* [PATCH] ALSA: hda/realtek: Limit Line In boost and ADC gain on DECENTA MATX03
@ 2026-10-13 12:03 wangdich9700
2026-10-14 9:22 ` Takashi Iwai
0 siblings, 1 reply; 2+ messages in thread
From: wangdich9700 @ 2026-10-13 12:03 UTC (permalink / raw)
To: tiwai, wangdich9700, perex; +Cc: linux-sound, linux-kernel, wangdicheng
From: wangdicheng <wangdicheng@kylinos.cn>
The DECENTA MATX03 (ALC897, SSID 1e0b:1297) exposes a high gain
range on both the Line In pin boost and the ADC input amp. Pin 0x1a
reports nsteps=0x03 with stepsize=0x27, that is 0, 9.75, 19.5 and
29.25 dB. ADC 0x08 reports nsteps=0x3f with stepsize=0x02, that is
0 to 31.5 dB in 0.5 dB steps, and defaults to 0x3f (maximum). The
combined 60.75 dB analogue cascade overloads the front-end when the
source level is high, producing audible harmonic distortion during
Line In monitoring.
Add a new fixup alc897_fixup_limit_linein_boost that overrides the amp
capabilities of the Line In and external Mic pins to nsteps=1 (max
~11.75 dB), and also limits the ADC input amp to nsteps=1 (max ~0.5 dB).
The total gain drops from 60.75 dB to ~12.25 dB, eliminating the
clipping. Internal and dock mic pins are excluded to avoid affecting
built-in microphones.
Tested on the affected machine: with the fixup applied, Line In
recording at maximum source volume is clean, and dmesg confirms the
fixup is picked:
ALC897: picked fixup alc897-limit-linein-boost (SSID)
Signed-off-by: wangdicheng <wangdicheng@kylinos.cn>
---
sound/hda/codecs/realtek/alc662.c | 50 +++++++++++++++++++++++++++++++
1 file changed, 50 insertions(+)
diff --git a/sound/hda/codecs/realtek/alc662.c b/sound/hda/codecs/realtek/alc662.c
index eff7c83b9c59..a140891d51fa 100644
--- a/sound/hda/codecs/realtek/alc662.c
+++ b/sound/hda/codecs/realtek/alc662.c
@@ -198,6 +198,49 @@ static void alc897_fixup_lenovo_headset_mode(struct hda_codec *codec,
}
}
+static void alc897_fixup_limit_linein_boost(struct hda_codec *codec,
+ const struct hda_fixup *fix, int action)
+{
+ struct alc_spec *spec = codec->spec;
+ struct auto_pin_cfg *cfg = &spec->gen.autocfg;
+ int i;
+
+ if (action != HDA_FIXUP_ACT_PROBE)
+ return;
+
+ for (i = 0; i < cfg->num_inputs; i++) {
+ hda_nid_t nid = cfg->inputs[i].pin;
+ unsigned int defcfg;
+
+ if (cfg->inputs[i].type != AUTO_PIN_LINE_IN &&
+ cfg->inputs[i].type != AUTO_PIN_MIC)
+ continue;
+
+ defcfg = snd_hda_codec_get_pincfg(codec, nid);
+ if (snd_hda_get_input_pin_attr(defcfg) == INPUT_PIN_ATTR_INT ||
+ snd_hda_get_input_pin_attr(defcfg) == INPUT_PIN_ATTR_DOCK)
+ continue;
+
+ snd_hda_override_amp_caps(codec, nid, HDA_INPUT,
+ (0x00 << AC_AMPCAP_OFFSET_SHIFT) |
+ (0x01 << AC_AMPCAP_NUM_STEPS_SHIFT) |
+ (0x2f << AC_AMPCAP_STEP_SIZE_SHIFT) |
+ (0 << AC_AMPCAP_MUTE_SHIFT));
+ }
+
+ for (i = 0; i < spec->gen.num_all_adcs; i++) {
+ hda_nid_t adc = spec->gen.all_adcs[i];
+ unsigned int caps;
+
+ caps = query_amp_caps(codec, adc, HDA_INPUT);
+ if ((caps & AC_AMPCAP_NUM_STEPS) >> AC_AMPCAP_NUM_STEPS_SHIFT > 1) {
+ caps &= ~AC_AMPCAP_NUM_STEPS;
+ caps |= (0x01 << AC_AMPCAP_NUM_STEPS_SHIFT);
