Creative Recon3d & Sound Blaster Z (CA0132), No sound at all

[Michael Murphy]

[WORKAROUND] Run the following command:

echo "options snd-hda-intel position_fix=1" | sudo tee -a /etc/modprobe.d/alsa-base.conf

[EDIT]
Workaround doesn't work anymore for me at least. I assume it only has a small possibility of working due to a firmware bug.

[ORIGINAL REPORT]
No sound, no jacks; It is detected by Linux however.

ProblemType: Bug
DistroRelease: Ubuntu 13.04
Package: alsa-base 1.0.25+dfsg-0ubuntu4
ProcVersionSignature: Ubuntu 3.8.0-16.26-generic 3.8.5
Uname: Linux 3.8.0-16-generic x86_64
NonfreeKernelModules: fglrx
ApportVersion: 2.9.2-0ubuntu5
Architecture: amd64
Date: Mon Apr 8 18:56:44 2013
InstallationDate: Installed on 2013-04-07 (1 days ago)
InstallationMedia: Ubuntu 13.04 "Raring Ringtail" - Alpha amd64 (20130402.1)
MarkForUpload: True
PackageArchitecture: all
ProcEnviron:
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=en_US.UTF-8
 SHELL=/bin/bash
SourcePackage: alsa-driver
Symptom: audio
Symptom_AlsaPlaybackTest: ALSA playback test through plughw:Creative failed
Symptom_Card: HDA Creative - HDA Creative
Symptom_Type: No sound at all
Title: [To be filled by O.E.M., Creative CA0132, Green Line Out, Rear] No sound at all
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 03/12/2012
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 1102
dmi.board.asset.tag: To be filled by O.E.M.
dmi.board.name: SABERTOOTH 990FX
dmi.board.vendor: ASUSTeK COMPUTER INC.
dmi.board.version: Rev 1.xx
dmi.chassis.asset.tag: To Be Filled By O.E.M.
dmi.chassis.type: 3
dmi.chassis.vendor: To Be Filled By O.E.M.
dmi.chassis.version: To Be Filled By O.E.M.
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr1102:bd03/12/2012:svnTobefilledbyO.E.M.:pnTobefilledbyO.E.M.:pvrTobefilledbyO.E.M.:rvnASUSTeKCOMPUTERINC.:rnSABERTOOTH990FX:rvrRev1.xx:cvnToBeFilledByO.E.M.:ct3:cvrToBeFilledByO.E.M.:
dmi.product.name: To be filled by O.E.M.
dmi.product.version: To be filled by O.E.M.
dmi.sys.vendor: To be filled by O.E.M.

[seyden (seyden-linux-kernel-bugs)]

Good day.

CA0132 is support, my soundcard Creative Soundblaster Z detected, but no sound.

With last dev-kernel (3.9-r3) I enable this .config:

CONFIG_SND_HDA_CODEC_CA0132=y
CONFIG_SND_HDA_CODEC_CA0132_DSP=y
CONFIG_EXTRA_FIRMWARE="ctefx.bin ctspeq.bin"

These firmware included when compiling:
  MK_FW firmware/ctefx.bin.gen.S
  MK_FW firmware/ctspeq.bin.gen.S
  AS firmware/ctefx.bin.gen.o
  AS firmware/ctspeq.bin.gen.o

Output "lsmod | grep ca0132":

snd_hda_codec_ca0132 35296 1
snd_hda_codec 154312 4 snd_hda_codec_realtek,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec_ca0132
snd 69106 33 snd_hda_codec_realtek,snd_usb_audio,snd_hwdep,snd_timer,snd_hda_codec_hdmi,snd_pcm,snd_rawmidi,snd_usbmidi_lib,snd_hda_codec,snd_hda_intel,snd_seq_device,snd_pcsp,snd_hda_codec_ca0132

But Output "dmesg | grep ca0132":
[ 3.817296] ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING.

Where the problem?
Why is there no sound on this soundcard (Creative Soundblaster Z)?

Thanks for the help.

[ninex (ninex-linux-kernel-bugs)]

Same problem with Creative Sound Blaster ZXR
No sound at all.
rmmod and modprobe again loads DSP but still no sound

[ 719.874101] hda_intel: Disabling MSI
[ 719.874114] hda-intel 0000:03:00.0: Force to non-snoop mode
[ 720.459053] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
[ 721.267711] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
[ 721.468298] hda_codec: cannot build controls for #2 (error -16)
[ 722.055575] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
[ 722.276650] input: HDA Creative Line as /devices/pci0000:00/0000:00:1c.3/0000:03:00.0/sound/card1/input14
[ 722.276688] input: HDA Creative Front Headphone as /devices/pci0000:00/0000:00:1c.3/0000:03:00.0/sound/card1/input15
[ 722.276722] input: HDA Creative Line Out as /devices/pci0000:00/0000:00:1c.3/0000:03:00.0/sound/card1/input16
[ 744.783426] hda-intel: IRQ timing workaround is activated for card #1. Suggest a bigger bdl_pos_adj.

[p_narciso (pnarciso-linux-kernel-bugs)]

Another Creative Sound blaster Z user here, and with the same issue.

I'm running Xubuntu 13.04 with the latest hda driver and I can't get any sound from it.

I also tried kernel 3.9 and it's the same behavior, the card is completely mute.

[p_narciso (pnarciso-linux-kernel-bugs)]

Here's my alsa-info : http://www.alsa-project.org/db/?f=24542f6f9e68119362e1df65926ad15554f0f968

[ninex (ninex-linux-kernel-bugs)]

Here is my alsa info - Sound Blaster ZXR
http://www.alsa-project.org/db/?f=8975fef56feb7efcbd5ba01a71915ee9fcfedd85

[Acidphase (acidphase-linux-kernel-bugs)]

Add me to the crowd lol

Gentoo x86_64 3.9.2-gentoo

Sound Blaster Z

[ 3.369657] snd_hda_intel 0000:07:00.0: enabling device (0000 -> 0002)
[ 3.371425] hda_intel: Disabling MSI
[ 3.373131] hda-intel 0000:07:00.0: Force to non-snoop mode
-
[ 5.937591] ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING.

I have also have the external firmware built in.

CONFIG_SND_HDA_CODEC_CA0132=y
CONFIG_SND_HDA_CODEC_CA0132_DSP=y
CONFIG_EXTRA_FIRMWARE="ctefx.bin ctspeq.bin"

[Acidphase (acidphase-linux-kernel-bugs)]

From what little research that I have done as it seems the biggest difference between he Z series cards and the Recon 3D series is the new DAC

"24-bit/192 kHz Cirrus Logic CS4398" Info here:
http://www.cirrus.com/en/pubs/proDatasheet/CS4398_F1.pdf
http://www.cirrus.com/en/products/cs4398.html

It uses the same sound processor The quad-core SoundCore3D chip (ca0132).

Not sure if this info helps or I'm just spamming here, if there is anything the devs needs and I can help out in anyway feel free to let me know.

-Ron

[Michael Murphy (mmstick)]

Problem still persists with kernel 3.9 in Ubuntu Saucy Salamander.

[Acidphase (acidphase-linux-kernel-bugs)]

Is anything moving forward on this ?

I have no sound in Linux now for quite some time..

[mmstickman (mmstickman-linux-kernel-bugs)]

I suggest changing the name of the title to: "Creative Recon3D + Sound Blaster Z (CA0132) Linux Support"

I bought a Creative Recon3D back in late 2011 when it released. I have tried every Linux kernel since then and numerous workarounds in attempt to get this to function. However, even with the latest RC of kernel 3.10 I still cannot get any sound out of my Recon3D at all.

[Marc Bourgoin (jagged-jimmy-j)]

Heyyo,

I get sound with my SoundBlaster Recon 3d Fatal1ty Professional when I use any version of the 32bit kernel (3.8.x and 3.9.4 thoroughly tested)on any version of Linux (Ubuntu, Xubuntu, Kubuntu... can't install Fedora for some reason though... just hangs) where-as 64bit works only if I modify the rc.local file with the fix_position... and that's not even a guarantee. There's definitely a bug in the 64bit Kernel only at least for me... give it a shot yourself mmstickman, see if it works on your Recon 3D.

[Michael Murphy (mmstick)]

Care to explain in more detail?

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Daniel Letzeisen (dtl131)]

[WORKAROUND] Run the following command:
echo "options snd-hda-intel position_fix=1" | sudo tee -a /etc/modprobe.d/alsa-base.conf

[Michael Murphy (mmstick)]

I see, I had fixed it in rc.local with the position_fix thing. It does indeed work now, quite well.

[Acidphase (acidphase-linux-kernel-bugs)]

"2013-03-21" almost 3 months later not even a single reply from a developer on this. I mean anything would be nice. "Not fixing, working on it, haven't had the time" Something.

[mmstickman (mmstickman-linux-kernel-bugs)]

I was able to get a Creative CA0132 (Recon3D + Sound Blaster Z) sound card to function within Linux. Someone noted that the 32-bit kernel plays just fine, but that the 64-bit kernel needs a special fix in /etc/rc.local. Add the following before exit 0:

rmmod snd_hda_intel
modprobe snd_hda_intel position_fix=1

Someone at ALSA needs to get this patched up ASAP.

[ninex (ninex-linux-kernel-bugs)]

(In reply to comment #10)
> I was able to get a Creative CA0132 (Recon3D + Sound Blaster Z) sound card to
> function within Linux. Someone noted that the 32-bit kernel plays just fine,
> but that the 64-bit kernel needs a special fix in /etc/rc.local. Add the
> following before exit 0:
>
> rmmod snd_hda_intel
> modprobe snd_hda_intel position_fix=1
>
> Someone at ALSA needs to get this patched up ASAP.

this works with Recon3D but not with SoundBlaster Z/XZ/ZXR

[Acidphase (acidphase-linux-kernel-bugs)]

What about the Z series cards? Any plans to fix them? Still no word from the Devs.

[Acidphase (acidphase-linux-kernel-bugs)]

By the way the rc.local equivalent in Gentoo is in /etc/local.d/

Configuration

All scripts in /etc/local.d/ with the suffix .start will be executed at boot time, all scripts with suffix .stop at shutdown time.
To e.g. output the text Hello world! at boot time, create the new file /etc/local.d/HelloWorld.start:

 [Collapse] File/etc/local.d/HelloWorld.start
#!/bin/sh
echo "Hello world!"
Afterwards mark the script as executable:
root #
chmod +x /etc/local.d/HelloWorld.start
To start the local.d scripts at boot time, add its init.d script to your default runlevel:
root #
rc-update add local default
Now you can start the service by making OpenRC check for stopped services in the default runlevel
root #
rc
or start it explicitly
root #
rc-service local start

Reference: http://wiki.gentoo.org/wiki/Local.d

I'm curious if someone with a Gentoo setup has used the work around mentioned in comments 10 and 11.

[mmstickman (mmstickman-linux-kernel-bugs)]

This is also good reference: https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/1166529

Someone finally assigned the bug there with an importance along with a different fix that does the same thing.

echo "options snd-hda-intel position_fix=1" | sudo tee -a /etc/modprobe.d/alsa-base.conf

[tiwai (tiwai-linux-kernel-bugs)]

(In reply to comment #12)
> What about the Z series cards? Any plans to fix them? Still no word from the
> Devs.

Try to ask Creative. The driver comes from them, and we've got no technical information available regarding the DSP on the board. We can try to fix a regression, but it's not the case, unfortunately.

[tiwai (tiwai-linux-kernel-bugs)]

(In reply to comment #10)
> I was able to get a Creative CA0132 (Recon3D + Sound Blaster Z) sound card to
> function within Linux. Someone noted that the 32-bit kernel plays just fine,
> but that the 64-bit kernel needs a special fix in /etc/rc.local. Add the
> following before exit 0:
>
> rmmod snd_hda_intel
> modprobe snd_hda_intel position_fix=1
>
> Someone at ALSA needs to get this patched up ASAP.

Is it with the recent kernel (3.9 and later)? Since CA0132 has a DSP support recently, we need to know whether the problem is in the DSP loading part or it's after DSP is running.

With a Recon3D board I tested earlier, the DSP loading failed at the first module load, for example. This resulted in the complete silence.

So, what happens if you didn't pass position_fix=1 option? Does the driver tell you that DSP is running or failed?

Also, could you confirm that the behavior really differs between 32bit and 64bit kernels even on the same machine?

[Paulo Narciso (p-narciso)]

Doesn't work for Z cards.

[Daniel Letzeisen (dtl131)]

It may be coincidence, but both of you have onboard or HDMI audio devices that force disabling of MSI (message signaled interrupts)

@Paulo: I would try disabling the hdmi audio:
sudo echo "blacklist snd-hda-codec-hdmi" > /etc/modprobe.d/blacklist-hdmi.conf
sudo depmod -a

See if sound works after rebooting

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to comment #16)
> (In reply to comment #10)
> > I was able to get a Creative CA0132 (Recon3D + Sound Blaster Z) sound card
> to
> > function within Linux. Someone noted that the 32-bit kernel plays just
> fine,
> > but that the 64-bit kernel needs a special fix in /etc/rc.local. Add the
> > following before exit 0:
> >
> > rmmod snd_hda_intel
> > modprobe snd_hda_intel position_fix=1
> >
> > Someone at ALSA needs to get this patched up ASAP.
>
> Is it with the recent kernel (3.9 and later)? Since CA0132 has a DSP support
> recently, we need to know whether the problem is in the DSP loading part or
> it's after DSP is running.
>
> With a Recon3D board I tested earlier, the DSP loading failed at the first
> module load, for example. This resulted in the complete silence.
>
> So, what happens if you didn't pass position_fix=1 option? Does the driver
> tell you that DSP is running or failed?
>
> Also, could you confirm that the behavior really differs between 32bit and
> 64bit kernels even on the same machine?

Takashi, first thank you for your replies it's greatly appreciated.

1. It's with the 3.9 and 3.10-rcX kernels.

2. Yes the position fix loads the DSP:
~ $ dmesg | grep DSP
[ 9.900606] ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING.
[ 22.294978] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

3. I don't have a 32bit kernel setup at the moment to let you know if it really differs.

I and other have/are tried/trying to reach out to creative. I personally sent a email to their Public relations and Developers with not surprise haven't received a response. We also reached out on to the discussion forums one of the threads had to get closed because of the Windows Fanboys starting a flame war on the thread.

I had feeling it was because you were missing documentation on this chip and DAC. I/we have even pointed out that all we need is some basic driver or information for this purpose and the community would take it from there.
It's like pulling teeth dealing with the trolls on forums and trying to get a intelligent dialogue going with and actual creative developer.

Forum Links:
http://forums.creative.com/showthread.php?t=699670 Closed
http://forums.creative.com/showthread.php?t=699748 Open

Creative Labs Connect has tumble weed blowing around since 2008:
http://connect.creativelabs.com/linux/default.aspx

In short no one seems to care @ Ceative thus far, which will be remembered the next time I'm in the market for a sound card lol.

In the interim is there any sort of technical docs that would help? I know I mentioned a link to the white papers of the DAC on the Blaster Z cards straight from Cirrus Logic back in comment #6.

[Acidphase (acidphase-linux-kernel-bugs)]

Takashi I forwarded the email I sent out to Creative to you.

[tiwai (tiwai-linux-kernel-bugs)]

(In reply to comment #17)
> > So, what happens if you didn't pass position_fix=1 option? Does the driver
> > tell you that DSP is running or failed?
> >
> > Also, could you confirm that the behavior really differs between 32bit and
> > 64bit kernels even on the same machine?
>
> Takashi, first thank you for your replies it's greatly appreciated.
>
> 1. It's with the 3.9 and 3.10-rcX kernels.
>
> 2. Yes the position fix loads the DSP:
> ~ $ dmesg | grep DSP
> [ 9.900606] ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING.
> [ 22.294978] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

Why two lines in the long time period? And are both of them the result *without* position_fix=1 option?

The second DSP load might be after the power-saving resume. But it's still unclear what you tested exactly...

I'm asking this because we need to understand why position_fix option fixes anything. The option is no magic.

[ninex (ninex-linux-kernel-bugs)]

>
> Why two lines in the long time period? And are both of them the result
> *without* position_fix=1 option?
>
> The second DSP load might be after the power-saving resume. But it's still
> unclear what you tested exactly...
>
> I'm asking this because we need to understand why position_fix option fixes
> anything. The option is no magic.

this optin dosn't fix any thing.
the magick is:

unload module
load it again
then dsp loads and i am getting ca0132 DOWNLOAD OK
:-) DSP IS RUNNING. message.

but card still don't play sounds

[tiwai (tiwai-linux-kernel-bugs)]

It's no surprise. The no sound on Z-series is possibly other reasons.

But rather the question is about Recon3D. For recon3D, do you still need this option at all? Or just reloading the module helps?

[perex (perex-linux-kernel-bugs)]

Note: I worked with an older kernel (2.6.18) and the initialization sequence must be called twice (the complete HDA controller & codec init) otherwise the Recon3D does not produce any sound after boot (rmmod & insmod sequence is sufficient to make things working even for the pre-DSP driver version). I was not able to trace the exact point-of-failure . It looks like a bad timing somewhere.

diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index b9a3f6a..b7fabf5 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -542,6 +542,7 @@ enum {
 #define AZX_DCAPS_ALIGN_BUFSIZE (1 << 22) /* buffer size alignment */
 #define AZX_DCAPS_4K_BDLE_BOUNDARY (1 << 23) /* BDLE in 4k boundary */
 #define AZX_DCAPS_POSFIX_COMBO (1 << 24) /* Use COMBO as default */
+#define AZX_DCAPS_RETRY2 (1 << 25) /* Do init twice */

 /* quirks for ATI SB / AMD Hudson */
 #define AZX_DCAPS_PRESET_ATI_SB \
@@ -558,7 +559,8 @@ enum {
   AZX_DCAPS_ALIGN_BUFSIZE)

 #define AZX_DCAPS_PRESET_CTHDA \
- (AZX_DCAPS_NO_MSI | AZX_DCAPS_POSFIX_LPIB | AZX_DCAPS_4K_BDLE_BOUNDARY)
+ (AZX_DCAPS_NO_MSI | AZX_DCAPS_POSFIX_LPIB | AZX_DCAPS_4K_BDLE_BOUNDARY |\
+ AZX_DCAPS_RETRY2)

 /*
  * VGA-switcher support
@@ -3148,6 +3150,7 @@ static int __devinit azx_probe(struct pci_dev *pci,
  static int dev;
  struct snd_card *card;
  struct azx *chip;
+ unsigned int caps_mask = 0;
  int err;

  if (dev >= SNDRV_CARDS)
@@ -3157,6 +3160,7 @@ static int __devinit azx_probe(struct pci_dev *pci,
   return -ENOENT;
  }

+retry:
  card = snd_card_new(index[dev], id[dev], THIS_MODULE, 0);
  if (!card) {
   snd_printk(KERN_ERR SFX "Error creating card!\n");
@@ -3166,10 +3170,16 @@ static int __devinit azx_probe(struct pci_dev *pci,
  /* set this here since it's referred in snd_hda_load_patch() */
  snd_card_set_dev(card, &pci->dev);

- err = azx_create(card, pci, dev, pci_id->driver_data, &chip);
+ err = azx_create(card, pci, dev,
+ pci_id->driver_data & ~caps_mask, &chip);
  if (err < 0)
   goto out_free;
  card->private_data = chip;
+ if (chip->driver_caps & AZX_DCAPS_RETRY2) {
+ caps_mask = AZX_DCAPS_RETRY2;
+ snd_card_free(card);
+ goto retry;
+ }

  if (!chip->disabled) {
   err = azx_probe_continue(chip);

[Michael Murphy (mmstick)]

Hmm, for some strange reason after I reboot, sound is no longer working once again with my recon3d in Ubuntu 13.04.

[Michael Murphy (mmstick)]

After purging, reinstalling, and reconfiguring alsa and pulse, still no sound at all from my Recon3D. The position_fix=1 doesn't seem to work anymore now.

[Raymond (superquad-vortex2)]

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/log/?qt=grep&q=ca0132

Try latest alsa driver with CONFIG_SND_HDA_CODEC_CA0132_DSP

[Michael Murphy (mmstick)]

I don't see how that helps. I already tried the latest ALSA Daily PPA and such. That patch doesn't really note anything about actual firmware bug fixes too. Nor do I know how to manually compile/install ALSA.

[Raymond (superquad-vortex2)]

you have to specify -with-card-options=hda-codec-ca0132-dsp and --with-debug=verbose during configure

ask the author where to get the firmware "ctefx.bin"

[Michael Murphy (mmstick)]

Your specification isn't specific enough to detail on what a person should do in order to get these cards running.

[Raymond (superquad-vortex2)]

you have to ask ppa maintainer since those daily snapshot build fail

make[2]: Entering directory `/usr/src/linux-headers-3.8.0-25-generic'
  CC [M] /build/buildd/oem-audio-hda-daily-lts-raring-dkms-0.201307011243~precise1/buildroot/driver-build/patch_analog.o
  CC [M] /build/buildd/oem-audio-hda-daily-lts-raring-dkms-0.201307011243~precise1/buildroot/driver-build/patch_ca0110.o
  CC [M] /build/buildd/oem-audio-hda-daily-lts-raring-dkms-0.201307011243~precise1/buildroot/driver-build/patch_ca0132.o
/build/buildd/oem-audio-hda-daily-lts-raring-dkms-0.201307011243~precise1/buildroot/driver-build/patch_ca0132.c:142:12: error: 'SNDRV_CTL_ELEM_ID_NAME_MAXLEN' undeclared here (not in a function)
/build/buildd/oem-audio-hda-daily-lts-raring-dkms-0.201307011243~precise1/buildroot/driver-build/patch_ca0132.c: In function 'add_fx_switch':
/build/buildd/oem-audio-hda-daily-lts-raring-dkms-0.201307011243~precise1/buildroot/driver-build/patch_ca0132.c:3938:7: warning: unused variable 'namestr' [-Wunused-variable]
make[3]: *** [/build/buildd/oem-audio-hda-daily-lts-raring-dkms-0.201307011243~precise1/buildroot/driver-build/patch_ca0132.o] Error 1

[Michael Murphy (mmstick)]

Well it worked for a while once I installed kernel 3.11-rc1. But then I messed up my system upgrading to 13.10 and so I reformatted and installed ubuntu 13.04 freshly and used my fresh installation script to get it configured exactly as it was before, even the 3.11-rc1 kernel, and it's not working. What gives? It's not the first time it has done this. The moment I think I have finally fixed it, a few reboots later or a reinstall and it will no longer work, ever. Something is wrong here and I can't figure it out.

[Acidphase (acidphase-linux-kernel-bugs)]

Well just as an FYI I received this email today which may explain why this whole codec/driver is a mess.

"Hello Ron,
Thank you for getting in touch, and many apologies for the delayed response.

I'm sorry to hear that you are having difficulty getting your SoundBlaster Z to work on your system under Linux. I was involved in the upstreaming of the Creative CA0132 codec to the ALSA repository, so I might be able to help provide some information on this.

The CA0132 codec was originally intended for and designed to support the Chromebook Pixel hardware architecture. I'm sorry to say that this means that the current codec isn't expected to work on other products that have different architectures, such as the Z series. Thank you very much for all the information and details you provided though, it is much appreciated, and I hope they may come in useful at a later date.

Best regards,
- Ian"

We really need to get someone rolling on this because it's been far to long already with no sound with these cards. Creative either needs to be harassed or we as a community have to pull together and try to provide the developers with as much information as possible.

Would Kernel mesages and /proc info on the Codec help at all or would it just produce a lot of spam for the Developers?

[richie (richie-linux-kernel-bugs)]

Hi,
i own a Creative Soundblaster ZXR and am also unable to get sound from my card.

I am not able to write a Kernel Module myself but am willing to help resolving this bug.
If I can help with testing something, please send an e-mail to me.

Greetings

[Acidphase (acidphase-linux-kernel-bugs)]

Created attachment 106952
Kernel Debug information & Codec output

Well in light of the latest response from the email I received from Ian Minett concerning the ca0132 portion of the snd_intel_hda kernel driver. I'm not sure what the plan of attack is from here. I figure though in the interim I can spam you with a bunch of kernel debug messages from ALSA as I set it to the highest verbosity level.

 Actually if there is more I can do let me know. I will also provide the output information from the codec produced by the ca0132 portion of intel sound driver via /proc/asound/card*/codec*.

I'm not sure any of this helps just trying to do something sitting around idle is driving me nuts with no sound support. So if it's useless information I apologize in advance.

[ninex (ninex-linux-kernel-bugs)]

(In reply to Acidphase from comment #23)
> Well just as an FYI I received this email today which may explain why this
> whole codec/driver is a mess.
>
> "Hello Ron,
> Thank you for getting in touch, and many apologies for the delayed response.
>
> I'm sorry to hear that you are having difficulty getting your SoundBlaster Z
> to work on your system under Linux. I was involved in the upstreaming of the
> Creative CA0132 codec to the ALSA repository, so I might be able to help
> provide some information on this.
>
> The CA0132 codec was originally intended for and designed to support the
> Chromebook Pixel hardware architecture. I'm sorry to say that this means
> that the current codec isn't expected to work on other products that have
> different architectures, such as the Z series. Thank you very much for all
> the information and details you provided though, it is much appreciated, and
> I hope they may come in useful at a later date.
>
> Best regards,
> - Ian"
>
> We really need to get someone rolling on this because it's been far to long
> already with no sound with these cards. Creative either needs to be harassed
> or we as a community have to pull together and try to provide the developers
> with as much information as possible.
>
> Would Kernel mesages and /proc info on the Codec help at all or would it
> just produce a lot of spam for the Developers?

