bugs.archlinux.org/task/12820
please signoff for both arches
greetings tpowa --
Systems booted and everything seems to be running fine. Signoff both arches. Eric
Tobias Powalowski wrote:
Hi guys, new kernel adresses the following things: - bump to latest version - fixed bugs: http://bugs.archlinux.org/task/12821 http://bugs.archlinux.org/task/12816 http://bugs.archlinux.org/task/12820
please signoff for both arches
greetings tpowa
FYI - I'm seeing a problem with the 2.6.28-3 kernel. For some reason there's no sound coming out of either the speakers or the headphones on my laptop (Dell Precision M4400) after the upgrade. Sound module is snd-hda-intel. I checked volume is up, mute is not on, yada yada, but no sound coming out. Problem goes away as soon as I downgrade the kernel. Anyone else seeing this and/or have any idea what the problem might be? Thanks, DR
2009/1/19 David Rosenstrauch <darose@darose.net>:
FYI - I'm seeing a problem with the 2.6.28-3 kernel. For some reason there's no sound coming out of either the speakers or the headphones on my laptop (Dell Precision M4400) after the upgrade.
Sound module is snd-hda-intel. I checked volume is up, mute is not on, yada yada, but no sound coming out. Problem goes away as soon as I downgrade the kernel.
Anyone else seeing this and/or have any idea what the problem might be?
Maybe there's a regression in your soundcard autodetection? Do you pass any "model=" parameter to snd-hda-intel in /etc/modprobe.conf? Anyway, there have been a couple of commits in that driver, if you want you can track them searching for "alsa: hda" in the changelog (http://kernel.org/pub/linux/kernel/v2.6/ChangeLog-2.6.28.1). Corrado
bardo wrote:
2009/1/19 David Rosenstrauch <darose@darose.net>:
FYI - I'm seeing a problem with the 2.6.28-3 kernel. For some reason there's no sound coming out of either the speakers or the headphones on my laptop (Dell Precision M4400) after the upgrade.
Sound module is snd-hda-intel. I checked volume is up, mute is not on, yada yada, but no sound coming out. Problem goes away as soon as I downgrade the kernel.
Anyone else seeing this and/or have any idea what the problem might be?
Maybe there's a regression in your soundcard autodetection? Do you pass any "model=" parameter to snd-hda-intel in /etc/modprobe.conf? Anyway, there have been a couple of commits in that driver, if you want you can track them searching for "alsa: hda" in the changelog (http://kernel.org/pub/linux/kernel/v2.6/ChangeLog-2.6.28.1).
Corrado
Thanks for the info. No, I don't pass any model parameter, partly because I didn't know about this parm until recently, and partly because I'm not sure what the correct value should be. ("head -n 1 /proc/asound/card0/codec#0" says "IDT 92HD71B7X", but I'm not sure where to go to look up the model number from that.) Will check the change log and see what's of relevance. Thanks, DR
2009/1/19 David Rosenstrauch <darose@darose.net>:
No, I don't pass any model parameter, partly because I didn't know about this parm until recently, and partly because I'm not sure what the correct value should be. ("head -n 1 /proc/asound/card0/codec#0" says "IDT 92HD71B7X", but I'm not sure where to go to look up the model number from that.)
hda-intel, like the infamous ac97, is a widely-used codec, and the kernel module supports a whole lot of different devices. I don't have a lot of time to check right now, but a quick grep on the sources shows your card's support is included in sound/pci/hda/patch_sigmatel.c. From a quick search, the list of possible options can be found here: http://free-electrons.com/kerneldoc/latest/sound/alsa/ALSA-Configuration.txt and your best bet could be dell-m44 (similar to 4400, but marked as Inspiron) or dell-m43 (near to 44 but marked as Precision). HTH, Corrado
bardo wrote:
2009/1/19 David Rosenstrauch <darose@darose.net>:
No, I don't pass any model parameter, partly because I didn't know about this parm until recently, and partly because I'm not sure what the correct value should be. ("head -n 1 /proc/asound/card0/codec#0" says "IDT 92HD71B7X", but I'm not sure where to go to look up the model number from that.)
hda-intel, like the infamous ac97, is a widely-used codec, and the kernel module supports a whole lot of different devices. I don't have a lot of time to check right now, but a quick grep on the sources shows your card's support is included in sound/pci/hda/patch_sigmatel.c. From a quick search, the list of possible options can be found here: http://free-electrons.com/kerneldoc/latest/sound/alsa/ALSA-Configuration.txt and your best bet could be dell-m44 (similar to 4400, but marked as Inspiron) or dell-m43 (near to 44 but marked as Precision).
