Update MAINTAINERS for ALSA SoC
Subsystems (4)
PATCH 1/17 Update MAINTAINERS for ALSA SoC
Add myself as a point of contact for the ALSA SoC subsystem and add a reference to the development GIT tree.
--- MAINTAINERS | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)
MAINTAINERS +3 -0
@@ -3514,6 +3514,9 @@ S: Maintained SOUND - SOC LAYER / DYNAMIC AUDIO POWER MANAGEMENT P: Liam Girdwood M: [email protected] +P: Mark Brown +M: [email protected] +T: git opensource.wolfsonmicro.com/linux-2.6-asoc L: [email protected] (subscribers-only) S: Supported
Signature · 1 line
1.5.3.8
PATCH 2/17 Fix inverted Bass, Treble, Mono volume and Phone volume WM9712 mixer controls.
From: Joe Sauer <[email protected]>
--- sound/soc/codecs/wm9712.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-)
sound/soc/codecs/wm9712.c +4 -4
@@ -102,7 +102,7 @@ SOC_SINGLE("Speaker Playback ZC Switch", AC97_MASTER, 7, 1, 0),SOC_SINGLE ("Speaker Playback Invert Switch" ,AC97_MASTER ,6 ,1 ,0 ),SOC_SINGLE ("Headphone Playback ZC Switch" ,AC97_HEADPHONE ,7 ,1 ,0 ),SOC_SINGLE ("Mono Playback ZC Switch" ,AC97_MASTER_MONO ,7 ,1 ,0 ), -SOC_SINGLE ("Mono Playback Volume" ,AC97_MASTER_MONO ,0 ,31 ,0 ), +SOC_SINGLE ("Mono Playback Volume" ,AC97_MASTER_MONO ,0 ,31 ,1 ),SOC_SINGLE ("ALC Target Volume" ,AC97_CODEC_CLASS_REV ,12 ,15 ,0 ), SOC_SINGLE("ALC Hold Time", AC97_CODEC_CLASS_REV, 8, 15, 0),
@@ -131,7 +131,7 @@ SOC_SINGLE("Aux Playback Headphone Volume", AC97_CD, 12, 7, 1),SOC_SINGLE ("Aux Playback Speaker Volume" ,AC97_CD ,8 ,7 ,1 ),SOC_SINGLE ("Aux Playback Phone Volume" ,AC97_CD ,4 ,7 ,1 ), -SOC_SINGLE ("Phone Volume" ,AC97_PHONE ,0 ,15 ,0 ), +SOC_SINGLE ("Phone Volume" ,AC97_PHONE ,0 ,15 ,1 ),SOC_DOUBLE ("Line Capture Volume" ,AC97_LINE ,8 ,0 ,31 ,1 ), SOC_SINGLE("Capture 20dB Boost Switch", AC97_REC_SEL, 14, 1, 0),
@@ -145,8 +145,8 @@ SOC_ENUM("Bass Control", wm9712_enum[5]),SOC_SINGLE ("Bass Cut-off Switch" ,AC97_MASTER_TONE ,12 ,1 ,1 ),SOC_SINGLE ("Tone Cut-off Switch" ,AC97_MASTER_TONE ,4 ,1 ,1 ),SOC_SINGLE ("Playback Attenuate (-6dB) Switch" ,AC97_MASTER_TONE ,6 ,1 ,0 ), -SOC_SINGLE ("Bass Volume" ,AC97_MASTER_TONE ,8 ,15 ,0 ), -SOC_SINGLE ("Treble Volume" ,AC97_MASTER_TONE ,0 ,15 ,0 ), +SOC_SINGLE ("Bass Volume" ,AC97_MASTER_TONE ,8 ,15 ,1 ), +SOC_SINGLE ("Treble Volume" ,AC97_MASTER_TONE ,0 ,15 ,1 ),SOC_SINGLE ("Capture ADC Switch" ,AC97_REC_GAIN ,15 ,1 ,1 ), SOC_ENUM("Capture Volume Steps", wm9712_enum[6]),
Signature · 1 line
1.5.3.8
PATCH 3/17 Add device level DAPM event
From: Liam Girdwood <[email protected]>
Added a device level dapm event so that both the machine and codec are informed when dapm events occur.
--- include/sound/soc-dapm.h | 1 + include/sound/soc.h | 3 +++ sound/soc/soc-core.c | 23 +++++++++++------------ sound/soc/soc-dapm.c | 23 +++++++++++++++++++++++ 4 files changed, 38 insertions(+), 12 deletions(-)
include/sound/soc-dapm.h +1 -0
@@ -199,6 +199,7 @@ void snd_soc_dapm_free(struct snd_soc_device *socdev);/* dapm events */ int snd_soc_dapm_stream_event (struct snd_soc_codec *codec,char *stream,int event); +int snd_soc_dapm_device_event (struct snd_soc_device *socdev,int event);/* dapm sys fs - used by the core */ int snd_soc_dapm_sys_add (struct device *dev);
include/sound/soc.h +3 -0
@@ -426,6 +426,9 @@ struct snd_soc_machine {int (*resume_pre)(struct platform_device *pdev);int (*resume_post)(struct platform_device *pdev); +/* callbacks */ +int (*dapm_event)(struct snd_soc_machine *,int event); +/* CPU <--> Codec DAI links */ struct snd_soc_dai_link *dai_link;int num_links;
sound/soc/soc-core.c +11 -12
@@ -289,15 +289,16 @@ static void close_delayed_work(struct work_struct *work)if (codec_dai->pop_wait ==1 ) { codec_dai->pop_wait =0 ; -snd_soc_dapm_stream_event (codec, codec_dai->playback.stream_name, +snd_soc_dapm_stream_event (codec, + codec_dai->playback.stream_name,SND_SOC_DAPM_STREAM_STOP );/* power down the codec power domain if no longer active */ if (codec->active ==0 ) {dbg ("pop wq D3 %s %s\n" , codec->name, codec_dai->playback.stream_name); -if (codec->dapm_event) - codec->dapm_event (codec, SNDRV_CTL_POWER_D3hot); +snd_soc_dapm_device_event (socdev, + SNDRV_CTL_POWER_D3hot); } } }
@@ -353,12 +354,12 @@ static int soc_codec_close(struct snd_pcm_substream *substream) }else {/* capture streams can be powered down now */ snd_soc_dapm_stream_event (codec, - codec_dai->capture.stream_name,SND_SOC_DAPM_STREAM_STOP ); + codec_dai->capture.stream_name, +SND_SOC_DAPM_STREAM_STOP ); -if (codec->active ==0 && codec_dai->pop_wait ==0 ){ -if (codec->dapm_event) - codec->dapm_event (codec, SNDRV_CTL_POWER_D3hot); - } +if (codec->active ==0 && codec_dai->pop_wait ==0 ) +snd_soc_dapm_device_event (socdev, + SNDRV_CTL_POWER_D3hot); }mutex_unlock (&pcm_mutex);
@@ -433,8 +434,7 @@ static int soc_pcm_prepare(struct snd_pcm_substream *substream)/* no delayed work - do we need to power up codec */ if (codec->dapm_state !=SNDRV_CTL_POWER_D0 ) { -if (codec->dapm_event) - codec->dapm_event (codec,SNDRV_CTL_POWER_D1 ); +snd_soc_dapm_device_event (socdev,SNDRV_CTL_POWER_D1 );if (substream->stream ==SNDRV_PCM_STREAM_PLAYBACK ) snd_soc_dapm_stream_event(codec,
@@ -445,8 +445,7 @@ static int soc_pcm_prepare(struct snd_pcm_substream *substream) codec_dai->capture.stream_name, SND_SOC_DAPM_STREAM_START); -if (codec->dapm_event) - codec->dapm_event (codec,SNDRV_CTL_POWER_D0 ); +snd_soc_dapm_device_event (socdev,SNDRV_CTL_POWER_D0 );if (codec_dai->dai_ops.digital_mute) codec_dai->dai_ops.digital_mute (codec_dai,0 );
sound/soc/soc-dapm.c +23 -0
@@ -1280,6 +1280,29 @@ int snd_soc_dapm_stream_event(struct snd_soc_codec *codec,EXPORT_SYMBOL_GPL (snd_soc_dapm_stream_event);/** +* snd_soc_dapm_device_event - send a device event to the dapm core +* @socdev: audio device +* @event: device event +* +* Sends a device event to the dapm core. The core then makes any +* necessary machine or codec power changes.. +* +* Returns 0 for success else error. +*/ +int snd_soc_dapm_device_event (struct snd_soc_device *socdev,int event) +{ +struct snd_soc_codec *codec = socdev->codec; +struct snd_soc_machine *machine = socdev->machine; + +if (machine->dapm_event) + machine->dapm_event (machine, event); +if (codec->dapm_event) + codec->dapm_event (codec, event); +return 0 ; +} +EXPORT_SYMBOL_GPL (snd_soc_dapm_device_event); + +/** * snd_soc_dapm_set_endpoint - set audio endpoint status * @codec: audio codec * @endpoint: audio signal endpoint (or start point)
Signature · 1 line
1.5.3.8
PATCH 4/17 ASoC TLV support.
From: Philipp Zabel <[email protected]>
--- include/sound/soc-dapm.h | 24 ++++++++++++++--- include/sound/soc.h | 58 +++++++++++++++++++++++++++++++++------- sound/soc/soc-core.c | 66 ++++++++++++++++++++++++++------------------- sound/soc/soc-dapm.c | 14 +++++---- 4 files changed, 114 insertions(+), 48 deletions(-)
include/sound/soc-dapm.h +20 -4
@@ -131,18 +131,34 @@ .shift = wshift, .invert = winvert}/* dapm kcontrol types */ -#define SOC_DAPM_SINGLE (xname, reg, shift, mask, invert) \ +#define SOC_DAPM_SINGLE (xname, reg, shift, max, invert) \ { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, \ .info = snd_soc_info_volsw, \ .get = snd_soc_dapm_get_volsw, .put = snd_soc_dapm_put_volsw, \ - .private_value = SOC_SINGLE_VALUE(reg, shift, mask, invert) } -#define SOC_DAPM_DOUBLE (xname, reg, shift_left, shift_right, mask, invert, \ + .private_value = SOC_SINGLE_VALUE(reg, shift, max, invert) } +#define SOC_DAPM_DOUBLE (xname, reg, shift_left, shift_right, max, invert, \ power) \ { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = (xname), \ .info = snd_soc_info_volsw, \ .get = snd_soc_dapm_get_volsw, .put = snd_soc_dapm_put_volsw, \ .private_value = (reg) | ((shift_left) << 8) | ((shift_right) << 12) |\ - ((mask) << 16) | ((invert) << 24) } + ((max) << 16) | ((invert) << 24) } +#define SOC_DAPM_SINGLE_TLV (xname, reg, shift, max, invert, tlv_array) \ +{ .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, \ + .info = snd_soc_info_volsw, \ + .access = SNDRV_CTL_ELEM_ACCESS_TLV_READ | SNDRV_CTL_ELEM_ACCESS_READWRITE,\ + .tlv.p = (tlv_array), \ + .get = snd_soc_dapm_get_volsw, .put = snd_soc_dapm_put_volsw, \ + .private_value = SOC_SINGLE_VALUE(reg, shift, max, invert) } +#define SOC_DAPM_DOUBLE_TLV (xname, reg, shift_left, shift_right, max, invert, \ + power, tlv_array) \ +{ .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = (xname), \ + .access = SNDRV_CTL_ELEM_ACCESS_TLV_READ | SNDRV_CTL_ELEM_ACCESS_READWRITE,\ + .tlv.p = (tlv_array), \ + .info = snd_soc_info_volsw, \ + .get = snd_soc_dapm_get_volsw, .put = snd_soc_dapm_put_volsw, \ + .private_value = (reg) | ((shift_left) << 8) | ((shift_right) << 12) |\ + ((max) << 16) | ((invert) << 24) }#define SOC_DAPM_ENUM (xname, xenum) \ { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, \ .info = snd_soc_info_enum_double, \
include/sound/soc.h +48 -10
84 lines not on this page · raw message
sound/soc/soc-core.c +38 -28
159 lines not on this page · raw message
sound/soc/soc-dapm.c +8 -6
@@ -1019,8 +1019,9 @@ int snd_soc_dapm_get_volsw(struct snd_kcontrol *kcontrol,int reg = kcontrol->private_value &0xff ;int shift = (kcontrol->private_value >>8 ) &0x0f ;int rshift = (kcontrol->private_value >>12 ) &0x0f ; -int mask = (kcontrol->private_value >>16 ) &0xff ; +int max = (kcontrol->private_value >>16 ) &0xff ;int invert = (kcontrol->private_value >>24 ) &0x01 ; +int mask = (1 <<fls (max)) -1 ;/* return the saved value if we are powered down */ if (widget->id == snd_soc_dapm_pga && !widget->power) {
@@ -1035,10 +1036,10 @@ int snd_soc_dapm_get_volsw(struct snd_kcontrol *kcontrol, (snd_soc_read (widget->codec, reg) >> rshift) & mask;if (invert) { ucontrol->value.integer.value[0 ] = - mask - ucontrol->value.integer.value[0 ]; + max - ucontrol->value.integer.value[0 ];if (shift != rshift) ucontrol->value.integer.value[1 ] = - mask - ucontrol->value.integer.value[1 ]; + max - ucontrol->value.integer.value[1 ]; }return 0 ;
@@ -1061,7 +1062,8 @@ int snd_soc_dapm_put_volsw(struct snd_kcontrol *kcontrol,int reg = kcontrol->private_value &0xff ;int shift = (kcontrol->private_value >>8 ) &0x0f ;int rshift = (kcontrol->private_value >>12 ) &0x0f ; -int mask = (kcontrol->private_value >>16 ) &0xff ; +int max = (kcontrol->private_value >>16 ) &0xff ; +int mask = (1 <<fls (max)) -1 ;int invert = (kcontrol->private_value >>24 ) &0x01 ;unsigned short val, val2, val_mask;int ret;
@@ -1069,13 +1071,13 @@ int snd_soc_dapm_put_volsw(struct snd_kcontrol *kcontrol, val = (ucontrol->value.integer.value[0 ] & mask);if (invert) - val = mask - val; + val = max - val; val_mask = mask << shift; val = val << shift;if (shift != rshift) { val2 = (ucontrol->value.integer.value[1 ] & mask);if (invert) - val2 = mask - val2; + val2 = max - val2; val_mask |= mask << rshift; val |= val2 << rshift; }
Signature · 1 line
1.5.3.8
PATCH 5/17 Add D1 power event to power down event sequence
From: Liam Girdwood <[email protected]>
--- sound/soc/soc-core.c | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-)
sound/soc/soc-core.c +8 -0
@@ -288,6 +288,14 @@ static void close_delayed_work(struct work_struct *work)/* are we waiting on this codec DAI stream */ if (codec_dai->pop_wait ==1 ) { +/* power down the codec to D1 if no longer active */ +if (codec->active ==0 ) { +dbg ("pop wq D1 %s %s\n" , codec->name, + codec_dai->playback.stream_name); +snd_soc_dapm_device_event (socdev, +SNDRV_CTL_POWER_D1 ); + } + codec_dai->pop_wait =0 ; snd_soc_dapm_stream_event(codec, codec_dai->playback.stream_name,
Signature · 1 line
1.5.3.8
PATCH 6/17 Ensure PCMs are suspended
From: Liam Girdwood <[email protected]>
This fixes a bug whereby PCMs were not being suspended when the rest of the audio subsystem was suspended.
