At Fri, 05 Jul 2013 14:00:45 +0200, Tormen wrote:
Hi,
Thanks for your answer! :))
On 05/07/13 07:29, Takashi Iwai wrote:
At Fri, 05 Jul 2013 00:53:48 +0200, Tormen wrote:
Hi again,
I digged again into google and stumbled across this thread from a year ago: http://mailman.alsa-project.org/pipermail/alsa-devel/2012-March/050504.html
I have the same Notebook than Adam.
The thread stopped there. I contacted Adam and he confirmed me something I also had noticed (but forgotten): The sound output via the speakers broke (went silent) between 3.1 and 3.2. It was working though for sometime between 3.2 and 3.9.6 - I confirmed it to be working in 3.7! And now it is broken again.
So, the speaker worked on *your* machine with 3.7 kernel?
Yes.
Then I need alsa-info.sh output in the working state.
*WORKING* sound (speakers play sound and everything else (from what I've seen so far) seems as it should) uname -a: Linux seven 3.7-trunk-amd64 #1 SMP Debian 3.7.8-1~experimental.1 x86_64 GNU/Linux http://www.alsa-project.org/db/?f=3012ed619f55c49b279c24b74a8192f7a3ccd3e3
This was without pulseaudio, without /etc/asound.conf and without any kernel module parameters to snd and snd-intel-hda. I had also tried 3.9.6 under this conditions, but same problem.
I guess you need more debug? Unfortunately alsa-compile complained that the linux-source-3.7 package vanished in the meantime from experimental. But I guess I can figure that out if you need more debug output (as in the debian 3.7 kernel image I am using CONFIG_SND_DEBUG is not set).
Didn't you read my previous reply...?
Sorry, I had missed it.
Give alsa-info.sh outputs (don't paste, use attachments). At best, give the outputs from both working and non-working kernels. Without these, I cannot debug further. thanks, Takashi
*NON-working* sound (speakers don't play any sound, headphone out works fine) uname -a: Linux seven 3.9-0.bpo.1-amd64 #1 SMP Debian 3.9.6-1~bpo70+1 x86_64 GNU/Linux http://www.alsa-project.org/db/?f=af6f9823cfa865872e84f9d4da569b2a72b071b0
This is the debian 3.9.6 kernel + alsa-compile to get "debug=2" for snd.ko module to work as by default the CONFIG_SND_DEBUG is not set in debian kernels.
OK, looking at two outputs, the difference is that the recent kernel allows the 5.1 output configuration with two mic jacks. But, this looks rather strange as a laptop. Does the machine have really two microphone jacks? The BIOS sets pins 0x18 and 0x19 are mic jacks.
You have three items in "Capture Source", Internal Mic, Mic and Mic 1. Which inputs are working actually? If one of the last two is bogus, we can disable it, then we can the unused pin. For example, a patch to disable the pin 0x19 would be like below.
Takashi
--- diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 14ac9b0..cb749b5 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -1713,6 +1713,7 @@ enum { ALC889_FIXUP_IMAC91_VREF, ALC882_FIXUP_INV_DMIC, ALC882_FIXUP_NO_PRIMARY_HP, + ALC882_FIXUP_VAIO_Z, };
static void alc889_fixup_coef(struct hda_codec *codec, @@ -2034,6 +2035,15 @@ static const struct hda_fixup alc882_fixups[] = { .type = HDA_FIXUP_FUNC, .v.func = alc882_fixup_no_primary_hp, }, + [ALC882_FIXUP_VAIO_Z] = { + .type = HDA_FIXUP_PINS, + .v.pins = (const struct hda_pintbl[]) { + { 0x19, 0x411111f0 }, /* disable pin 0x19 */ + { } + }, + .chained = true, + .chain_id = ALC882_FIXUP_NO_PRIMARY_HP, + }, };
static const struct snd_pci_quirk alc882_fixup_tbl[] = { @@ -2068,7 +2078,7 @@ static const struct snd_pci_quirk alc882_fixup_tbl[] = { SND_PCI_QUIRK(0x1043, 0x1971, "Asus W2JC", ALC882_FIXUP_ASUS_W2JC), SND_PCI_QUIRK(0x1043, 0x835f, "Asus Eee 1601", ALC888_FIXUP_EEE1601), SND_PCI_QUIRK(0x104d, 0x9047, "Sony Vaio TT", ALC889_FIXUP_VAIO_TT), - SND_PCI_QUIRK(0x104d, 0x905a, "Sony Vaio Z", ALC882_FIXUP_NO_PRIMARY_HP), + SND_PCI_QUIRK(0x104d, 0x905a, "Sony Vaio Z", ALC882_FIXUP_VAIO_Z), SND_PCI_QUIRK(0x104d, 0x9043, "Sony Vaio VGC-LN51JGB", ALC882_FIXUP_NO_PRIMARY_HP),
/* All Apple entries are in codec SSIDs */