HTH, Corrado
Hmmm .... IDT 92HD71B7X actually sounds like it matches more closely to one of these: STAC92HD71B* ref Reference board dell-m4-1 Dell desktops dell-m4-2 Dell desktops dell-m4-3 Dell desktops I'll check them out - thanks. DR
David Rosenstrauch wrote:
bardo wrote:
hda-intel, like the infamous ac97, is a widely-used codec, and the kernel module supports a whole lot of different devices. I don't have a lot of time to check right now, but a quick grep on the sources shows your card's support is included in sound/pci/hda/patch_sigmatel.c. From a quick search, the list of possible options can be found here: http://free-electrons.com/kerneldoc/latest/sound/alsa/ALSA-Configuration.txt
and your best bet could be dell-m44 (similar to 4400, but marked as Inspiron) or dell-m43 (near to 44 but marked as Precision).
HTH, Corrado
Hmmm .... IDT 92HD71B7X actually sounds like it matches more closely to one of these:
STAC92HD71B* ref Reference board dell-m4-1 Dell desktops dell-m4-2 Dell desktops dell-m4-3 Dell desktops
I'll check them out - thanks.
DR
Hmmm ... I tried testing this again, loading the module with each of the dell-m4-* model parameters, in turn, but no joy. *Some*thing's broken here. But I'm not enough of a kernel hacker to be able to figure out what or how to fix it. There's both a kernel bug and an ALSA bug about this; I guess I'll follow up there. DR
On 19 Jan 2009, at 10:26 am -0500, David Rosenstrauch wrote:
For some reason there's no sound coming out of either the speakers or the headphones on my laptop (Dell Precision M4400) after the upgrade.
Sound module is snd-hda-intel. I checked volume is up, mute is not on, yada yada, but no sound coming out.
I had a similar problem a couple of days ago. Is the module snd_pcsp getting loaded? If so, it's possible it may be grabbing your sound card before your real driver has a chance to do so. The fix that worked for me was just to blacklist it. I believe there's also a way to make it load after the correct sound modules (if you like the audible bell), which some judicious googling should turn up. Good luck getting that fixed! Ivy -- If I Ever Become An Evil Villainess... 20. Despite its proven stress-relieving effect, I will not indulge in maniacal laughter. When so occupied, it's too easy to miss unexpected developments that a more attentive individual could adjust to accordingly.
joyfulgirl@archlinux.us wrote:
On 19 Jan 2009, at 10:26 am -0500, David Rosenstrauch wrote:
For some reason there's no sound coming out of either the speakers or the headphones on my laptop (Dell Precision M4400) after the upgrade.
Sound module is snd-hda-intel. I checked volume is up, mute is not on, yada yada, but no sound coming out.
I had a similar problem a couple of days ago. Is the module snd_pcsp getting loaded? If so, it's possible it may be grabbing your sound card before your real driver has a chance to do so. The fix that worked for me was just to blacklist it. I believe there's also a way to make it load after the correct sound modules (if you like the audible bell), which some judicious googling should turn up.
Good luck getting that fixed!
Ivy
I'll try that too. Tnx much for the suggestion. DR
2009/1/19 David Rosenstrauch <darose@darose.net>:
FYI - I'm seeing a problem with the 2.6.28-3 kernel. For some reason there's no sound coming out of either the speakers or the headphones on my laptop (Dell Precision M4400) after the upgrade.
Sound module is snd-hda-intel. I checked volume is up, mute is not on, yada yada, but no sound coming out. Problem goes away as soon as I downgrade the kernel.
Anyone else seeing this and/or have any idea what the problem might be?
Fun thing: I already upgraded the kernel when I answered your mail, but I hadn't rebooted. Today I did, and same thing happened :) I'm using snd-hda-intel too, on a 13" macbook, latest gen. My mixer is much thinner than with 2.6.28 (5 entries vs ~10), and there's no way to get some sound. Downgrading fixed. Did you find out anything? C.
bardo wrote:
2009/1/19 David Rosenstrauch <darose@darose.net>:
FYI - I'm seeing a problem with the 2.6.28-3 kernel. For some reason there's no sound coming out of either the speakers or the headphones on my laptop (Dell Precision M4400) after the upgrade.
Sound module is snd-hda-intel. I checked volume is up, mute is not on, yada yada, but no sound coming out. Problem goes away as soon as I downgrade the kernel.
Anyone else seeing this and/or have any idea what the problem might be?