+ snd_hda_override_amp_caps(codec, adc, HDA_INPUT, caps);
+ }
+ }
+}
+
static const struct coef_fw alc668_coefs[] = {
WRITE_COEF(0x01, 0xbebe), WRITE_COEF(0x02, 0xaaaa), WRITE_COEF(0x03, 0x0),
WRITE_COEF(0x04, 0x0180), WRITE_COEF(0x06, 0x0), WRITE_COEF(0x07, 0x0f80),
@@ -326,6 +369,7 @@ enum {
ALC897_FIXUP_HEADSET_MIC_PIN3,
ALC662_FIXUP_CSL_GPIO,
ALC897_FIXUP_BEELINK_SER6_AMP,
+ ALC897_FIXUP_LIMIT_LINEIN_BOOST,
};
static const struct hda_fixup alc662_fixups[] = {
@@ -787,6 +831,10 @@ static const struct hda_fixup alc662_fixups[] = {
.type = HDA_FIXUP_FUNC,
.v.func = alc_fixup_gpio4,
},
+ [ALC897_FIXUP_LIMIT_LINEIN_BOOST] = {
+ .type = HDA_FIXUP_FUNC,
+ .v.func = alc897_fixup_limit_linein_boost,
+ },
};
static const struct hda_quirk alc662_fixup_tbl[] = {
@@ -859,6 +907,7 @@ static const struct hda_quirk alc662_fixup_tbl[] = {
SND_PCI_QUIRK(0x1c6c, 0x1239, "Compaq N14JP6-V2", ALC897_FIXUP_HP_HSMIC_VERB),
SND_PCI_QUIRK(0x1e63, 0x6d9a, "F+ FLAPTOP r", ALC897_FIXUP_HP_HSMIC_VERB),
SND_PCI_QUIRK(0x1f66, 0x0202, "Beelink SER6 Max 6900", ALC897_FIXUP_BEELINK_SER6_AMP),
+ SND_PCI_QUIRK(0x1e0b, 0x1297, "DECENTA MATX03", ALC897_FIXUP_LIMIT_LINEIN_BOOST),
#if 0
/* Below is a quirk table taken from the old code.
@@ -954,6 +1003,7 @@ static const struct hda_model_fixup alc662_fixup_models[] = {
{.id = ALC662_FIXUP_LENOVO_MULTI_CODECS, .name = "dual-codecs"},
{.id = ALC669_FIXUP_ACER_ASPIRE_ETHOS, .name = "aspire-ethos"},
{.id = ALC897_FIXUP_UNIS_H3C_X500S, .name = "unis-h3c-x500s"},
+ {.id = ALC897_FIXUP_LIMIT_LINEIN_BOOST, .name = "alc897-limit-linein-boost"},
{}
};
--
2.25.1
^ permalink raw reply related [flat|nested] 2+ messages in thread* Re: [PATCH] ALSA: hda/realtek: Limit Line In boost and ADC gain on DECENTA MATX03
2026-10-13 12:03 [PATCH] ALSA: hda/realtek: Limit Line In boost and ADC gain on DECENTA MATX03 wangdich9700
@ 2026-10-14 9:22 ` Takashi Iwai
0 siblings, 0 replies; 2+ messages in thread
From: Takashi Iwai @ 2026-10-14 9:22 UTC (permalink / raw)
To: wangdich9700; +Cc: tiwai, perex, linux-sound, linux-kernel, wangdicheng
On Tue, 13 Oct 2026 14:03:27 +0200,
wangdich9700@163.com wrote:
>
> From: wangdicheng <wangdicheng@kylinos.cn>
>
> The DECENTA MATX03 (ALC897, SSID 1e0b:1297) exposes a high gain
> range on both the Line In pin boost and the ADC input amp. Pin 0x1a
> reports nsteps=0x03 with stepsize=0x27, that is 0, 9.75, 19.5 and
> 29.25 dB. ADC 0x08 reports nsteps=0x3f with stepsize=0x02, that is
> 0 to 31.5 dB in 0.5 dB steps, and defaults to 0x3f (maximum). The
> combined 60.75 dB analogue cascade overloads the front-end when the
> source level is high, producing audible harmonic distortion during
> Line In monitoring.
>
> Add a new fixup alc897_fixup_limit_linein_boost that overrides the amp
> capabilities of the Line In and external Mic pins to nsteps=1 (max
> ~11.75 dB), and also limits the ADC input amp to nsteps=1 (max ~0.5 dB).
> The total gain drops from 60.75 dB to ~12.25 dB, eliminating the
> clipping. Internal and dock mic pins are excluded to avoid affecting
> built-in microphones.
