[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <372f49bfa3cf45838ffa994a14e844c0@realtek.com>
Date: Fri, 17 Dec 2021 02:25:07 +0000
From: Derek [方德義] <derek.fang@...ltek.com>
To: Ajit Kumar Pandey <AjitKumar.Pandey@....com>,
Shuming [范書銘] <shumingf@...ltek.com>,
"broonie@...nel.org" <broonie@...nel.org>,
"alsa-devel@...a-project.org" <alsa-devel@...a-project.org>
CC: Oder Chiou <oder_chiou@...ltek.com>,
"Sunil-kumar.Dommati@....com" <Sunil-kumar.Dommati@....com>,
open list <linux-kernel@...r.kernel.org>,
"Basavaraj.Hiregoudar@....com" <Basavaraj.Hiregoudar@....com>,
Takashi Iwai <tiwai@...e.com>,
Liam Girdwood <lgirdwood@...il.com>,
"Vijendar.Mukunda@....com" <Vijendar.Mukunda@....com>,
"Alexander.Deucher@....com" <Alexander.Deucher@....com>,
Jack Yu <jack.yu@...ltek.com>,
"Flove(HsinFu)" <flove@...ltek.com>,
Albert Chen <albertchen@...ltek.com>
Subject: RE: [PATCH v2 1/1] ASoC: rt5682s: Add dapm switch to mute/unmute HP playback output
> Subject: Re: [PATCH v2 1/1] ASoC: rt5682s: Add dapm switch to mute/unmute
> HP playback output
>
> On 12/9/2021 7:55 AM, Shuming [范書銘] wrote:
> > [CAUTION: External Email]
> >
> >> Subject: [PATCH v2 1/1] ASoC: rt5682s: Add dapm switch to mute/unmute HP
> >> playback output
> >>
> >> Add dapm switch in playback path to mute or unmute HP output data.
> >> We will set and reset MUTE_SFT bit in RT5682S_HP_CTRL_1 register based on
> >> switch value to mute or unmute respective channel.
> >>
> >> Signed-off-by: Ajit Kumar Pandey <AjitKumar.Pandey@....com>
> >> ---
> >> sound/soc/codecs/rt5682s.c | 19 +++++++++++++++++--
> >> 1 file changed, 17 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/sound/soc/codecs/rt5682s.c b/sound/soc/codecs/rt5682s.c index
> >> d49a4f68566d..21ab5f7df422 100644
> >> --- a/sound/soc/codecs/rt5682s.c
> >> +++ b/sound/soc/codecs/rt5682s.c
> >> @@ -1573,6 +1573,14 @@ static const char * const
> >> rt5682s_adcdat_pin_select[] = {
> >> "ADCDAT1", "ADCDAT2",
> >> };
> >>
> >> +/* Out Switch */
> >> +static const struct snd_kcontrol_new hpol_switch =
> >> + SOC_DAPM_SINGLE_AUTODISABLE("Switch", RT5682S_HP_CTRL_1,
> >> + RT5682S_L_MUTE_SFT, 1, 1);
> >> +static const struct snd_kcontrol_new hpor_switch =
> >> + SOC_DAPM_SINGLE_AUTODISABLE("Switch", RT5682S_HP_CTRL_1,
> >> + RT5682S_R_MUTE_SFT, 1, 1);
> >> +
> >> static SOC_VALUE_ENUM_SINGLE_DECL(rt5682s_adcdat_pin_enum,
> >> RT5682S_GPIO_CTRL_1, RT5682S_GP4_PIN_SFT,
> >> RT5682S_GP4_PIN_MASK,
> >> rt5682s_adcdat_pin_select, rt5682s_adcdat_pin_values); @@
> -1746,6
> >> +1754,11 @@ static const struct snd_soc_dapm_widget
> >> rt5682s_dapm_widgets[] = {
> >> SND_SOC_DAPM_PGA_S("HP Amp", 1, SND_SOC_NOPM, 0, 0,
> >> rt5682s_hp_amp_event,
> >> SND_SOC_DAPM_POST_PMD |
> SND_SOC_DAPM_POST_PMU),
> >>
> >> + SND_SOC_DAPM_SWITCH("HPOL Playback", SND_SOC_NOPM, 0, 0,
> >> + &hpol_switch),
> >> + SND_SOC_DAPM_SWITCH("HPOR Playback", SND_SOC_NOPM, 0, 0,
> >> + &hpor_switch),
> >> +
> >> /* CLK DET */
> >> SND_SOC_DAPM_SUPPLY("CLKDET SYS", RT5682S_CLK_DET,
> >> RT5682S_SYS_CLK_DET_SFT, 0, NULL, 0), @@ -1895,8
> +1908,10 @@
> >> static const struct snd_soc_dapm_route rt5682s_dapm_routes[] = {
> >> {"HP Amp", NULL, "CLKDET SYS"},
> >> {"HP Amp", NULL, "SAR"},
> >>
> >> - {"HPOL", NULL, "HP Amp"},
> >> - {"HPOR", NULL, "HP Amp"},
> >> + {"HPOL Playback", "Switch", "HP Amp"},
> >> + {"HPOR Playback", "Switch", "HP Amp"},
> >> + {"HPOL", NULL, "HPOL Playback"},
> >> + {"HPOR", NULL, "HPOR Playback"},
> >> };
> >
> > RT5682S enables the 1 bit control for HP output.
> > Therefore, the setting of RT5682S_HP_CTRL_1[15][7] will be not effective.
> >
> >
> Ok, it was actually effective and muted audio when we test in our
> platform, possible because i'm tearing down dapm path based on switch
> value. If we don't have register bits available to mute i'll resubmit
> the change with virtual switch widget to disable DAPM and stop audio.
>
What purpose do you want to add an additional switch to power up/down
the dapm in the codec driver instead of using existing controls?
And we thought the virtual switch you talked about could be implemented
in the machine driver.
Powered by blists - more mailing lists