Maybe we can ask Ian for technical specification of all variants of SoundCore 3d chip aka ca0132?
If we get specs we can write driver.

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to Gregory Krzystek from comment #26)
> (In reply to Acidphase from comment #23)
> > Well just as an FYI I received this email today which may explain why this
> > whole codec/driver is a mess.
> >
> > "Hello Ron,
> > Thank you for getting in touch, and many apologies for the delayed
> response.
> >
> > I'm sorry to hear that you are having difficulty getting your SoundBlaster
> Z
> > to work on your system under Linux. I was involved in the upstreaming of
> the
> > Creative CA0132 codec to the ALSA repository, so I might be able to help
> > provide some information on this.
> >
> > The CA0132 codec was originally intended for and designed to support the
> > Chromebook Pixel hardware architecture. I'm sorry to say that this means
> > that the current codec isn't expected to work on other products that have
> > different architectures, such as the Z series. Thank you very much for all
> > the information and details you provided though, it is much appreciated,
> and
> > I hope they may come in useful at a later date.
> >
> > Best regards,
> > - Ian"
> >
> > We really need to get someone rolling on this because it's been far to long
> > already with no sound with these cards. Creative either needs to be
> harassed
> > or we as a community have to pull together and try to provide the
> developers
> > with as much information as possible.
> >
> > Would Kernel mesages and /proc info on the Codec help at all or would it
> > just produce a lot of spam for the Developers?
>
> Maybe we can ask Ian for technical specification of all variants of
> SoundCore 3d chip aka ca0132?
> If we get specs we can write driver.

It's worth a shot I'll shoot him an email see if I can get him in contact with Takashi as he's the man for the job lol.

[Acidphase (acidphase-linux-kernel-bugs)]

Created attachment 107072
alsa-info Kernel 3.8.11

I have Sound with my Sound Blaster Z!

So for kicks and giggles I decided to setup and compile an older kernel which was already in my /usr/src directory.

3.8.11-gentoo #3 SMP PREEMPT Thu Aug 1 02:32:53 EDT 2013 x86_64 AMD FX(tm)-8150 Eight-Core Processor AuthenticAMD GNU/Linux

I used the position fix as well:
rmmod snd_hda_intel
modprobe snd_hda_intel position_fix=1

I'm scared to reboot lol The sound is working fine clear and no crackle/hiss.

[mmstickman (mmstickman-linux-kernel-bugs)]

If you lose sound after a reboot, you will have to boot into Windows and then boot back into Linux to regain sound. I would also set your favourite settings in the control panel in Windows as it carries the same parameters, even EQ settings, to Linux.

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to mmstick from comment #29)
> If you lose sound after a reboot, you will have to boot into Windows and
> then boot back into Linux to regain sound. I would also set your favourite
> settings in the control panel in Windows as it carries the same parameters,
> even EQ settings, to Linux.

No need still have sound after reboot :D Sound system is still intact and working as intended. Hopefully they can take my information and go back to 3.8.11 to begin a regression process.

I'm not sure at what exact version sound stopped working and I'm not about to configure and compile all kernel versions leading up to 3.10/11 lol

[mmstickman (mmstickman-linux-kernel-bugs)]

(In reply to Acidphase from comment #30)
> (In reply to mmstick from comment #29)
> > If you lose sound after a reboot, you will have to boot into Windows and
> > then boot back into Linux to regain sound. I would also set your favourite
> > settings in the control panel in Windows as it carries the same parameters,
> > even EQ settings, to Linux.
>
> No need still have sound after reboot :D Sound system is still intact and
> working as intended. Hopefully they can take my information and go back to
> 3.8.11 to begin a regression process.
>
> I'm not sure at what exact version sound stopped working and I'm not about
> to configure and compile all kernel versions leading up to 3.10/11 lol

I'm certain that if you did that, you'd really lose sound. I lost sound on my Recon3D every time I upgraded the kernel and rebooted. It was only when I booted into Windows and then booted back into Ubuntu did sound return.

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to mmstick from comment #31)
> (In reply to Acidphase from comment #30)
> > (In reply to mmstick from comment #29)
> > > If you lose sound after a reboot, you will have to boot into Windows and
> > > then boot back into Linux to regain sound. I would also set your favorite
> > > settings in the control panel in Windows as it carries the same
> parameters,
> > > even EQ settings, to Linux.
> >
> > No need still have sound after reboot :D Sound system is still intact and
> > working as intended. Hopefully they can take my information and go back to
> > 3.8.11 to begin a regression process.
> >
> > I'm not sure at what exact version sound stopped working and I'm not about
> > to configure and compile all kernel versions leading up to 3.10/11 lol
>
> I'm certain that if you did that, you'd really lose sound. I lost sound on
> my Recon3D every time I upgraded the kernel and rebooted. It was only when I
> booted into Windows and then booted back into Ubuntu did sound return.

I think what you're trying to convey is that if I boot back into one of the newer kernels where I would lose sound then when I go back to the working kernel(3.8.11) I wouldn't get sound back unless I booted into windows?

If so this isn't the case I booted into 2 newer kernels 3.10.4 and 3.11-rc3 as expected so sound in either. Then I rebooted once more with 3.8.11 and I had sound again. Which tells me the sound driver in 3.8.11 is pretty stable which is a good thing in our scenario. (I haven't been in windows in days).

You have to keep in mind that our systems are most likely entirely different for one so is our cards I have a Blaster Z so the BIOS in them is different, I know our system BIOS is different. The sound device is being re-initialized with no problems for me. I also have a UEFI setup which Gentoo is installed with efivars present and working as well.

[Acidphase (acidphase-linux-kernel-bugs)]

Sigh just wanted to inform people that yes I have sound via Digital but not analogue.

I have sound via digital with kernels:
linux-3.4.55-gentoo
linux-3.8.11-gentoo
linux-3.8.13-gentoo

Kernels I've also triedI haven't tested analogue on all of them)

linux-3.9.11-gentoo-r1
linux-3.10.
linux-3.10-rc1
linux-3.10-rc2
linux-3.10-rc6
linux-3.10.1-gentoo
linux-3.10.2-gentoo
linux-3.10.3-gentoo-r1
linux-3.10.4-gentoo
linux-3.10.5-gentoo
linux-3.11-rc2
linux-3.11-rc3
linux-3.11-rc4

I may start downloading some vanilla kernels via git and go from there to try and ascertain the exact version where I lose digital sound. I just don't know how much time I have to mess with this today.

Sorry for the bad news...

[Acidphase (acidphase-linux-kernel-bugs)]

Okay after testing a ridiculous amount of kernels I pretty much have Digital Audio from 3.1.1(when ca0132 was introduced) to 3.8.13 and then 3.9.1 where the DSP firmware was added in I lose Digital Audio. Not sure if this info helps but it's what I have determined.

I obviously didn't test all the incremental releases if someone has that much time on them enjoy lol

[Akovia Voka (akovia1)]

I'm having the same trouble with the Recon 3D card in 12.04LTS. Is there any hope with the posted fixes to try for my older distro? I also have HDMI sound but the command posted didn't work for me to disable it.

[Michael Murphy (mmstick)]

As far as I know you need to upgrade your kernel to something like 3.10 and then I've only been able to get sound out of the headphone port on the back. No sound from any other port.

[Michael Murphy (mmstick)]

If you lose sound you need to boot into Windows and then boot back into Linux. The control panel settings in Windows also applies in Linux.

[Raymond (superquad-vortex2)]

>> As far as I know you need to upgrade your kernel to something like 3.10 and then I've only been able to get sound out of the headphone port on the back. No sound from any other port.

can you post the output of alsa-info.sh, pactl list and pactl stat ?

[Raymond (superquad-vortex2)]

but you are still using Driver version: k3.8.0-27-generic

pulseaudio expect those Jack detection kcontrol for the headphone Jack which return true when headphone is plugged

iface CARD
  name 'Front Headphone Jack'
  value false
  comment {
   access read
   type BOOLEAN
   count 1

[Michael Murphy (mmstick)]

Nothing changes when using 3.10 which I've used before, and I don't really understand the second thing you said.

[Raymond (superquad-vortex2)]

pulseaudio expect front headphone Jack kcontrol similar to alc892

Node 0x1b [Pin Complex] wcaps 0x40058f: Stereo Amp-In Amp-Out
  Control: name="Headphone Playback Switch", index=0, device=0
    ControlAmp: chs=3, dir=Out, idx=0, ofs=0
  Control: name="Front Headphone Jack", index=0, device=0
  Amp-In caps: ofs=0x00, nsteps=0x03, stepsize=0x27, mute=0
  Amp-In vals: [0x00 0x00]
  Amp-Out caps: ofs=0x00, nsteps=0x00, stepsize=0x00, mute=1
  Amp-Out vals: [0x00 0x00]
  Pincap 0x0001373e: IN OUT HP EAPD Detect Trigger
    Vref caps: HIZ 50 GRD 80 100
  EAPD 0x2: EAPD
  Pin Default 0x02214c20: [Jack] HP Out at Ext Front
    Conn = 1/8, Color = Green
    DefAssociation = 0x2, Sequence = 0x0
  Pin-ctls: 0xc0: OUT HP VREF_HIZ
  Unsolicited: tag=01, enabled=1
  Power states: D0 D1 D2 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 5
     0x0c 0x0d 0x0e 0x0f 0x26*

but CA0132 did not have those Jack detection kcontrol and have speaker playback switch at the green line out jack

Node 0x0b [Pin Complex] wcaps 0x400581: Stereo
  Control: name="Speaker Playback Switch", index=0, device=0
    ControlAmp: chs=1, dir=Out, idx=0, ofs=0
  Pincap 0x00010014: OUT EAPD Detect
  EAPD 0x0:
  Pin Default 0x01014010: [Jack] Line Out at Ext Rear
    Conn = 1/8, Color = Green
    DefAssociation = 0x1, Sequence = 0x0
  Pin-ctls: 0x40: OUT
  Unsolicited: tag=06, enabled=1
  Power states: D0 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 1
     0x02

Node 0x0f [Pin Complex] wcaps 0x400581: Stereo
  Control: name="Headphone Playback Switch", index=0, device=0
    ControlAmp: chs=1, dir=Out, idx=0, ofs=0
  Control: name="Headphone Playback Volume", index=0, device=0
    ControlAmp: chs=3, dir=Out, idx=0, ofs=0
  Pincap 0x0000001c: OUT HP Detect
  Pin Default 0x0221401f: [Jack] HP Out at Ext Front
    Conn = 1/8, Color = Green
    DefAssociation = 0x1, Sequence = 0xf
  Pin-ctls: 0xc0: OUT HP
  Unsolicited: tag=06, enabled=1
  Power states: D0 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 1
     0x02
Node 0x10 [Pin Complex] wcaps 0x400581: Stereo
  Pincap 0x0000001c: OUT HP Detect
  Pin Default 0x02216011: [Jack] HP Out at Ext Front
    Conn = 1/8, Color = Orange
    DefAssociation = 0x1, Sequence = 0x1
  Pin-ctls: 0x00:
  Unsolicited: tag=06, enabled=1
  Power states: D0 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 1
     0x03
Node 0x11 [Pin Complex] wcaps 0x40058b: Stereo Amp-In
  Amp-In caps: ofs=0x00, nsteps=0x03, stepsize=0x27, mute=0
  Amp-In vals: [0x00 0x00]
  Pincap 0x00003734: IN OUT Detect
    Vref caps: HIZ 50 GRD 80 100
  Pin Default 0x02012014: [Jack] Line Out at Ext Front
    Conn = 1/8, Color = Grey
    DefAssociation = 0x1, Sequence = 0x4
  Pin-ctls: 0x24: IN VREF_80
  Unsolicited: tag=06, enabled=1
  Power states: D0 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 1
     0x04

[Michael Murphy (mmstick)]

So the solution is just wait until such controls are coded for CA0132? From what I heard, CA0132 was primarily coded for the Chromebook Pixel hardware rather than the Recon3D or Sound Blaster Z chips.

[Raymond (superquad-vortex2)]

try hda-jack-sense-test for your card 1 to find out which node is the headphone Jack , line out, Michael and line in jacks

[Michael Murphy (mmstick)]

How do I get hda-jack-sense-test?

[Raymond (superquad-vortex2)]

https://launchpad.net/~diwic/+archive/hda

install snd-hda-tools

[Michael Murphy (mmstick)]

mmstick@mmstick-PC:~$ sudo hda-jack-sense-test
Pin 0x03 (Unknown Digital Out): present = Yes
Pin 0x05 (Unknown Digital Out): present = No
Pin 0x07 (Unknown Digital Out): present = No
Pin 0x09 (Unknown Digital Out): present = No
Pin 0x0b (Unknown Digital Out): present = No
Pin 0x0d (Unknown Digital Out): present = No

[Raymond (superquad-vortex2)]

did you specify card since you have three hda-intel cards ?

Card hw:2 'Creative'/'HDA Creative at 0xfe504000 irq 47'
  Mixer name : 'Creative CA0132'
  Components : 'HDA:11020011,11020013,00100918'
  Controls : 21
  Simple ctrls : 8
Simple mixer control 'Headphone',0
  Capabilities: pvolume pswitch pswitch-joined penum
  Playback channels: Front Left - Front Right
  Limits: Playback 0 - 31
  Mono:
  Front Left: Playback 2 [6%] [-4.00dB] [on]
  Front Right: Playback 2 [6%] [-4.00dB] [on]
Simple mixer control 'Speaker',0
  Capabilities: pswitch pswitch-joined penum
  Playback channels: Mono
  Mono: Playback [off]

[Michael Murphy (mmstick)]

I disabled onboard audio but I don't know how to specify a card with this program because it does not have a man page.

[Raymond (superquad-vortex2)]

def parseoptions():
    from optparse import OptionParser
    parser = OptionParser()
    parser.add_option("-c", "--card", dest="cardindex", default=0, metavar="CARD",
        help="card index (as can be seen in /proc/asound/cards)")
    parser.add_option("-d", "--codec", dest="codecindex", default=0, metavar="CODEC",
        help="codec device index (as can be seen in /proc/asound/cardX/codecY)")
    parser.add_option("-a", "--allpins", dest="tryallpins", default=False,
        action="store_true",
        help="also check pins which (probably) are not physical jacks")
    (options, args) = parser.parse_args()
    return int(options.cardindex), int(options.codecindex), options.tryallpins

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/?id=825315bc5b5c33e5af5124ff100ef05a30ad722f

seem have three capture devices

info++;
+ info->name = "CA0132 Analog Mic-In2";
+ info->stream[SNDRV_PCM_STREAM_CAPTURE] = ca0132_pcm_analog_capture;
+ info->stream[SNDRV_PCM_STREAM_CAPTURE].substreams = 1;
+ info->stream[SNDRV_PCM_STREAM_CAPTURE].nid = spec->adcs[1];
+ codec->num_pcms++;
+
+ info++;
+ info->name = "CA0132 What U Hear";
+ info->stream[SNDRV_PCM_STREAM_CAPTURE] = ca0132_pcm_analog_capture;
+ info->stream[SNDRV_PCM_STREAM_CAPTURE].substreams = 1;
+ info->stream[SNDRV_PCM_STREAM_CAPTURE].nid = spec->adcs[2];
+ codec->num_pcms++;

[Michael Murphy (mmstick)]

mmstick@mmstick-PC:~$ sudo hda-jack-sense-test -c 1 -d 1 -a
Pin 0x0b (Green Line Out)
Pin 0x0c (Purple SPDIF Out)
Pin 0x0d (Yellow SPDIF Out)
Pin 0x0e (Blue SPDIF In)
Pin 0x0f (Green HP Out)
Pin 0x10 (Orange HP Out)
Pin 0x11 (Grey Line Out)
Pin 0x12 (Pink Mic)
Pin 0x13 (Internal Line In)
Pin 0x18 (Not connected)

Green HP Out is the silver headphone port on the back of the card. Even if they say Present = Yes when plugged in to other ports, there is no sound.

[mmstickman (mmstickman-linux-kernel-bugs)]

I'm not sure about digital audio for mine, but I've only been able to get audio from the silver headphone out port on the back of the sound card and nowhere else. The system detects that something gets plugged into other ports, but there is no sound being directed to it over than silent static from the port.

mmstick@mmstick-PC:~$ sudo hda-jack-sense-test -c 1 -d 1 -a
Pin 0x0b (Green Line Out): present = No
Pin 0x0c (Purple SPDIF Out): present = No
Pin 0x0d (Yellow SPDIF Out): present = No
Pin 0x0e (Blue SPDIF In): present = No
Pin 0x0f (Green HP Out): present = Yes
Pin 0x10 (Orange HP Out): present = No
Pin 0x11 (Grey Line Out): present = No
Pin 0x12 (Pink Mic): present = No
Pin 0x13 (Internal Line In): present = No
Pin 0x18 (Not connected): present = No

Green HP Out of course is the silver headphone port on the back.

[Raymond (superquad-vortex2)]

have you sent email to the author

ALSA: hda/ca0132: Add PCM enhancements
Change the max channels of analog playback to 6.

 .channels_min = 2,
- .channels_max = 2,
+ .channels_max = 6,
.ops = {

but max channels still 2

 info->name = "CA0132 Analog";
 info->stream[SNDRV_PCM_STREAM_PLAYBACK] = ca0132_pcm_analog_playback;
 info->stream[SNDRV_PCM_STREAM_PLAYBACK].nid = spec->dacs[0];
 info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max =
  spec->multiout.max_channels;

spec->multiout.dac_nids = spec->dacs;
 spec->multiout.num_dacs = 3;
 spec->multiout.max_channels = 2;

[Raymond (superquad-vortex2)]

using hda-emu , it create 4 capture devices and digital capture is not device 1 (defined in hda-intel.conf)

autoconfig: line_outs=2 (0xb/0x11/0x0/0x0/0x0) type:line
   speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
   hp_outs=2 (0x10/0xf/0x0/0x0/0x0)
   mono: mono_out=0x0
   dig-out=0xc/0xd
   inputs:
     Mic=0x12
     Line=0x13
   dig-in=0xe
# Building PCMs...

Attach PCM dev 0, name CA0132 Analog, type audio, play #1, capture #1

Attach PCM dev 1, name CA0132 Analog Mic-In2, type audio, play #0, capture #1

Attach PCM dev 2, name CA0132 What U Hear, type audio, play #0, capture #1

Attach PCM dev 3, name CA0132 Digital, type SPDIF, play #1, capture #1
# Init and building controls...

[Raymond (superquad-vortex2)]

you are using driver later than this patch since line-in and Mic-in not exist

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/?id=55cf87fe0e9783e25f442be1e48b8319d86131ea

simple mixer control 'Line',0
  Capabilities: cvolume cswitch penum
  Capture channels: Front Left - Front Right
  Limits: Capture 0 - 99
  Front Left: Capture 0 [0%] [-90.00dB] [off]
  Front Right: Capture 0 [0%] [-90.00dB] [off]
Simple mixer control 'Mic',0
  Capabilities: cvolume cswitch penum
  Capture channels: Front Left - Front Right
  Limits: Capture 0 - 99
  Front Left: Capture 30 [30%] [-60.00dB] [on]
  Front Right: Capture 30 [30%] [-60.00dB] [on]

[Michael Murphy (mmstick)]

I haven't sent emails to anyone. Considering I'm using kernel 3.8 (Ubuntu 13.04) it is understandable that I'm using a newer driver. I don't really understand anything else you are talking about.

[Raymond (superquad-vortex2)]

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1038651

take a look at comment 14 and 15

https://launchpadlibrarian.net/112961024/Card0.Codecs.codec.0.txt

Node 0x07 [Audio Input] wcaps 0x100591: Stereo
  Control: name="Mic-In Capture Switch", index=0, device=0
    ControlAmp: chs=3, dir=In, idx=0, ofs=0
  Control: name="Mic-In Capture Volume", index=0, device=0
    ControlAmp: chs=3, dir=In, idx=0, ofs=0
  Device: name="CA0132 Analog", type="Audio", device=0
  Converter: stream=0, channel=0
  SDI-Select: 0
  PCM:
    rates [0x1ec]: 16000 22050 44100 48000 88200 96000
    bits [0x1f]: 8 16 20 24 32
    formats [0x1]: PCM
  Unsolicited: tag=00, enabled=0
  Power states: D0 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 1
     0x12
Node 0x08 [Audio Input] wcaps 0x10059b: Stereo Amp-In
  Control: name="Line-In Capture Switch", index=0, device=0
    ControlAmp: chs=3, dir=In, idx=0, ofs=0
  Control: name="Line-In Capture Volume", index=0, device=0
    ControlAmp: chs=3, dir=In, idx=0, ofs=0
  Amp-In caps: ofs=0x5a, nsteps=0x63, stepsize=0x03, mute=1
  Amp-In vals: [0x00 0x00]
  Converter: stream=0, channel=0
  SDI-Select: 0
  PCM:
    rates [0x1ec]: 16000 22050 44100 48000 88200 96000
    bits [0x1f]: 8 16 20 24 32
    formats [0x1]: PCM
  Unsolicited: tag=00, enabled=0
  Power states: D0 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 1
     0x11

[Michael Murphy (mmstick)]

Which means? I don't know much, if anything, about ALSA or sound drivers/codecs.

[Raymond (superquad-vortex2)]

http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/refs/tags

http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/log/sound/pci/hda/patch_ca0132.c

you have to ask David whether CA0132 support DSP firmware in dkms ?

did you find the firmware file in creative windows driver CD ?

[Michael Murphy (mmstick)]

I haven't done anything with firmware files nor do I have a driver CD.

[Raymond (superquad-vortex2)]

http:/git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/patch_ca0132.c?id=01ef7dbffb411d9d78d1150b268d9c757f9f2f93

part of DSP code was added in 2012-11-28

but 3.8-RC1 was 2012-12-22

so you have to ask Ubuntu audio team to check whether k3.8.0-27-generic contain part of the DSP code

[Acidphase (acidphase-linux-kernel-bugs)]

Well I was able to get analog audio out of the front HD Audio pin connector which makes me happy at least I have sound via analog now.

This tells me the drivers is definitely working I have a feeling the jack assignments may be the issue we're having.

I'm uses the current kernel 3.11.0

[tiwai (tiwai-linux-kernel-bugs)]

Oh good to hear you got something! What was the workaround you did exactly?

[p_narciso (pnarciso-linux-kernel-bugs)]

I'm running Ubuntu 13.10 + SoundBlaster Z with the latest kernel 3.11 and I can't have sound with any analog connectors including the front panel ones.

[mmstickman (mmstickman-linux-kernel-bugs)]

(In reply to Paulo Narciso from comment #38)
> I'm running Ubuntu 13.10 + SoundBlaster Z with the latest kernel 3.11 and I
> can't have sound with any analog connectors including the front panel ones.

Have you tried booting into Windows and then booting back into Ubuntu? I have to do that sometimes with my Recon3d.

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to Takashi Iwai from comment #37)
> Oh good to hear you got something! What was the workaround you did exactly?

Nothing really just installed the latest kernel and built the audio drivers as modules then the usual position fix we have been using.

3.11.0-gentoo #1 SMP PREEMPT Fri Sep 6 15:41:02 EDT 2013 x86_64 AMD FX(tm)-8150 Eight-Core Processor AuthenticAMD GNU/Linux

rmmod snd_hda_intel
modprobe snd_hda_intel position_fix=1

[madbiologist (me-again)]

Firmware files for the Creative CA0132 HD-audio codec were added in linux-firmware 1.107. Ubuntu 13.04 "Raring Ringtail" is using the older linux-firmware 1.106 package.

[Michael Murphy (mmstick)]

With firmware 1.107 and kernel 3.10, the green speaker out and silver headphone out work, but the front headphone port still does not work at all. I don't have any equipment for testing digital out though.

[Raymond (superquad-vortex2)]

https://wiki.ubuntu.com/Audio/AlsaInfo

post the output of alsa-info.sh in kernel 3.10 with the green speaker out and silver headphone out work ‎

[Raymond (superquad-vortex2)]

can you post the system log (dmesg） since some info is missing (e.g. those HP and input pins after autoconfig :line_outs) ?