Fun thing: I already upgraded the kernel when I answered your mail, but I hadn't rebooted. Today I did, and same thing happened :) I'm using snd-hda-intel too, on a 13" macbook, latest gen. My mixer is much thinner than with 2.6.28 (5 entries vs ~10), and there's no way to get some sound. Downgrading fixed. Did you find out anything?
C.
Not really. In theory this should be a matter of doing like bardo said: find out which chipset your machine uses, and then find the correct "model" parameter to supply for that chipset (via this page: http://free-electrons.com/kerneldoc/latest/sound/alsa/ALSA-Configuration.txt). (Well, technically, in theory you shouldn't have to specify a model at all, and the module should just autodetect it.) But that doesn't appear to be working, as I tried all 4 models listed for my chipset. So I have to figure then that this is a kernel or ALSA bug, and so monitor the upstream bug(s) to see when it'll be fixed. These 2 bug entries appear to be the relevant ones: http://bugzilla.kernel.org/show_bug.cgi?id=12336 https://bugtrack.alsa-project.org/alsa-bug/view.php?id=4306 So might be a while before this one gets fixed. HTH, DR
2009/1/20 David Rosenstrauch <darose@darose.net>:
Not really. In theory this should be a matter of doing like bardo said:
Bardo... which would always mean me =)
find out which chipset your machine uses, and then find the correct "model" parameter to supply for that chipset (via this page: http://free-electrons.com/kerneldoc/latest/sound/alsa/ALSA-Configuration.txt). (Well, technically, in theory you shouldn't have to specify a model at all, and the module should just autodetect it.) But that doesn't appear to be working, as I tried all 4 models listed for my chipset.
Did you reboot before every try? Best way is either blacklisting the module and loading manually or changing the model parameter in modprobe.conf before rebooting. This is because the BIOS needs to be reset too. It is also mentioned in the kernel bug you linked.
So I have to figure then that this is a kernel or ALSA bug, and so monitor the upstream bug(s) to see when it'll be fixed. These 2 bug entries appear to be the relevant ones:
http://bugzilla.kernel.org/show_bug.cgi?id=12336 https://bugtrack.alsa-project.org/alsa-bug/view.php?id=4306
If we both have the same problem, I'm afraid these bug reports aren't related: I have sound working nicely with 2.6.28, problems started to appear with 2.6.28.1. Did you go straight from 2.6.27.10 to 2.6.28.1? If so, try testing 2.6.28. C.
bardo wrote:
2009/1/20 David Rosenstrauch <darose@darose.net>:
Not really. In theory this should be a matter of doing like bardo said: find out which chipset your machine uses, and then find the correct "model" parameter to supply for that chipset (via this page: http://free-electrons.com/kerneldoc/latest/sound/alsa/ALSA-Configuration.txt). (Well, technically, in theory you shouldn't have to specify a model at all, and the module should just autodetect it.) But that doesn't appear to be working, as I tried all 4 models listed for my chipset.
Did you reboot before every try? Best way is either blacklisting the module and loading manually or changing the model parameter in modprobe.conf before rebooting. This is because the BIOS needs to be reset too. It is also mentioned in the kernel bug you linked.
Egads! I have to reboot too?!?!? Sigh. Alright, I'll try that too when I get a moment.
So I have to figure then that this is a kernel or ALSA bug, and so monitor the upstream bug(s) to see when it'll be fixed. These 2 bug entries appear to be the relevant ones:
http://bugzilla.kernel.org/show_bug.cgi?id=12336 https://bugtrack.alsa-project.org/alsa-bug/view.php?id=4306
If we both have the same problem, I'm afraid these bug reports aren't related: I have sound working nicely with 2.6.28, problems started to appear with 2.6.28.1. Did you go straight from 2.6.27.10 to 2.6.28.1? If so, try testing 2.6.28.
C.
I haven't tried 2.6.28.1. The module was working in 2.6.27.10, and stopped working in the 2.6.28(-3) kernel package in the core repository. Thanks, DR
I haven't tried 2.6.28.1. The module was working in 2.6.27.10, and stopped working in the 2.6.28(-3) kernel package in the core repository.
Try blacklisting the PC speaker module and rebooting. This "fixed" it for me. Dale
Dale Blount wrote:
I haven't tried 2.6.28.1. The module was working in 2.6.27.10, and stopped working in the 2.6.28(-3) kernel package in the core repository.
Try blacklisting the PC speaker module and rebooting. This "fixed" it for me.
Dale
I'm doing this in my rc.conf: MODULES=(snd-pcm-oss snd-hda-intel !snd-pcsp !pcspkr tun fuse loop) Is that not sufficient? DR
On Tue, 2009-01-20 at 14:53 -0500, David Rosenstrauch wrote:
Dale Blount wrote:
I haven't tried 2.6.28.1. The module was working in 2.6.27.10, and stopped working in the 2.6.28(-3) kernel package in the core repository.
