[alsa-devel] No sound with Realtek ALC298 on a Samsung Galaxy book
Dear ALSA community,
I just bought a Samsung Galaxy book 12" laptop and I installed the latest Fedora version (31) on it. Unfortunately when I try to listen something there is absolutely no sound from speaker nor headsets. I checked that volumes of the different channels were max. I also checked for muting system but I didn't found anything. Of course it works perfectly under Windows. I saw on internet that many ALC298 users had troubles with it. I tried different options to the intel hda driver (especially the model one) but it changes nothing.
Here is the alsa-info.sh output: http://alsa-project.org/db/? f=871881e295972b9ecf252b25e90d659e38d939b8 I would appreciate some help in order to find a solution. Tell me if you need more information. Thanks in advance.
Aurélien
On Sat, 15 Feb 2020 17:56:58 +0100, Aurélien Croc wrote:
Dear ALSA community,
I just bought a Samsung Galaxy book 12" laptop and I installed the latest Fedora version (31) on it. Unfortunately when I try to listen something there is absolutely no sound from speaker nor headsets. I checked that volumes of the different channels were max. I also checked for muting system but I didn't found anything. Of course it works perfectly under Windows. I saw on internet that many ALC298 users had troubles with it. I tried different options to the intel hda driver (especially the model one) but it changes nothing.
Here is the alsa-info.sh output: http://alsa-project.org/db/? f=871881e295972b9ecf252b25e90d659e38d939b8 I would appreciate some help in order to find a solution. Tell me if you need more information. Thanks in advance.
The BIOS seems broken, it just sets a single output pin. You need to try to figure out the pin connections, e.g. via hdajackretask program. The headphone might work with that. But the speaker output might need another special handling, and it's nothing but trial-and-test with the existing quirks for similar devices. Take a look at sound/pci/hda/patch_realtek.c.
Takashi
The BIOS seems broken, it just sets a single output pin. You need to try to figure out the pin connections, e.g. via hdajackretask program. The headphone might work with that. But the speaker output might need another special handling, and it's nothing but trial-and-test with the existing quirks for similar devices. Take a look at sound/pci/hda/patch_realtek.c.
Thank you for your answer. I tried hdajackretask but I didn't succeed to get sound from headsets.. Maybe I didn't tried all the advanced features of this tool.. I also tried some quirks from your file but it changes nothing. There are too many possibilities. It seems to be hard to test them all.
Don't you know a solution to get what is bugged or to get information about which channel is linked to headets or internal speakers from Windows drivers? I think about a tool to sniff communication between the driver and the sound card? or read information from the sound card under windows? Reverse engineering the Windows driver would lead to the solution but will be very long for me since I'm not familiar with the Intel HDA card and how this Realtek chip internally works. Don't hesitate if you have any hints which could help me (and all the other users of this laptop..) Cheers,
Aurélien
Le lundi 17 février 2020, 09:00:42 CET Takashi Iwai a écrit :
On Sat, 15 Feb 2020 17:56:58 +0100,
Aurélien Croc wrote:
Dear ALSA community,
I just bought a Samsung Galaxy book 12" laptop and I installed the latest Fedora version (31) on it. Unfortunately when I try to listen something there is absolutely no sound from speaker nor headsets. I checked that volumes of the different channels were max. I also checked for muting system but I didn't found anything. Of course it works perfectly under Windows. I saw on internet that many ALC298 users had troubles with it. I tried different options to the intel hda driver (especially the model one) but it changes nothing.
Here is the alsa-info.sh output: http://alsa-project.org/db/? f=871881e295972b9ecf252b25e90d659e38d939b8 I would appreciate some help in order to find a solution. Tell me if you need more information. Thanks in advance.
The BIOS seems broken, it just sets a single output pin. You need to try to figure out the pin connections, e.g. via hdajackretask program. The headphone might work with that. But the speaker output might need another special handling, and it's nothing but trial-and-test with the existing quirks for similar devices. Take a look at sound/pci/hda/patch_realtek.c.
Takashi
On Mon, 17 Feb 2020 15:03:16 +0100, Aurélien Croc wrote:
The BIOS seems broken, it just sets a single output pin. You need to try to figure out the pin connections, e.g. via hdajackretask program. The headphone might work with that. But the speaker output might need another special handling, and it's nothing but trial-and-test with the existing quirks for similar devices. Take a look at sound/pci/hda/patch_realtek.c.