3.945714] hda-intel 0000:06:00.0: Using LPIB position fix
[ 3.945716] hda-intel 0000:06:00.0: Force to non-snoop mode
[ 3.966439] autoconfig: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
--
[ 6.510248] ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING.
[ 6.520474] input: HDA Creative Front Headphone as /devices/pci0000:00/0000:00:0a.0/0000:06:00.0/sound/card1/input14
[ 6.520548] input: HDA Creative Line Out as /devices/pci0000:00/0000:00:0a.0/0000:06:00.0/sound/card1/input15
[ 6.520609] input: HDA Creative Line as /devices/pci0000:00/0000:00:0a.0/0000:06:00.0/sound/card1/input16
[ 6.520666] input: HDA Creative Mic as /devices/pci0000:00/0000:00:0a.0/0000:06:00.0/sound/card1/input17
[ 6.537660] EXT4-fs (sdc1): mounted filesystem with ordered data mode. Opts: (null)
--
[ 12.016154] <6>[fglrx] Reserved FB block: Unshared offset:bfff4000, size:c000
[ 12.230498] HDMI: ELD buf size is 0, force 128
[ 12.230553] HDMI: invalid ELD data byte 0
[ 12.230618] HDMI: ELD buf size is 0, force 128
[ 12.230632] HDMI: invalid ELD data byte 0
[ 12.526917] HDMI: ELD buf size is 0, force 128
[ 12.526955] HDMI: invalid ELD data byte 0
[ 12.630891] hda-intel 0000:01:00.1: Handle VGA-switcheroo audio client
[ 12.630895] hda-intel 0000:01:00.1: Force to non-snoop mode
[ 12.630928] snd_hda_intel 0000:01:00.1: irq 89 for MSI/MSI-X
[ 12.633529] hda-intel 0000:01:00.1: Enable sync_write for stable communication
[ 12.635716] HDMI: ELD buf size is 0, force 128
[ 12.635743] HDMI: invalid ELD data byte 0
[ 12.636067] input: HDA ATI HDMI HDMI/DP,pcm=11 as /devices/pci0000:00/0000:00:02.0/0000:01:00.1/sound/card0/input18
[ 12.636136] input: HDA ATI HDMI HDMI/DP,pcm=10 as /devices/pci0000:00/0000:00:02.0/0000:01:00.1/sound/card0/input19
[ 12.636197] input: HDA ATI HDMI HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:02.0/0000:01:00.1/sound/card0/input20
[ 12.636262] input: HDA ATI HDMI HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:02.0/0000:01:00.1/sound/card0/input21
[ 12.636325] input: HDA ATI HDMI HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:02.0/0000:01:00.1/sound/card0/input22
[ 12.636389] input: HDA ATI HDMI HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:02.0/0000:01:00.1/sound/card0/input23
[ 12.639339] hda_intel: Disabling MSI
[ 12.639342] hda-intel 0000:06:00.0: Using LPIB position fix
[ 12.639344] hda-intel 0000:06:00.0: Force to non-snoop mode
[ 12.655213] autoconfig: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
--
[ 13.206689] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
[ 13.428262] input: HDA Creative Front Headphone as /devices/pci0000:00/0000:00:0a.0/0000:06:00.0/sound/card1/input24
[ 13.428330] input: HDA Creative Line Out as /devices/pci0000:00/0000:00:0a.0/0000:06:00.0/sound/card1/input25
[ 13.428401] input: HDA Creative Line as /devices/pci0000:00/0000:00:0a.0/0000:06:00.0/sound/card1/input26

[Michael Murphy (mmstick)]

After I rebooted with my headphones plugged into the green speaker out, I could not get sound out of my silver headphone port. I tried booting into Windows 7 and then back to Ubuntu to bring the sound back, which usually works, but it didn't. Instead, booting into kernel 3.8 brought sound back to my silver headphone ports, and thus booting back into 3.10 I now have sound in the port again.

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

http://www.alsa-project.org/main/index.php/Detailed_changes_v1.0.26_v1.0.27

Sound Blaster HDA CA0132 Firmware
 - Update ca0132 firmware files
 ctefx.bin:
 Update the CA0132 DSP image firmware file. This version:
 - Includes key-click reduction feature.
 - Fixes noisy output when GraphicEQ and SpeakerEQ are both turned on.
 ctspeq.bin:
 Add firmware file containing coefficient preset for SpeakerEQ
 DSP feature. This preset has been tuned for Chromebook Pixel hardware.
 Other CA0132 systems will run without loading this binary.
 - Add firmware for CA0132 DSP

does it mean that only chromebook pixel need the firmware and other should not use the firmware ?

[mmstickman (mmstickman-linux-kernel-bugs)]

(In reply to Raymond from comment #41)
> http://www.alsa-project.org/main/index.php/Detailed_changes_v1.0.26_v1.0.27
>
>
> Sound Blaster HDA CA0132 Firmware
> - Update ca0132 firmware files
> ctefx.bin:
> Update the CA0132 DSP image firmware file. This version:
> - Includes key-click reduction feature.
> - Fixes noisy output when GraphicEQ and SpeakerEQ are both turned on.
> ctspeq.bin:
> Add firmware file containing coefficient preset for SpeakerEQ
> DSP feature. This preset has been tuned for Chromebook Pixel hardware.
> Other CA0132 systems will run without loading this binary.
> - Add firmware for CA0132 DSP
>
> does it mean that only chromebook pixel need the firmware and other should
> not use the firmware ?

Using Ubuntu 13.04 here with kernel 3.10 and linux-firmware-1.07 which has this and it works fine with my Recon3D, sort of. I can only get sound out of my silver headphone port and green speaker out now, and if I lose sound after a reboot, I find that if I boot into kernel 3.8 and back into kernel 3.10 I regain sound.

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

Simple mixer control 'HP/Speaker',0
  Capabilities: pswitch pswitch-joined penum
  Playback channels: Mono
  Mono: Playback [off]
Simple mixer control 'HP/Speaker Auto Detect',0
  Capabilities: pswitch pswitch-joined penum
  Playback channels: Mono
  Mono: Playback [off]

how about the 10 band equalizer ?

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/?id=ef6b2eada3b8c1b21f6479d7480ea7030183fe1d

+/* Enable this to see controls for tuning purpose. */
+/*#define ENABLE_TUNING_CONTROLS*/
+

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

Simple mixer control 'Equalizer',0
  Capabilities: pswitch pswitch-joined penum
  Playback channels: Mono
  Mono: Playback [on]

should user allow to turn on equalizer when the driver does not load ctspeq.bin

[mmstickman (mmstickman-linux-kernel-bugs)]

I haven't seen a 10-band equaliser in alsamixer. There are entries displaying names of features found in the control panel in Windows, but there is nothing you can do with them.

[p_narciso (pnarciso-linux-kernel-bugs)]

Just to inform that I had sound one time, through the front panel headphone connector. But when I restarted or shutdown the machine, sound is gone.
There's some issues with ca0132 driver.

First issue is that the firmware file most of the time won't get loaded even with the position fix 1. When the firmware is not loaded, there's no sound.

Second issue seems to come from the fact that none of the output connectors are recognized, the only connectors recognized are the front panel ones.

Here's my dmesg:
[ 5.071768] autoconfig: line_outs=1 (0xd/0x0/0x0/0x0/0x0) type:line
[ 5.071769] speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 5.071770] hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 5.071771] mono: mono_out=0x0
[ 5.071771] dig-out=0xc/0x0
[ 5.071772] inputs:
[ 5.071773] Mic=0x12
[ 5.071773] dig-in=0xe
[ 7.352799] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx
[ 7.352825] IPv6: ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
[ 7.671704] ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING.
[ 7.696258] input: HDA Creative Front Headphone as /devices/pci0000:00/0000:00:1c.4/0000:04:00.0/sound/card2/input15
[ 7.696321] input: HDA Creative Mic as /devices/pci0000:00/0000:00:1c.4/0000:04:00.0/sound/card2/input16

I'm no programmer, so I can't fix this by myself, but considering that Creative released the source code for the driver, it shouldn't be that hard to fix this by an experienced programmer.

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

https://launchpadlibrarian.net/149825222/rear-silver-headphone-out.txt

it is strange that unsolicited event tag at node 0x10 is 0x1 instead of 0x10

seem snd_hda_jack overwrite the event tag

#define UNSOL_TAG_HP 0x10
#define UNSOL_TAG_AMIC1 0x12
#define UNSOL_TAG_DSP 0x16

static void ca0132_init_unsol(struct hda_codec *codec)
{
 snd_hda_jack_detect_enable(codec, UNSOL_TAG_HP, UNSOL_TAG_HP);
 snd_hda_jack_detect_enable(codec, UNSOL_TAG_AMIC1, UNSOL_TAG_AMIC1);
}

Node 0x10 [Pin Complex] wcaps 0x400581: Stereo
  Control: name="Front Headphone Jack", index=0, device=0
  Pincap 0x0000001c: OUT HP Detect
  Pin Default 0x02216011: [Jack] HP Out at Ext Front
    Conn = 1/8, Color = Orange
    DefAssociation = 0x1, Sequence = 0x1
  Pin-ctls: 0x00:
  Unsolicited: tag=01, enabled=1
  Power states: D0 D3 EPSS
  Power: setting=D0, actual=D0
  Connection: 1
     0x03

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

> I haven't seen a 10-band equaliser in alsamixer.

have you define ENABLE_TUNING_CONTROLS ?

need

#include <sound/tlv.h>

since those 10 band EQ controls have dB scale

[mmstickman (mmstickman-linux-kernel-bugs)]

(In reply to Raymond from comment #48)
> > I haven't seen a 10-band equaliser in alsamixer.
>
> have you define ENABLE_TUNING_CONTROLS ?
>
> need
>
> #include <sound/tlv.h>
>
>
> since those 10 band EQ controls have dB scale

I'm not a programmer

[mmstickman (mmstickman-linux-kernel-bugs)]

I've decided that I'm just going to stay with kernel 3.8 for now. Kernel 3.10 keeps causing me to lose sound every time I reboot with my Recon3D.

[p_narciso (pnarciso-linux-kernel-bugs)]

This arrived two day ago, what does it change?

ALSA: hda/ca0132: Staticize codec_send_command
'codec_send_command' is used only in this file. Make it static. Signed-off-by: Sachin Kamat <email address hidden> Signed-off-by: Takashi Iwai <email address hidden>
Diffstat
-rw-r--r-- sound/pci/hda/patch_ca0132.c 2

1 files changed, 1 insertions, 1 deletions
diff --git a/sound/pci/hda/patch_ca0132.c b/sound/pci/hda/patch_ca0132.c
index 6e9876f..54d1479 100644
--- a/sound/pci/hda/patch_ca0132.c
+++ b/sound/pci/hda/patch_ca0132.c
@@ -759,7 +759,7 @@ struct ca0132_spec {
/*
* CA0132 codec access
*/
-unsigned int codec_send_command(struct hda_codec *codec, hda_nid_t nid,
+static unsigned int codec_send_command(struct hda_codec *codec, hda_nid_t nid,
unsigned int verb, unsigned int parm, unsigned int *res)
{
unsigned int response;

[craftkiller (craftkiller-linux-kernel-bugs)]

(In reply to Paulo Narciso from comment #51)
> This arrived two day ago, what does it change?

In C marking something as "static" makes it only accessible locally within the file. This way other files can have their own "codec_send_command" without having name collisions. It won't change the functionality at all, they're just keeping the code base and the global namespace clean.

[mikhail.v.gavrilov (mikhail.v.gavrilov-linux-kernel-bugs)]

Creative Sound Blaster ZXR (SB1510) didn't play any sound with 3.11 kernel.

[Michael Murphy (mmstick)]

Using kernel 3.11, pretty much the same stuff happens. It works the first time, then after a reboot I won't be able to get sound until I do something like boot into Windows 7, reboot and boot into kernel 3.8, reboot and then boot into kernel 3.11.

[Denis Sultu (denissultu)]

Hi,

i'm having the same issue. At first i could get audio in Stereo from the spdif port on Ubuntu 13.04 with Kernel 3.8 but no 5.1. I tried install 3.10 and then 3.11 but both made me lose sound completely. I don't have a Windows partition so i couldnt try booting into windows.

I have a creative recon3d fatal1ty champion.

Do you think that if i install the latest version of Ubuntu 13.10 it will change anything?

What about using the latest version of Mint or Fedora? Would that have better chances of working?

Please i need help. I've been playing quite a bit with Linux these past few months and i understand well commands but when it comes to sound, eiditing files or pulseaudio im completely new and would need some intructions if i'm to try any suggestions.

thanks,

Denis

[Raymond (superquad-vortex2)]

> With firmware 1.107 and kernel 3.10, the green speaker out and silver headphone out work, but the front headphone port still does not work at all. I don't have any equipment for testing digital out though.

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/?id=95c6e9cb774979c270f0ecb9ec819d02592ec89f

do you mean the grey Jack since the driver hardcode the node of headphone to 0x0f ?

+
+static void ca0132_config(struct hda_codec *codec)
+{
+ struct ca0132_spec *spec = codec->spec;
+ struct auto_pin_cfg *cfg = &spec->autocfg;
+
+ /* line-outs */
+ cfg->line_outs = 1;
+ cfg->line_out_pins[0] = 0x0b; /* front */
+ cfg->line_out_type = AUTO_PIN_LINE_OUT;
+
+ spec->dacs[0] = 0x02;
+ spec->out_pins[0] = 0x0b;
+ spec->multiout.dac_nids = spec->dacs;
+ spec->multiout.num_dacs = 1;
+ spec->multiout.max_channels = 2;
+
+ /* headphone */
+ cfg->hp_outs = 1;
+ cfg->hp_pins[0] = 0x0f;
+
+ spec->hp_dac = 0;
+ spec->multiout.hp_nid = 0;
+
+ /* inputs */
+ cfg->num_inputs = 2; /* Mic-in and line-in */
+ cfg->inputs[0].pin = 0x12;
+ cfg->inputs[0].type = AUTO_PIN_MIC;
+ cfg->inputs[1].pin = 0x11;
+ cfg->inputs[1].type = AUTO_PIN_LINE_IN;
+
+ /* Mic-in */
+ spec->input_pins[0] = 0x12;
+ spec->input_labels[0] = "Mic-In";
+ spec->adcs[0] = 0x07;
+
+ /* Line-In */
+ spec->input_pins[1] = 0x11;
+ spec->input_labels[1] = "Line-In";
+ spec->adcs[1] = 0x08;
+ spec->num_inputs = 2;
+}

[Raymond (superquad-vortex2)]

http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/?id=a73d511c4867c5aa75a9ab50f7e73d5086c48cda

HP is expected at node 0x10

#define UNSOL_TAG_HP 0x10
#define UNSOL_TAG_AMIC1 0x12
#define UNSOL_TAG_DSP 0x16

+static void ca0132_init_unsol(struct hda_codec *codec)
+{
+ snd_hda_jack_detect_enable(codec, UNSOL_TAG_HP, UNSOL_TAG_HP);
+ snd_hda_jack_detect_enable(codec, UNSOL_TAG_AMIC1, UNSOL_TAG_AMIC1);
+}
+
 static void refresh_amp

[Michael Murphy (mmstick)]

The jack is very much a shiny silver colour physically. It's the jack with the 600 ohm headphone amp.

Getting CA0132 cards to work nicely in Linux is rather tricky. If you lose sound from a newer kernel, it's kind of complicated to get it back. I noticed that kernel 3.10 had the worst support, which caused some rather awful issues so I would avoid that when using CA0132 sound cards. Apparently using 3.10 would essentially kill your sound after a reboot which may or may not come back if you boot into kernel 3.8 and Windows. Every time I would lose sound, I would just boot into Windows and then boot into kernel 3.8 just to be safe. I don't have much extensive testing of kernel 3.11, but using the liquorix kernel (based on 3.11) right now, I haven't had any problems with sound after a reboot so far. I've rebooted 3 times and still have sound.

[Raymond (superquad-vortex2)]

https://www.google.com/intl/en/chrome/devices/chromebook-pixel/#pixel-specs

The Chormebook has a Combo headphone/mic jack which is different from those pcie sound cards

[Michael Murphy (mmstick)]

After a fourth reboot with the Liquorix 3.11 kernel, I've lost sound again with my headphones. I can get sound from the green speaker out port, but cannot get sound from the headphone port.

[mmstickman (mmstickman-linux-kernel-bugs)]

With kernel 3.11, I get sound from the headphone port the first 2-3 times I boot into it, but after a reboot I will definitely lose sound from the headphone port. If I plug my headphones into the rear green speaker port, I get sound, even if lower quality. Still can't get sound from the front ports though.

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

if the coded is designed for chormium pixel, you should ask the two authors from chrome.org , who submited patches ,to obtain the alsa-info of the chromium pixel,

[Michael Murphy (mmstick)]

I found a fix for the front headphone port. It seems by default, CA0132 has HP/Speaker and HP/Speaker Auto Detect muted. Using alsamixer and navigating to the control and pressing M unmutes them.

[mmstickman (mmstickman-linux-kernel-bugs)]

Found a solution for my problems with the front headphone port. By default CA0132 has the HP/Speaker and HP/Speaker Auto Detect muted. Unmuting them by going into alsamixer and pressing M on their controls unmutes them and sound works.

[p_narciso (pnarciso-linux-kernel-bugs)]

Is there any news about this issue? Is someone working on this? Has anyone received some feeback from Creative?
With SteamOS being released it doesn't make sense not having support for this gaming sound card.

[mmstickman (mmstickman-linux-kernel-bugs)]

The lack of communication speaks volumes in itself. There's incompetence all over the world, so it's not too surprising that Creative isn't too bright.

[p_narciso (pnarciso-linux-kernel-bugs)]

Creative released driver support for Ca0132 chip, and there's available source code. It's just a matter of configuring the output pins of discrete cards.
But it seems no one seems to care.

[mmstickman (mmstickman-linux-kernel-bugs)]

I doubt many developers capable of fixing it have a Creative card since they aren't exactly the target audience of Creative. It'd be better if Creative just had at least a single developer finish the code.

[Michael Murphy (mmstick)]

After an upgrade from Ubuntu 13.04 to Ubuntu 13.10 I've completely lost all sound for unknown reasons and cannot get it to return no matter what sort of workarounds I attempt. I'm using the same kernel, so the only thing difference is the packages that come with Ubuntu 13.04.

[Raymond (superquad-vortex2)]

https://code.launchpad.net/ubuntu/+source/alsa-driver

the modified date seem freeze at Feb/2013

[Michael Murphy (mmstick)]

After reformatting and using the Ubuntu 13.04 Minimal CD, immediately running sudo sed -i 's/raring/saucy/g' /etc/apt/sources.list and reinstalling all of my stuff, I have sound in Ubuntu 13.10. Still, I'm not sure what caused me to lose sound when upgrading from Ubuntu 13.04 to Ubuntu 13.10.

[Michael Murphy (mmstick)]

Perhaps I spoke too soon. After rebooting I lost sound completely again. I tried:

* Booting into Windows then booting into kernel 3.8
* Booting with kernel 3.8 with headphones plugged into the rear headphone jack.
* Purging and reconfiguring alsa and delete asound's from /var/lib/alsa and rebooting
* Booting with position fix off and on
* Booting into kernel 3.11 with position fix off and on
* Booting into kernel 3.11 with headphones plugged into the rear and front headphone jacks.

Nothing works. Furthermore, apparently if I boot with kernel 3.11 there are about 6 error messages, two of which say permission error, and the screen is black with just the cursor for about 5 seconds before Unity decides to load. Half the time Ubuntu hangs when booting, so that's a major problem. Then if you use unity --replace & disown, unity will completely explode, then sometimes Qt programs just look horrible. I guess I'm going to have to stick with Ubuntu 13.04 on this machine.

[Paulo Narciso (p-narciso)]

Creative won't care about Linux, Ubuntu devs also won't care. And Alsa dev's don't have much interess fixing this cards issue.

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

I just setup a new computer with a ASROCK extreame 11 and it has the creative 3d Sound chip as part of its HDA-Intell setup. I have sound only out of the front L/R speaker port on the back pannel and not the headphone port from the hdaudio header. I have tried the alsa mixer un muting hp/speaker and hp/sense , but it does not seem to have any effect. I think the pin mapping may be part of the issue but I have also seen that the driver is only picking up the simple controls wich is 25 of the 40 or 45 possible so I am wondering if that is another part.

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

(In reply to J Pollitt from comment #61)
> I just setup a new computer with a ASROCK extreame 11 and it has the
> creative 3d Sound chip as part of its HDA-Intell setup. I have sound only
> out of the front L/R speaker port on the back pannel and not the headphone
> port from the hdaudio header. I have tried the alsa mixer un muting
> hp/speaker and hp/sense , but it does not seem to have any effect. I think
> the pin mapping may be part of the issue but I have also seen that the
> driver is only picking up the simple controls wich is 25 of the 40 or 45
> possible so I am wondering if that is another part.

sorry should have mentioned i am on arch linux with 3.11.6 kernel

[mmstickman (mmstickman-linux-kernel-bugs)]

You're pretty much SOL because Creative doesn't care. The pin mappings have all been wrong for over a year now. Sometimes unmuting works, sometimes it doesn't.

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to J Pollitt from comment #62)
> (In reply to J Pollitt from comment #61)
> > I just setup a new computer with a ASROCK extreame 11 and it has the
> > creative 3d Sound chip as part of its HDA-Intell setup. I have sound only
> > out of the front L/R speaker port on the back pannel and not the headphone
> > port from the hdaudio header. I have tried the alsa mixer un muting
> > hp/speaker and hp/sense , but it does not seem to have any effect. I think
> > the pin mapping may be part of the issue but I have also seen that the
> > driver is only picking up the simple controls wich is 25 of the 40 or 45
> > possible so I am wondering if that is another part.
>
> sorry should have mentioned i am on arch linux with 3.11.6 kernel

I'm guessing you're referring to the Front Panel Audio Header (HD_AUDIO1) depicted in the user manual under section 1.3 Motherboard Layout item #41?
If so for kicks and giggles do you have a means to try #42 the HDMI_SPDIF Header?

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to Acidphase from comment #64)
> (In reply to J Pollitt from comment #62)
> > (In reply to J Pollitt from comment #61)
> > > I just setup a new computer with a ASROCK extreame 11 and it has the
> > > creative 3d Sound chip as part of its HDA-Intell setup. I have sound only
> > > out of the front L/R speaker port on the back pannel and not the
> headphone
> > > port from the hdaudio header. I have tried the alsa mixer un muting
> > > hp/speaker and hp/sense , but it does not seem to have any effect. I
> think
> > > the pin mapping may be part of the issue but I have also seen that the
> > > driver is only picking up the simple controls wich is 25 of the 40 or 45
> > > possible so I am wondering if that is another part.
> >
> > sorry should have mentioned i am on arch linux with 3.11.6 kernel
>
> I'm guessing you're referring to the Front Panel Audio Header (HD_AUDIO1)
> depicted in the user manual under section 1.3 Motherboard Layout item #41?
> If so for kicks and giggles do you have a means to try #42 the HDMI_SPDIF
> Header?

I forgot to add check your BIOS settings as well some motherboards have settings for choosing the function of those headers such as AC97,HD,SPDIF, and HDMI

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

(In reply to Acidphase from comment #65)
> I'm guessing you're referring to the Front Panel Audio Header (HD_AUDIO1)
> depicted in the user manual under section 1.3 Motherboard Layout item #41?
> If so for kicks and giggles do you have a means to try #42 the HDMI_SPDIF
> Header?
>
> I forgot to add check your BIOS settings as well some motherboards have
> settings for choosing the function of those headers such as AC97,HD,SPDIF,
> and HDMI

Yes, I was refereing to the HD_Audio1 Header, I checked the pin outs for my case adn they all matched up, also the UEFI setup tool shows the jack sense state on the header if I connect a jack to either headphones or mic it is detected at uefi level. THe HDaudion is enbled, I am not sure if that is the whole audio system or just the header, yet.

for my audio stack I have kernel-alsa-pulse, just for reference, and this is a clean arch install. not cruft from the past 5 years of rolling upgrades or other hardware.

So, I do not have any thing that can take in SPDIF, but that is possibly good as I do have a DMM and when I connected that to header 42 the HMDI-SPDIF port it stays a 0V until I enter alsamixer. If I toggle the SPDIFF settings nothing happens if I toggle the hp/SPeaker and HP/Speaker Auto Detect I get changees.
HP/Speaker HP/Spk Auto Detect Voltage (Audio Out FrontR/L)
1 1 0V yes
1 0 3.44V No
0 1 0V yes
0 0 0V yes

No combination gives me audio out of any other analog port Center/Sub, RearR/L , or HDAudio1 Headphone. The front mic on HDAudio1 seems to work, but I have not tested the other inputs yet.

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

@Jaroslav Kysela, is this Bug still open for issues witht hte CA0132 sound path, or should I file a seperate one. I have not really done alot with Linux development but I think I have most of the tools to start debugging issues. Like I stated earlier this dsp is on my main board, so I would like work out the issues, and restore sound to all the analog ports. Any pionters on places in the code to start to jumpstart work or traces that are needed let me know.

[ytardy (ytardy-linux-kernel-bugs)]

I did a fresh install of Ubuntu 13.10 (64 bits) with a brand new Sound Blaster Z. No sound at all from any output... AlsaMixer shows Creative CA0132 chipset with various analog and digital controls, but no results. May I help?

[mmstickman (mmstickman-linux-kernel-bugs)]

Good luck with that. No matter what I did I could not get any sound at all from Ubuntu 13.10. I had to revert back to Ubuntu 13.04 just to get sound functioning and it has nothing to do with the kernel either since I'm using kernel 3.12 in Ubuntu 13.04 without any problems.

[mmstickman (mmstickman-linux-kernel-bugs)]

It is also a great idea if you are lucky enough to get sound working to backup your asound.state in /var/lib/alsa. This didn't help with Ubuntu 13.10 though. Sometimes when I reboot, I will completely lose sound and the only way to get it back is to delete the asound.state and put the backed up one in there as a replacement. Currently that seems to happen every time I open alsamixer and reboot.

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

Ok Jaroslav Kysela, wrote this tool hda_analyzer.py. Has anyone played with it? If not please download it so maybe we can see if there is variance in waht it is being listed ont he hardware front. I believe it extracts the board ID and revision but it also details of the pin usage detected. I can see evidance for why the position fix workaround was needed. Has any one tried the other position fix values? Still digging sorry if this is all old news.