Try blacklisting the PC speaker module and rebooting. This "fixed" it for me.
Dale
I'm doing this in my rc.conf:
MODULES=(snd-pcm-oss snd-hda-intel !snd-pcsp !pcspkr tun fuse loop)
Is that not sufficient?
Did you reboot afterward? I could not unload mine as it was "in use". I'll check the syntax/order of my rc.conf when I get home as I haven't been able to WoL it in a while. Dale
Dale Blount wrote:
On Tue, 2009-01-20 at 14:53 -0500, David Rosenstrauch wrote:
Dale Blount wrote:
I haven't tried 2.6.28.1. The module was working in 2.6.27.10, and stopped working in the 2.6.28(-3) kernel package in the core repository.
Try blacklisting the PC speaker module and rebooting. This "fixed" it for me.
Dale I'm doing this in my rc.conf:
MODULES=(snd-pcm-oss snd-hda-intel !snd-pcsp !pcspkr tun fuse loop)
Is that not sufficient?
Did you reboot afterward? I could not unload mine as it was "in use".
Yep. I'll have to try bardo's suggestion of rebooting after trying each of the models too. Thanks, DR
David Rosenstrauch wrote:
Dale Blount wrote:
On Tue, 2009-01-20 at 14:53 -0500, David Rosenstrauch wrote:
Dale Blount wrote:
Try blacklisting the PC speaker module and rebooting. This "fixed" it for me.
Dale I'm doing this in my rc.conf:
MODULES=(snd-pcm-oss snd-hda-intel !snd-pcsp !pcspkr tun fuse loop)
Is that not sufficient?
Did you reboot afterward? I could not unload mine as it was "in use".
Yep.
I'll have to try bardo's suggestion of rebooting after trying each of the models too.
Thanks,
DR
I tried everything that everyone suggested in this thread, but no dice. Definitely a bug in 2.6.28. :-( DR
2009/1/21 David Rosenstrauch <darose@darose.net>:
I tried everything that everyone suggested in this thread, but no dice. Definitely a bug in 2.6.28. :-(
I managed to get a partial success (most of my mixer got back, but sound still doesn't come out, and I think it's because of a missing switch) by using the latest alsa-1.0.19. This, be warned, means recompiling the kernel with some PKGBUILD modifications, nothing really difficult, but takes some time. Instructions are here: http://alsa.opensrc.org/index.php/AlsaBuild2.6#Overwrite_alsa_in_the_kernel_... Corrado
2009/1/21 bardo <ilbardo@gmail.com>:
2009/1/21 David Rosenstrauch <darose@darose.net>:
I tried everything that everyone suggested in this thread, but no dice. Definitely a bug in 2.6.28. :-(
I managed to get a partial success (most of my mixer got back, but sound still doesn't come out, and I think it's because of a missing switch) by using the latest alsa-1.0.19. This, be warned, means recompiling the kernel with some PKGBUILD modifications, nothing really difficult, but takes some time. Instructions are here: http://alsa.opensrc.org/index.php/AlsaBuild2.6#Overwrite_alsa_in_the_kernel_...
I made it! Maybe I was too tired yesterday, don't know what I did, but it seems that the old settings work with alsa 1.0.19, so i have my sound back by passing model=mbp3. An interesting feature of the new alsa-driver is the experimental capability to (finally!) reset the card's status when unloading the driver, and I found out it works well for me. If you need 2.6.28, then it's worth to give it a shot, but I think the new alsa will only be merged in 2.6.29... C.
bardo wrote:
2009/1/21 bardo <ilbardo@gmail.com>:
2009/1/21 David Rosenstrauch <darose@darose.net>:
I tried everything that everyone suggested in this thread, but no dice. Definitely a bug in 2.6.28. :-( I managed to get a partial success (most of my mixer got back, but sound still doesn't come out, and I think it's because of a missing switch) by using the latest alsa-1.0.19. This, be warned, means recompiling the kernel with some PKGBUILD modifications, nothing really difficult, but takes some time. Instructions are here: http://alsa.opensrc.org/index.php/AlsaBuild2.6#Overwrite_alsa_in_the_kernel_...
I made it! Maybe I was too tired yesterday, don't know what I did, but it seems that the old settings work with alsa 1.0.19, so i have my sound back by passing model=mbp3. An interesting feature of the new alsa-driver is the experimental capability to (finally!) reset the card's status when unloading the driver, and I found out it works well for me. If you need 2.6.28, then it's worth to give it a shot, but I think the new alsa will only be merged in 2.6.29...