Thank you for your answer. I tried hdajackretask but I didn't succeed to get sound from headsets.. Maybe I didn't tried all the advanced features of this tool.. I also tried some quirks from your file but it changes nothing. There are too many possibilities. It seems to be hard to test them all.
Yeah, that's the problem -- there are way too many device-specific implementations and workarounds :-<
Don't you know a solution to get what is bugged or to get information about which channel is linked to headets or internal speakers from Windows drivers? I think about a tool to sniff communication between the driver and the sound card? or read information from the sound card under windows?
Sometimes you have some data in *.INF file or such. At least, if the standard Windows driver (not the h/w vendor's one) works, the extra configuration is usually put there.
Also, when you test the sound, don't use PulseAudio but test directly with ALSA native apps (e.g. aplay with -Dhw:0 or -Dplughw:0, etc). You can use speaker-test program, too.
And always try a headphone output at first. The headset mic might not work, but the headphone output is usually the easiest one to get working.
Takashi
Reverse engineering the Windows driver would lead to the solution but will be very long for me since I'm not familiar with the Intel HDA card and how this Realtek chip internally works. Don't hesitate if you have any hints which could help me (and all the other users of this laptop..) Cheers,
Aurélien
Le lundi 17 février 2020, 09:00:42 CET Takashi Iwai a écrit :
On Sat, 15 Feb 2020 17:56:58 +0100,
Aurélien Croc wrote:
Dear ALSA community,
I just bought a Samsung Galaxy book 12" laptop and I installed the latest Fedora version (31) on it. Unfortunately when I try to listen something there is absolutely no sound from speaker nor headsets. I checked that volumes of the different channels were max. I also checked for muting system but I didn't found anything. Of course it works perfectly under Windows. I saw on internet that many ALC298 users had troubles with it. I tried different options to the intel hda driver (especially the model one) but it changes nothing.
Here is the alsa-info.sh output: http://alsa-project.org/db/? f=871881e295972b9ecf252b25e90d659e38d939b8 I would appreciate some help in order to find a solution. Tell me if you need more information. Thanks in advance.
The BIOS seems broken, it just sets a single output pin. You need to try to figure out the pin connections, e.g. via hdajackretask program. The headphone might work with that. But the speaker output might need another special handling, and it's nothing but trial-and-test with the existing quirks for similar devices. Take a look at sound/pci/hda/patch_realtek.c.
Takashi
Sometimes you have some data in *.INF file or such. At least, if the standard Windows driver (not the h/w vendor's one) works, the extra configuration is usually put there.
The Windows driver is the one provided by Realtek. The manufacturer ID is "INTELAUDIO\FUNC_01&VEN_10EC&DEV_0298&SUBSYS_144DC14F". I looked into the INF file which corresponds to this driver but I'm not sure to understand it. There is a specific section attached to this device which looks like: [CusEffectSettingsMaxim.AddReg] HKR,SSTPPCfg{C75061F3-F2B2-4DCC-8F9F-82ABB4131E66}\SPK\EFX,16777217, 1, 1,0,0,1,1,0,0,0 HKR,SSTPPCfg{C75061F3-F2B2-4DCC-8F9F-82ABB4131E66}\SPK\EFX,16777218, 1, 2,0,0,1,0,0,80,2 ....
I don't know if the solution is in these lines or not..
Also, when you test the sound, don't use PulseAudio but test directly with ALSA native apps (e.g. aplay with -Dhw:0 or -Dplughw:0, etc). You can use speaker-test program, too.
I tried again all the channels with headphone only by using only ALSA and not pulseaudio. There is no sound at all. Note that the internal microphone and the microphone of my headphone when it's plugged in works very well..
And always try a headphone output at first. The headset mic might not work, but the headphone output is usually the easiest one to get working.
Thank you for your help. I hope we'll find the problem..
On Mon, 17 Feb 2020 20:01:05 +0100, Aurélien Croc wrote:
Sometimes you have some data in *.INF file or such. At least, if the standard Windows driver (not the h/w vendor's one) works, the extra configuration is usually put there.
The Windows driver is the one provided by Realtek. The manufacturer ID is "INTELAUDIO\FUNC_01&VEN_10EC&DEV_0298&SUBSYS_144DC14F". I looked into the INF file which corresponds to this driver but I'm not sure to understand it. There is a specific section attached to this device which looks like: [CusEffectSettingsMaxim.AddReg] HKR,SSTPPCfg{C75061F3-F2B2-4DCC-8F9F-82ABB4131E66}\SPK\EFX,16777217, 1, 1,0,0,1,1,0,0,0 HKR,SSTPPCfg{C75061F3-F2B2-4DCC-8F9F-82ABB4131E66}\SPK\EFX,16777218, 1, 2,0,0,1,0,0,80,2 ....