My out put from hda_analyzer

Codec: 0x11020011
Address: 0
AFG Function Id: 0x1 (unsol 1)
Vendor Id: 0x11020011
Subsystem Id: 0x11020010
Revision Id: 0x100918
No Modem Function Group found
Default PCM:
    rates [0x0]:
    bits [0x0]:
    formats [0x0]:
Default Amp-In caps: N/A
Default Amp-Out caps: N/A
GPIO: io=0, o=0, i=0, unsolicited=1, wake=1

Node 0x02 [Audio Output] wcaps 0x49d: Stereo Amp-Out
  Device: name="CA0132 Analog", type="Audio", device=0
  Amp-Out caps: ofs=0x5a, nsteps=0x63, stepsize=0x03, mute=1
  Amp-Out vals: [0x63 0x63]
  Converter: stream=8, channel=0
  PCM:
    rates [0x5ec]: 16000 22050 44100 48000 88200 96000 192000
    bits [0x1f]: 8 16 20 24 32
    formats [0x1]: PCM
  Unsolicited: tag=0x00, enabled=0
  Power: setting=D0, actual=D0

Node 0x03 [Audio Output] wcaps 0x49d: Stereo Amp-Out
  Amp-Out caps: ofs=0x5a, nsteps=0x63, stepsize=0x03, mute=1
  Amp-Out vals: [0x5a 0x5a]
  Converter: stream=0, channel=0
  PCM:
    rates [0x5ec]: 16000 22050 44100 48000 88200 96000 192000
    bits [0x1f]: 8 16 20 24 32
    formats [0x1]: PCM
  Unsolicited: tag=0x00, enabled=0
  Power: setting=D0, actual=D0

Node 0x04 [Audio Output] wcaps 0x49d: Stereo Amp-Out
  Amp-Out caps: ofs=0x5a, nsteps=0x63, stepsize=0x03, mute=1
  Amp-Out vals: [0x5a 0x5a]
  Converter: stream=0, channel=0
  PCM:
    rates [0x5ec]: 16000 22050 44100 48000 88200 96000 192000
    bits [0x1f]: 8 16 20 24 32
    formats [0x1]: PCM
  Unsolicited: tag=0x00, enabled=0
  Power: setting=D0, actual=D0

Node 0x05 [Audio Output] wcaps 0x691: Stereo Digital
  Device: name="CA0132 Digital", type="SPDIF", device=1
  Control: name="IEC958 Playback Con Mask", index=0, device=0
  Control: name="IEC958 Playback Pro Mask", index=0, device=0
  Control: name="IEC958 Playback Default", index=0, device=0
  Control: name="IEC958 Playback Switch", index=0, device=0
  Control: name="IEC958 Default PCM Playback Switch", index=0, device=0
  Converter: stream=0, channel=0
  Digital: Enabled
  Digital category: 0x0
  PCM:
    rates [0x5e0]: 44100 48000 88200 96000 192000
    bits [0x1e]: 16 20 24 32
    formats [0x5]: PCM AC3
  Unsolicited: tag=0x00, enabled=0
  Power: setting=D0, actual=D0

Node 0x06 [Audio Output] wcaps 0x691: Stereo Digital
  Converter: stream=0, channel=0
  Digital: Enabled Pro
  Digital category: 0x0
  PCM:
    rates [0x5e0]: 44100 48000 88200 96000 192000
    bits [0x1e]: 16 20 24 32
    formats [0x5]: PCM AC3
  Unsolicited: tag=0x00, enabled=0
  Power: setting=D0, actual=D0
Node 0x07 [Audio Input] wcaps 0x10059b: Stereo Amp-In
  Device: name="CA0132 Analog", type="Audio", device=0
  Amp-In caps: ofs=0x5a, nsteps=0x63, stepsize=0x03, mute=1
  Amp-In vals: [0x5a 0x5a]
  Converter: stream=0, channel=0
  SDI-Select: 0
  ...

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

Okay I have tried the position_fix=1 and 4 , also I have tried expanding it out as it is an array. All the options have had the same effect of switing my out put to center/bass analog output while all other outputs are not functional. In pulse audio it now shows up as line out, adn while the analog output is still present, there is no sound on it.

[mmstickman (mmstickman-linux-kernel-bugs)]

As I said, you're pretty much SOL. You can complain on Creative's forums if you want, but they simply won't care. http://forums.creative.com/showthread.php?t=699748&page=12

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

Has any one ever had more than two analog audio outputs working? Has anyone had more than one digital output working? The reason I am asking is that if creative put out a puposely busted dsp bin file for the chrome book then the it would be limited to the internal speaker , headphone, and hdmi out only, and may account for the issues. I am noticing in hda analyzer that wcaps = 0x49d for all the analog out amps, I have the first amp node 0x02 hooked up to pins 0x0b and 0x0f id wcaps an address the same is true for all the pins wcapps are all identical, for the two that work and are conected to audio out 2 this seems okay but I would have expected the others to be different. I realize creative will not offer much in the way support, but I am hopeful to get the basics a front/rear ananlog and headphone jack.

[mmstickman (mmstickman-linux-kernel-bugs)]

Pretty much: no. I never had more than one port work, if even one port. The only port I was ever able to get functioning with newer kernels is the front audio headphone port. Lately, however, I've completely lost sound with my Recon3D PCIE and no matter what I do, I can't get sound back in Linux. Therefore, today I finally got fed up with dealing with this issue ever since I bought the card back in 2011 and am going back to onboard audio and just going to use qpaeq for my EQ needs until I can afford an ASUS Essence ST/X.

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

>
> Using Ubuntu 13.04 here with kernel 3.10 and linux-firmware-1.07 which has
> this and it works fine with my Recon3D, sort of. I can only get sound out of
> my silver headphone port and green speaker out now, and if I lose sound
> after a reboot, I find that if I boot into kernel 3.8 and back into kernel
> 3.10 I regain sound.

http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/patch_ca0132.c?id=c41999a23929f30808bae6009d8065052d4d73fd

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1038651

as reporter mention that " On Dell Alienware 14x Ubuntu 12.04 sound was working"

does it mean that only alienware and recon3d works before adding DSP firmware ?

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

I am attempting to go through my main board manufacturer, it may pan out it may not, but worth a go. I have asked for a firmware file that is not limited to google pixile which I believe I have read this is used in. From the initital performance out of the box with position fix it seams that the chip is only enable with a single audio out connected to two outs and two inputs conected adn posibly the one digital out.

I have the cloned the git tree and started looking but it will take me a while to confirm weather I am correct or there is undocumented controls on the firmware that will enable all outputs.

The alienware 14x is another laptop so that may be the default working state,
I have not seen any mentions of the chip working witout the firmware

Below may be a lead maybe not.

I recieved this email today

Hello,

We do not have any Linux driver for this Creative audio chip so far. My colleague in Taiwan will check with Creative.

Thanks.

Regards

ASRock Support

[Acidphase (acidphase-linux-kernel-bugs)]

Creative is going to comeback and say "No" lol

[mmstickman (mmstickman-linux-kernel-bugs)]

(In reply to Raymond from comment #76)
> >
> > Using Ubuntu 13.04 here with kernel 3.10 and linux-firmware-1.07 which has
> > this and it works fine with my Recon3D, sort of. I can only get sound out
> of
> > my silver headphone port and green speaker out now, and if I lose sound
> > after a reboot, I find that if I boot into kernel 3.8 and back into kernel
> > 3.10 I regain sound.
>
>
> http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/
> hda/patch_ca0132.c?id=c41999a23929f30808bae6009d8065052d4d73fd
>
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1038651
>
>
> as reporter mention that " On Dell Alienware 14x Ubuntu 12.04 sound was
> working"
>
> does it mean that only alienware and recon3d works before adding DSP
> firmware ?

It works after the DSP firmware is added, sort of. After the DSP firmware I could only get audio out of the front headphone port and the controls for the DSP features work, but in Ubuntu 13.10 and currently in Ubuntu 13.04 I'm no longer able to get audio at all. It may have to do with how I'm using linux-firmware-1.116 right now from the Oibaf PPA for the latest radeon firmware and drivers.

[Acidphase (acidphase-linux-kernel-bugs)]

As an FYI I have sound with the front pin out using Kernel 3.12 and all since they added the DSP firmware.

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

On the bonus side, though ASROCK asking when they are shipping thier product is more noticeable than just a linux user, get a steamdev to care adn try as well you might get some headway there as well.

As a side note I have two simultanios outputs as mentioned before and two mic inputs working.

My curent postiion fix is 4

more /etc/modprobe.d/ca0132.conf
options snd_hda_intel position_fix=4

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

http://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/log/sound/pci/hda/patch_ca0132.c

Aug 2012 ALSA: hda - Add missing SPDIF I/O setup for CA0132
Nov 2012 ALSA: hda - Add CA0132 register definitions file
Jan 2013 ALSA: hda/ca0132: Add new definitions and structs for DSP
Feb 2013 ALSA: hda - Add new Kconfig CONFIG_SND_HDA_CODEC_CA0132_DSP

do you mean sound still working when CONFIG_SND_HDA_CODEC_CA0132_DSP is added ?

[Acidphase (acidphase-linux-kernel-bugs)]

It is for me, with the front connector pin out. The DSP options work as well such as things like Dialogue Plus and X-Bass etc..

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

do you mean those 10 equalizer controls also work if you define ENABLE_TUNING_CONTROLS ?

[Acidphase (acidphase-linux-kernel-bugs)]

I do not have this defined so that will be no, I guess I could give it a shot later. I was referring to the DSP features of the Sound Blaster Z card.

[Marc Bourgoin (jagged-jimmy-j)]

Heyyo,

Now on the latest versions of the Linux Kernel? All of the Sound Blaster Recon 3D sound cards all work perfectly fine in 32bit. I believe it was kernel 3.9 that added a compatible driver... sadly though the kernel's built in driver doesn't seem to work in 64bit. :(

Tested with Ubuntu, Kubuntu, Fedora, Linux Mint. All the same results. 32bit works perfect but 64bit doesn't work properly if at all. It's definitely the kernel that makes the difference.

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

http://mailman.alsa-project.org/pipermail/alsa-devel/2013-November/069046.html

" The CA0132 is very dependent on
the firmware that is loaded into the DSP that resides in the codec.

To get your system to work you'll have to obtain the firmware that was
designed for your motherboard, then change the Linux driver to be able
to talk to it, without a spec for how it works."

does this mean that the firmware loaded by windows driver still can be used after warm boot for those Recon3d , but lost sound when the alsa driver loaded the firmware which is not designed for Recon3d ?

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to Raymond from comment #86)
> http://mailman.alsa-project.org/pipermail/alsa-devel/2013-November/069046.
> html
>
> " The CA0132 is very dependent on
> the firmware that is loaded into the DSP that resides in the codec.
>
> To get your system to work you'll have to obtain the firmware that was
> designed for your motherboard, then change the Linux driver to be able
> to talk to it, without a spec for how it works."
>
>
> does this mean that the firmware loaded by windows driver still can be used
> after warm boot for those Recon3d , but lost sound when the alsa driver
> loaded the firmware which is not designed for Recon3d ?

Um not really this was two developers talking about reverse engineering the driver. Basically this explains all the reason why it's not working. The driver/codec is written in such a way that the instruction set is very low level and being Takashi does not have nay reference for the pin out/instructions/memory stack this make it quite a challenge.

[mmstickman (mmstickman-linux-kernel-bugs)]

Seems I can get my sound card to work with 32-bit Ubuntu, but I can't get it to work at all in 64-bit, regardless of how or what position_fix I apply.

[mmstickman (mmstickman-linux-kernel-bugs)]

After putting my Recon3D in a different computer, it works with Ubuntu 13.10 64-bit without needing to apply a position_fix. The only difference between this machine and my desktop is that my desktop uses an ASUS UEFI motherboard and this one has a MSI UEFI motherboard.

Everything seems to work besides the equalizer which does nothing. I recompiled the kernel to enable the equalizer and it works but the bands aren't properly named and it's missing the 11th band of the equalizer which controls the pre-amp level. Currently the bands are named: {EQ Band0, EQ Band1, EQ Band2, EQ Band3, EQ Band4, EQ Band5, EQ Band6, EQ Band7, EQ Band8, EQ Band9} but they should instead be named {31Hz, 62Hz, 125Hz, 250Hz, 500Hz, 1KHz, 2KHz, 4KHz, 8KHz, 16KHz}.

I'm wondering if it would be possible to add something like:

{
          .name = "LEVEL",
   .parent_nid = EQUALIZER,
   .nid = EQUALIZER_BAND_10,
   .mid = 0x96,
   .req = 21,
   .direct = EFX_DIR_OUT,
   .def_val = 0x00000000
}

To also get the preamp band.

I'm also wondering why the equalizer is not enabled by default if it works.

[andreas.reis (andreas.reis-linux-kernel-bugs)]

>Everything seems to work besides the equalizer

Including the mic? That's what bugs me the most (I wrote that report on alsa-devel linked by Raymond).

I've contacted Gigabyte and asked them to at least request fixes from Creative (like ASRock seems to have done) for the CA0132 on board of my G1.Sniper M5, this is their reply:

>Thank you for emailing GIGABYTE.
>We are delighted with your interest in our products.
>
>This Creative chip is a Windows design only.
>There will be no Linux support.
>
>Kind regards

Either they've already asked, or they won't even try.

[mmstickman (mmstickman-linux-kernel-bugs)]

They've essentially asked and Creative has told them that it is 'Windows-only'. Although Creative won't directly come out and admit it, they don't care about Linux at all. The ALSA team essentially needs to just reverse engineer their chip, which luckily most of the chip is functional in Linux, just pins need to be mapped correctly and issues with position_fix and 32-bit vs 64-bit need to be fixed.

As for whether or not microphones work with my Recon3D on my MSI machine, the answer no.

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

(In reply to Andreas Reis from comment #90)
> >Everything seems to work besides the equalizer
>
> Including the mic? That's what bugs me the most (I wrote that report on
> alsa-devel linked by Raymond).
Andreas,

   My mics work on my asrock extream11, Creative is definantly not helping out, out much, the Asrock support, has asked them questions from me and creative is not answereing them on what parts of thier integration work can be released or weather the binary blob provided currently contains support for more channels than Speaker and headphone out and 2 mic inputs.

I agree with Andreas that we face an issue here for each system, that it will be dependant on the vendor implementation for the firmware, and with out creative okaying it it will make it much harder to get the firmware from the vendors that we can load, AS rock offered me the verb table but they were concerned that Creative, might not like that I will let everyone know if I get a different firmware adn vertable from as rock to implement. I asked ASROCK to go to creative and get a released firmware that at least implements all inputs and outputs as I rfear from creative initial impimatation the firmware we have now does not.

The other posibility would be to try and take the loaded firmware from the windows driver and dump it to load from linux. Either way it will be a long uphill battle to get support, Not somethink I have had to think about in Linux for a very long time.

[michael.schuchlenz (michael.schuchlenz-linux-kernel-bugs)]

if it helps my creative soundblaster Z on 64 bit lubuntu gives a veeeeery silent sound (from video) when everything is turned all the way up!

[mmstickman (mmstickman-linux-kernel-bugs)]

I had my Recon3D working perfectly without a position fix in my system with a MSI motherboard, and when I plugged the Recon3D into my ASUS Sabertooth 990FX system with the hard drive from the MSI system, sound worked perfectly there too. However, when I accidentally booted into the OS installed natively from the Sabertooth 990FX, I lost sound, and could not regain sound no matter what I tried when booting back into the OS that was installed from the MSI system. I simply don't understand why this is.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

I have a Sound Blaster Recon 3D Fatal1ty edition PCI-Express.

I found the analog stereo output works perfectly without issue since kernel 3.9 for me ONLY in 32bit.. 64bit is hit-or-miss and I always need that position_fix.

Microphone does not work in either 32bit or 64bit and my front audio ports have never worked either.

My motherboard is a Gigabyte Z68X-UD7 B3. Sound card is in the PCI-E 1x slot.

[Marc Bourgoin (jagged-jimmy-j)]

Heyyo,

I have a Sound Blaster Recon 3D Fatal1ty edition PCI-Express.

Microphone does not work in either 32bit or 64bit and my front audio ports have never worked either. I've tried kernels 3.9 up to 3.12.6... no luck so far.

My motherboard is a Gigabyte Z68X-UD7 B3. Sound card is in the PCI-E 1x slot.

[Michael Murphy (mmstick)]

You're SOL. I only had it work perfectly with 64-bit on a MSI motherboard.

[zlisiecki (zlisiecki-linux-kernel-bugs)]

Hi, my card is Sound Blaster ZxR on PCI-Express and the motherboard is ASUS P8H61-MX with Intel i5 processor in which I switched it's original HD Audio Controller VIA VT1708S out. On this hardware I installed OpenSuSE 13.1 Linux with 3.11.6 kernel configured as 64bit by myself. The sound definitely won't go with position fix or without it.

As the problem is known since at least a half a year I'd like to know if anybody is working on it, or should I better sell this card and forget it.

[mmstickman (mmstickman-linux-kernel-bugs)]

The problem has been known since 2011 pretty much. There's no fix that I have found other than using 32-bit or buying something else, like an external USB DAC or a supported ASUS card. When you do get it to work in 32-bit, you can't use the equalizer unless you manually patch the patch_ca0132.c file yourself. I have no idea why the issue with 64-bit hasn't been fixed yet, nor why the equalizer is not enabled by default.

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

can you post the lspci output of your ca0132 ?

lspci -vvvv

chromebox pixel seem work on 64 bit

https://bugzilla.redhat.com/show_bug.cgi?id=1045784

[mmstickman (mmstickman-linux-kernel-bugs)]

Chromebook Pixel is the only CA0132 chip officially supported by an actual company so it isn't surprising that it works. Having to manually enable HP/Speaker with alsamixer is also a known bug with all CA0132 chips.

[zlisiecki (zlisiecki-linux-kernel-bugs)]

Hi, this is the output, sorry for the lenght:

vega:/usr/src/linux # lspci -vv
00:00.0 Host bridge: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor DRAM Controller (rev 09)
        Subsystem: ASUSTeK Computer Inc. Device 844d
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort+ >SERR- <PERR- INTx-
        Latency: 0
        Capabilities: [e0] Vendor Specific Information: Len=0c <?>

00:01.0 PCI bridge: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor PCI Express Root Port (rev 09) (prog-if 00 [Normal decode])
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Bus: primary=00, secondary=01, subordinate=01, sec-latency=0
        Secondary status: 66MHz- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- <SERR- <PERR-
        BridgeCtl: Parity- SERR- NoISA- VGA- MAbort- >Reset- FastB2B-
                PriDiscTmr- SecDiscTmr- DiscTmrStat- DiscTmrSERREn-
        Capabilities: [88] Subsystem: ASUSTeK Computer Inc. Device 844d
        Capabilities: [80] Power Management version 3
                Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
                Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
        Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
                Address: fee0f00c Data: 4171
        Capabilities: [a0] Express (v2) Root Port (Slot+), MSI 00
                DevCap: MaxPayload 256 bytes, PhantFunc 0, Latency L0s <64ns, L1 <1us
                        ExtTag- RBE+ FLReset-
                DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported-
                        RlxdOrd- ExtTag- PhantFunc- AuxPwr- NoSnoop-
                        MaxPayload 256 bytes, MaxReadReq 128 bytes
                DevSta: CorrErr- UncorrErr- FatalErr- UnsuppReq- AuxPwr- TransPend-
                LnkCap: Port #2, Speed 8GT/s, Width x16, ASPM L0s L1, Latency L0 <1us, L1 <8us
                        ClockPM- Surprise- LLActRep- BwNot+
                LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- Retrain- CommClk-
                        ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
                LnkSta: Speed 2.5GT/s, Width x0, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
                SltCap: AttnBtn- PwrCtrl- MRL- AttnInd- PwrInd- HotPlug- Surprise-
                        Slot #1, PowerLimit 75.000W; Interlock- NoCompl+
                SltCtl: Enable: AttnBtn- PwrFlt- MRL- PresDet- CmdCplt- HPIrq- LinkChg-
                        Control: AttnInd Unknown, PwrInd Unknown, Power- Interlock-
                SltSta: Status: AttnBtn- PowerFlt- MRL- CmdCplt- PresDet- Interlock-
                        Changed: MRL- PresDet- LinkState-
                RootCtl: ErrCorrectable- ErrNon-Fatal- ErrFatal- PMEIntEna- CRSVisible-
                RootCap: CRSVisible-
                RootSta: PME ReqID 0000, PME...

[John R. Boutiette (boutiettej)]

I also have the Recon3D card with CA0132. I'm running linux mint 16 from a clean install. Sound will work on every other boot with the "sound fix=1" added. This also applies to windows 7. I use the program Pithos( a web client based player) for Pandora to record audio off the player through a program called Audio Recorder. I set the recorder to record audio from stereo out. I set the recorder to start recording when it sensed audio from Stereo out. It did record audio even when the nothing was being outputed from the speakers. I rebooted the system and checked the file it had recorded and it played perfectly like the speakers were working. I also noticed in mt Dmesg file in /var/log that sometimes it said that ca0132 didn't load or never even said anything about this card. Next boot "Dsp CA0132 loaded". Hope this helps in some way.

[ninex (ninex-linux-kernel-bugs)]

I have good and bad news.
The bad one is that creative are silently removed all forum threads about Z/ZX/ZXR don't working under linux.

The good one is: there is an opensource hda driver working with ca0132 (z/zx/zxr) cards. So some one can try to find why linux driver don't work.
I recently isntalled PC-BSD and found that i FreeBSD kernel detects my card, so tried to play sound. basic stereo output work "from a box" with smal reconfiguration i get 5.1 sound too. FreeBSD code is open, so someone can try to analyze FreeBSD hda driver and fix linux driver. I know that FreeBSD sound system is different than linux, but if there is a working driver that we can analyze. There is a Hope ;)

[abb802 (abb802-linux-kernel-bugs)]

I have a sound card sb recon3d card and i can't tell us with ubuntu 14.04 and without touching anything sometimes it work and sometimes not, the 99% times don't work. I use a usb steelseries siberia headset and i was surprised that the other headset was working when i was disconnect the steelseries. I've been reading the thread and I suppose the error is when the module is loaded. Now i'm using Arch linux, this distribution don't use rc.local and i don't know where i can put the option "position_fix=1". Anyone can help me?

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

http://wiki.antergos.com/ca0132

That guide is the one i made on Antergos but Arch is the same thing.

(In reply to arboschbu from comment #102)
> I have a sound card sb recon3d card and i can't tell us with ubuntu 14.04
> and without touching anything sometimes it work and sometimes not, the 99%
> times don't work. I use a usb steelseries siberia headset and i was
> surprised that the other headset was working when i was disconnect the
> steelseries. I've been reading the thread and I suppose the error is when
> the module is loaded. Now i'm using Arch linux, this distribution don't use
> rc.local and i don't know where i can put the option "position_fix=1".
> Anyone can help me?

[abb802 (abb802-linux-kernel-bugs)]

(In reply to Marc Bourgoin from comment #103)
> Heyyo,
>
> http://wiki.antergos.com/ca0132
>
> That guide is the one i made on Antergos but Arch is the same thing.
>
> (In reply to arboschbu from comment #102)
> > I have a sound card sb recon3d card and i can't tell us with ubuntu 14.04
> > and without touching anything sometimes it work and sometimes not, the 99%
> > times don't work. I use a usb steelseries siberia headset and i was
> > surprised that the other headset was working when i was disconnect the
> > steelseries. I've been reading the thread and I suppose the error is when
> > the module is loaded. Now i'm using Arch linux, this distribution don't use
> > rc.local and i don't know where i can put the option "position_fix=1".
> > Anyone can help me?

Hi,

It works, thanks a lot for the solution.

Greetings.

[sojhung (sojhung-linux-kernel-bugs)]

Hello,

I own the notorious (in the context of what you have been discussing so far) Sound Blaster Z plugged in an AMD hexa-core running Ubuntu 14.04. No sound, of course.

I don't know if my experience would bring any value to the table, but if anyone is interested in making use of the SB Z under Linux, just to enjoy media playback such as music and movie, here's what I have done.

Since my primary objective was to listen to MP3 files under Linux while I was working on code, I took the liberty of addressing the issue from a different perspective: PCI passthrough using KVM.

Using qemu and IOMMU feature from the Gigabyte motherboard, I was able to passthrough Sound Blaster Z to the guest running Windows 7 which played my favorite Def Leppard songs loud and clear.

But there is a catch: the audio performance of the guest will be directly affected by the stress level of the host system. For example, if you have several tabs opened in your web browser, and you tried moving around the browser within desktop, the music playback from the guest will slowdown noticeably.

I have rather limited experience with PCI passthrough although I played with Xen and Qemu for some time, so I don't have a solution to this yet.

Cheers,
Steve

[tiwai (tiwai-linux-kernel-bugs)]

Actually position_fix=1 option should be a myth for the recent kernel. At least, the controller with PCI ID 1102:0010 and 1102:0012 take the equivalent as position_fix=1 as default. These also disables MSI and snooping, too.

If the recent kernel still doesn't *probe* the card (not considering about the functionality), you may try the patch below. This applies similar workarounds that are used for old Create X-Fi boards.

The story so far is about the HD-audio controller. Yet the more difficult story is about HD-audio codec on the board. The CA0132 codec driver was largely rewritten to match with Chromebook Pixel. Unfortunately, this seems to have broken many other CA0132 boards.

For skipping CA0132-specific driver code, you can try to pass model=generic. This will deploy the generic parser instead of CA0132-specific driver code.
With the generic parser, some boards might work in some level, but some won't work well. IIRC, the digital I/O and some jack detection would be malfunctional. Basically these are because the board isn't fully compliant to HD-audio spec.

In anyway, please check the above with the recent kernel, 3.15 or 3.16. No matter whether it works or not, please give alsa-info.sh outputs, at best, with and without model option. Run the script with --no-upload option, and attach the files to Bugzilla. If the same model is already found in Bugzilla, you can skip, of course.

[tiwai (tiwai-linux-kernel-bugs)]

Created attachment 145841
Test patch for CA0132 HD-audio controller

[ganthore (ganthore-linux-kernel-bugs)]

Created attachment 146171
alsa-info for CA0132 (ZxR)

Same issue as the rest. I'm using a Creative ZxR branded card.

The "model=generic" is not an effective workaround for my card...

I have attached my alsa-info.sh output to this ticket. (alsa-info.txt.CRS4j3hhWT)

Kernel logs show that the DSP is loaded and running correctly:
[ 14.648519] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

LSPCI:
04:00.0 Audio device: Creative Labs Device 0012 (rev 01)
        Subsystem: Creative Labs Device 0033
        Flags: bus master, fast devsel, latency 0, IRQ 18
        Memory at f7a04000 (64-bit, non-prefetchable) [size=16K]
        Memory at f7a00000 (64-bit, non-prefetchable) [size=16K]
        Capabilities: [40] Power Management version 3
        Capabilities: [50] MSI: Enable- Count=1/1 Maskable+ 64bit+
        Capabilities: [70] Express Endpoint, MSI 00
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [140] Virtual Channel
        Capabilities: [170] Device Serial Number 00-00-00-00-00-00-00-00
        Capabilities: [180] Power Budgeting <?>
        Kernel driver in use: snd_hda_intel
        Kernel modules: snd_hda_intel

If I can provide you any more info just let me know.

[tiwai (tiwai-linux-kernel-bugs)]

(In reply to ganthore from comment #108)
> Created attachment 146171 [details]
> alsa-info for CA0132 (ZxR)
>
> Same issue as the rest. I'm using a Creative ZxR branded card.
>
> The "model=generic" is not an effective workaround for my card...

Since you have two HD-audio devices and ca0132 is the second one, you need to pass like model=,generic

If you still see something about ca0132 DSP, it means that the model option isn't passed properly to the device by some reason.

[tiwai (tiwai-linux-kernel-bugs)]

In doubt, you can try the patch below. It will use the generic parser forcibly for CA0132 boards except for Chromebook Pixel.

[tiwai (tiwai-linux-kernel-bugs)]

Created attachment 146341
Patch to use generic parser for non-Pixel ca0132 boards

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

Err... I'm a bit of a n00b when it comes to building custom kernels... but I'll definitely have to try this out tonight once I'm home! Thank you for looking into this long-running issue Takashi Iwai!!! :)

I know before I didn't have issues with the DSP / CA0132 working in stereo for my PCI-Express Sound Blaster Recon 3D (FATAL1TY edition if that makes a difference) in 32bit... but I will definitely try it in both 32bit and 64bit on kernel 3.16. Currently I can't get it to work in 64bit without using...

sudo kill `ps uax |grep pulseaudio | grep -v grep | awk '{print $2}'` ; sudo rmmod snd-hda-intel; sleep 3; sudo modprobe snd-hda-intel

and that only gets the stereo sound working... no headphone ports from my PC Tower... so hopefully this patch will correct that issue as well. ;)

[tiwai (tiwai-linux-kernel-bugs)]

Note that using the generic parser code means to take the HD-audio-compliant mode of CA0132 boards. This is a subset, and is known to be imperfect on some boards. At least, Recon 3D was buggy with this mode (e.g. Master volume doesn't work properly, SPDIF is broken, etc). There might be different Recon 3D, so we need to check.

I asked and provided the patch just because I was told that freebsd works. AFAIK, freebsd driver handles it just like the normal HD-audio. For the board that works on freebsd would work with the generic parser as well. For others, the generic parser might be worse than the current ca0132 driver that has lots of static quirks.

[ganthore (ganthore-linux-kernel-bugs)]

I took your advice but I had to make a couple of changes...

Probing the device as you suggested did not work:
options snd-hda-intel model=,generic

If I probe the sound device with these options, I get sound:
options snd-hda-intel model=,,generic position_fix=1

The interesting part is, the left channel output is clean but the right channel is full of static but I know you already advised that this would be a problem.

[subshain (subshain-linux-kernel-bugs)]

Created attachment 146821
Creative SB ZxR - No Sound

[ninex (ninex-linux-kernel-bugs)]

(In reply to Lars Ulrich from comment #115)
> Created attachment 146821 [details]
> Creative SB ZxR - No Sound

i can confirm. ZxR , no sound on linux.

[subshain (subshain-linux-kernel-bugs)]

So, what we gonna do about it?
Try to reach the Creative devs again?
And hope that they help us to program the ca0132 codec?

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

(In reply to Lars Ulrich from comment #117)
> So, what we gonna do about it?
> Try to reach the Creative devs again?
> And hope that they help us to program the ca0132 codec?

Heyyo,

Maybe do some reading before you post...

(In reply to Takashi Iwai from comment #111)
> Created attachment 146341 [details]
> Patch to use generic parser for non-Pixel ca0132 boards

I'm currently doing the makepkg on my Arch 64bit install using kernel 3.16... I suggest if you want to help? try making your own custom kernel and see what happens.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Created attachment 146881
alsa-info ca0132 Recon 3D

test of "Attachment #146341" with antergos (arch) 64bit 3.16.1-1.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Comment on attachment 146881
alsa-info ca0132 Recon 3D

here's my test of "Attachment #146341" using Arch 64bit 3.16.1-1 kernel. no sound.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

Wow... trying 64bit "3.16.1-1-ARCH" kernel with no patches or anything... breaks fix at boot that worked on 3.15 and below... using "rmmod snd-hda-intel" and "modprobe snd-hda-intel"...

Whatever changes were done to the Kernel in 3.16? It made it worse on my CA0132 Recon 3D PCI-E... anyone else having this problem?

[dimgl (dimiguel)]

Anyone have any luck with this? This is driving me crazy and I don't want to have a brick on my computer. I have the Sound Blaster Z and I'm running Ubuntu 14.04.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

More tinkering... yet again, 32bit 3.16.2 generic kernel on Ubuntu works flawlessly with my CA0132 Sound Blaster Recon 3D without the need of any rmmod or any other commands.

I think it's because the ported Chromebook Pixel drivers are 32bit native... maybe there's a problem with the firmware loader in 64bit which causes the "ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING." error.

here's a photo of the loader I managed to capture on my smartphone when she stalled on boot:

http://imgur.com/phxRZEK

It's times like these that I wish I did get more into learning how to code... I swear, all the problem is in 64bit is somewhere when the driver was ported from the Chromebook Pixel that something may have been lost in translation to 64bit.

[mikhail.v.gavrilov (mikhail.v.gavrilov-linux-kernel-bugs)]

Created attachment 149911
alsa info for ZXR (SB1510) - no sound - 3.16.2 kernel

[thomas.lindroth (thomas.lindroth-linux-kernel-bugs)]

Created attachment 149921
/proc/asound/PCH/codec#2

I recently got a new Gigabyte Z97X-Gaming G1 mother board with onboard CA0132 connected to the mobo chipset [8086:8ca0]. I couldn't get any sound at all from it with kernel 3.16 and changing the alsamixer settings didn't make any difference.
I was able to get sound by using the HDA Analyzer tool and enabling OUT on the 0x0f pin complex node.

Node 0x02 is analog audio out named "CA0132 Analog". It's connected to pin complex 0x0b (Line Out Green rear) and 0x0f (HP Out Green Front). Node 0x03 is another analog audio that's nameless. It's connected to pin complex 0x10 (HP Out Orange Front). When I toggle the switch "HP/Speaker" in alsamixer the OUT property will toggle between the 0x0b and 0x10 pin complex. The code probably assumes 0x0b is rear speakers and 0x10 front headphones. Even though 0x0f is described as a headphone front it's actually speaker rear.

The sound is ok but it sounds different from what I'm used to. It's like there's an artificial stereo separation. Perhaps some DSP block is enabled per default. I don't have the DSP firmware loaded so that's unlikely.

[salinasv (salinasv-linux-kernel-bugs)]

Thomas I just got the same motherboard (Gigabyte z97X-Gaming G1) and I have got sound from the front headphone. Make sure you have enabled the HP/Speaker Auto D and play with the HP/Speaker control (on/off, it should be on if the auto detect is off).

Problems:
* The sound goes on and off if the output volume (the actual wave, not the setting) goes higher than some arbitrary threshold. If you get your Master control down enough, you can have continued sound.
* Even though the left channel have a lower volume than the right.
* Most of the controls does not work.

As a general information, hoping to be useful. I have tried the following options:
* position_fix=1; As expected from Iwai comment, does nothing.
* position_fix=1; Same as above.
* model=generic; Does not show "ca0132 DOWNLOAD OK :-) DSP IS RUNNING." and nothing works.

I need to try the HDA Analyzer tool and see what can I get.

[h (h-linux-kernel-bugs)]

Got my Sound Blaser Zx (separate PCI-E card) working. Only front-panel headphone out works, recording works with the back panel Mic in.
I load snd_hda_intel on boot with model=,generic (set in /etc/modprobe.d/ca0132fix.conf), then in rc.local:

rmmod snd_hda_intel
modprobe snd_hda_intel model=,auto

With kernel 3.17.1 and alsa-firmware-1.0.28 I get: "ca0132 DOWNLOAD OK :-) DSP IS RUNNING.". Fedora 20, x86_64.
In alsamixer I enable "HP/Speaker" (I'm using alsactl to restore mixer settings after login for comfort).

[Jonathan (jadit2)]

Sound Blaster Z: device detected but no sound. Confirmed that the device is unmuted.

System: Debian Jessie x64
Kernel: 3.16-2-amd64
Advanced Linux Sound Architecture Driver Version k3.16-2-amd64.

Anyone want to start a bounty on this bug?

[alex.mashin (alex.mashin-linux-kernel-bugs)]

This bug is still there under kernel 3.16 and Alsa 1.0.28.

[hastegeldhastewas (hastegeldhastewas-linux-kernel-bugs)]

Hi,
is there somebody who knows how to contact one of the linux developers to get a statement why exactly there is no one working on our problem?

What about "Jaroslav Kysela" to whom the task is assigned to?

Do the devs have no time?
Is there a misunderstanding and the devs think there is no problem?
Can the people who have the hardware test something out to take the things further?

Regards
Richard

[vovochka13 (vovochka13-linux-kernel-bugs)]

(In reply to hastegeldhastewas from comment #128)
> Hi,
> is there somebody who knows how to contact one of the linux developers to
> get a statement why exactly there is no one working on our problem?
>
> What about "Jaroslav Kysela" to whom the task is assigned to?
>
> Do the devs have no time?
> Is there a misunderstanding and the devs think there is no problem?
> Can the people who have the hardware test something out to take the things
> further?
>
> Regards
> Richard

Dear Richard, this is open source, get used to it.
Even if this developers are paied to work on kernel, may be theirs priorities currently not creative cards.

If you realy want to do something, may be you should first find a professional alsa/kernel developer who can investigate this, orgonize a kickstarter (or similar) company to collect some money, to hire a developer to work on this specific bug.

Post the link to such company here, on phoronix. And may be some one will support this company.

But currently, when spcification of creative cards is a dark forest... Don't wait that there will be to many enthusiast to work on this problem :(

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

(In reply to Vladimir Perepechin from comment #129)
>
> Dear Richard, this is open source, get used to it.
> Even if this developers are paied to work on kernel, may be theirs
> priorities currently not creative cards.
>
> If you realy want to do something, may be you should first find a
> professional alsa/kernel developer who can investigate this, orgonize a
> kickstarter (or similar) company to collect some money, to hire a developer
> to work on this specific bug.
>
> Post the link to such company here, on phoronix. And may be some one will
> support this company.
>
> But currently, when spcification of creative cards is a dark forest... Don't
> wait that there will be to many enthusiast to work on this problem :(

Heyyo,

Now I can't fully agree with this since it works flawlessly in 32bit... Check my post above....

(In reply to Marc Bourgoin from comment #95)
> Heyyo,
>
> I have a Sound Blaster Recon 3D Fatal1ty edition PCI-Express.
>
> I found the analog stereo output works perfectly without issue since kernel
> 3.9 for me ONLY in 32bit.. 64bit is hit-or-miss and I always need that
> position_fix.
>
> Microphone does not work in either 32bit or 64bit and my front audio ports
> have never worked either.
>
> My motherboard is a Gigabyte Z68X-UD7 B3. Sound card is in the PCI-E 1x slot.

So yeah... why does this bug only exsist in the 64bit kernel version but works flawlessly in 32bit? To me? It's a bug and not a lack of information.

[mdrogers (mdrogers-linux-kernel-bugs)]

OK so using HDA analyzer you can toggle between the rear headphones and speakers.
I'm using gnome alsa-mixer but that doesn't really matter.
all the OUT toggling is on the 0x0f pin
HD/Speaker in Alsamixer.
I have HP/Speaker Auto Detect off

for just speakers(green rear jack, pin 0x0b)
OUT unchecked, HP/speaker unchecked(muted)

for rear headphone(silver rear jack, pin 0x0f)
OUT checked, HP/speaker checked(unmuted)

for both
OUT checked, HP/speaker unchecked(muted)

Anybody know how to set up commands to these so we can set each up as its own script and toggle easily?

Also has anybody been able to get

sudo kill `ps uax |grep pulseaudio | grep -v grep | awk '{print $2}'` ; sudo rmmod snd-hda-intel; sleep 3; sudo modprobe snd-hda-intel

to run automatically on startup or preferably login? I keep running into the issue of it not running(I think because it's a sudo command) even though I've modified sudoers(I'm a linux noob).

[perex (perex-linux-kernel-bugs)]

(In reply to Michael R from comment #131)
> Anybody know how to set up commands to these so we can set each up as its
> own script and toggle easily?

The HDA Analyzer can generate a Python script (use 'Exp' button which means export)..

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

Hmm, interesting! The bug might be in that "ctefx.bin" file! Maybe it's only setup for 32bit?

Trying Debian Jessie 64bit I got this from dmesg:

[ 9.075566] snd_hda_intel 0000:05:00.0: firmware: failed to load ctefx.bin (-2)
[ 9.075575] snd_hda_intel 0000:05:00.0: Direct firmware load failed with error -2
[ 9.075575] snd_hda_intel 0000:05:00.0: Falling back to user helper
[ 9.093817] input: HDA Creative Mic as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input8
[ 9.093869] input: HDA Creative Line as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input9
[ 9.093960] input: HDA Creative Line Out as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input10
[ 9.094055] input: HDA Creative Front Headphone as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input11

[h (h-linux-kernel-bugs)]

(In reply to Marc Bourgoin from comment #133)
> Heyyo,
>
> Hmm, interesting! The bug might be in that "ctefx.bin" file! Maybe it's only
> setup for 32bit?
>
> Trying Debian Jessie 64bit I got this from dmesg:
>
> [ 9.075566] snd_hda_intel 0000:05:00.0: firmware: failed to load
> ctefx.bin (-2)
> [ 9.075575] snd_hda_intel 0000:05:00.0: Direct firmware load failed with
> error -2
> [ 9.075575] snd_hda_intel 0000:05:00.0: Falling back to user helper
> [ 9.093817] input: HDA Creative Mic as
> /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/
> sound/card2/input8
> [ 9.093869] input: HDA Creative Line as
> /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/
> sound/card2/input9
> [ 9.093960] input: HDA Creative Line Out as
> /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/
> sound/card2/input10
> [ 9.094055] input: HDA Creative Front Headphone as
> /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/
> sound/card2/input11

That looks like alsa-firmware is not installed.

It does work on x86_64 when you rmmod snd_hda_intel and then modprobe it again.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

@Julain Weissgerber, Debian Jessie comes with alsa_base "1.0.27+1" but the description for it says "dummy package to ease purging of obsolete conffiles

This is an empty dummy package with the only purpose of ensuring
obsolete kmod files will be removed from the system. This functionality
is now shipped in kmod."

yet I do get this error on boot showing with dmesg:

[ 11.054516] sound hdaudioC2D1: autoconfig: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 11.054520] sound hdaudioC2D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 11.054522] sound hdaudioC2D1: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 11.054523] sound hdaudioC2D1: mono: mono_out=0x0
[ 11.054530] sound hdaudioC2D1: dig-out=0xc/0x0
[ 11.054531] sound hdaudioC2D1: inputs:
[ 11.054536] sound hdaudioC2D1: Mic=0x12
[ 11.054538] sound hdaudioC2D1: Line=0x11
[ 11.054539] sound hdaudioC2D1: dig-in=0xe
[ 11.085872] snd_hda_intel 0000:05:00.0: firmware: failed to load ctefx.bin (-2)
[ 11.085947] snd_hda_intel 0000:05:00.0: Direct firmware load failed with error -2
[ 11.085949] snd_hda_intel 0000:05:00.0: Falling back to user helper
[ 11.106801] input: HDA Creative Mic as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input8
[ 11.106878] input: HDA Creative Line as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input9
[ 11.106978] input: HDA Creative Line Out as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input10
[ 11.107183] input: HDA Creative Front Headphone as /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:03.0/0000:05:00.0/sound/card2/input11

so now I'm scratching my head cause even with that error? 32bit kernel 3.16.1 on debian I get stereo sound without issues...

alsamixer does load and I can tweak settings in there for the mic to get it to work.

So why is it on a vanilla Debian Jessie Netboot install in 32bit that gets the same errors as 64bit that the sound only works in 32bit... and in 64bit requiring the rmmod and modprobe on every reboot???

I'd LOVE to use arch Linux but I can't seem to find ANY good way to setting it up on boot for rmmod and modprobe... making a service setup similar to rc.local in arch similar to Debian or Ubuntu just causes a boot loop with audio failing to load... kernel 3.15 and down didn't have this problem in Arch Linux...

The more and more I try to fix this issue in Arch Linux? The more and more I think about giving up and just waiting for Windows 10 which is a step in the right direction... so far I'm loving the windows 10 tech preview cause everything just "works" and Arch Linux hates trying to use rmmod and modprobe before that damn alsa firmware loader in 64bit... same goes with any Linux Distro... so I know it's either the Linux Kernel or Alsa or the ctefx.bin file...

[UPDATE 1]
So out of curiosity I tried installing alsa_firmware_loaders and I got this now...

[ 10.391088] snd_hda_intel 0000:04:00.1: Disabling MSI
[ 10.391094] snd_hda_intel 0000:04:00.1: Handle VGA-switcheroo audio client
[ 10.391126] snd_hda_intel 0000:05:00.0: Disabling MSI
[ 10.391127] s...

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to Marc Bourgoin from comment #135)
>
>
> .... nope! Nothing... alsa-firmware 1.0.28 is installed and all that jazz in
> Debian Jessie 64bit and I get this error... I got that ctefx.bin error first
> boot so I did that cp ctefx.bin and ctspeq.bin commands from above and
> they're both in "/lib/firmware" just like I did in 32bit... nope.
> alsa-firmware 1.0.28 is the culprit of everyone in this thread's problems.
> It just doesn't work in 64bit kernels... maybe it never has? I haven't tried
> it with kernel 3.9... it seems that alsa-firmware and the ctefx.bin and
> ctspeq.bin files only work perfectly in 32bit kernels...

It's working with my 64-bit kernel 3.17.7-gentoo. Using the method I posted last year via position fix.
Actually I just moved to this newly marked stable kernel and I don't think I need the position fix anymore because the DSP is loading fine now.

dmesg | grep -i ca0132
[ 4.129762] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
[ 29.385805] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

You clearly see early on in the boot it loaded so I may remove my script from the /etc/local.d and see how it goes on next reboot. (whenever that is)

Also as far as the windows reference , Microsoft is supported by creative these drivers were put together by reverse engineering because creative as of the ca0132 DSP dropped support for Linux all together.
Which is the reason I'll never buy Creative again as it's really their fault and not the developers at alsa for the kernel drivers.

[LaurentP (l5d)]

Sound Blaster Z: device detected but no sound. Confirmed that the device is unmuted.

System: Linux Mint 17 Qiana Cinnamon x64
Kernel: 3.13.0-24-generic x86_64
inxi -M
Machine: Mobo: ASRock model: 980DE3/U3S3 Bios: American Megatrends version: P1.50 date: 12/03/2013
inxi -A :
 Card-1: NVIDIA GK107 HDMI Audio Controller driver: snd_hda_intel Sound: ALSA ver: k3.13.0-24-generic
 Card-2: Creative Labs Device 0012 driver: snd_hda_intel
 Card-3: Advanced Micro Devices [AMD/ATI] SBx00 Azalia (Intel HDA) driver: snd_hda_intel
inxi -C
CPU: Octa core AMD FX-8320 Eight-Core (-MCP-) cache: 16384 KB flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm)
          Clock Speeds: 1: 1400.00 MHz 2: 1400.00 MHz 3: 1400.00 MHz 4: 1400.00 MHz 5: 1400.00 MHz 6: 3500.00 MHz 7: 1400.00 MHz 8: 1400.00 MHz
Do you Want any other info ?
in case, tell me how to get them, in French or in very simple english, there are some terms i don't understand in this subject.
Regards,
LaurentP.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

(In reply to Acidphase from comment #136)
> (In reply to Marc Bourgoin from comment #135)
> >
> >
> > .... nope! Nothing... alsa-firmware 1.0.28 is installed and all that jazz
> in
> > Debian Jessie 64bit and I get this error... I got that ctefx.bin error
> first
> > boot so I did that cp ctefx.bin and ctspeq.bin commands from above and
> > they're both in "/lib/firmware" just like I did in 32bit... nope.
> > alsa-firmware 1.0.28 is the culprit of everyone in this thread's problems.
> > It just doesn't work in 64bit kernels... maybe it never has? I haven't
> tried
> > it with kernel 3.9... it seems that alsa-firmware and the ctefx.bin and
> > ctspeq.bin files only work perfectly in 32bit kernels...
>
> It's working with my 64-bit kernel 3.17.7-gentoo. Using the method I posted
> last year via position fix.
> Actually I just moved to this newly marked stable kernel and I don't think I
> need the position fix anymore because the DSP is loading fine now.
>
> dmesg | grep -i ca0132
> [ 4.129762] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
> [ 29.385805] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
>
> You clearly see early on in the boot it loaded so I may remove my script
> from the /etc/local.d and see how it goes on next reboot. (whenever that is)
>
> Also as far as the windows reference , Microsoft is supported by creative
> these drivers were put together by reverse engineering because creative as
> of the ca0132 DSP dropped support for Linux all together.
> Which is the reason I'll never buy Creative again as it's really their fault
> and not the developers at alsa for the kernel drivers.

Heyyo,

Hmm I've tried even 3.18.0 on both Manjaro and Debian Jessie 64bit and no sound... Download always fails for me... I'll try downloading Gentoo maybe they have it setup differently or something...

as for your init.d scripts? Can you post the exact scripts you have please? Earlier you quoted the gentoo wiki I think for the "hello world" script.

[Acidphase (acidphase-linux-kernel-bugs)]

(In reply to Marc Bourgoin from comment #137)
> (In reply to Acidphase from comment #136)
> > (In reply to Marc Bourgoin from comment #135)
> > >
> > >
> > > .... nope! Nothing... alsa-firmware 1.0.28 is installed and all that jazz
> in
> > > Debian Jessie 64bit and I get this error... I got that ctefx.bin error
> first
> > > boot so I did that cp ctefx.bin and ctspeq.bin commands from above and
> > > they're both in "/lib/firmware" just like I did in 32bit... nope.
> > > alsa-firmware 1.0.28 is the culprit of everyone in this thread's
> problems.
> > > It just doesn't work in 64bit kernels... maybe it never has? I haven't
> tried
> > > it with kernel 3.9... it seems that alsa-firmware and the ctefx.bin and
> > > ctspeq.bin files only work perfectly in 32bit kernels...
> >
> > It's working with my 64-bit kernel 3.17.7-gentoo. Using the method I posted
> > last year via position fix.
> > Actually I just moved to this newly marked stable kernel and I don't think
> I
> > need the position fix anymore because the DSP is loading fine now.
> >
> > dmesg | grep -i ca0132
> > [ 4.129762] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
> > [ 29.385805] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
> >
> > You clearly see early on in the boot it loaded so I may remove my script
> > from the /etc/local.d and see how it goes on next reboot. (whenever that
> is)
> >
> > Also as far as the windows reference , Microsoft is supported by creative
> > these drivers were put together by reverse engineering because creative as
> > of the ca0132 DSP dropped support for Linux all together.
> > Which is the reason I'll never buy Creative again as it's really their
> fault
> > and not the developers at alsa for the kernel drivers.
>
> Heyyo,
>
> Hmm I've tried even 3.18.0 on both Manjaro and Debian Jessie 64bit and no
> sound... Download always fails for me... I'll try downloading Gentoo maybe
> they have it setup differently or something...
>
> as for your init.d scripts? Can you post the exact scripts you have please?
> Earlier you quoted the gentoo wiki I think for the "hello world" script.

It's all Kernel based setups. Did you setup the kernel correctly?

CONFIG_SND_HDA_CODEC_CA0132=y
CONFIG_SND_HDA_CODEC_CA0132_DSP=y
CONFIG_EXTRA_FIRMWARE="ctefx.bin ctspeq.bin"

You have to tell it to load the firmware manually. You may want to try the position fix too.

 rmmod snd_hda_intel
 modprobe snd_hda_intel position_fix=1

[snostormjp+linux (snostormjp+linux-linux-kernel-bugs)]

@Marc Bourgoin

Just for reference I am on Arch Linux 64 bit and my Mainboard has a CA0132 on it ,
wich is not great as the DSP load outthere is not fully functional, but it does worrk an I do have sound. If you would like I can go over my setup and wal you through minimal functionality. I have not real special fixes loaded not rmmod, modprobe I get sould on every boot, 2 mic inputs and 2 audio outputs
I did place a /etc/modprobe.d/ca0132.conf file with
options snd_hda_intel position_fix=1 in it, but I do not think taht was needed just preference as to which outputs I wanted to be on.