--- include/sound/soc.h | 3 +++ sound/soc/soc-core.c | 5 +++++ 2 files changed, 8 insertions(+), 0 deletions(-)
include/sound/soc.h +3 -0
@@ -448,6 +448,9 @@ struct snd_soc_dai_link {/* codec/machine specific init - e.g. add machine controls */ int (*init)(struct snd_soc_codec *codec); + +/* DAI pcm */ +struct snd_pcm *pcm; };/* SoC machine */
sound/soc/soc-core.c +5 -0
@@ -646,6 +646,10 @@ static int soc_suspend(struct platform_device *pdev, pm_message_t state) dai->dai_ops.digital_mute (dai,1 ); } +/* suspend all pcms */ +for (i =0 ; i < machine->num_links; i++) +snd_pcm_suspend_all (machine->dai_link[i].pcm); +if (machine->suspend_pre) machine->suspend_pre (pdev, state);
@@ -880,6 +884,7 @@ static int soc_new_pcm(struct snd_soc_device *socdev,return ret; } + dai_link->pcm = pcm; pcm->private_data = rtd; soc_pcm_ops.mmap = socdev->platform->pcm_ops->mmap; soc_pcm_ops.pointer = socdev->platform->pcm_ops->pointer;
Signature · 1 line
1.5.3.8
PATCH 7/17 Fix power switching support for DAPM_SWITCH widgets.
From: Milan plzik <[email protected]>
--- sound/soc/soc-dapm.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-)
sound/soc/soc-dapm.c +3 -2
@@ -692,7 +692,7 @@ static int dapm_mux_update_power(struct snd_soc_dapm_widget *widget,return 0 ; } -/* test and update the power status of a mixer widget */ +/* test and update the power status of a mixer or switch widget */ static int dapm_mixer_update_power (struct snd_soc_dapm_widget *widget,struct snd_kcontrol *kcontrol,int reg,int val_mask,int val,int invert)
@@ -700,7 +700,8 @@ static int dapm_mixer_update_power(struct snd_soc_dapm_widget *widget,struct snd_soc_dapm_path *path;int found =0 ; -if (widget->id != snd_soc_dapm_mixer) +if (widget->id != snd_soc_dapm_mixer && + widget->id != snd_soc_dapm_switch)return -ENODEV ;if (!snd_soc_test_bits (widget->codec, reg, val_mask, val))
Signature · 1 line
1.5.3.8
PATCH 8/17 Clean up tabs.
From: Liam Girdwood <[email protected]>
--- sound/soc/soc-dapm.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
sound/soc/soc-dapm.c +2 -2
@@ -1298,9 +1298,9 @@ int snd_soc_dapm_device_event(struct snd_soc_device *socdev, int event)struct snd_soc_machine *machine = socdev->machine;if (machine->dapm_event) - machine->dapm_event (machine, event); + machine->dapm_event (machine, event);if (codec->dapm_event) - codec->dapm_event (codec, event); + codec->dapm_event (codec, event);return 0 ; }EXPORT_SYMBOL_GPL (snd_soc_dapm_device_event);
Signature · 1 line
1.5.3.8
PATCH 9/17 Add support for passing kcontrols with events.
From: Liam Girdwood <[email protected]>
--- include/sound/soc-dapm.h | 2 +- sound/soc/soc-dapm.c | 40 +++++++++++++++++++++++++++------------- 2 files changed, 28 insertions(+), 14 deletions(-)
include/sound/soc-dapm.h +1 -1
@@ -289,7 +289,7 @@ struct snd_soc_dapm_widget {/* external events */ unsigned short event_flags;/* flags to specify event types */ -int (*event)(struct snd_soc_dapm_widget *,int ); +int (*event)(struct snd_soc_dapm_widget *,struct snd_kcontrol *,int );/* kcontrols that relate to this widget */ int num_kcontrols;
sound/soc/soc-dapm.c +27 -13
@@ -524,11 +524,13 @@ static int dapm_power_widgets(struct snd_soc_codec *codec, int event)continue ;if (event ==SND_SOC_DAPM_STREAM_START ) { - ret = w->event (w,SND_SOC_DAPM_PRE_PMU ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_PRE_PMU );if (ret <0 )return ret; }else if (event ==SND_SOC_DAPM_STREAM_STOP ) { - ret = w->event (w,SND_SOC_DAPM_PRE_PMD ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_PRE_PMD );if (ret <0 )return ret; }
@@ -539,11 +541,13 @@ static int dapm_power_widgets(struct snd_soc_codec *codec, int event)continue ;if (event ==SND_SOC_DAPM_STREAM_START ) { - ret = w->event (w,SND_SOC_DAPM_POST_PMU ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_POST_PMU );if (ret <0 )return ret; }else if (event ==SND_SOC_DAPM_STREAM_STOP ) { - ret = w->event (w,SND_SOC_DAPM_POST_PMD ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_POST_PMD );if (ret <0 )return ret; }
@@ -567,26 +571,30 @@ static int dapm_power_widgets(struct snd_soc_codec *codec, int event)if (power) {/* power up event */ if (w->event_flags &SND_SOC_DAPM_PRE_PMU ) { - ret = w->event (w,SND_SOC_DAPM_PRE_PMU ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_PRE_PMU );if (ret <0 )return ret; }dapm_update_bits (w);if (w->event_flags &SND_SOC_DAPM_POST_PMU ){ - ret = w->event (w,SND_SOC_DAPM_POST_PMU ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_POST_PMU );if (ret <0 )return ret; } }else {/* power down event */ if (w->event_flags &SND_SOC_DAPM_PRE_PMD ) { - ret = w->event (w,SND_SOC_DAPM_PRE_PMD ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_PRE_PMD );if (ret <0 )return ret; }dapm_update_bits (w);if (w->event_flags &SND_SOC_DAPM_POST_PMD ) { - ret = w->event (w,SND_SOC_DAPM_POST_PMD ); + ret = w->event (w, +NULL ,SND_SOC_DAPM_POST_PMD );if (ret <0 )return ret; }
@@ -1096,13 +1104,17 @@ int snd_soc_dapm_put_volsw(struct snd_kcontrol *kcontrol,dapm_mixer_update_power (widget, kcontrol, reg, val_mask, val, invert);if (widget->event) {if (widget->event_flags &SND_SOC_DAPM_PRE_REG ) { - ret = widget->event (widget,SND_SOC_DAPM_PRE_REG ); -if (ret <0 ) + ret = widget->event (widget, kcontrol, +SND_SOC_DAPM_PRE_REG ); +if (ret <0 ) { + ret =1 ; goto out; + } } ret =snd_soc_update_bits (widget->codec, reg, val_mask, val);if (widget->event_flags &SND_SOC_DAPM_POST_REG ) - ret = widget->event (widget,SND_SOC_DAPM_POST_REG ); + ret = widget->event (widget, kcontrol, +SND_SOC_DAPM_POST_REG ); }else ret =snd_soc_update_bits (widget->codec, reg, val_mask, val);
@@ -1177,13 +1189,15 @@ int snd_soc_dapm_put_enum_double(struct snd_kcontrol *kcontrol,dapm_mux_update_power (widget, kcontrol, mask, mux, e);if (widget->event) {if (widget->event_flags &SND_SOC_DAPM_PRE_REG ) { - ret = widget->event (widget,SND_SOC_DAPM_PRE_REG ); + ret = widget->event (widget, + kcontrol,SND_SOC_DAPM_PRE_REG );if (ret <0 ) goto out; } ret =snd_soc_update_bits (widget->codec, e->reg, mask, val);if (widget->event_flags &SND_SOC_DAPM_POST_REG ) - ret = widget->event (widget,SND_SOC_DAPM_POST_REG ); + ret = widget->event (widget, + kcontrol,SND_SOC_DAPM_POST_REG ); }else ret =snd_soc_update_bits (widget->codec, e->reg, mask, val);
Signature · 1 line
1.5.3.8
PATCH 10/17 Don't lock the codec list in snd_soc_dapm_new_widgets()
snd_soc_dapm_new_widgets() takes the codec lock when adding new widgets, causing lockdep warnings when applications later call down through ALSA to adjust controls. Since widgets are only added during probe this lock should be unneeded so don't take it.
Thanks to Dmitry Baryshkov <[email protected]> for reporting this issue.
--- sound/soc/soc-dapm.c | 2 -- 1 files changed, 0 insertions(+), 2 deletions(-)
sound/soc/soc-dapm.c +0 -2
@@ -972,7 +972,6 @@ int snd_soc_dapm_new_widgets(struct snd_soc_codec *codec) {struct snd_soc_dapm_widget *w; -mutex_lock (&codec->mutex);list_for_each_entry (w, &codec->dapm_widgets, list) {if (w->new)
@@ -1007,7 +1006,6 @@ int snd_soc_dapm_new_widgets(struct snd_soc_codec *codec) }dapm_power_widgets (codec,SND_SOC_DAPM_STREAM_NOP ); -mutex_unlock (&codec->mutex);return 0 ; }EXPORT_SYMBOL_GPL (snd_soc_dapm_new_widgets);
Signature · 1 line
1.5.3.8
PATCH 11/17 Bump ASoC core version number
--- include/sound/soc.h | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
include/sound/soc.h +1 -1
@@ -22,7 +22,7 @@#include <sound/control.h> #include <sound/ac97_codec.h> -#define SND_SOC_VERSION "0.13.1" +#define SND_SOC_VERSION "0.13.2"/* * Convenience kcontrol builders
Signature · 1 line
1.5.3.8
PATCH 12/17 Initial WM8753 TLV support for capture mixer.
From: Liam Girdwood <[email protected]>
--- sound/soc/codecs/wm8753.c | 7 +++++-- 1 files changed, 5 insertions(+), 2 deletions(-)
sound/soc/codecs/wm8753.c +5 -2
@@ -48,6 +48,7 @@#include <sound/soc.h> #include <sound/soc-dapm.h> #include <sound/initval.h> +#include <sound/tlv.h> #include <asm/div64.h> #include "wm8753.h"
@@ -258,6 +259,8 @@ static int wm8753_set_dai(struct snd_kcontrol *kcontrol,return 1 ; } +static const DECLARE_TLV_DB_LINEAR(rec_mix_tlv, -1500, 600); +static const struct snd_kcontrol_new wm8753_snd_controls[] = {SOC_DOUBLE_R ("PCM Volume" ,WM8753_LDAC ,WM8753_RDAC ,0 ,255 ,0 ),
@@ -287,8 +290,8 @@ SOC_SINGLE("Bass Volume", WM8753_BASS, 0, 15, 1),SOC_SINGLE ("Treble Volume" ,WM8753_TREBLE ,0 ,15 ,1 ),SOC_ENUM ("Treble Cut-off" , wm8753_enum[2 ]), -SOC_DOUBLE ("Sidetone Capture Volume" ,WM8753_RECMIX1 ,0 ,4 ,7 ,1 ), -SOC_SINGLE ("Voice Sidetone Capture Volume" ,WM8753_RECMIX2 ,0 ,7 ,1 ), +SOC_DOUBLE_TLV ("Sidetone Capture Volume" ,WM8753_RECMIX1 ,0 ,4 ,7 ,1 , rec_mix_tlv), +SOC_SINGLE_TLV ("Voice Sidetone Capture Volume" ,WM8753_RECMIX2 ,0 ,7 ,1 , rec_mix_tlv),SOC_DOUBLE_R ("Capture Volume" ,WM8753_LINVOL ,WM8753_RINVOL ,0 ,63 ,0 ), SOC_DOUBLE_R("Capture ZC Switch", WM8753_LINVOL, WM8753_RINVOL, 6, 1, 0),
Signature · 1 line
1.5.3.8
PATCH 13/17 Reinitialise DMA on every resume
From: Graeme Gregory <[email protected]>
This one changes the DMA initialisation as it turns out the DMA driver in s3c24xx doesnt store registers between suspend/resume so you have to re-initialise the channels on every resume.
--- sound/soc/s3c24xx/s3c24xx-pcm.c | 48 ++++++++++++++++++++------------------ 1 files changed, 25 insertions(+), 23 deletions(-)
sound/soc/s3c24xx/s3c24xx-pcm.c +25 -23
@@ -49,7 +49,9 @@ static const struct snd_pcm_hardware s3c24xx_pcm_hardware = { .info = SNDRV_PCM_INFO_INTERLEAVED |SNDRV_PCM_INFO_BLOCK_TRANSFER |SNDRV_PCM_INFO_MMAP | - SNDRV_PCM_INFO_MMAP_VALID, +SNDRV_PCM_INFO_MMAP_VALID | +SNDRV_PCM_INFO_PAUSE | + SNDRV_PCM_INFO_RESUME, .formats =SNDRV_PCM_FMTBIT_S16_LE |SNDRV_PCM_FMTBIT_U16_LE | SNDRV_PCM_FMTBIT_U8 |
@@ -176,28 +178,6 @@ static int s3c24xx_pcm_hw_params(struct snd_pcm_substream *substream, } } -/* channel needs configuring for mem=>device, increment memory addr, -* sync to pclk, half-word transfers to the IIS-FIFO. */ -if (substream->stream ==SNDRV_PCM_STREAM_PLAYBACK ) { -s3c2410_dma_devconfig (prtd->params->channel, -S3C2410_DMASRC_MEM ,S3C2410_DISRCC_INC | -S3C2410_DISRCC_APB , prtd->params->dma_addr); - -s3c2410_dma_config (prtd->params->channel, - prtd->params->dma_size, -S3C2410_DCON_SYNC_PCLK | -S3C2410_DCON_HANDSHAKE ); - }else { -s3c2410_dma_config (prtd->params->channel, - prtd->params->dma_size, -S3C2410_DCON_HANDSHAKE | -S3C2410_DCON_SYNC_PCLK ); - -s3c2410_dma_devconfig (prtd->params->channel, -S3C2410_DMASRC_HW ,0x3 , - prtd->params->dma_addr); - } -s3c2410_dma_set_buffdone_fn (prtd->params->channel, s3c24xx_audio_buffdone);
@@ -246,6 +226,28 @@ static int s3c24xx_pcm_prepare(struct snd_pcm_substream *substream)if (!prtd->params)return 0 ; +/* channel needs configuring for mem=>device, increment memory addr, +* sync to pclk, half-word transfers to the IIS-FIFO. */ +if (substream->stream ==SNDRV_PCM_STREAM_PLAYBACK ) { +s3c2410_dma_devconfig (prtd->params->channel, +S3C2410_DMASRC_MEM ,S3C2410_DISRCC_INC | +S3C2410_DISRCC_APB , prtd->params->dma_addr); + +s3c2410_dma_config (prtd->params->channel, + prtd->params->dma_size, +S3C2410_DCON_SYNC_PCLK | +S3C2410_DCON_HANDSHAKE ); + }else { +s3c2410_dma_config (prtd->params->channel, + prtd->params->dma_size, +S3C2410_DCON_HANDSHAKE | +S3C2410_DCON_SYNC_PCLK ); + +s3c2410_dma_devconfig (prtd->params->channel, +S3C2410_DMASRC_HW ,0x3 , + prtd->params->dma_addr); + } +/* flush the DMA channel */ s3c2410_dma_ctrl (prtd->params->channel,S3C2410_DMAOP_FLUSH ); prtd->dma_loaded =0 ;
Signature · 1 line
1.5.3.8
PATCH 14/17 Support suspend and resume of the I2S interface on s3c24xx.
From: Graeme Gregory <[email protected]>
--- sound/soc/s3c24xx/s3c24xx-i2s.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 files changed, 38 insertions(+), 0 deletions(-)
sound/soc/s3c24xx/s3c24xx-i2s.c +38 -0
@@ -75,6 +75,10 @@ static struct s3c24xx_pcm_dma_params s3c24xx_i2s_pcm_stereo_in = {struct s3c24xx_i2s_info {void __iomem *regs;struct clk *iis_clk; +u32 iiscon; +u32 iismod; +u32 iisfcon; +u32 iispsr; };static struct s3c24xx_i2s_info s3c24xx_i2s;
@@ -405,6 +409,38 @@ static int s3c24xx_i2s_probe(struct platform_device *pdev)return 0 ; } +#ifdef CONFIG_PM +int s3c24xx_i2s_suspend (struct platform_device *pdev, +struct snd_soc_cpu_dai *cpu_dai) +{ + s3c24xx_i2s.iiscon =readl (s3c24xx_i2s.regs +S3C2410_IISCON ); + s3c24xx_i2s.iismod =readl (s3c24xx_i2s.regs +S3C2410_IISMOD ); + s3c24xx_i2s.iisfcon =readl (s3c24xx_i2s.regs +S3C2410_IISFCON ); + s3c24xx_i2s.iispsr =readl (s3c24xx_i2s.regs +S3C2410_IISPSR ); + +clk_disable (s3c24xx_i2s.iis_clk); + +return 0 ; +} + +int s3c24xx_i2s_resume (struct platform_device *pdev, +struct snd_soc_cpu_dai *cpu_dai) +{ +clk_enable (s3c24xx_i2s.iis_clk); + +writel (s3c24xx_i2s.iiscon, s3c24xx_i2s.regs +S3C2410_IISCON ); +writel (s3c24xx_i2s.iismod, s3c24xx_i2s.regs +S3C2410_IISMOD ); +writel (s3c24xx_i2s.iisfcon, s3c24xx_i2s.regs +S3C2410_IISFCON ); +writel (s3c24xx_i2s.iispsr, s3c24xx_i2s.regs +S3C2410_IISPSR ); + +return 0 ; +} +#else +#define s3c24xx_i2s_suspend NULL +#define s3c24xx_i2s_resume NULL +#endif + +#define S3C24XX_I2S_RATES \ (SNDRV_PCM_RATE_8000 | SNDRV_PCM_RATE_11025 | SNDRV_PCM_RATE_16000 | \ SNDRV_PCM_RATE_22050 | SNDRV_PCM_RATE_32000 | SNDRV_PCM_RATE_44100 | \
@@ -415,6 +451,8 @@ struct snd_soc_cpu_dai s3c24xx_i2s_dai = { .id = 0, .type = SND_SOC_DAI_I2S, .probe = s3c24xx_i2s_probe, + .suspend = s3c24xx_i2s_suspend, + .resume = s3c24xx_i2s_resume, .playback = { .channels_min = 2, .channels_max = 2,
Signature · 1 line
1.5.3.8
PATCH 15/17 ASoC: S3C2412 IIS driver
From: Ben Dooks <[email protected]>
S3C2412 SoC IIS support for ALSA/ASoC
--- sound/soc/s3c24xx/Kconfig | 3 + sound/soc/s3c24xx/Makefile | 2 + sound/soc/s3c24xx/s3c2412-i2s.c | 682 +++++++++++++++++++++++++++++++++++++++ sound/soc/s3c24xx/s3c2412-i2s.h | 38 +++ 4 files changed, 725 insertions(+), 0 deletions(-) create mode 100644 sound/soc/s3c24xx/s3c2412-i2s.c create mode 100644 sound/soc/s3c24xx/s3c2412-i2s.h
sound/soc/s3c24xx/Kconfig +3 -0
@@ -10,6 +10,9 @@ config SND_S3C24XX_SOC config SND_S3C24XX_SOC_I2S tristate +config SND_S3C2412_SOC_I2S + tristate + config SND_S3C2443_SOC_AC97 tristate select AC97_BUS
sound/soc/s3c24xx/Makefile +2 -0
@@ -1,11 +1,13 @@ # S3c24XX Platform Support snd-soc-s3c24xx-objs := s3c24xx-pcm.o snd-soc-s3c24xx-i2s-objs := s3c24xx-i2s.o +snd-soc-s3c2412-i2s-objs := s3c2412-i2s.o snd-soc-s3c2443-ac97-objs := s3c2443-ac97.o obj-$(CONFIG_SND_S3C24XX_SOC) += snd-soc-s3c24xx.o obj-$(CONFIG_SND_S3C24XX_SOC_I2S) += snd-soc-s3c24xx-i2s.o obj-$(CONFIG_SND_S3C2443_SOC_AC97) += snd-soc-s3c2443-ac97.o +obj-$(CONFIG_SND_S3C2412_SOC_I2S) += snd-soc-s3c2412-i2s.o # S3C24XX Machine Support snd-soc-neo1973-wm8753-objs := neo1973_wm8753.o
sound/soc/s3c24xx/s3c2412-i2s.c new file +682 -0
682 lines not on this page · raw message
sound/soc/s3c24xx/s3c2412-i2s.h new file +38 -0
@@ -0,0 +1,38 @@ +/* sound/soc/s3c24xx/s3c2412-i2s.c +* +* ALSA Soc Audio Layer - S3C2412 I2S driver +* +* Copyright (c) 2007 Simtec Electronics +* http://armlinux.simtec.co.uk/ +* Ben Dooks <[email protected]> +* +* 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 +* option) any later version. +*/ + +#ifndef __SND_SOC_S3C24XX_S3C2412_I2S_H +#define __SND_SOC_S3C24XX_S3C2412_I2S_H __FILE__ + +#define S3C2412_DIV_BCLK (1) +#define S3C2412_DIV_RCLK (2) +#define S3C2412_DIV_PRESCALER (3) + +#define S3C2412_CLKSRC_PCLK (0) +#define S3C2412_CLKSRC_I2SCLK (1) + +extern struct clk *s3c2412_get_iisclk (void ); + +extern struct snd_soc_cpu_dai s3c2412_i2s_dai; + +struct s3c2412_rate_calc { +unsigned int clk_div;/* for prescaler */ +unsigned int fs_div;/* for root frame clock */ +}; + +extern int s3c2412_iis_calc_rate (struct s3c2412_rate_calc *info, +unsigned int *fstab, +unsigned int rate,struct clk *clk); + +#endif /* __SND_SOC_S3C24XX_S3C2412_I2S_H */
Signature · 1 line
1.5.3.8
PATCH 16/17 S3C2412: suspend and resume support
From: Ben Dooks <[email protected]>
Support for suspend/resume for the S3C2412 ASoC IIS core driver.
--- sound/soc/s3c24xx/s3c2412-i2s.c | 63 +++++++++++++++++++++++++++++++++++++++ 1 files changed, 63 insertions(+), 0 deletions(-)
sound/soc/s3c24xx/s3c2412-i2s.c +63 -0
@@ -80,6 +80,10 @@ struct s3c2412_i2s_info {struct clk *iis_clk;struct clk *iis_pclk;struct clk *iis_cclk; + +u32 suspend_iismod; +u32 suspend_iiscon; +u32 suspend_iispsr; };static struct s3c2412_i2s_info s3c2412_i2s;
@@ -642,6 +646,63 @@ static int s3c2412_i2s_probe(struct platform_device *pdev)return 0 ; } +#ifdef CONFIG_PM +static int s3c2412_i2s_suspend (struct platform_device *dev, +struct snd_soc_cpu_dai *dai) +{ +struct s3c2412_i2s_info *i2s = &s3c2412_i2s; +u32 iismod; + +if (dai->active) { + i2s->suspend_iismod =readl (i2s->regs +S3C2412_IISMOD ); + i2s->suspend_iiscon =readl (i2s->regs +S3C2412_IISCON ); + i2s->suspend_iispsr =readl (i2s->regs +S3C2412_IISPSR ); + +/* some basic suspend checks */ + + iismod =readl (i2s->regs +S3C2412_IISMOD ); + +if (iismod &S3C2412_IISCON_RXDMA_ACTIVE ) +dev_warn (&dev->dev,"%s: RXDMA active?\n" , __func__); + +if (iismod &S3C2412_IISCON_TXDMA_ACTIVE ) +dev_warn (&dev->dev,"%s: TXDMA active?\n" , __func__); + +if (iismod &S3C2412_IISCON_IIS_ACTIVE ) +dev_warn (&dev->dev,"%s: IIS active\n" , __func__); + } + +return 0 ; +} + +static int s3c2412_i2s_resume (struct platform_device *pdev, +struct snd_soc_cpu_dai *dai) +{ +struct s3c2412_i2s_info *i2s = &s3c2412_i2s; + +dev_info (&pdev->dev,"dai_active %d, IISMOD %08x, IISCON %08x\n" , + dai->active, i2s->suspend_iismod, i2s->suspend_iiscon); + +if (dai->active) { +writel (i2s->suspend_iiscon, i2s->regs +S3C2412_IISCON ); +writel (i2s->suspend_iismod, i2s->regs +S3C2412_IISMOD ); +writel (i2s->suspend_iispsr, i2s->regs +S3C2412_IISPSR ); + +writel (S3C2412_IISFIC_RXFLUSH |S3C2412_IISFIC_TXFLUSH , + i2s->regs +S3C2412_IISFIC ); + +ndelay (250 ); +writel (0x0 , i2s->regs +S3C2412_IISFIC ); + + } + +return 0 ; +} +#else +#define s3c2412_i2s_suspend NULL +#define s3c2412_i2s_resume NULL +#endif /* CONFIG_PM */ +#define S3C2412_I2S_RATES \ (SNDRV_PCM_RATE_8000 | SNDRV_PCM_RATE_11025 | SNDRV_PCM_RATE_16000 | \ SNDRV_PCM_RATE_22050 | SNDRV_PCM_RATE_32000 | SNDRV_PCM_RATE_44100 | \
@@ -652,6 +713,8 @@ struct snd_soc_cpu_dai s3c2412_i2s_dai = { .id = 0, .type = SND_SOC_DAI_I2S, .probe = s3c2412_i2s_probe, + .suspend= s3c2412_i2s_suspend, + .resume = s3c2412_i2s_resume, .playback = { .channels_min = 2, .channels_max = 2,
Signature · 1 line
1.5.3.8
PATCH 17/17 Preliminary ac97 drivers for Toshiba e800 PDAs
From: Ian Molton <[email protected]>
Currently only the AUX channel is used (touchscreen)
--- sound/soc/pxa/Kconfig | 9 ++++ sound/soc/pxa/Makefile | 2 + sound/soc/pxa/e800_wm9712.c | 90 +++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 101 insertions(+), 0 deletions(-) create mode 100644 sound/soc/pxa/e800_wm9712.c
sound/soc/pxa/Kconfig +9 -0
@@ -53,3 +53,12 @@ config SND_PXA2XX_SOC_TOSA help Say Y if you want to add support for SoC audio on Sharp Zaurus SL-C6000x models (Tosa). + +config SND_PXA2XX_SOC_E800 + tristate "SoC AC97 Audio support for e800" + depends on SND_PXA2XX_SOC && MACH_E800 + select SND_SOC_WM9712 + select SND_PXA2XX_SOC_AC97 + help + Say Y if you want to add support for SoC audio on the + Toshiba e800 PDA
sound/soc/pxa/Makefile +2 -0
@@ -11,10 +11,12 @@ obj-$(CONFIG_SND_PXA2XX_SOC_I2S) += snd-soc-pxa2xx-i2s.o snd-soc-corgi-objs := corgi.o snd-soc-poodle-objs := poodle.o snd-soc-tosa-objs := tosa.o +snd-soc-e800-objs := e800_wm9712.o snd-soc-spitz-objs := spitz.o obj-$(CONFIG_SND_PXA2XX_SOC_CORGI) += snd-soc-corgi.o obj-$(CONFIG_SND_PXA2XX_SOC_POODLE) += snd-soc-poodle.o obj-$(CONFIG_SND_PXA2XX_SOC_TOSA) += snd-soc-tosa.o +obj-$(CONFIG_SND_PXA2XX_SOC_E800) += snd-soc-e800.o obj-$(CONFIG_SND_PXA2XX_SOC_SPITZ) += snd-soc-spitz.o
sound/soc/pxa/e800_wm9712.c new file +90 -0
90 lines not on this page · raw message
Signature · 1 line
1.5.3.8
Re: [PATCH 10/17] Don't lock the codec list in snd_soc_dapm_new_widgets()
In reply to Mark BrownHi,
I'm sorry, but I tested this patch only now. And I just got another message from lockdep:
=======================================================
[ INFO: possible circular locking dependency detected ] 2.6.24-rc7-dirty #79
-------------------------------------------------------
alsactl/953 is trying to acquire lock:
(&codec->mutex){--..}, at: [<c01d03f0>] snd_soc_dapm_put_volsw+0xc4/0x218
but task is already holding lock:
(&card->controls_rwsem){----}, at: [<c01bcfac>] snd_ctl_elem_write+0x24/0x13c
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (&card->controls_rwsem){----}:
[<c0058d84>] lock_acquire+0x6c/0x84
[<c0243eb4>] down_write+0x30/0x40
[<c01bd28c>] snd_ctl_add+0x54/0x1dc
[<c01d2b78>] tosa_ac97_init+0x60/0xec
[<c01ce5a0>] snd_soc_register_card+0x58/0x210
[<c01d193c>] wm9712_soc_probe+0x1f0/0x280
[<c01cf46c>] soc_probe+0x98/0x188
[<c015e998>] platform_drv_probe+0x20/0x24
[<c015d1dc>] driver_probe_device+0x100/0x1b8
[<c015d2a4>] __device_attach+0x10/0x14
[<c015c3a0>] bus_for_each_drv+0x48/0x84
[<c015d344>] device_attach+0x70/0x9c
[<c015c308>] bus_attach_device+0x38/0x88
[<c015b068>] device_add+0x274/0x4bc
[<c015ee40>] platform_device_add+0x100/0x158
[<c0016e04>] tosa_init+0x60/0x90
[<c00087e4>] kernel_init+0xc8/0x298
[<c0039cf0>] do_exit+0x0/0x760-> #0 (&codec->mutex){--..}:
[<c0058d84>] lock_acquire+0x6c/0x84
[<c0243b48>] mutex_lock_nested+0xf8/0x2bc
[<c01d03f0>] snd_soc_dapm_put_volsw+0xc4/0x218
[<c01bd080>] snd_ctl_elem_write+0xf8/0x13c
[<c01bde2c>] snd_ctl_ioctl+0x720/0xbfc
[<c0099314>] do_ioctl+0x38/0x98
[<c0099620>] vfs_ioctl+0x2ac/0x2dc
[<c0099690>] sys_ioctl+0x40/0x64
[<c001cf60>] ret_fast_syscall+0x0/0x2cother info that might help us debug this:
2 locks held by alsactl/953:
#0: (&card->power_lock){--..}, at: [<c01bde08>] snd_ctl_ioctl+0x6fc/0xbfc #1: (&card->controls_rwsem){----}, at: [<c01bcfac>]
snd_ctl_elem_write+0x24/0x13c stack backtrace: [<c00217d4>] (dump_stack+0x0/0x14) from [<c0056848>] (print_circular_bug_tail+0x78/0x94) [<c00567d0>] (print_circular_bug_tail+0x0/0x94) from [<c0058520>] (__lock_acquire+0x970/0xcf4) r6:c3f02538 r5:00000001 r4:c3f02560 [<c0057bb0>] (__lock_acquire+0x0/0xcf4) from [<c0058d84>] (lock_acquire+0x6c/0x84) [<c0058d18>] (lock_acquire+0x0/0x84) from [<c0243b48>] (mutex_lock_nested+0xf8/0x2bc) r7:c3f02220 r6:60000013 r5:c3d6eda8 r4:c3f9bd04 [<c0243a50>] (mutex_lock_nested+0x0/0x2bc) from [<c01d03f0>] (snd_soc_dapm_put_volsw+0xc4/0x218) [<c01d032c>] (snd_soc_dapm_put_volsw+0x0/0x218) from [<c01bd080>] (snd_ctl_elem_write+0xf8/0x13c) [<c01bcf88>] (snd_ctl_elem_write+0x0/0x13c) from [<c01bde2c>] (snd_ctl_ioctl+0x720/0xbfc) [<c01bd70c>] (snd_ctl_ioctl+0x0/0xbfc) from [<c0099314>] (do_ioctl+0x38/0x98) [<c00992dc>] (do_ioctl+0x0/0x98) from [<c0099620>] (vfs_ioctl+0x2ac/0x2dc) r6:bed526d8 r5:c3ed8288 r4:c3e93140 [<c0099374>] (vfs_ioctl+0x0/0x2dc) from [<c0099690>] (sys_ioctl+0x40/0x64) r9:c3f9a000 r8:c001d108 r6:c2c85513 r5:fffffff7 r4:c3e93140 [<c0099650>] (sys_ioctl+0x0/0x64) from [<c001cf60>] (ret_fast_syscall+0x0/0x2c) r6:00000002 r5:00000000 r4:00000000
2008/1/10, Mark Brown <[email protected]>:
Quoted 37 lines
snd_soc_dapm_new_widgets()takes the codec lock when adding new widgets, causing lockdep warnings when applications later call down through ALSA to adjust controls. Since widgets are only added during probe this lock should be unneeded so don't take it.Thanks to Dmitry Baryshkov <[email protected]> for reporting this issue.
Signed-off-by Mark Brown <[email protected]>Cc Dmitry Baryshkov <[email protected]>--- sound/soc/soc-dapm.c | 2 -- 1 files changed, 0 insertions(+), 2 deletions(-)sound/soc/soc-dapm.c +0
-3@@ -972,7 +972,6 @@ int snd_soc_dapm_new_widgets(struct snd_soc_codec *codec) {struct snd_soc_dapm_widget *w; -mutex_lock (&codec->mutex);list_for_each_entry (w, &codec->dapm_widgets, list) {if (w->new)@@ -1007,7 +1006,6 @@ int snd_soc_dapm_new_widgets(struct snd_soc_codec *codec) }dapm_power_widgets (codec,SND_SOC_DAPM_STREAM_NOP ); -mutex_unlock (&codec->mutex);return 0 ; }EXPORT_SYMBOL_GPL (snd_soc_dapm_new_widgets); --1.5.3.8
Signature · 2 lines
With best wishes Dmitry
Re: [PATCH 10/17] Don't lock the codec list in snd_soc_dapm_new_widgets()
Quoted 2 lines · Dmitry
I'm sorry, but I tested this patch only now. And I just got another message from lockdep:
No problem, I managed to reproduce your original case so I was confident of the fix. This one looks at first glance to be a very similar problem due to unneeded locking during initialisation so it shouldn't cause any issues in practise other than the warning from lockdep. I'll have a proper look on Thursday (I won't be able to tomorrow).
