Bugzilla – Bug 115253
No sound on Amilo M1437G notebook (hda-intel)
Last modified: 2007-02-26 12:04:57 UTC
Setup ===== Laptop Fujitsu Siemens Amilo M1437G lscpi: 00:01.0 Class 0403: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) High Definition Audio Controller (rev 04) lspci -n: 00:01.0 Class 0403 8086:2668 Problem ======= I can't hear anything. More detailed: - Audio has been set up, the audio hardware is detected and initialized. - Alsamixer is more or less turned on, but I only get the following elements: Headphone, PCM, CD, Mic, Caller ID, Input Source, Input Source 1, Input source 2, Internal Speaker, Off-Hook. So no master volume or anything like that. Connecting a headphone didn't help either.
Please attach /proc/asound/card0/codec#* files.\ If it's ALC880 codec, you can try "model" module option for snd-hda-intel with different values. See /usr/src/linux/Documentation/sound/ALSA-Configuration.txt for details.
I'm hearing light noise with model=z71v, 6stack and 6stack-digout. With the two 6stack models, it seems the mic is turned on and mirrored to the left, builtin speaker.
Just found the following link: http://www.amilo-forum.de/topic,2939,-amilo-1437g.html (search for sound). The author used a current cvs checkout and it worked. I'm afraid I probably won't be able to test this myself today. I'll try during the next few days and report back.
Hmm, the beta4 kernel has almost the same code with the current ALSA CVS regarding hda-intel. So, cvs version will unlikely help, I guess. If the earlier version of ALSA CVS worked, it means that "3stack" model did work for your device, though. Could you provide the requested files? Also, let me know the PCI subsystem ID, e.g. output of "lspci -nv".
Created attachment 48969 [details] codec#1, codec#2, lspci -vn Oops, I was so busy playing with the model lines that I forgot to post them. Just as an info: The laptop has two visible jacks, one with a mic and a line in symbol, the other with the text spdif and a headphones symbol. The laptop also has a builtin microphone and built in stereo speakers
And I meant codec#0 and codec#1 of course
This might be the problem of hda-modem. SInce the modem codec has the first id, this is assigned as the first device. Could you try the patch below together with ALSA 1.0.10rc1?
Created attachment 49025 [details] Patch to ALSA 1.0.10rc1 release
Created attachment 49026 [details] Equivalent patch to SUSE kernel tree
It looks like you might be on the right track: The modem was not detected during the installer and I was wondering whether to open another bug for it. I've tested your patch but thing are still not right: ------------ snd_page_alloc: module not supported by Novell, setting U taint flag. snd: module not supported by Novell, setting U taint flag. snd_timer: module not supported by Novell, setting U taint flag. snd_pcm: module not supported by Novell, setting U taint flag. snd_hda_codec: module not supported by Novell, setting U taint flag. snd_hda_intel: module not supported by Novell, setting U taint flag. ACPI-0212: *** Warning: Device is not power manageable ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10 PCI: Setting latency timer of device 0000:00:1b.0 to 64 ALSA /root/jmayer/kernel/alsa-driver-1.0.10rc1/pci/hda/../../alsa-kernel/pci/hda/hda_intel.c:1207: hda-intel: Too many modem PCMs ACPI: PCI interrupt for device 0000:00:1b.0 disabled HDA Intel: probe of 0000:00:1b.0 failed with error -22 snd_mixer_oss: module not supported by Novell, setting U taint flag. snd_pcm_oss: module not supported by Novell, setting U taint flag. ------------ This is what I did to test the patch: I installed todays kotd (kernel-2.6.13-20050906125922-default) including nongpl and sources. Then I did a make cloneconfig and make prepare. After that I unpacked the alsa-drivers 10.0.10rc1, applied the alsa patch and ran ./configure && make && make install Then I rebooted and ran alsaconf, which returned quickly instead of playing the testsound. alsamixer fails with "... No such device"
Oh it's a typo. Try the additional patch below.
Created attachment 49056 [details] Additional patch for hda-intel
BTW, the modem can't be detected by yast2 (yet) although it might work on your machine with slmodemd if you define a proper modem PCM.
OK, the patch is working for audio out: Alsaconf did not produce any audible output but the time it took to finish made clear that it was playing the sound ant that it was just the volume that was missing. I needed to manually unmute the "internal" element and set the volume from 0 to e.g. 75. After that aplay /usr/share/sounds/alsa/test.wav played the stereo signal properly (ok, I don't know whether channels may be swapped). Then I typed "arecord >/tmp/test.audio" Now the box is hung completey, not even changing vts works (I'm in runlevel 3).
Hmm, that's bad. To be sure, try to run "arecord -fdat -Dhw foo.wav".