C.
Hmmm ... Well, I imagine the 2.6.28 series will be with us for a while, so I imagine I'll have to build the drivers locally as you suggest, else I'll be pinned to 2.6.27. (Running my desktop machine without sound is definitely NOT an option, as I need some good pumping trance music on while I'm coding.) Thanks, DR
bardo schrieb:
2009/1/21 David Rosenstrauch <darose@darose.net>:
I tried everything that everyone suggested in this thread, but no dice. Definitely a bug in 2.6.28. :-(
I managed to get a partial success (most of my mixer got back, but sound still doesn't come out, and I think it's because of a missing switch) by using the latest alsa-1.0.19. This, be warned, means recompiling the kernel with some PKGBUILD modifications
It doesn't: You can download the standalone alsa-driver package, compile it and install the modules to /lib/modules/2.6.28-ARCH/updates (I think you can just replace /kernel/ with /updates/ in a Makefile, you'll have to figure out the details). The module in /updates/ will override the ones delivered with our kernel package, but you won't have file conflicts.
On Tue, 2009-01-20 at 14:56 -0500, Dale Blount wrote:
On Tue, 2009-01-20 at 14:53 -0500, David Rosenstrauch wrote:
Dale Blount wrote:
I haven't tried 2.6.28.1. The module was working in 2.6.27.10, and stopped working in the 2.6.28(-3) kernel package in the core repository.
Try blacklisting the PC speaker module and rebooting. This "fixed" it for me.
Dale
I'm doing this in my rc.conf:
MODULES=(snd-pcm-oss snd-hda-intel !snd-pcsp !pcspkr tun fuse loop)
Is that not sufficient?
Did you reboot afterward? I could not unload mine as it was "in use".
I'll check the syntax/order of my rc.conf when I get home as I haven't been able to WoL it in a while.
MODULES=(!snd_pcsp) is my setting. Not sure if the underscore vs the dash is causing any issues. Dale
On Dienstag, 20. Januar 2009 20:53 David Rosenstrauch wrote:
MODULES=(snd-pcm-oss snd-hda-intel !snd-pcsp !pcspkr tun fuse loop)
I suggest to do this MODULES=(!pcspkr tun fuse loop) but from my view the most problem with this "very usefull" -)) pcspkr is that he wants to grab the first slot. You can avoid this with a file and mine looks so: # cat /etc/modprobe.d/sound options snd-hda-intel model=6stack-dig power_save=60 enable=1 index=0 options snd-pcsp enable=1 index=2 alias snd-card-0 snd-hda-intel alias sound-slot-0 snd-hda-intel For me this works as the best and perhaps for you too. See you, Attila Source: http://osdir.com/ml/linux.suse.kernel/2008-06/msg00010.html
2009/1/20 Attila <attila@invalid.invalid>:
I suggest to do this
MODULES=(!pcspkr tun fuse loop)
but from my view the most problem with this "very usefull" -)) pcspkr is that he wants to grab the first slot.
You can avoid this with a file and mine looks so:
# cat /etc/modprobe.d/sound options snd-hda-intel model=6stack-dig power_save=60 enable=1 index=0 options snd-pcsp enable=1 index=2 alias snd-card-0 snd-hda-intel alias sound-slot-0 snd-hda-intel
For me this works as the best and perhaps for you too.
pcspkr and snd-pcsp are two different modules. The former provides you with the usual "beep" sounds, the latter tries to use the speaker as a sound card when your sound card is not supported or you don't have one at all. As far as I know, only snd-pcsp, sharing the alsa backend, can conflict with other modules. So there should be no need to blacklist pcspkr if it is wanted. It's ok to blacklist snd-pcsp or use the index=2 option, though. Corrado
On Dienstag, 20. Januar 2009 23:27 bardo wrote:
pcspkr and snd-pcsp are two different modules. The former provides you with the usual "beep" sounds, the latter tries to use the speaker as a sound card when your sound card is not supported or you don't have one at all. As far as I know, only snd-pcsp, sharing the alsa backend, can conflict with other modules. So there should be no need to blacklist pcspkr if it is wanted. It's ok to blacklist snd-pcsp or use the index=2 option, though.
Yes you be right and your explanation is better than mine. I blacklist pcspkr only because i don't need it and i want to to suppress the messages in the logfile. See you, Attila
participants (8)
-
Attila
-
bardo
-
Dale Blount
-
David Rosenstrauch
-
Eric Bélanger
-
joyfulgirl＠archlinux.us
-
Thomas Bächler
-
Tobias Powalowski