>
> Tested on the affected machine: with the fixup applied, Line In
> recording at maximum source volume is clean, and dmesg confirms the
> fixup is picked:
>
> ALC897: picked fixup alc897-limit-linein-boost (SSID)
>
> Signed-off-by: wangdicheng <wangdicheng@kylinos.cn>
> ---
> sound/hda/codecs/realtek/alc662.c | 50 +++++++++++++++++++++++++++++++
> 1 file changed, 50 insertions(+)
>
> diff --git a/sound/hda/codecs/realtek/alc662.c b/sound/hda/codecs/realtek/alc662.c
> index eff7c83b9c59..a140891d51fa 100644
> --- a/sound/hda/codecs/realtek/alc662.c
> +++ b/sound/hda/codecs/realtek/alc662.c
> @@ -198,6 +198,49 @@ static void alc897_fixup_lenovo_headset_mode(struct hda_codec *codec,
> }
> }
>
> +static void alc897_fixup_limit_linein_boost(struct hda_codec *codec,
> + const struct hda_fixup *fix, int action)
> +{
> + struct alc_spec *spec = codec->spec;
> + struct auto_pin_cfg *cfg = &spec->gen.autocfg;
> + int i;
> +
> + if (action != HDA_FIXUP_ACT_PROBE)
> + return;
> +
> + for (i = 0; i < cfg->num_inputs; i++) {
> + hda_nid_t nid = cfg->inputs[i].pin;
> + unsigned int defcfg;
> +
> + if (cfg->inputs[i].type != AUTO_PIN_LINE_IN &&
> + cfg->inputs[i].type != AUTO_PIN_MIC)
> + continue;
> +
> + defcfg = snd_hda_codec_get_pincfg(codec, nid);
> + if (snd_hda_get_input_pin_attr(defcfg) == INPUT_PIN_ATTR_INT ||
> + snd_hda_get_input_pin_attr(defcfg) == INPUT_PIN_ATTR_DOCK)
> + continue;
> +
> + snd_hda_override_amp_caps(codec, nid, HDA_INPUT,
> + (0x00 << AC_AMPCAP_OFFSET_SHIFT) |
> + (0x01 << AC_AMPCAP_NUM_STEPS_SHIFT) |
> + (0x2f << AC_AMPCAP_STEP_SIZE_SHIFT) |
> + (0 << AC_AMPCAP_MUTE_SHIFT));
> + }
> +
> + for (i = 0; i < spec->gen.num_all_adcs; i++) {
> + hda_nid_t adc = spec->gen.all_adcs[i];
> + unsigned int caps;
> +
> + caps = query_amp_caps(codec, adc, HDA_INPUT);
> + if ((caps & AC_AMPCAP_NUM_STEPS) >> AC_AMPCAP_NUM_STEPS_SHIFT > 1) {
> + caps &= ~AC_AMPCAP_NUM_STEPS;
> + caps |= (0x01 << AC_AMPCAP_NUM_STEPS_SHIFT);
> + snd_hda_override_amp_caps(codec, adc, HDA_INPUT, caps);
> + }
> + }
> +}
> +
I guess there aren't so many pins and ADCs, and the quirk is very
specific to the model, so you can just write a few
snd_hda_override_amp_caps() for the fixed hard-coded NIDs.
The alc269_fixup_limit_int_mic_boost() is applied to lots of different
models, hence it handles in a generic way, OTOH.
> static const struct hda_quirk alc662_fixup_tbl[] = {
> @@ -859,6 +907,7 @@ static const struct hda_quirk alc662_fixup_tbl[] = {
> SND_PCI_QUIRK(0x1c6c, 0x1239, "Compaq N14JP6-V2", ALC897_FIXUP_HP_HSMIC_VERB),
> SND_PCI_QUIRK(0x1e63, 0x6d9a, "F+ FLAPTOP r", ALC897_FIXUP_HP_HSMIC_VERB),
> SND_PCI_QUIRK(0x1f66, 0x0202, "Beelink SER6 Max 6900", ALC897_FIXUP_BEELINK_SER6_AMP),
> + SND_PCI_QUIRK(0x1e0b, 0x1297, "DECENTA MATX03", ALC897_FIXUP_LIMIT_LINEIN_BOOST),
The table is sorted in PCI SSID order, so please put at the right
position.
> @@ -954,6 +1003,7 @@ static const struct hda_model_fixup alc662_fixup_models[] = {
> {.id = ALC662_FIXUP_LENOVO_MULTI_CODECS, .name = "dual-codecs"},
> {.id = ALC669_FIXUP_ACER_ASPIRE_ETHOS, .name = "aspire-ethos"},
> {.id = ALC897_FIXUP_UNIS_H3C_X500S, .name = "unis-h3c-x500s"},
> + {.id = ALC897_FIXUP_LIMIT_LINEIN_BOOST, .name = "alc897-limit-linein-boost"},
> {}
> };
Nowadays we don't want a model name addition unless it's really
needed. The existing quirk can be applied with an alias instead of a
model name string.
thanks,
Takashi
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-10-14 9:22 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-10-13 12:03 [PATCH] ALSA: hda/realtek: Limit Line In boost and ADC gain on DECENTA MATX03 wangdich9700
2026-10-14 9:22 ` Takashi Iwai
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox