From mboxrd@z Thu Jan 1 00:00:00 1970 From: Charles Keepax Subject: Re: [PATCHv5 2/9] ASOC: wm8971: modify the control of deemphasis Date: Wed, 29 Oct 2014 09:46:39 +0000 Message-ID: <20141029094638.GH23178@opensource.wolfsonmicro.com> References: <1414130435-14183-1-git-send-email-xavier.hsu@linaro.org> <1414130435-14183-2-git-send-email-xavier.hsu@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from opensource.wolfsonmicro.com (opensource.wolfsonmicro.com [80.75.67.52]) by alsa0.perex.cz (Postfix) with ESMTP id 8825B2604CE for ; Wed, 29 Oct 2014 10:46:39 +0100 (CET) Content-Disposition: inline In-Reply-To: <1414130435-14183-2-git-send-email-xavier.hsu@linaro.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: alsa-devel-bounces@alsa-project.org Sender: alsa-devel-bounces@alsa-project.org To: Xavier Hsu Cc: Andy Green , alsa-devel@alsa-project.org, patches@opensource.wolfsonmicro.com, lars@metafoo.de, patches@linaro.org List-Id: alsa-devel@alsa-project.org On Fri, Oct 24, 2014 at 02:00:28PM +0800, Xavier Hsu wrote: > We add a switch for automatic updating > the frequency of wm8971. > > Signed-off-by: Xavier Hsu > Signed-off-by: Andy Green > --- > sound/soc/codecs/wm8971.c | 77 +++++++++++++++++++++++++++++++++++++++++++-- > 1 file changed, 75 insertions(+), 2 deletions(-) > mode change 100644 => 100755 sound/soc/codecs/wm8971.c > > diff --git a/sound/soc/codecs/wm8971.c b/sound/soc/codecs/wm8971.c > old mode 100644 > new mode 100755 > index 01c76a4..6df0a3b > --- a/sound/soc/codecs/wm8971.c > +++ b/sound/soc/codecs/wm8971.c > @@ -7,6 +7,10 @@ > * > * Based on wm8753.c by Liam Girdwood > * > + * WM8971 Improve Copyright (C) 2014 Linaro, Ltd > + * Author: Xavier Hsu > + * Andy Green > + * > * This program is free software; you can redistribute it and/or modify it > * under the terms of the GNU General Public License as published by the > * Free Software Foundation; either version 2 of the License, or (at your > @@ -36,6 +40,11 @@ static struct workqueue_struct *wm8971_workq; > /* codec private data */ > struct wm8971_priv { > unsigned int sysclk; > + > + /* De-emphasis */ > + struct mutex deemph_mutex; > + int playback_fs; > + bool deemph; > }; > > /* > @@ -100,7 +109,6 @@ static const char const *wm8971_alc_func[] = {"Off", "Right", > "Left", "Stereo"}; > static const char const *wm8971_ng_type[] = {"Constant PGA Gain", > "Mute ADC Output"}; > -static const char const *wm8971_deemp[] = {"None", "32kHz", "44.1kHz", "48kHz"}; > static const char const *wm8971_mono_mux[] = {"Stereo", "Mono (Left)", > "Mono (Right)", "Digital Mono"}; > static const char const *wm8971_dac_phase[] = {"Non Inverted", "Inverted"}; > @@ -114,6 +122,66 @@ static const char const *wm8971_rpga_sel[] = {"Line", "Mic", "NC", > "Differential"}; > static const char const *wm8971_adcpol[] = {"Normal", "L Invert", > "R Invert", "L + R Invert"}; > +static const char const *wm8971_deemp[] = {"None", "32kHz", > + "44.1kHz", "48kHz"}; > + > +static int wm8971_set_deemph(struct snd_soc_codec *codec) > +{ > + struct wm8971_priv *wm8971 = snd_soc_codec_get_drvdata(codec); > + int val = 0, i, best = 0; > + > + /* If we're using deemphasis select the nearest available sample > + * rate. > + */ > + if (wm8971->deemph) { > + best = 1; > + for (i = 2; i < ARRAY_SIZE(wm8971_deemp); i++) { > + if (abs(wm8971_deemp[i] - wm8971->playback_fs) < > + abs(wm8971_deemp[best] - wm8971->playback_fs)) I am not quite clear on how this is working, we seem to be subtracting an int (playback_fs) from a char * (wm8971_deemp[i])? Also playback_fs is never set anywhere in the driver so will always be zero? Has this patch been tested? > + best = i; > + } > + val = best << 1; > + } > + > + dev_dbg(codec->dev, "Set deemphasis %d (%dHz)\n", > + best, wm8971_deemp[best]); > + > + return snd_soc_update_bits(codec, WM8971_ADCDAC, 0x6, val); > +} > + > +static int wm8971_get_deemph(struct snd_kcontrol *kcontrol, > + struct snd_ctl_elem_value *ucontrol) > +{ > + struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol); > + struct wm8971_priv *wm8971 = snd_soc_codec_get_drvdata(codec); > + > + ucontrol->value.enumerated.item[0] = wm8971->deemph; > + > + return 0; > +} > + > +static int wm8971_put_deemph(struct snd_kcontrol *kcontrol, > + struct snd_ctl_elem_value *ucontrol) > +{ > + struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol); > + struct wm8971_priv *wm8971 = snd_soc_codec_get_drvdata(codec); > + int deemph = ucontrol->value.enumerated.item[0]; > + int ret = 0; > + > + if (deemph > 1) > + return -EINVAL; > + > + mutex_lock(&wm8971->deemph_mutex); > + if (wm8971->deemph != deemph) { > + wm8971->deemph = deemph; > + wm8971_set_deemph(codec); > + > + ret = 1; > + } > + mutex_unlock(&wm8971->deemph_mutex); > + > + return ret; > +} > > static const struct soc_enum wm8971_enum[] = { > SOC_ENUM_SINGLE(WM8971_BASS, 7, 2, wm8971_bass), /* 0 */ > @@ -184,7 +252,8 @@ static const struct snd_kcontrol_new wm8971_snd_controls[] = { > SOC_SINGLE("Capture 6dB Attenuate", WM8971_ADCDAC, 8, 1, 0), > SOC_SINGLE("Playback 6dB Attenuate", WM8971_ADCDAC, 7, 1, 0), > > - SOC_ENUM("Playback De-emphasis", wm8971_enum[5]), You are removing the only user of wm8971_enum[5], probably should remove that entry in the array as well. Ah I see you then remove it in patch 2 perhaps reorder those two patches and explicitly remove it in this patch, would be more clear. > + SOC_SINGLE_BOOL_EXT("Playback De-emphasis Switch", 0, > + wm8971_get_deemph, wm8971_put_deemph), > SOC_ENUM("Playback Function", wm8971_enum[6]), > SOC_ENUM("Playback Phase", wm8971_enum[7]), > > @@ -534,6 +603,8 @@ static int wm8971_pcm_hw_params(struct snd_pcm_substream *substream, > break; > } > > + wm8971_set_deemph(codec); > + > /* set iface & srate */ > snd_soc_write(codec, WM8971_IFACE, iface); > if (coeff >= 0) > @@ -727,6 +798,8 @@ static int wm8971_i2c_probe(struct i2c_client *i2c, > if (wm8971 == NULL) > return -ENOMEM; > > + mutex_init(&wm8971->deemph_mutex); > + > regmap = devm_regmap_init_i2c(i2c, &wm8971_regmap); > if (IS_ERR(regmap)) > return PTR_ERR(regmap); > -- > 1.7.9.5 > Also I tried to apply this against the current ASoC for next and it fails to apply. Is this generated against the right tree? Thanks, Charles