[stefan-oltmanns (stefan-oltmanns-linux-kernel-bugs)]

I own a mainboard with a CA0132 (Gigybyte G1.Sniper M3 Z77). I´m running Ubuntu 14.04 LTS 64-Bit with standard kernel 3.13. Until now only analog stereo/headphone out was working, but I was able to fix S/PDIF out: In file patch_ca0132.c in function ca0132_config I changed the headphone out pin (spec->out_pins[1]) from 0x10 to 0x0a. This maybe breaks something else, but for me both headphone and stereo out are still working.

I cannot get analog multichannel out working: I tried setting spec->multiout.max_channels to 6, add additional outputs and connect them to the other DACs in ca0132_init, the result is when running speaker-test that the channels are incorrectly mapped/mixed to stereo/headphone out for 4/6 channels, 3/5 channels no output at all. I tried different values for the output pins, does not change anything (only values above 0xC break S/PDIF, so they are used somehow, that´s how I accidentally fixed S/PDIF).

This are the kernel messages during boot (some could be HDMI (nvidia)):

kernel: [ 1.670502] snd_hda_intel 0000:00:1b.0: enabling device (0000 -> 0002)
kernel: [ 1.670689] snd_hda_intel 0000:00:1b.0: irq 46 for MSI/MSI-X
kernel: [ 1.683121] autoconfig: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
kernel: [ 1.683126] speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
kernel: [ 1.683129] hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
kernel: [ 1.683130] mono: mono_out=0x0
kernel: [ 1.683132] dig-out=0xc/0x0
kernel: [ 1.683133] inputs:
kernel: [ 1.683135] Mic=0x12
kernel: [ 1.683137] Line=0x11
...
kernel: [ 2.279042] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
...
kernel: [ 2.532358] input: HDA Intel PCH Front Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input8
kernel: [ 2.532451] input: HDA Intel PCH Line Out as /devices/pci0000:00/0000:00:1b.0/sound/card0/input7
kernel: [ 2.532530] input: HDA Intel PCH Line as /devices/pci0000:00/0000:00:1b.0/sound/card0/input6
kernel: [ 2.532638] input: HDA Intel PCH Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input5
kernel: [ 2.533187] snd_hda_intel 0000:02:00.1: enabling device (0000 -> 0002)

It still lists only 1 line-out. Is there some other place where I need to change the number of channels? Any ideas what I could/should try? This is the first time I ever modified something in the kernel.

[stefan-oltmanns (stefan-oltmanns-linux-kernel-bugs)]

Got something, in patch_ca0132 after snd_hda_parse_pin_def_config is called I added:

 spec->autocfg.line_outs = 3;
 spec->autocfg.line_out_pins[0] = 0xb;
 spec->autocfg.line_out_pins[1] = 0xc;
 spec->autocfg.line_out_pins[2] = 0xd;

The result is now that if I run speaker-test the output is mixed/mapped over 4 channels: the stereo out (0x0b) and the front-panel out (0x10). While headphone out used to be the same signal as stereo out, there is no output anymore. That means stereo-out/headphone out are not the same outputs. Interestingly now exactly those two connections work mentioned in the kernel message (Front Headphone and Line Out). How/where are those input devices created and why are the other outputs not created?

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

(In reply to J Pollitt from comment #139)
> @Marc Bourgoin
>
> Just for reference I am on Arch Linux 64 bit and my Mainboard has a CA0132
> on it ,
> wich is not great as the DSP load outthere is not fully functional, but it
> does worrk an I do have sound. If you would like I can go over my setup and
> wal you through minimal functionality. I have not real special fixes loaded
> not rmmod, modprobe I get sould on every boot, 2 mic inputs and 2 audio
> outputs
> I did place a /etc/modprobe.d/ca0132.conf file with
> options snd_hda_intel position_fix=1 in it, but I do not think taht was
> needed just preference as to which outputs I wanted to be on.

Heyyo,

I tried the options snd_hda_intel position_fix=1 and that didn't work for me on boot, then I tried options snd_hda_intel model=generic and that didn't work either... so then I tried options snd_hda_intel model=,generic and same thing...

BUT... if I open the terminal and use..
sudo kill `ps uax |grep pulseaudio | grep -v grep | awk '{print $2}'` ; sudo rmmod snd-hda-intel; sleep 3; sudo modprobe snd-hda-intel

I get sound working in 64bit with kernel 3.17... There doesn't seem to be a way I can use Arch unless every time I boot I run that command cause it seems like whenever I try something with a .conf in modprobe.d it doesn't work or load or something... sigh.

There's also no developers looking at fixing either the alsa firmware loader or the "ctefx.bin" or "ctspeq.bin" cause it always gives an error in 64bit only for me on boot.

Screw it, the main issue I have with Linux is the lack of NVIDIA SLI support anyways... I feel that issue stems more from the limitations of OpenGL's current state as AMD's CrossFire is also barely present/functioning... so the gaming performance on Linux for me is always lesser than Windows since Linux won't make use of my 2nd GPU unless I'm playing Doom 3 or another ID Tech 4 Engine game... seems to be the only game engine on Linux that supports frame buffer flipping instead of swapping.. heck, ID Tech 5 Engine on Windows doesn't support Multi-GPU for the same issue.. yet "The Evil Within" on ID Tech 5 Engine does... but it uses DirectX 11 only and marks the first time an ID Software and the ID Tech Engine released a game without any OpenGL support.

As much as I love how open and interesting Linux is? it's only an interesting adventure for me but now it's bordering on frustrating... maybe I'll try tinkering with Linux again in a few years and see if things improve, maybe if a SteamOS machine manufacturer pushes for multi-GPU setups we'll start seeing a change maybe in OpenGL and the way game devs code games.

To the others on here? good luck and take care. I'm out for now...

[stefanspr94 (stefanspr94-linux-kernel-bugs)]

Is this bug even still being worked on? This thread has not been updated for over a month. I would like to hear from Jaroslav Kysela whats the current state of this.

[beck (beck-linux-kernel-bugs)]

I confirm that changing spec->out_pins[1] to 0x0A will activate the optical S/PDIF on my ca0132 PCIE card.
As this is the only feature i am interested in, i did not check other outputs/inputs.
I applied the change to a clean 3.19.1 on a 64bit Debian 8 system and have sound working through optical out with no other fixes or tricks.
The possible SFX listed by ALSA such as X-Bass however do not seem have any audible impact, may be just my ears, but the only interest i have in those would be to make sure they are disabled.
My thanks go out to Stefan O. for pointing this out.

Sound is clean, no clicks or blips during boot or playback.

Finally - no more ground-loops humming through my hifi.

[slobodan.kandic (slobodan.kandic-linux-kernel-bugs)]

Hello, can you say me where i can find the patch_ca0132.c file? i searched everywhere on my system and i can't find it.
i also have the same problem with you and iam using my soundblaster z to feed my dac through S/PDIF.

Thanks.

[beck (beck-linux-kernel-bugs)]

Created attachment 173111
attachment-29674-0.html

You need the kernel source, get it from kernel.org. it is in folder sound/pci/hda. Change and compile changed sources. There will be howtos for your distro on this.

On April 4, 2015 12:30:03 PM CEST, <email address hidden> wrote:
>
>https://bugzilla.kernel.org/show_bug.cgi?id=55541
>
>ikomiko <email address hidden> changed:
>
> What |Removed |Added
>----------------------------------------------------------------------------
> CC| |<email address hidden>
>
>--- Comment #145 from ikomiko <email address hidden> ---
>Hello, can you say me where i can find the patch_ca0132.c file? i
>searched
>everywhere on my system and i can't find it.
>i also have the same problem with you and iam using my soundblaster z
>to feed
>my dac through S/PDIF.
>
>Thanks.
>
>--
>You are receiving this mail because:
>You are on the CC list for the bug.

[iamgrief (iamgrief-linux-kernel-bugs)]

I am frustrating. The issue occurs with the pretty common ca0132 chipset when 64bit kernel (everybody use 64 bit instead of 32 bit nowadays) is installed. On the other hand, the issue is not applicable to 32bit version of the kernel which means that there is a working code already. Kernel team has nobody who can investigate, address and fix the issue for 2 years. Really?

[stefanspr94 (stefanspr94-linux-kernel-bugs)]

@Grief I do absolutely feel the same. I am very disappointed about the situation too. @Kernel-Devs I am not a developer and do just have minor C knowledge but I think the problem is 64bit memory addressing. Another firmware loading routine would have to be made in order to solve the problem.

[xcariba (xcariba-linux-kernel-bugs)]

I don`t have problems with firmware loading, front headphones and microphone works fine with simple script, but I want to use rear outputs (I have SB Zx with external heaphone output). I`ve tried to make it work by editing patch_ca0132.c file and recompiling mainline kernel, but still rear outputs don`t work. Is there any docs on ca0132? I don`t have a lot of experience in alsa drivers, I only worked with simple linux kernel modules.

[pgdm (paulogdemitri)]

Hello. Kernel 4.0 here and no sound on HP jack. What a mess...

[fabianrensen (fabianrensen-linux-kernel-bugs)]

@Alexander Really? I always need to boot Windows first to "initialize" the card and after reboot I can get (front) headphones to work with the script. Also, rebooting from Linux back to Windows breaks the microphone, so I have to completely shut down first.

Any other news so far?

[xcariba (xcariba-linux-kernel-bugs)]

@fabianrensen I've got microphone bug, but I don't need to boot windows before Linix.

There is one interesting commit in sound tree
https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/patch_ca0132.c?id=fe14f39e88c8ac16d0a051f8444a2294f8cb358c
In new alienware_pincfgs[] array there are several pin ids with N/A comment. Only speakers, hp and mic commented since laptop only has speakers and combined jack. Maybe they can be used to configure all outputs on Z series.

I've tried several months ago to add extra out pins with different ids, but I don't really understand what is going on in patch_ca0132.c file.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

Wow... 64bit Kernel 4.2.0 made things worse!!! :(

Trying it out on Ubuntu Mate 15.10 64bit... 4.2.0-16 btw... tried booting with /etc/rc.local set with:

rmmod snd_hda_intel
modprobe snd_hda_intel position_fix=1

Gives me sound yeah... but doesn't work in Chrome or FireFox at all... only the desktop...

So then I tried booting with /etc/rc.local set with:

rmmod snd_hda_intel
modprobe snd_hda_intel model=generic

and that gives me sound... but randomly sound stops and starts working!

gonna try the latest alsa-firmware 1.0.29 files now from the AlsaProject website and dump them into /usr/firmware

.... NOPE! Same issue where sound randomly stops working when it wants to...

dmesg | grep -i ca0132
[ 8.218428] snd_hda_codec_ca0132 hdaudioC2D1: autoconfig for CA0132: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 8.218432] snd_hda_codec_ca0132 hdaudioC2D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 8.218433] snd_hda_codec_ca0132 hdaudioC2D1: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 8.218434] snd_hda_codec_ca0132 hdaudioC2D1: mono: mono_out=0x0
[ 8.218435] snd_hda_codec_ca0132 hdaudioC2D1: dig-out=0xc/0x0
[ 8.218436] snd_hda_codec_ca0132 hdaudioC2D1: inputs:
[ 8.218437] snd_hda_codec_ca0132 hdaudioC2D1: Mic=0x12
[ 8.218438] snd_hda_codec_ca0132 hdaudioC2D1: Line=0x11
[ 8.218438] snd_hda_codec_ca0132 hdaudioC2D1: dig-in=0xe
[ 11.186450] ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING.
[ 19.911889] snd_hda_codec_ca0132 hdaudioC2D1: autoconfig for CA0132: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 19.911891] snd_hda_codec_ca0132 hdaudioC2D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 19.911892] snd_hda_codec_ca0132 hdaudioC2D1: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 19.911893] snd_hda_codec_ca0132 hdaudioC2D1: mono: mono_out=0x0
[ 19.911893] snd_hda_codec_ca0132 hdaudioC2D1: dig-out=0xc/0xd
[ 19.911894] snd_hda_codec_ca0132 hdaudioC2D1: inputs:
[ 19.911895] snd_hda_codec_ca0132 hdaudioC2D1: Mic=0x12
[ 19.911896] snd_hda_codec_ca0132 hdaudioC2D1: Line=0x11
[ 19.911897] snd_hda_codec_ca0132 hdaudioC2D1: dig-in=0xe
[ 20.495197] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
[ 21.322538] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

uname -a
Linux themard-mate-64bit 4.2.0-16-generic #19-Ubuntu SMP Thu Oct 8 15:35:06 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux

.... is there any chance @Jaroslav Kysela can check into this more??? or someone???

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

After more digging around it looks like the Alsa Project makes changes to the ctefx.bin and ctspeq.bin files...

http://www.alsa-project.org/main/index.php/Main_Page

the Bug Tracker portion of the website is down and the "help to debug" doesn't tell you how to submit bugs lol agh! :P

Ok, for the Alsa project it turns out there's an actual dude who handles Sound Blasters! Maybe he can bring us salvation for 64bit firmware loading of ca0132! :)

http://www.alsa-project.org/main/index.php/Alsa_Team

I'm going to email James Courtier-Dutton and hopefully he can fix this bug that has plagued us for years... I'm gonna submit him my alsa info from their uploader that showed me "/tmp/alsa-info.txt.sLncDWamsd"

[superquad.vortex2 (superquad.vortex2-linux-kernel-bugs)]

https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/commit/sound/pci/hda/hda_intel.c?id=cadd16ea33a938d49aee99edd4758cc76048b399

[iamgrief (iamgrief-linux-kernel-bugs)]

As soon as I saw this interesting patch in previous message, I tried to to test it and know what? On kernel 4.2.3 I hear the sound from green out on the back even without that patch. Nothing else works but it's a progress anyway for me, but I guess this is the situation where other people in the tread are at.

Regarding the patch, it didn't work for me, on kernel 4.2.3 I have one working out with and without it.

Very sad that microphone is not working...

Relevant piece of dmesg:

[ 8.220066] snd_hda_codec_ca0132 hdaudioC0D0: autoconfig for CA0132: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 8.220067] snd_hda_codec_ca0132 hdaudioC0D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 8.220067] snd_hda_codec_ca0132 hdaudioC0D0: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 8.220068] snd_hda_codec_ca0132 hdaudioC0D0: mono: mono_out=0x0
[ 8.220068] snd_hda_codec_ca0132 hdaudioC0D0: dig-out=0xc/0x0
[ 8.220068] snd_hda_codec_ca0132 hdaudioC0D0: inputs:
[ 8.220069] snd_hda_codec_ca0132 hdaudioC0D0: Mic=0x12
[ 8.220069] snd_hda_codec_ca0132 hdaudioC0D0: Line=0x11

[ 8.787492] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

[ 9.624422] ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo, @Grief , to get microphone working? I've always had to do this, even in the 32bit kernel:

1. Open up "Terminal"

2. Type in "alsamixer" and hit enter
3. Press "F6" to Select Sound Card

4. Select "HDA Creative" and hit enter

5. Press "F4" for Capture

6. Use the left and right arrows to navigate the selections and use "Insert" key to enable/disable selections

7. Select the "AMic1/DMic" and click your "Insert" key to enable it. Do the same with "AMic1/DMic Auto Detect".

8. If the input is quiet, select "Mic1-Boost" and click your "Insert" key to enable mic boost.

[iamgrief (iamgrief-linux-kernel-bugs)]

I am soooo stupid :(

Marc, thank you a lot! AMic1/DMic Auto Detect seems was not needed for me, I had only to enable "Capture" and "AMic1/DMic". But all functions work well exactly as they work on windows, so I enabled "Mic1-Boost (30 dB)", "Echo Cancellation", "Noise Reduction" as they make the sound quality of my cheap headset much better.

[h (h-linux-kernel-bugs)]

With the patch from Takashi Iwai applied to kernel 4.2.5 64-bit I finally get "DOWNLOAD OK :-) DSP IS RUNNING." without having to reload snd_hda_intel. Still can only use the front panel jack for output regardless of toggling HP/Speaker in alsamixer. Mic still works with the jack on the back of the card.

Something still seems wrong with the buffer though - playback only works for less than 1 second and then goes silent. This also happens without the patch.

$ speaker-test -Dhw:1,0 -c2

Sound for 1 sec, then goes silent.

$ speaker-test -Dhw:1,0 -c2 -b 21666

...works just fine. Increasing the buffer size to 21667 and it will be back to going silent.

For Pulseaudio I had to add
load-module module-alsa-card device_id=1 tsched_buffer_size=4159
to /etc/pulseaudio/default.pa.
Again, increasing buffer size to just 4160 and it stops working.

[tiwai (tiwai-linux-kernel-bugs)]

The stall is a regression since 4.2 kernel and fixed today in sound git tree by the commit de1ab6af5c3d92c0a031083962a7ff270cf301b7
    ALSA: hda - Fix lost 4k BDL boundary workaround
Found in for-linus or for-next branch.

[h (h-linux-kernel-bugs)]

Thanks a lot for working on this Takashi. Unfortunately just applying the change in that commit to 4.2.5 or 4.3.0 and trying to run speaker-test without restricting the buffer size results in "Unable to set hw params for playback: Invalid argument
Setting of hwparams failed: Invalid argument", and for every attempt this shows up in dmesg:
[ 137.330619] snd_hda_intel 0000:06:00.0: Too many BDL entries: buffer=4194304, period=1048576
Pulseaudio works fine though without tsched_buffer_size=4159...

[tiwai (tiwai-linux-kernel-bugs)]

Yes, the restriction makes the driver requiring more BDL entries because it has to align 4k bytes. Depending on the parameters, it can't work as is -- it's a hardware limitation. The behavior existed up to 4.1, but this was overseen during the code change for 4.2, sorry.

So, simply reduce the buffer size to pass to speaker-test. Or reduce the periods (but then speaker-test will behave badly, too). PA basically doesn't use periods, that's why it works with the same large buffer size.

[jadit2 (jadit2-linux-kernel-bugs)]

Super excited about the progress as I've been watching this issue since 2013. :]

[iamgrief (iamgrief-linux-kernel-bugs)]

Is there a chance to make other jacks working in addition to back green and pink ones?

[tiwai (tiwai-linux-kernel-bugs)]

(In reply to Grief from comment #163)
> Is there a chance to make other jacks working in addition to back green and
> pink ones?

I don't know what are other jacks and what are for. You need to list up and figure out the following by yourself:
- The position and the color of each jack, and its purpose.
- The pin NID corresponding to each jack.

For checking the latter, you can try hda-verb or hdajackretask. With hda-verb, something like
  hda-verb /dev/snd/hwC1D1 0x0b GET_PIN_SENSE 0

where /dev/snd/hwC1D1 is the hwdep device file corresponding to your Creative sound card. In this case, it's card 1, codec address 1.
0x0b is the widget NID. On CA0132 (at least on Recon3D), the NIDs from 0x0b to 0x13 and NID 0x18 are pin widgets, so look through all these.

When a pin is detected to the given NID, the command above should return the value with bit 31 set, e.g. 0x80000000. Plug/unplug each jack, and find out which NID corresponds to which jack.

After figuring out all these, give the result together with alsa-info.sh output.

[ganthore (ganthore-linux-kernel-bugs)]

Hey Takashi,

I'm still getting consistently nothing out of ZxR card (currently running 4.3.0).

I followed your instructions but I got a return of 0x0. I wrote a quick script to run hda-verb on an range from 0x0b to 0x18. I consistently got a return of value = 0x0. If I did this right just let me know and I'll attach the script to this ticket so others can quickly run it.

In my case there is no "green" jack. The card has 2 RCA out ports that hook up to cable that converts to 3.5mm out. I also have a daughter board that comes with this card but I don't have that connected right now because it only supplies optic out and aux in.

/dev/snd/hwC3D1:
nid = 0xb, verb = 0xf09, param = 0x0
value = 0x0
nid = 0xc, verb = 0xf09, param = 0x0
value = 0x0
nid = 0xd, verb = 0xf09, param = 0x0
value = 0x0
nid = 0xe, verb = 0xf09, param = 0x0
value = 0x0
nid = 0xf, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x10, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x11, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x12, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x13, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x14, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x15, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x16, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x17, verb = 0xf09, param = 0x0
value = 0x0
nid = 0x18, verb = 0xf09, param = 0x0
value = 0x0

From dmesg:
[ 3.554859] snd_hda_codec_ca0132 hdaudioC3D1: autoconfig for CA0132: line_outs=1 (0xd/0x0/0x0/0x0/0x0) type:speaker
[ 3.554861] snd_hda_codec_ca0132 hdaudioC3D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 3.554863] snd_hda_codec_ca0132 hdaudioC3D1: hp_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 3.554864] snd_hda_codec_ca0132 hdaudioC3D1: mono: mono_out=0x0
[ 3.554864] snd_hda_codec_ca0132 hdaudioC3D1: inputs:
[ 3.554866] snd_hda_codec_ca0132 hdaudioC3D1: Mic=0x12

The following link is my current alsa-info (please ignore much older one attached to this ticket):
http://www.alsa-project.org/db/?f=a9217c4765524689ebfbbc92da791d38d9506a0a

Please let me know if I'm doing something wrong or if you'd like any more data.

[ganthore (ganthore-linux-kernel-bugs)]

Well, I do get some incredibly awful static along with system sound if I keep my model set to generic. My ZxR is the 4th card that comes up in the device list right now so I have my modprobe.d config set to:

% cat /etc/modprobe.d/sound.conf
options snd-hda-intel position_fix=1 model=,,,generic

[tiwai (tiwai-linux-kernel-bugs)]

(In reply to ganthore from comment #165)
> Hey Takashi,
>
> I'm still getting consistently nothing out of ZxR card (currently running
> 4.3.0).
>
> I followed your instructions but I got a return of 0x0. I wrote a quick
> script to run hda-verb on an range from 0x0b to 0x18. I consistently got a
> return of value = 0x0. If I did this right just let me know and I'll attach
> the script to this ticket so others can quickly run it.
>
> In my case there is no "green" jack. The card has 2 RCA out ports that hook
> up to cable that converts to 3.5mm out. I also have a daughter board that
> comes with this card but I don't have that connected right now because it
> only supplies optic out and aux in.

OK, then it doesn't work, as RCA jack doesn't have a proper detection mechanism.

As you can see in patch_ca0132.c ca0132_config() function, it has static mapping of pins. So far, it assumes only two models: the generic one and Alienware. The former uses pins 0x0b and 0x10 for outputs, 0x10 is the headphone jack. Meanwhile the latter uses the pins 0x10 and 0x0f as outputs, and 0x0f is the headphone jack.

Similarly, you need to find out two actual outputs (might be one, I don't know the spec of this board) by trial and error, i.e. just hard-coding the number in patch_ca0132.c. The BIOS reports it's 0x0d, so this might be one.

[fabianrensen (fabianrensen-linux-kernel-bugs)]

(In reply to Takashi Iwai from comment #167)
> (In reply to ganthore from comment #165)
> > Hey Takashi,
> >
> > I'm still getting consistently nothing out of ZxR card (currently running
> > 4.3.0).
> >
> > I followed your instructions but I got a return of 0x0. I wrote a quick
> > script to run hda-verb on an range from 0x0b to 0x18. I consistently got a
> > return of value = 0x0. If I did this right just let me know and I'll attach
> > the script to this ticket so others can quickly run it.
> >
> > In my case there is no "green" jack. The card has 2 RCA out ports that hook
> > up to cable that converts to 3.5mm out. I also have a daughter board that
> > comes with this card but I don't have that connected right now because it
> > only supplies optic out and aux in.
>
> OK, then it doesn't work, as RCA jack doesn't have a proper detection
> mechanism.
>
> As you can see in patch_ca0132.c ca0132_config() function, it has static
> mapping of pins. So far, it assumes only two models: the generic one and
> Alienware. The former uses pins 0x0b and 0x10 for outputs, 0x10 is the
> headphone jack. Meanwhile the latter uses the pins 0x10 and 0x0f as
> outputs, and 0x0f is the headphone jack.
>
> Similarly, you need to find out two actual outputs (might be one, I don't
> know the spec of this board) by trial and error, i.e. just hard-coding the
> number in patch_ca0132.c. The BIOS reports it's 0x0d, so this might be one.

I am not a kernel dev, nor do I have an advanced knowledge about it, but if you say the pin values need to be hard coded, that certainly means recompiling something. So does that mean we have to recompile the whole kernel for every test or is it possible to only recompile the kernel module?

[tiwai (tiwai-linux-kernel-bugs)]

You can compile only the target kernel module, in this case, the ones in sound/pci/hda/ subdirectory, without rebuilding the whole tree, too.

[fabianrensen (fabianrensen-linux-kernel-bugs)]

Okay, thx.

I just compiled ALSA from source via the alsa-compile.sh script and included your 32-bit patch.

I can confirm that this indeed fixes the DSP load error. I get a

ca0132 DOWNLOAD OK :-) DSP IS RUNNING.

on boot now and front headphones are working after boot.

[fabianrensen (fabianrensen-linux-kernel-bugs)]

(In reply to fabianrensen from comment #170)
> Okay, thx.
>
> I just compiled ALSA from source via the alsa-compile.sh script and included
> your 32-bit patch.
>
> I can confirm that this indeed fixes the DSP load error. I get a
>
> ca0132 DOWNLOAD OK :-) DSP IS RUNNING.
>
> on boot now and front headphones are working after boot.

And btw I'm on Mint 17.2 x64 KDE, Kernel 3.16.0-38-generic and I have a Soundblaster Z.

[faffo665 (faffo665-linux-kernel-bugs)]

Download and installed linux kernel 4.4-rc1 for arch linux (via aur)
the DSP il loading and front headphone jack works (i do not have a spare mic to test front mic jack but connecting a pair of headphones gives no pop or noise, so I assume it's not working)

Rear jacks not working at all.

I'm on arch linux x86_64, kernel 4.4.0-rc1-g8005c49 (pulled directly from git so i assume it's unpatched) and my soundcard is a Creative SoundBlaster Z.

[tiwai (tiwai-linux-kernel-bugs)]

As already mentioned, you must figure out the pins by yourself. It can be done only by the owner of the board. Use hdajackretask or whatever to identify which pin corresponds to which I/O jack. The normal 3.5 cinch jacks should be detectable, so they can be identified easily.

The rest is another trial-and-error, to map the identified pins in the arrays in ca0132_config() function.

[faffo665 (faffo665-linux-kernel-bugs)]

Fast way to do so?
I'm new to this kernel-patching-thing... I can get by a download, compile and install of whole kernels... but I haven't grasped yet how to patch.

I get i have to edit patch patch_ca0132.c (I'm assuming the one on you pushed to Linus' repo) but then? how do i patch just the file?
Sorry if i'm annoying but i'm trying to learn (no offense taken if you have no time to spare to teach me ;) )

[fabianrensen (fabianrensen-linux-kernel-bugs)]

(In reply to Takashi Iwai from comment #173)
> As already mentioned, you must figure out the pins by yourself. It can be
> done only by the owner of the board. Use hdajackretask or whatever to
> identify which pin corresponds to which I/O jack. The normal 3.5 cinch
> jacks should be detectable, so they can be identified easily.
>
> The rest is another trial-and-error, to map the identified pins in the
> arrays in ca0132_config() function.

Unfortunately, hdsjackretask always shows 5 pins as connected, regardless of what is connected.

[dhiru.kholia (dhiru.kholia-linux-kernel-bugs)]

Created attachment 194731
FreeBSD 11.0-CURRENT verbose boot logs

The rear jack works very well on FreeBSD 11.0-CURRENT.

[dhiru.kholia (dhiru.kholia-linux-kernel-bugs)]

Please see "Original pins configuration" in the FreeBSD verbose boot logs. I suspect that it can help us in getting the rear jacks to work well.

[tiwai (tiwai-linux-kernel-bugs)]

Grrr, there are way too many boards and this bug got too long, so I have no idea at all who writes for what card model for which problem on which kernel version with which distro.

The original problem, the firmware loading issue, was already fixed in 4.4-rc1 and will be backported to stable kernels. So, I close this bug for now.

For the rest problems, please open a new bug for *each* different board.
Please follow the rule below:
- Put the exact board name and PCI SSID (e.g. 1102:0023) in the subject line to make clear.
- Each bug entry is dedicated only for the given board model. Don't mix up.
- Give alsa-info.sh output in each bug report.
  Take alsa-info.sh output with --no-upload option, and attach the output file to Bugzilla (don't paste).
- Describe exactly which function doesn't work.

Thanks.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

Takashi, I tried sending a PayPal donation to your suse.de email on November 2nd but you never accepted it. Is there a better way to send you a donation? I'm seriously grateful on your work squashing this bug.

[tiwai (tiwai-linux-kernel-bugs)]

Marc, no need for donation, although I thank for your offer. Fixing the issues is just my task as a maintainer, so don't take it too seriously :)

Rather please advertise the fix (and how to report the rest issues) on blog or whatever so that people may know of it.

[andrei03.costache (andrei03.costache-linux-kernel-bugs)]

Hello Guys,

I also have a Sound Blaster Z running under Ubuntu 15.10. (the MB is a Gigabyte GA-Z97X-Gaming 7, but all onboard stuff is disabled in BIOS - sound, video, network, ...).

I am really new to Linux, but I am not computer illiterate and want to learn. Nevertheless, the whole discussion on this bug, which I have read several times (trying myself to do whatever it was that you guys were doing) is just too advanced for me.

Could you please make just a step by step guide for me to to follow stupidly, so that I can also apply the fix which apparently is now available? Just consider that I have a fresh installation of Ubuntu 15.10. (no recompiled kernels, no mods, no patches, no nothing...)

Also, I own a SoundBlaster ZxR, and in the interest of contributing and learning, I am also willing to go out and buy a Sound Blaster Zx, if you are willing to walk me through providing to you all the info that you need and testing all the stuff that you need to finally clean-up all the remaining problems with the pins and whatnot.

Just let me know what you think.

Thanks

[dhiru.kholia (dhiru.kholia-linux-kernel-bugs)]

Hi Andrei,

Installing, and running http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.4-rc2+cod1-wily/ is enough to get the front headphone jack working.

Good Luck!

[andrei03.costache (andrei03.costache-linux-kernel-bugs)]

Hi Dhiru,

I finally managed to install and run the 4.4.0 kernel from where you indicated. But there is no sound neither from the headphones jack in front of the case, nor from the back of the soundcard.

Also, dmesg now says that the DSP is downloaded and running...

But still no sound...

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo,

I used the Phoronix build which works good since there's problems with Ubuntu's 4.4 RC kernel builds.

http://www.phoronix.com/scan.php?page=news_item&px=Linux-4.4-rc2-Spun-For-Ubuntu

Using those Phoronix builds? 4.4 RC2 is giving me sound on boot without any tweaking! It works good Thanks again Takashi Iwai!!!! :)

Tested on 15.10 Ubuntu Mate

[andrei03.costache (andrei03.costache-linux-kernel-bugs)]

(In reply to Marc Bourgoin from comment #184)
> Heyyo,
>
> I used the Phoronix build which works good since there's problems with
> Ubuntu's 4.4 RC kernel builds.
>
> http://www.phoronix.com/scan.php?page=news_item&px=Linux-4.4-rc2-Spun-For-
> Ubuntu
>
> Using those Phoronix builds? 4.4 RC2 is giving me sound on boot without any
> tweaking! It works good Thanks again Takashi Iwai!!!! :)
>
> Tested on 15.10 Ubuntu Mate

Hi Marc,

No, I tested the ubuntu mainline kernel builds, like Dhiru sugegsted.
 - With the version 4.4.0rc2 the "DSP is downloaded and running"; but I get no sound.
 - With the version 4.4.0rc3 the "DSP is downloaded and running" as well, but still no sound.

In the mean time, I have enabled the onboard sound and also inserted in the PCI slot available on the MB an old Audigy 2 (SB0240) card I had lying around.

Here is the DMESG output I get with kernel 4.4.0rc3 from the ubuntu mainline:

andrei@andrei-pc:~$ dmesg |grep snd
[ 2.676441] snd_hda_intel 0000:00:1b.0: enabling device (0000 -> 0002)
[ 2.676676] snd_hda_intel 0000:01:00.1: Disabling MSI
[ 2.676680] snd_hda_intel 0000:01:00.1: Handle vga_switcheroo audio client
[ 2.676744] snd_hda_intel 0000:08:00.0: enabling device (0000 -> 0002)
[ 2.676751] snd_hda_intel 0000:08:00.0: Disabling MSI
[ 2.676752] snd_hda_intel 0000:08:00.0: Force to non-snoop mode
[ 2.699959] snd_hda_codec_ca0132 hdaudioC2D1: autoconfig for CA0132: line_outs=1 (0xd/0x0/0x0/0x0/0x0) type:line
[ 2.699961] snd_hda_codec_ca0132 hdaudioC2D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 2.699963] snd_hda_codec_ca0132 hdaudioC2D1: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 2.699963] snd_hda_codec_ca0132 hdaudioC2D1: mono: mono_out=0x0
[ 2.699964] snd_hda_codec_ca0132 hdaudioC2D1: dig-out=0xc/0x0
[ 2.699965] snd_hda_codec_ca0132 hdaudioC2D1: inputs:
[ 2.699966] snd_hda_codec_ca0132 hdaudioC2D1: Mic=0x12
[ 2.699967] snd_hda_codec_ca0132 hdaudioC2D1: dig-in=0xe
[ 2.704506] snd_hda_codec_realtek hdaudioC0D2: ALC1150: SKU not ready 0x00000000
[ 2.705020] snd_hda_codec_realtek hdaudioC0D2: autoconfig for ALC1150: line_outs=3 (0x14/0x15/0x16/0x0/0x0) type:line
[ 2.705021] snd_hda_codec_realtek hdaudioC0D2: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 2.705022] snd_hda_codec_realtek hdaudioC0D2: hp_outs=1 (0x1b/0x0/0x0/0x0/0x0)
[ 2.705023] snd_hda_codec_realtek hdaudioC0D2: mono: mono_out=0x0
[ 2.705024] snd_hda_codec_realtek hdaudioC0D2: dig-out=0x11/0x1e
[ 2.705024] snd_hda_codec_realtek hdaudioC0D2: inputs:
[ 2.705026] snd_hda_codec_realtek hdaudioC0D2: Front Mic=0x19
[ 2.705027] snd_hda_codec_realtek hdaudioC0D2: Rear Mic=0x18
[ 2.705028] snd_hda_codec_realtek hdaudioC0D2: Line=0x1a
[ 2.715352] snd_emu10k1 0000:05:00.0: enabling device (0000 -> 0001)
[ 2.730576] snd_emu10k1 0000:05:00.0: Installing spdif_bug patch: SB Audigy 2 [SB0240]
[ 3.282317] snd_hda_codec_ca0132 hdaudioC2D1: ca0132 DSP downloaded and running
[ 4.550260] snd_hda_codec_ca0132 hdaudioC2D1: ca0132 DSP downloaded and running

Some things I notice right away:
- ca0132 downloaded and running, but no sound
- the ALC 1150 is det...

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo Andrei, hmm just to confirm you are using the line-out port on the sound card and not the headphone port correct? I just tested with the headphone port (the one next to the mic) and it does not work even if I select headphone in Pulse Audio (tbh selecting line out or headphone seems to not change anything)... yet if I plug into the line out (the one next to the headphone port) it works fine for me with pulse audio set to either headphone or line out.

As of Ubuntu 4.4 rc3 kernel builds? They did fix that bug so you should be able to run their build of that kernel without issues.

I use a Sound Blaster Recon3D FALTAL1TY Professional PCI-E btw.

the_mard@themard-mate64:~$ dmesg |grep snd
[ 9.648981] snd_hda_intel 0000:03:00.1: Disabling MSI
[ 9.648985] snd_hda_intel 0000:03:00.1: Handle vga_switcheroo audio client
[ 9.649044] snd_hda_intel 0000:04:00.1: Disabling MSI
[ 9.649046] snd_hda_intel 0000:04:00.1: Handle vga_switcheroo audio client
[ 9.649092] snd_hda_intel 0000:05:00.0: Disabling MSI
[ 9.649093] snd_hda_intel 0000:05:00.0: Force to non-snoop mode
[ 9.677161] snd_hda_codec_ca0132 hdaudioC2D1: autoconfig for CA0132: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 9.677163] snd_hda_codec_ca0132 hdaudioC2D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 9.677164] snd_hda_codec_ca0132 hdaudioC2D1: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 9.677164] snd_hda_codec_ca0132 hdaudioC2D1: mono: mono_out=0x0
[ 9.677165] snd_hda_codec_ca0132 hdaudioC2D1: dig-out=0xc/0x0
[ 9.677166] snd_hda_codec_ca0132 hdaudioC2D1: inputs:
[ 9.677167] snd_hda_codec_ca0132 hdaudioC2D1: Mic=0x12
[ 9.677168] snd_hda_codec_ca0132 hdaudioC2D1: Line=0x11
[ 9.677168] snd_hda_codec_ca0132 hdaudioC2D1: dig-in=0xe
[ 11.504281] snd_hda_codec_ca0132 hdaudioC2D1: ca0132 DSP downloaded and running
[ 14.095938] snd_hda_codec_ca0132 hdaudioC2D1: ca0132 DSP downloaded and running

Here's what I mean for ports on the back:
http://images17.newegg.com/is/image/newegg/29-102-048-Z09?$S640$

The top one is microphone, the one next to it is headphones and the third is the line out that works.

[andrei03.costache (andrei03.costache-linux-kernel-bugs)]

Created attachment 196841
3 sound cards (onboard + Audigy SB0220 + Creative SoundBlaster Z) - only onboard audio works

[andrei03.costache (andrei03.costache-linux-kernel-bugs)]

Hello Marc,

Thanks for the reply. You are right, I was using the headphone jack, but I checked and the line out jack does not work either.

So, to sum up, my SW and HW configuration is the following:
HW:
- onboard audio - works
- NVIDIA digital audio - could not test (no DAC available)
- Creative Sound Blaster Audigy 2 (SB0220) - does not work
- Creative Sound Blaster Z - does not work
SW:
- fresh install of Ubuntu 15.10
- perform a kernel upgrade to 4.4.0rc3 from Ubuntu mainline
- install bcmwl driver (via apt)
- enable nvidia 352.xx driver (via "additional drivers" tab)
- no audio settings changed - everything default
- no special audio SW installed from SWCenter
- everything is pretty much default Ubuntu config.

How I test sound:
I go to SystemSettings -> Sound;
I select the output where I have some speakers/headphones connected
I press "Test Sound" button.
No sound comes out, except from the onboard audio.

My dmesg outut is the one from the post above (186)
I attached my alsa-info output.

Please give me a hint what to do to get the Sound Blaster Z working. I want very badly to switch to Linux, and the sound topic is the only one keeping me from completely giving up on Windows.

[alexey12424323 (alexey12424323-linux-kernel-bugs)]

Hi, i can't get my Sound Blaster Z to work on linux either. No sound comes out from it not from Headphone Front Panel nor from Rear Line Out. I tried kernel 4.4rc but that didn't change anything. I even tried hardcoding pin confgs in ca0132.c and recompile kernel but that didn't hep either. It seems like for some people it works and for some it doesn't. Can someone plsease post your alsa-info.sh of WORKING Sound Blaster Z? I need to see the pin config and maybe i can get it working. Thanks. Just in case i will attach my alsa-info.sh here: http://pastebin.com/YRMGrzXf

[louiz (louiz-linux-kernel-bugs)]

(In reply to Takashi Iwai from comment #178)
>
> For the rest problems, please open a new bug for *each* different board.
> Please follow the rule below:
>
> Thanks.

I’ve opened this bug for the specific Sound Blaster Z issue:

https://bugzilla.kernel.org/show_bug.cgi?id=109191

[LaurentP (l5d)]

Hello, Happy new year for 2016.

With Kernel 3.19.0-32-generic x86_64, the Creative Sound Blaster Z [1102:0012] still don't play sound.
is someone testing with newer kernel, like 4.2.0.22 ? this is the newer i can found in Linux mint 17.3 Rosa, which is derivative from Ubuntu 14.04.

[thatsallurspaceships (csub)]

SB Recon3D [1102:0012]
Currently with Kernel 4.3.3-2r no sound aka. ca0132 Download Failed

[mikhail.v.gavrilov (mikhail.v.gavrilov-linux-kernel-bugs)]

(In reply to Takashi Iwai from comment #178)
> Grrr, there are way too many boards and this bug got too long, so I have no
> idea at all who writes for what card model for which problem on which kernel
> version with which distro.
>
> The original problem, the firmware loading issue, was already fixed in
> 4.4-rc1 and will be backported to stable kernels. So, I close this bug for
> now.
>
> For the rest problems, please open a new bug for *each* different board.
> Please follow the rule below:
> - Put the exact board name and PCI SSID (e.g. 1102:0023) in the subject line
> to make clear.
> - Each bug entry is dedicated only for the given board model. Don't mix up.
> - Give alsa-info.sh output in each bug report.
> Take alsa-info.sh output with --no-upload option, and attach the output
> file to Bugzilla (don't paste).
> - Describe exactly which function doesn't work.
>
> Thanks.

Takashi, please look into this: https://bugzilla.kernel.org/show_bug.cgi?id=112851
Creative Soundblaster ZxR [1102:0012]

[mdrogers (mdrogers-linux-kernel-bugs)]

(In reply to Takashi Iwai from comment #180)
> Marc, no need for donation, although I thank for your offer. Fixing the
> issues is just my task as a maintainer, so don't take it too seriously :)
>
> Rather please advertise the fix (and how to report the rest issues) on blog
> or whatever so that people may know of it.

I've tried doing the modification to the ca0132 patch file and recompiling the kernel and can't seem to get the audio to work still. In fact after installing the kernel it never appeared to have changed the value of the output.
Takashi, do you think it's possible you or another person could post a step by step tutorial as to the 0x0F port change modification in the patch file?

I know some have spoke just modifying the kernel to include the compiled patch file instead of recompiling the whole kernel with it, how does one do this exactly?

I've had very little exposure to Linux and am trying to start using it full time to learn more but this sound error is stopping me. Currently running mint 17.3 with I believe kernel 3.19 and I'm trying to use a SoundBlaster Recon3D Fatal1ty.

[jagged_jimmy_j (jaggedjimmyj-linux-kernel-bugs)]

Heyyo Michael R,

No need for fancy patching as it is part of kernel 4.4 meow. Just find a guide online on how to install kernel 4.4 for Linux Mint and you will be set. :)

[Michal (fishmanos79)]

Ubuntu 15.10
kernel 4.2.0-34-generic

Gigabyte GA-X99-Gaming G1
Creative Recon3d (CA0132)

aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: PCH [HDA Intel PCH], device 0: CA0132 Analog [CA0132 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 1: CA0132 Digital [CA0132 Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: NVidia [HDA NVidia], device 3: HDMI 0 [HDMI 0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: NVidia [HDA NVidia], device 7: HDMI 1 [HDMI 1]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: NVidia [HDA NVidia], device 8: HDMI 2 [HDMI 2]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: NVidia [HDA NVidia], device 9: HDMI 3 [HDMI 3]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

No sound, no analog output in system settings (see attached)
Alsamixes shows no mute/unmute possibility (see attached)

[Michal (fishmanos79)]

Sorry, couldn't find a way to add another attachment to previous post.

[Mephi (mephi-mephi)]

I'm another one.

Ubuntu 16.04
Kernel 4.4.0-15-generic

Gigabyte Sniper m5
Creative CA0132

mephi@mephi-desktop:~$ dmesg | grep audio
[ 1.459862] snd_hda_intel 0000:01:00.1: Handle vga_switcheroo audio client
[ 1.474284] snd_hda_codec_ca0132 hdaudioC1D2: autoconfig for CA0132: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 1.474286] snd_hda_codec_ca0132 hdaudioC1D2: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 1.474287] snd_hda_codec_ca0132 hdaudioC1D2: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 1.474288] snd_hda_codec_ca0132 hdaudioC1D2: mono: mono_out=0x0
[ 1.474289] snd_hda_codec_ca0132 hdaudioC1D2: dig-out=0xc/0x0
[ 1.474290] snd_hda_codec_ca0132 hdaudioC1D2: inputs:
[ 1.474291] snd_hda_codec_ca0132 hdaudioC1D2: Mic=0x12
[ 1.474292] snd_hda_codec_ca0132 hdaudioC1D2: Line=0x11
[ 2.089162] snd_hda_codec_ca0132 hdaudioC1D2: ca0132 DSP downloaded and running
[ 2.101041] snd_hda_intel 0000:00:03.0: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
[ 36.721197] snd_hda_codec_ca0132 hdaudioC1D2: ca0132 DSP downloaded and running
[ 37.580974] snd_hda_codec_hdmi hdaudioC2D0: HDMI: invalid ELD data byte 4
[ 57.877294] snd_hda_codec_hdmi hdaudioC2D0: HDMI: invalid ELD data byte 50
mephi@mephi-desktop:~$ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: HDMI [HDA Intel HDMI], device 3: HDMI 0 [HDMI 0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: HDMI [HDA Intel HDMI], device 7: HDMI 1 [HDMI 1]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: HDMI [HDA Intel HDMI], device 8: HDMI 2 [HDMI 2]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: PCH [HDA Intel PCH], device 0: CA0132 Analog [CA0132 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: PCH [HDA Intel PCH], device 1: CA0132 Digital [CA0132 Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 2: NVidia [HDA NVidia], device 3: HDMI 0 [HDMI 0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 2: NVidia [HDA NVidia], device 7: HDMI 1 [HDMI 1]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 2: NVidia [HDA NVidia], device 8: HDMI 2 [HDMI 2]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 2: NVidia [HDA NVidia], device 9: HDMI 3 [HDMI 3]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

No sound, no options to use the Creative card in the Sound Settings. Kernel seems to find it (according to dmesg). My alsamixer output is the same as Michal's (above)

[Mephi (mephi-mephi)]

I don't know if I'm getting any closer with this, but if I run:
sudo apt-get install alsa-tools-gui
hdajackretask

Then I can select Creative CA0132 and override the Green and Orange headphone ports (And set them to headphone)
I have to install boot override and then reboot.

Then I get the headphones as an option in the Ubuntu sound settings.

Still no sound though...

[Mephi (mephi-mephi)]

I have sound!!!

The line out port now works. I'm not completely sure which of the changes caused it, but the line out port wasn't working to start with.

It'd be nice to get the headphone output working but this is a good start :-)

[Islam Alshaikh إسلام الشيخ (islam-alshaikh)]

Hello,

Sound is not working in 16.04 kubuntu

Motherbord is MSI

Creative SoundBlaster zx
ca0132

dmesg | grep audio
[ 2.540685] snd_hda_intel 0000:00:03.0: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
[ 2.553922] snd_hda_codec_realtek hdaudioC1D0: autoconfig for ALC887-VD: line_outs=1 (0x14/0x0/0x0/0x0/0x0) type:line
[ 2.553925] snd_hda_codec_realtek hdaudioC1D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 2.553926] snd_hda_codec_realtek hdaudioC1D0: hp_outs=1 (0x1b/0x0/0x0/0x0/0x0)
[ 2.553927] snd_hda_codec_realtek hdaudioC1D0: mono: mono_out=0x0
[ 2.553928] snd_hda_codec_realtek hdaudioC1D0: inputs:
[ 2.553930] snd_hda_codec_realtek hdaudioC1D0: Front Mic=0x19
[ 2.553932] snd_hda_codec_realtek hdaudioC1D0: Rear Mic=0x18
[ 2.553933] snd_hda_codec_realtek hdaudioC1D0: Line=0x1a
[ 2.554789] snd_hda_codec_ca0132 hdaudioC2D1: autoconfig for CA0132: line_outs=1 (0xd/0x0/0x0/0x0/0x0) type:line
[ 2.554792] snd_hda_codec_ca0132 hdaudioC2D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 2.554794] snd_hda_codec_ca0132 hdaudioC2D1: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 2.554796] snd_hda_codec_ca0132 hdaudioC2D1: mono: mono_out=0x0
[ 2.554797] snd_hda_codec_ca0132 hdaudioC2D1: dig-out=0xc/0x0
[ 2.554799] snd_hda_codec_ca0132 hdaudioC2D1: inputs:
[ 2.554801] snd_hda_codec_ca0132 hdaudioC2D1: Mic=0x12
[ 2.554802] snd_hda_codec_ca0132 hdaudioC2D1: dig-in=0xe
[ 3.123032] snd_hda_codec_ca0132 hdaudioC2D1: ca0132 DSP downloaded and running
[ 4.259145] snd_hda_codec_ca0132 hdaudioC2D1: ca0132 DSP downloaded and running
[ 4.733580] snd_hda_intel 0000:00:03.0: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915])
[ 4.742177] snd_hda_codec_realtek hdaudioC1D0: autoconfig for ALC887-VD: line_outs=1 (0x14/0x0/0x0/0x0/0x0) type:line
[ 4.742179] snd_hda_codec_realtek hdaudioC1D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 4.742180] snd_hda_codec_realtek hdaudioC1D0: hp_outs=1 (0x1b/0x0/0x0/0x0/0x0)
[ 4.742181] snd_hda_codec_realtek hdaudioC1D0: mono: mono_out=0x0
[ 4.742181] snd_hda_codec_realtek hdaudioC1D0: inputs:
[ 4.742183] snd_hda_codec_realtek hdaudioC1D0: Front Mic=0x19
[ 4.742184] snd_hda_codec_realtek hdaudioC1D0: Rear Mic=0x18
[ 4.742184] snd_hda_codec_realtek hdaudioC1D0: Line=0x1a
[ 4.750465] snd_hda_codec_generic hdaudioC0D0: ignore pin 0x7, too many assigned pins
[ 4.750470] snd_hda_codec_generic hdaudioC0D0: autoconfig for Generic: line_outs=0 (0x0/0x0/0x0/0x0/0x0) type:line
[ 4.750471] snd_hda_codec_generic hdaudioC0D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 4.750472] snd_hda_codec_generic hdaudioC0D0: hp_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 4.750473] snd_hda_codec_generic hdaudioC0D0: mono: mono_out=0x0
[ 4.750474] snd_hda_codec_generic hdaudioC0D0: dig-out=0x5/0x6
[ 4.750475] snd_hda_codec_generic hdaudioC0D0: inputs:
[ 4.755579] snd_hda_codec_ca0132 hdaudioC2D1: autoconfig for CA0132: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 4.755582] snd_hda_codec_ca0132 hdaudioC2D1: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 4.755583] snd_hda_codec_ca0...

[Islam Alshaikh إسلام الشيخ (islam-alshaikh)]

It is 5.1

but detected as 2.1 without any sound

[Egbert van der Wal (eggie)]

I have the same issue: no sound on a Recon3D / CA0132. The sound chip is an onboard chip labelled Sound Core 3D, the motherboard is a GA-X170-Extreme ECC. This is on Ubuntu 16.04.