I don't know if the solution is in these lines or not..
Also, when you test the sound, don't use PulseAudio but test directly with ALSA native apps (e.g. aplay with -Dhw:0 or -Dplughw:0, etc). You can use speaker-test program, too.
I tried again all the channels with headphone only by using only ALSA and not pulseaudio. There is no sound at all. Note that the internal microphone and the microphone of my headphone when it's plugged in works very well..
Aha, so the input works but only the output doesn't? Then it's not about the basic stuff like pin configuration but rather the missing COEF setup or such that is specific to Realtek codec (and/or the Samsung platform).
Kailang, do you have any clue?
thanks,
Takashi
And always try a headphone output at first. The headset mic might not work, but the headphone output is usually the easiest one to get working.
Thank you for your help. I hope we'll find the problem..
-----Original Message----- From: Takashi Iwai tiwai@suse.de Sent: Tuesday, February 18, 2020 6:05 AM To: Aurélien Croc aurelien@ap2c.com Cc: alsa-devel@alsa-project.org; Kailang kailang@realtek.com Subject: Re: [alsa-devel] No sound with Realtek ALC298 on a Samsung Galaxy book
On Mon, 17 Feb 2020 20:01:05 +0100, Aurélien Croc wrote:
Sometimes you have some data in *.INF file or such. At least, if the standard Windows driver (not the h/w vendor's one) works, the extra configuration is usually put there.
The Windows driver is the one provided by Realtek. The manufacturer ID is "INTELAUDIO\FUNC_01&VEN_10EC&DEV_0298&SUBSYS_144DC14F". I looked into the INF file which corresponds to this driver but I'm not sure to understand it. There is a specific section attached to this device which looks like: [CusEffectSettingsMaxim.AddReg]
HKR,SSTPPCfg{C75061F3-F2B2-4DCC-8F9F-82ABB4131E66}\SPK\EFX,16777217 ,
1, 1,0,0,1,1,0,0,0
HKR,SSTPPCfg{C75061F3-F2B2-4DCC-8F9F-82ABB4131E66}\SPK\EFX,16777218 ,
1, 2,0,0,1,0,0,80,2 ....
I don't know if the solution is in these lines or not..
Also, when you test the sound, don't use PulseAudio but test directly with ALSA native apps (e.g. aplay with -Dhw:0 or -Dplughw:0, etc). You can use speaker-test program, too.
I tried again all the channels with headphone only by using only ALSA and not pulseaudio. There is no sound at all. Note that the internal microphone and the microphone of my headphone when it's plugged in works very well..
Aha, so the input works but only the output doesn't? Then it's not about the basic stuff like pin configuration but rather the missing COEF setup or such that is specific to Realtek codec (and/or the Samsung platform).
Kailang, do you have any clue?
Sorry. Samsung didn't request this platform need to support Linux.
thanks,
Takashi
And always try a headphone output at first. The headset mic might not work, but the headphone output is usually the easiest one to get working.
Thank you for your help. I hope we'll find the problem..
------Please consider the environment before printing this e-mail.
Hi,
Sorry. Samsung didn't request this platform need to support Linux.
I can understand that you can't work on my problem if your client (Samsung) didn't paid for that. However is it possible to get ALC298 specs in order to understand how to configure it (in particular a document which explains its COEF registers and their meaning) and I will do the job.
Thank you in advance. Regards,
Aurélien
P.S. : sorry for those who received this email twice: I had trouble with the SMTP server..
Hi,
Aha, so the input works but only the output doesn't? Then it's not about the basic stuff like pin configuration but rather the missing COEF setup or such that is specific to Realtek codec (and/or the Samsung platform).
Kailang, do you have any clue?
While waiting for Kailang's answer (and I hope he'll get the solution!!), isn't it possible to grab information from Windows? I'm sorry I don't know very well how Realtek chip works but if the problem would comes from some registers which are not well initialized isn't it possible to dump the state of the realtek chip and its registers under Windows and compare it with the one under Linux? Because I'm not sure that revers engineer the Windows driver would really easily lead to the solution.
Thank you again for your help.
Aurélien
participants (3)
-
Aurélien Croc -
Kailang -
Takashi Iwai