If the problem still happens, try to reinstall ALSA modules in the kernel package again. You may hear sounds via "aplay -Dplughw:1 foo.wav". Then try to run "arecord -fdat -Dhw:1 foo.wav".
Runnig it doesn't work. Both, aplay and arecord fail with "No such device". Also adding module=6stack didn't make a difference - it still fails the same way.
Interesting, what shows /proc/asound/pcm ?
OK, back one step: The original alsa code installs some modules in acore, the suse kernel has the modules in core, so just reinstalling the kernel led to some duplicates. Now that the duplicates are removed, it still looks like this (without module=6stack): aplay and arecord still complain about "No such device", but now I have a /proc/asound/pcm file: 00-00 Si3054 Modem : Si3054 Modem : playback 1 : capture 1 00-01 ALC880 Analog : ALC Analog : playback 1 : capture 2
Did you add "-Dhw:1" option to aplay/arecord? Do you have /dev/snd/pcmC0D1[pc] files?
I'm sorry but it looks like I messed up the modules more than I thought. I've now done the following: I've removed all the /lib/modules/2.6... stuff, Reinstalled kernel+modules from comment #10. Built a kernel including the two patches and installed the modules from the kernel. That way I didn't have any differences in the path names between kernel and the original alsa. After doing that, arecord didn't hang the system any more. So maybe I was running a mixture of old and new modules during my last test. Sorry for the wasted time :-( The test now just doesn't seem to record anything (neither via the built-in mic nor via an externally connected mic). All in all, it's a big step forward when compared to the original kernel :-)
Nice to hear that it's a non-critical bug now :) I guess the silent recording is a problem of mixer setting. Please raise "Capture" volume and turn on "Capture" switch (push space on alsamixer), then choose an apporpriate source for two "Input Source" controls. The patch will go into rc2.
MIC was already unmuted and raised to 75 and 100. Neither produced any result. I've also run kmix and krecord but didn't find anything different than with alsamixer and arecord.
Do you have any ideas how to get the microphone working or how to find out why it isn't working?
Possibly a wrong pin is assigned. You can try model=test option, and adjust mixer settings to find something matching. Note that you have to raise "Capture" volume and turn on "Capture Switch" to record something. The "Mic" volume etc play no role for recording. They are only for (loopback) playback.
Sorry for the delay, but I a) was on a business trip and b) had to return the laptop to its owner. I now have time and laptop until the weekend, so let's play some more ;-) Loading the module with model=test fails, alsamixer doesn't find a device. The problem occurs during module load, as can be seen from the following dmesg output: snd_hda_codec: module not supported by Novell, setting U taint flag. snd_hda_intel: module not supported by Novell, setting U taint flag. ACPI-0212: *** Warning: Device is not power manageable ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10 PCI: Setting latency timer of device 0000:00:1b.0 to 64 ALSA sound/pci/hda/hda_codec.c:1551: hda_codec: model 'test' is selected ALSA sound/core/control.c:324: control 2:0:0:CLFE Playback Volume:0 is already present ACPI: PCI interrupt for device 0000:00:1b.0 disabled HDA Intel: probe of 0000:00:1b.0 failed with error -16 Is there a newer kernel i could/should try (I'm still using the kernel from comment#10)
OK, it's a bug of driver. The patch below fixes this.
Created attachment 50720 [details] Patch to fix test model
OK, one fixed and at least one to go: Sep 23 23:32:27 guinevere kernel: snd_hda_codec: module not supported by Novell, setting U taint flag. Sep 23 23:32:27 guinevere kernel: snd_hda_intel: module not supported by Novell, setting U taint flag. Sep 23 23:32:27 guinevere kernel: ACPI: PCI Interrupt 0000:00:1b.0[A] -> Link [LNKA] -> GSI 10 (level, low) -> IRQ 10 Sep 23 23:32:27 guinevere kernel: PCI: Setting latency timer of device 0000:00:1b.0 to 64 Sep 23 23:32:27 guinevere kernel: ALSA sound/pci/hda/hda_codec.c:1551: hda_codec: model 'test' is selected Sep 23 23:32:27 guinevere kernel: ALSA sound/core/control.c:324: control 2:0:0:Capture Volume:0 is already present Sep 23 23:32:27 guinevere kernel: ACPI: PCI interrupt for device 0000:00:1b.0 disabled Sep 23 23:32:27 guinevere kernel: HDA Intel: probe of 0000:00:1b.0 failed with error -16
Oh, I forgot: After it didn't work I updated to kernel-default-2.6.13-SL100_BRANCH_20050922170017 but it didn't help :)
My fault. Please try the second patch.
Created attachment 50820 [details] Fix for test model, take 2
This should have been fixed on upstream. Resolved to FIXED for now.