Re: [PATCH] ALSA: hda/realtek: Limit Internal Microphone boost for ALC3246

[Chester A. Unal]

On 09/03/2025 12:39, Chester A. Unal via B4 Relay wrote:
> From: "Chester A. Unal" <chester.a.unal@xxxxxxxxxx>
>
> Internal Microphone is too loud on Dell Latitude 5490, Dell Device
> [1028:0816], which has got ALC3246. I suppose this is the case for all
> devices with ALC3246 which is ALC256 for Dell devices. Therefore, limit the
> Internal Microphone boost for ALC3246. Change
> ALC255_FIXUP_DELL1_MIC_NO_PRESENCE to
> ALC255_FIXUP_DELL1_LIMIT_INT_MIC_BOOST which includes the former fixup.
>
> Signed-off-by: Chester A. Unal <chester.a.unal@xxxxxxxxxx>
> ---
> I don't know ALSA or the Intel HDA specification very well so feel free to
> correct any incorrect information in the patch log.
>
> I experience another issue with this laptop that is also there on ALC3204.
> Headphones output loops back to Headset Microphone. So if I record with
> this microphone, sysystem audio is recorded as well. I have tried a fair
> share of fixups by compiling snd-hda-codec-realtek.ko with different fixups
> applied for "0x10ec0256, 0x1028", replacing the existing module in
> /lib/modules, and then rebooting.
>
> ALC2XX_FIXUP_HEADSET_MIC makes Headset Microphone appear unplugged. But I
> can see on the OS that it detects voice, so, it works. When it appears
> unplugged, Headphones output won't loop back to it. This fixup also makes
> Speakers appear unavailable and makes Microphone disappear, as
> ALC255_FIXUP_DELL1_MIC_NO_PRESENCE won't apply anymore.
>
> Another issue is when I toggle capture on Headphone Mic on alsamixer,
> physically no audio will be played on Headphone. Speaker is unaffected. If
> the capture is toggled on anything else, Headphone will play audio.
>
> Tested on Debian Trixie Alpha, kernel 6.12.17. My headphones are
> Tip-Ring-Ring-Sleeve. Here's how I test both issues:
>
> systemctl --user stop pipewire.socket && systemctl --user stop pipewire
> alsamixer -c 0
>
> speaker-test -Dhw:0,0 -c2 -r48000 -FS16_LE -t sine
>
> arecord -Dhw:0,0 -c2 -r48000 -fS16_LE -d30 test.wav
> aplay -Dhw:0,0 test.wav
>
> Attached is test.wav. The timeline of the recording is:
> - I run speaker-test. Then I run arecord. After that I say: Speaker-test is
>    running. Capture is toggled on Headphone Mic.
> - After about five seconds I say: I am switching capture to Headset Mic.
> - I toggle capture on Headset Mic. After that I say: This is Headset Mic.
> - I kill speaker-test. After that I say: I have killed speaker-test.

Here's the link to the recording.

https://github.com/thesofproject/linux/issues/4292#issuecomment-2708834717

Chester A.

> Maintainers, please let me know if you have an idea how to fix these two
> issues.
>
> Chester A.
> ---
>   sound/pci/hda/patch_realtek.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
> index d2a1f836dbbf714c26be682133b44cc0bd173ad9..379a325e96641232485a2fc8e6de03798461f466 100644
> --- a/sound/pci/hda/patch_realtek.c
> +++ b/sound/pci/hda/patch_realtek.c
> @@ -11799,7 +11799,7 @@ static const struct snd_hda_pin_quirk alc269_fallback_pin_fixup_tbl[] = {
>   	SND_HDA_PIN_QUIRK(0x10ec0295, 0x1028, "Dell", ALC269_FIXUP_DELL4_MIC_NO_PRESENCE_QUIET,
>   		{0x19, 0x40000000},
>   		{0x1b, 0x40000000}),
> -	SND_HDA_PIN_QUIRK(0x10ec0256, 0x1028, "Dell", ALC255_FIXUP_DELL1_MIC_NO_PRESENCE,
> +	SND_HDA_PIN_QUIRK(0x10ec0256, 0x1028, "Dell", ALC255_FIXUP_DELL1_LIMIT_INT_MIC_BOOST,
>   		{0x19, 0x40000000},
>   		{0x1a, 0x40000000}),
>   	SND_HDA_PIN_QUIRK(0x10ec0236, 0x1028, "Dell", ALC255_FIXUP_DELL1_LIMIT_INT_MIC_BOOST,
>
> ---
> base-commit: b7c90e3e717abff6fe06445b98be306b732bbd2b
> change-id: 20250308-for-sound-alc256-dell-61967f99ce45
>
> Best regards,