Alsa / Pulseaudio claim everything is good, but there's no sound coming out of it at all.

dmesg | grep hda shows:

[ 7.200088] snd_hda_intel 0000:00:1f.3: enabling device (0000 -> 0002)
[ 7.200197] snd_hda_intel 0000:01:00.1: enabling device (0000 -> 0002)
[ 7.200226] snd_hda_intel 0000:01:00.1: Handle vga_switcheroo audio client
[ 7.241033] snd_hda_intel 0000:00:1f.3: failed to add i915_bpo component master (-19)
[ 7.252621] snd_hda_codec_ca0132 hdaudioC0D0: autoconfig for CA0132: line_outs=1 (0xb/0x0/0x0/0x0/0x0) type:line
[ 7.252623] snd_hda_codec_ca0132 hdaudioC0D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 7.252625] snd_hda_codec_ca0132 hdaudioC0D0: hp_outs=1 (0x10/0x0/0x0/0x0/0x0)
[ 7.252626] snd_hda_codec_ca0132 hdaudioC0D0: mono: mono_out=0x0
[ 7.252627] snd_hda_codec_ca0132 hdaudioC0D0: dig-out=0xc/0x0
[ 7.252628] snd_hda_codec_ca0132 hdaudioC0D0: inputs:
[ 7.252629] snd_hda_codec_ca0132 hdaudioC0D0: Mic=0x12
[ 7.252630] snd_hda_codec_ca0132 hdaudioC0D0: Line=0x11
[ 7.503386] snd_hda_codec_generic hdaudioC1D0: ignore pin 0x6, too many assigned pins
[ 7.511395] snd_hda_codec_generic hdaudioC1D0: ignore pin 0x7, too many assigned pins
[ 7.511398] snd_hda_codec_generic hdaudioC1D0: autoconfig for Generic: line_outs=0 (0x0/0x0/0x0/0x0/0x0) type:line
[ 7.511399] snd_hda_codec_generic hdaudioC1D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 7.511400] snd_hda_codec_generic hdaudioC1D0: hp_outs=0 (0x0/0x0/0x0/0x0/0x0)
[ 7.511400] snd_hda_codec_generic hdaudioC1D0: mono: mono_out=0x0
[ 7.511401] snd_hda_codec_generic hdaudioC1D0: dig-out=0x4/0x5
[ 7.511402] snd_hda_codec_generic hdaudioC1D0: inputs:
[ 7.847710] snd_hda_codec_ca0132 hdaudioC0D0: ca0132 DSP downloaded and running
[ 8.711805] snd_hda_codec_ca0132 hdaudioC0D0: ca0132 DSP downloaded and running

I tried blacklisting snd-hda-codec-hdmi, which didn't work so I ended up removing the file from lib/modules/*/ altogether. Didn't help.

I also tried adding enable_msi=1 and/or position_fix=1 to /etc/modprobe.d/alsa-base.conf, neither of them worked. Even though the dmesg now no longer states that MSI is being disabled for snd_hda_intel.

Maybe relevant: the motherboard has a HDMI-out (which doesn't work because I don't have a iGPU on my processor), and the videocard (Gigabyte NVidia GTX 1070 G1 Gaming) also has a HDMI-output which is also bound to snd-hda-intel. (Relevant) output of lspci -k:

00:1f.3 Audio device: Intel Corporation Sunrise Point-H HD Audio (rev 31)
        Subsystem: Gigabyte Technology Co., Ltd Sunrise Point-H HD Audio
        Kernel driver in use: snd_hda_intel
        Kernel modules: snd_hda_intel

01:00.0 VGA compatible controller: NVIDIA Corporation Device 1b81 (rev a1)
        Subsystem: Gigabyte Technology Co., Ltd Device 3701
        Kernel driver in use: nvidia
        Kernel modules: nvidiafb, nouveau, nvidia_367, nvidia_367_drm
01:00.1 Audio device: NVIDIA Corporation Device 10f0 (re...

[LaurentP (l5d)]

Hello, here the return of the command hdajacksensetest :

sudo hdajacksensetest -c 3 -d 1 -a
Pin 0x0b (Green Line Out, Rear side): present = Yes
Pin 0x0c (Purple SPDIF Out, Rear side): present = No
Pin 0x0d (Yellow SPDIF Out, Rear side): present = No
Pin 0x0e (Blue SPDIF In, Rear side): present = No
Pin 0x0f (Green Headphone, Front side): present = No
Pin 0x10 (Orange Headphone, Front side): present = No
Pin 0x11 (Grey Line Out, Front side): present = Yes
Pin 0x12 (Pink Mic, Mobile-In): present = Yes
Pin 0x13 (Internal Line In): present = No
Pin 0x18 (Not connected): present = No

and return of inxi -A
Audio: Card-1 Creative Labs SB Recon3D driver: snd_hda_intel Sound: ALSA v: k4.4.0-15-generic
           Card-2 NVIDIA Device 0fbc driver: snd_hda_intel
           Card-3 Advanced Micro Devices [AMD/ATI] SBx00 Azalia (Intel HDA) driver: snd_hda_intel
           Card-4 Guillemot driver: USB Audio

[Mephi (mephi-mephi)]

Just adding in that there's no change with an upgrade to 16.10
Also, I'm not sure if the link to the Linux Kernel bug is correct as that shows the bug as resolved in Kernels <=3.9-r3

[Marco (marko81)]

Still not working on 17.04 Zesty (kernel 4.8)

[Sean O'Donnell (seano96)]

This is affecting me.

Decided to give Ubuntu 16.04 a try and I cannot for the life of me get Optical or the green, black and orange/brown cables working either (The 3.5mm jacks).

Audio works in windows fine, I did the echo "options snd-hda-intel position_fix=1" | sudo tee -a /etc/modprobe.d/alsa-base.conf fix and it worked for a day, when I restarted from windows to Linux it broke again.

Everything shows up. It should be working but sound does not play at all.

HDMI Audio works fine via my monitor.

Motherboard: Gigabyte Gaming 7 Z170
Card: HDA Intel PCH
Chip: Creative CA0132

If I can do anything to help please let me know. I can't switch over from windows if I don't have audio.

[Egbert van der Wal (eggie)]

I just gave up and got me a Asus Xonar sound card and disabled the onboard. Works a treat!

The longer I worked with it, the more I got convinced it's a crappy chip. Even in Windows 10 it doesn't work properly - sometimes the volume levels are ignored or attached to different outputs than they should be.

Also, when I started Ubuntu 16.04 once, I needed to completely shut down the computer - if I just reboot to Windows it will also not work in Windows. After a shut down it usually works again.

I have no clue what this chip is doing and how it's interacting with the driver on Linux, but it's bad. Very bad.

[dakota pa lmer (dakotapa)]

SB Recon3D Line Out does not work.

I have a sound blaster zx running ubuntu 16.04 with kernerl 4.8.0 -46

My HDMI audio is working. I just plugged headphone in to the back of one of my monitors running and HDMI cord and sound is perfect...

Would love to have my sound card working though....

[Scooty Puff Jr. (sevenzig)]

Okay, I did a whole bunch of things tonight and I think I got mine working.

Added " options snd-hda-intel model=auto " to /etc/modprobe.d/alsa-base.conf
(https://forums.linuxmint.com/viewtopic.php?t=110209)

Added " sudo apt-get install dkms " and downloading the most recent ALSA daily for 16.04
(https://wiki.ubuntu.com/Audio/UpgradingAlsa/DKMS)

Added " rmmod snd_hda_intel " and " modprobe snd_hda_intel position_fix=1 " to /etc/rc.local
(https://github.com/ValveSoftware/SteamOS/issues/221)

Executed " sudo apt-get install --install-recommends linux-generic-lts-wily xserver-xorg-core-lts-wily xserver-xorg-lts-wily xserver-xorg-video-all-lts-wily xserver-xorg-input-all-lts-wily libwayland-egl1-mesa-lts-wily "
(https://ubuntuforums.org/showthread.php?t=2293912)

Installed hdajackretask [per mephi, #81 in this thread], selected Creative CA0132, and toggled Advanced override. Override Green Headphone, Front side. Set Device to Headphone and Channel (in group) to Front.

Also the simple stuff like making sure your desired audio output device is set to default and not muted. Make sure to run alsamixer after each reboot because I find my output muted after every restart.

Hope this helps someone!

[aidanjt (aidanjt)]

I'm having the same problem with PCI ID 1102:0012 & sub ID 1102:0010. It's a SB Z OEM (no fancy shroud).

[Zaszuś (zaszus)]

@sevenzig: Did you add the option " snd-hda-intel position_fix=1 " as well?

[Zaszuś (zaszus)]

@sevenzig: I have followed your advice, but unless I messed something up in or before the process, it didn't help. Nothing changed...

[Zaszuś (zaszus)]

Launchpad says alsa package was updated for the last time on 2013-10-20. Is it true that Linux/Ubuntu does not have any sound development for 4 years??

[Michael Murphy (mmstick)]

ALSA does not exist on it's own within a package. It's a part of the Linux kernel (because Linux is a macrokernel).

[Zaszuś (zaszus)]

Right, thanks Michael. I missed that part.

[LaurentP (l5d)]

hello,
i have got sound on Line out, rear jack, by applied the proccess documented in this forum : [url]https://ubuntuforums.org/showthread.php?t=2326126&p=13496676#post13496676[/url].

[ijml.team (ijml.team-linux-kernel-bugs)]

Gigabyte Z87 G1.Sniper M5 Creative CA0132 no surround sound (only stereo)

00:1b.0 Audio device [0403]: Intel Corporation 8 Series/C220 Series Chipset High Definition Audio Controller [8086:8c20] (rev 04)

Card: HDA Intel PCH
Chip: Creative CA0132

https://i.pinimg.com/564x/ae/64/9f/ae649f33532f1dc598e66c6936fe1353.jpg

[ijml.team (ijml.team-linux-kernel-bugs)]

Kernel 4.9.0-040900-generic
ubuntu 16.04 LTS

[Israel Jone (ijml-team)]

Gigabyte Z87 G1.Sniper M5 Creative CA0132 no surround 5.1 sound (only stereo) any idea?

ubuntu 16.04 LTS Kernel 4.9.0-040900-generic

00:1b.0 Audio device [0403]: Intel Corporation 8 Series/C220 Series Chipset High Definition Audio Controller [8086:8c20] (rev 04)

Card: HDA Intel PCH
Chip: Creative CA0132

https://i.pinimg.com/564x/ae/64/9f/ae649f33532f1dc598e66c6936fe1353.jpg

aplay -L
default
    Playback/recording through the PulseAudio sound server
null
    Discard all samples (playback) or generate zero samples (capture)
pulse
    PulseAudio Sound Server
hdmi:CARD=HDMI,DEV=0
    HDA Intel HDMI, HDMI 0
    HDMI Audio Output
hdmi:CARD=HDMI,DEV=1
    HDA Intel HDMI, HDMI 1
    HDMI Audio Output
hdmi:CARD=HDMI,DEV=2
    HDA Intel HDMI, HDMI 2
    HDMI Audio Output
dmix:CARD=HDMI,DEV=3
    HDA Intel HDMI, HDMI 0
    Direct sample mixing device
dmix:CARD=HDMI,DEV=7
    HDA Intel HDMI, HDMI 1
    Direct sample mixing device
dmix:CARD=HDMI,DEV=8
    HDA Intel HDMI, HDMI 2
    Direct sample mixing device
dsnoop:CARD=HDMI,DEV=3
    HDA Intel HDMI, HDMI 0
    Direct sample snooping device
dsnoop:CARD=HDMI,DEV=7
    HDA Intel HDMI, HDMI 1
    Direct sample snooping device
dsnoop:CARD=HDMI,DEV=8
    HDA Intel HDMI, HDMI 2
    Direct sample snooping device
hw:CARD=HDMI,DEV=3
    HDA Intel HDMI, HDMI 0
    Direct hardware device without any conversions
hw:CARD=HDMI,DEV=7
    HDA Intel HDMI, HDMI 1
    Direct hardware device without any conversions
hw:CARD=HDMI,DEV=8
    HDA Intel HDMI, HDMI 2
    Direct hardware device without any conversions
plughw:CARD=HDMI,DEV=3
    HDA Intel HDMI, HDMI 0
    Hardware device with all software conversions
plughw:CARD=HDMI,DEV=7
    HDA Intel HDMI, HDMI 1
    Hardware device with all software conversions
plughw:CARD=HDMI,DEV=8
    HDA Intel HDMI, HDMI 2
    Hardware device with all software conversions
sysdefault:CARD=PCH
    HDA Intel PCH, CA0132 Analog
    Default Audio Device
front:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    Front speakers
surround21:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    2.1 Surround output to Front and Subwoofer speakers
surround40:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    4.0 Surround output to Front and Rear speakers
surround41:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    4.1 Surround output to Front, Rear and Subwoofer speakers
surround50:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    5.0 Surround output to Front, Center and Rear speakers
surround51:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    5.1 Surround output to Front, Center, Rear and Subwoofer speakers
surround71:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    7.1 Surround output to Front, Center, Side, Rear and Woofer speakers
iec958:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Digital
    IEC958 (S/PDIF) Digital Audio Output
dmix:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    Direct sample mixing device
dmix:CARD=PCH,DEV=1
    HDA Intel PCH, CA0132 Digital
    Direct sample mixing device
dsnoop:CARD=PCH,DEV=0
    HDA Intel PCH, CA0132 Analog
    Direct sample snooping device
dsnoop:CARD=PCH,DEV=1
    HDA Intel PCH, CA0132 Digital
    Direct sample snooping device
hw:CARD=PCH,DEV=...

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Just found this thread, hopefully I'm not breaking any rules by posting on a few different ones. I've got full functionality on my Sound Blaster Z working now, and have the firmware and patch file uploaded here:
https://drive.google.com/drive/folders/1r-X8g6It5AO7SwgJc1JQfJ8KCOyqrr_d
You'll have to compile your own kernel and put the ctefx-sbz.bin firmware in your /lib/firmware folder. It's currently only tested on my Sound Blaster Z, which has an implementation ID of 0023. You can get this by sending the hda-verb 0xF20 to node 0x01. Hopefully it works for someone else.

[ijml.team (ijml.team-linux-kernel-bugs)]

(In reply to Connor M from comment #196)
> Just found this thread, hopefully I'm not breaking any rules by posting on a
> few different ones. I've got full functionality on my Sound Blaster Z
> working now, and have the firmware and patch file uploaded here:
> https://drive.google.com/drive/folders/1r-X8g6It5AO7SwgJc1JQfJ8KCOyqrr_d
> You'll have to compile your own kernel and put the ctefx-sbz.bin firmware in
> your /lib/firmware folder. It's currently only tested on my Sound Blaster Z,
> which has an implementation ID of 0023. You can get this by sending the
> hda-verb 0xF20 to node 0x01. Hopefully it works for someone else.

Surround output work on Sound Blaster Z??

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Yes, but the channel mapping is incorrect if you're using pulseaudio. If you use ALSA, it's not a problem. I'm not sure why this is, but I'll have to look into how Pulse detects ALSA's channel mapping. If you want to use surround, make sure to go into alsamixer, deselect 'HP/Speaker Auto Detect' and on 'Output Select' put 'Surround'. I just tested it myself to make sure. If you want to use it with pulse for now, you just have to swap the rear jacks. I'm not sure how long it'll take me to figure out what Pulse does or doesn't do properly.

[Conmanx360 (conmanx360-linux-kernel-bugs)]

You can fix the pulseaudio mapping issue by editing the file /usr/share/pulseaudio/alsa-mixer/profile-sets/default.conf, and setting the option:

[Mapping analog-surround-51]
device-strings = surround51:%f
channel-map = front-left,front-right,front-center,lfe,rear-left,rear-right
paths-output = analog-output analog-output-lineout analog-output-speaker
priority = 8
direction = output

Changing the channel-map to what I have above. By default, it's front-left,front-right,rear-left,rear-right,front-center,lfe. The order of pins on the SBZ is funky, 0x10 is center/lfe and 0x11 is rear surround. Most of the time, it's rear surround then center/lfe.

[ijml.team (ijml.team-linux-kernel-bugs)]

Okay, we now have all inputs and outputs working on the Recon3Di (Recon3D Integrated, seems to be on Gigabyte motherboards primarily) if anyone wants to try it, it's in the google drive link. Use the regular patch_ca0132.c and put the ctefx-r3di.bin in your /lib/firmware. We haven't tested output effects, but that shouldn't be much more difficult.

Like I said before, still looking for people with other Sound Blaster Z variants to help test, and people with the Recon3D non integrated. We can probably get all of them working on Linux, which would be nice, because they haven't worked for years.

could you do a tutorial on how to put these files in the kernel and compile?

I've tried but I do not know if I did everything right.

[Conmanx360 (conmanx360-linux-kernel-bugs)]

All you need to do is, before compiling the kernel, in the kernel source, replace the patch_ca0132.c file in /sound/pci/hda, and the in your /lib/firmware, not the kernel sources /lib/, add the ctefx-sbz.bin or ctefx-r3di.bin, whichever one you use.

I'm trying to speed up getting everything finished so I can submit it and try to get it approved. It's getting hard to do tech support at the same time as working on this.

[stefan-oltmanns (stefan-oltmanns-linux-kernel-bugs)]

Thank you very much for your great effort. I've got a Gigabyte G1.Sniper M3 with Recon3Di, I added one line to load the Recon3Di driver:
SND_PCI_QUIRK(0x1458, 0xA016, "Recon3Di", QUIRK_R3DI),

Most things seems to work, but one annoying bug: In surround mode the center/lfe signal is only at the front panel. In headphone mode the signal is at the frontpanel/back panel headphone connector. The center/lfe connector has no signal ever. I assume a different GPIO is used, but I have no idea how many GPIOs are available and for what other functions they are used. Any idea/help?

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Looking into it a little bit, having never messed with one with your ID, here's what is in the Windows driver inf:

[DriverSettings.AddReg]
HKR,DeviceSpecificOverwrite,SupportedMultiplexInputs, 0x00010001,0x00000005
HKR,DeviceSpecificOverwrite,SupportedMultiplexOutputs, 0x00010001,0x00000003
HKR,DeviceSpecificOverwrite,SupportDriveBay, 0x00010001,0x00000000
HKR,DeviceSpecificOverwrite,MPIO_GB_RearMicLine_FPMic, 0x00010001,0x00000001
HKR,DeviceSpecificOverwrite,MPIO_GB_FPHP_CtrLFE, 0x00010001,0x00000003
HKR,DeviceSpecificOverwrite,MPIO_GB_FPHP_CtrLFE_Mute, 0x00010001,0x00000004
HKR,DeviceSpecificOverwrite,MPIO_GB_HPAMP_Shutdown, 0x00010001,0x00000005
HKR,DeviceSpecificOverwrite,LineOutMaxVolume, 0x00010001,0x00000000

There is specific mention of "MPIO_GB_FPHP_CtrLFE" and then "MPIO_GB_FPHP_CtrLFE_Mute". Now, I don't exactly know what the numbers mean, but my guess is it may be the bitmask value for the GPIO data verb. So, 0x03 means pins 0 and 1 are on, which I'm pretty sure is what I have by default. Maybe check that 0x11 is set as output, you can do that by sending:

hda-verb /dev/snd/hwC*D* 0x11 0x707 0x44 .

That is pin 0x11, which is actually supposed to be surround, but that's the only one I've known to have problems. You could do the same to 0x10 as well, or just download hda-analyzer and see if it's set as out.

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Oh, just thought of this. Make sure the HP/Speaker Auto detect is off, then switch to surround using the "Output Source" control in alsamixer. If auto-detect is on, it'll never switch to surround, because it only auto-detects headphone and lineout.

[stefan-oltmanns (stefan-oltmanns-linux-kernel-bugs)]

Hi, thanks for your fast answer. I found out that reloading the module is not enough. After a restart it worked, probably after loading the stereo driver it was not completely resetted by loading the new driver.
I had auto-detect disabled. I'm not sure if that's normal, but with "OutFX" enabled it downmixes to stereo. Please add the ID to your patch, so my mainboard is supported out-of-the-box after your patch is committed to the kernel.

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Can do. I'll look into the OutFX downmixing to stereo, I haven't messed with that. Thanks for bringing it to my attention.
Full shutdowns are good with the ca0132 based chips, because their memory is kept upon reboots. They need a full power down to clear their internal memory.

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Also, just in case anyone is reading this, my GPIO assumptions above were incorrect. The numbers are the GPIO pin numbers, not the commands themselves. So 0x03 is FPHP_CTRLFE, 0x04 is CTRLFE_Mute, etc. These are set with verb 0x715, GPIO data.

[milkman6453 (milkman6453-linux-kernel-bugs)]

Hey connor, you mentioned your devid and the fixes are for that specific one.

my card is from a gigabyte g1 sniper 5

00:1b.0 0403: 8086:8c20 (rev 04)
 Subsystem: 1458:a026

Is there still no available fixes for this? Will your patch only work with your chip?

[Conmanx360 (conmanx360-linux-kernel-bugs)]

I have a few of the onboard gigabyte cards already included. You can test it. Add to the ca0132_quirks[] table this:
        SND_PCI_QUIRK(0x1458, 0xA026, "Recon3Di", QUIRK_R3DI),

And test it for yourself. I don't see why it shouldn't work.

[milkman6453 (milkman6453-linux-kernel-bugs)]

Will do. This would be under "compiling a kernel" just asking as I will be googling how to :)

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Yeah, just add that to the file 'patch_ca0132.c' which you should download from my google drive link, then put it into your kernel source. It'll be in the folder 'sound/pci/hda'. After that, you should be good to compile.

[milkman6453 (milkman6453-linux-kernel-bugs)]

Im having a hard time compiling a ubuntu-bionic kernel. Most guides i see are for much older kernels

[stefan-oltmanns (stefan-oltmanns-linux-kernel-bugs)]

For Ubuntu you do not need to compile the entire kernel (I did that at the beginning, on my system it took hours). It is possible to download the kernel source in ubuntu, copy the build settings of the currently running kernel and compile just that one directory and afterwards replacing that modules.
I pretty much follwed this guide: https://askubuntu.com/questions/515407/how-recipe-to-build-only-one-kernel-module
Note: This guide is for another kernel version at it's example is a modified SCSI driver, so you need to adapt it, but the steps are the same (I've done it on 18.04). Remember to run the make command inside the directory where the patch_ca0132.c file is, as you only want to build that directory. I did not create a special directory for the modified modules, I just replaced the existing ones.
Important issue: The CA0132 seems to remember stuff even after reboots as Connor said, so after building/installing the kernel modules shutdown the system and switch power supply off for one minute, than you should be safe. That was my mistake as I could not get the driver working, after that it works fine.

[milkman6453 (milkman6453-linux-kernel-bugs)]

I have succesfully compiled the 4.16.12 kernel with the patch and added the line of text stated.(which is identical to one already in the patch file)

My alsamixer does not have the new expected settings. No surround. I am testing optical out and neither the tolink or the coax spdif work.

[milkman6453 (milkman6453-linux-kernel-bugs)]

(In reply to Mike Milk from comment #214)
> I have succesfully compiled the 4.16.12 kernel with the patch and added the
> line of text stated.(which is identical to one already in the patch file)
>
> My alsamixer does not have the new expected settings. No surround. I am
> testing optical out and neither the tolink or the coax spdif work.

toslink

[Conmanx360 (conmanx360-linux-kernel-bugs)]

Hm, could you link your dmesg output? Did you put the new module in?

[milkman6453 (milkman6453-linux-kernel-bugs)]

Yes both module files are in /lib/firmware/

https://justpaste.it/2uo7i

[Conmanx360 (conmanx360-linux-kernel-bugs)]

It's not showing that it loaded with the quirk. I'm talking about putting the module into your /lib/modules/*kernel-version*/kernel/sound/pci/hda. The module is snd-hda-codec-ca0132.ko . You'll need to copy the compiled module to that folder.

[milkman6453 (milkman6453-linux-kernel-bugs)]

The file was already there. I did a full kernel compile with the patch. I assume it brought over the new module. If not i did move over the .ko module in my compiled folder. I will restart and dmesg again

[milkman6453 (milkman6453-linux-kernel-bugs)]

Rewriting the file with the one in the compiled folder did not change anything at alsa or sound output

https://justpaste.it/4563m

[milkman6453 (milkman6453-linux-kernel-bugs)]

The patch works well for 2 channel analog output. I am still unable to get 5.1 output.

[LaurentP (l5d)]

Hello : My latest tests for this sound card :

With Linux Mint LMDE4 64 (live session no config, no patch) the sound is ok 2.0
Kernel = 4.19.0-8-amd64 x86-64

With Linux Mint 19.3 Cinnamon 64 (live session no config, no patch) the sound is ok 5.1
Kernel = 5.0.0--32generic x86-64

With Linux Mint 20 Cinnamon 64 (live session) it is also OK.

[sunniegoldie (sunniegoldie-linux-kernel-bugs)]

It appears that the issue with your Creative Soundblaster Z soundcard is related to the DSP (Digital Signal Processor) not initializing correctly. Despite having the correct firmware and configuration enabled, the error message "ca0132 DOWNLOAD FAILED!!! DSP IS NOT RUNNING" suggests that the system is unable to load or start the DSP. To troubleshoot this, first, ensure that the firmware files (`ctefx.bin` and `ctspeq.bin`) are correctly placed and not corrupted. Additionally, check if the kernel version you're using (3.9-r3) might have compatibility issues with the soundcard, and consider testing with a stable or different kernel version. Verify your configuration settings in the `.config` file for any conflicts and ensure that your kernel modules are up-to-date and properly built. Consulting the documentation for your soundcard and kernel, as well as reaching out to relevant support forums, may also provide further assistance. Utilizing ERP software can aid in managing IT infrastructure by tracking hardware issues, handling firmware updates, and ensuring system compatibility to prevent similar problems in the future. More about ERP: https://www.impactfirst.co/id/c/software-erp

[gulabmalik118 (gulabmalik118-linux-kernel-bugs)]

Fantastic work right here on. I review a great deal of article, however I never ever heard a subject such as this. I Love this topic you made concerning the blogger’s pail listing.
https://housesitmatch.pro/increasing-natural-traffic/