Re: [PATCH 10/17] Don't lock the codec list in snd_soc_dapm_new_widgets()
In reply to DmitryQuoted 2 lines · Dmitry
I'm sorry, but I tested this patch only now. And I just got another message from lockdep:
Could you give this patch a try, please?
sound/soc/soc-core.c +5 -2
@@ -1131,7 +1131,6 @@ int snd_soc_register_card(struct snd_soc_device *socdev)struct snd_soc_machine *machine = socdev->machine;int ret =0 , i, ac97 =0 , err =0 ; -mutex_lock (&codec->mutex);for (i =0 ; i < machine->num_links; i++) {if (socdev->machine->dai_link[i].init) { err = socdev->machine->dai_link[i].init (codec);
@@ -1157,12 +1156,14 @@ int snd_soc_register_card(struct snd_soc_device *socdev) goto out; } +mutex_lock (&codec->mutex);#ifdef CONFIG_SND_SOC_AC97_BUS if (ac97) { ret =soc_ac97_dev_register (codec);if (ret <0 ) {printk (KERN_ERR "asoc: AC97 device register failed\n" );snd_card_free (codec->card); +mutex_unlock (&codec->mutex); goto out; } }
@@ -1175,8 +1176,10 @@ int snd_soc_register_card(struct snd_soc_device *socdev) err =device_create_file (socdev->dev, &dev_attr_codec_reg);if (err <0 )printk (KERN_WARNING "asoc: failed to add codec sysfs entries\n" ); -out: +mutex_unlock (&codec->mutex); + +out:return ret; }EXPORT_SYMBOL_GPL (snd_soc_register_card);
Juli - function of specific GPIOs?
Hi Takashi,
Sorry for distracting you from the constant struggle with HDA.
I would like to provide complete support for the ESI Juli@ card. Informing the ADC (AK5385) about current sample rate via the dedicated GPIOs is fairly simple, as well as the monitoring features (DIGOUT, DIGIN, ANAIN, MUTE + volume control of the remaining DACs).
However, I do not understand functions of the GPIO_FREQ_XXKHZ and GPIO_MULTI_XX GPIOs. The card's user manual says the card can detect incoming SPDIF rate. AK4114 can either read the rate from the input data in professional format only, or detect it by comparing external clock with the input rate. The AK4114 external clock pin does go to the Xilinx CPLD. The GPIOs lead to the CPLD too. Do they in some way control the CPLD to provide AK4114 with external clock so that AK4114 can provide correct input rate data through its registers?
But if so, where would CPLD get the independent clock when ICE1724 is in slave mode, clocked by AK4114? True, the PMCLK output is fed to the CPLD, but that would be slaved to AK4114 too. I guess the only independent clock during the slave mode would have to come directly from the 24.576MHz crystal of ICE, but I could not trace any connection to the CPLD.
Perhaps the mysterious AV73-1 helps, I could not google-out any information on this IC.
The automated detection of incoming rate is important for setting correct sample rates in codecs when running on external clock.
If resolved, I would add similar functionality to the MI/ODI/O card for Prodigy192 too.
I am sorry for bothering you, but the GPIOs named constants in juli.c suggest you had access to some documentation about the card. My tracing and beeping the card ends with the internal CPLD logic.
Thanks a lot for any information or suggestion.
Best regards,
Pavel Hofman.
-----------------
inSITE, s.r.o.
Rubesova 29, 326 00 Plzen, Czech Republic
Tel., fax: +420 - 37 - 74 493 58
GSM: +420 - 603 - 163 973
Email: [email protected]
Re: Juli - function of specific GPIOs?
At Fri, 18 Jan 2008 22:41:51 +0100,
Quoted 10 lines · Pavel Hofman
Hi Takashi,
Sorry for distracting you from the constant struggle with HDA.
I would like to provide complete support for the ESI Juli@ card. Informing the ADC (AK5385) about current sample rate via the dedicated GPIOs is fairly simple, as well as the monitoring features (DIGOUT, DIGIN, ANAIN, MUTE + volume control of the remaining DACs).
Quoted 8 lines
However, I do not understand functions of the
GPIO_FREQ_XXKHZandGPIO_MULTI_XXGPIOs. The card's user manual says the card can detect incoming SPDIF rate. AK4114 can either read the rate from the input data in professional format only, or detect it by comparing external clock with the input rate. The AK4114 external clock pin does go to the Xilinx CPLD. The GPIOs lead to the CPLD too. Do they in some way control the CPLD to provide AK4114 with external clock so that AK4114 can provide correct input rate data through its registers?
Sorry, I have little idea about Juli (and all ESI boards). The datasheet was given only to Jaroslav (cc'ed) at that time.
Takashi
Re: Juli - function of specific GPIOs?
Quoted 22 lines · Takashi Iwai
At Fri, 18 Jan 2008 22:41:51 +0100,
Quoted 9 lines · Pavel Hofman
Hi Takashi,
Sorry for distracting you from the constant struggle with HDA.
I would like to provide complete support for the ESI Juli@ card. Informing the ADC (AK5385) about current sample rate via the dedicated GPIOs is fairly simple, as well as the monitoring features (DIGOUT, DIGIN, ANAIN, MUTE + volume control of the remaining DACs).
Quoted 8 lines
However, I do not understand functions of the
GPIO_FREQ_XXKHZandGPIO_MULTI_XXGPIOs. The card's user manual says the card can detect incoming SPDIF rate. AK4114 can either read the rate from the input data in professional format only, or detect it by comparing external clock with the input rate. The AK4114 external clock pin does go to the Xilinx CPLD. The GPIOs lead to the CPLD too. Do they in some way control the CPLD to provide AK4114 with external clock so that AK4114 can provide correct input rate data through its registers?
Quoted 5 lines
Sorry, I have little idea about Juli (and all ESI boards). The datasheet was given only to Jaroslav (cc'ed) at that time.
Takashi
Takashi,
Thanks a lot for the info and sorry for overlooking the initial changetset author.
Perhaps Jaroslav will find a few minutes to recall the old story, pěkně prosím :)
Regards,
Pavel.
Signature · 8 lines
----------------- inSITE, s.r.o. Rubesova 29, 326 00 Plzen Tel., fax: +420 - 37 - 74 493 58 GSM: +420 - 603 - 163 973 Email: [email protected]
Re: [PATCH 10/17] Don't lock the codec list in snd_soc_dapm_new_widgets()
In reply to Mark BrownHi,
2008/1/18, Mark Brown <[email protected]>:
Quoted 6 lines · Dmitry
Quoted 3 lines
I'm sorry, but I tested this patch only now. And I just got another message from lockdep:
Could you give this patch a try, please?
Thanks, now I've got a clean boot log :)
Signature · 2 lines
With best wishes Dmitry
At Thu, 10 Jan 2008 13:16:59 +0000,
Quoted 6 lines · Mark Brown
Add myself as a point of contact for the ALSA SoC subsystem and add a reference to the development GIT tree.
Signed-off-by Mark Brown <[email protected]>Signed-off-by Liam Girdwood <[email protected]>
Thanks, all patches have been applied to HG tree now.
Takashi