November 2014 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH 0/2] add support for exynos7 I2S controller
by Padmavathi Venna 07 Nov '14

07 Nov '14

I2S on exynos7 includes the below changes - has got support for lower root clock sampling frequencies like 64, 128, 96, 192fs. I2S on previous SoCs supports only 256,512,384 and 768. - supports 7.1CH TDM mode for recording. exynos5 has only 7.1CH TDM mode for playback. - secondary dai works with only external DMA. on previous SoCs secondary dai can work with either both or only internal DMA. - I2S1 controller upgraded to 5.1CH with slighlty modified bit offsets for supporting lower RFS and more no.of BFS values. As internal dma is not mandatory on all SoCs this patchset introduced a quirk for the same and added a structure to hold all the modified bit offsets across all SoCs. This patchset is made based on linux-next master branch and tested with exynos7 base support from the below link. http://www.spinics.net/lists/linux-samsung-soc/msg37047.html Padmavathi Venna (2): ASoC: Samsung: Add quirk for internal DMA ASoC: samsung: add support for exynos7 I2S controller .../devicetree/bindings/sound/samsung-i2s.txt | 15 +- include/linux/platform_data/asoc-s3c.h | 1 + sound/soc/samsung/Kconfig | 2 +- sound/soc/samsung/i2s-regs.h | 10 +- sound/soc/samsung/i2s.c | 228 ++++++++++++++------ 5 files changed, 180 insertions(+), 76 deletions(-) -- 1.7.4.4

2 4

[alsa-devel] Serial MIDI driver for PL011 - as found on BCM2835 hardware
by Kim Taylor 07 Nov '14

07 Nov '14

Hi, I've written a driver for raw MIDI using the PL011. It is obviously based on the snd-serial-u16550 module, with a couple of improvements: - I've implemented a drain() callback. - I've also implemented a half duplex mode based on a timer callback which was necessary for reliable communication with my eMagic Unitor 8 hardware. (Testing with a serial link to other hardware shows that the full duplex mode is also working.) It can be found here: https://github.com/kmtaylor/rpi_patches/blob/master/serial-pl011.c If you think it should be included in the ALSA repository, please let me know what to do next. Regards, Kim

4 6

[alsa-devel] [PATCH 0/6] ASoC/ARM: move Marvell Armada 370 DB to simple-audio-card
by Thomas Petazzoni 07 Nov '14

07 Nov '14

Hello, This set of patches moves the Marvell Armada 370 DB audio complex description from a custom audio machine driver to the simple-audio-card DT binding. The first three patches are meant to go through the ASoC tree, the last three patches are meant to go through the mvebu maintainers tree. Quick description of the patches: - First patch fixes a small issue in the cs42l51 driver following its split in two parts. - Second patch makes the cs42l51 user-selectable by adding a prompt to its Kconfig option. - Third patch gets rid of the machine driver. - Fourth patch removes some now incorrect pin-muxing selection in the Armada 370 DB platform. - Fifth patch actually switches to use the simple-audio-card DT binding on Armada 370 DB. - Sixth patch updates mvebu_v7_defconfig to select the relevant options to get audio to work. Note: obviously, after PATCH 3 is applied, the audio will no longer work until PATCH 5 is applied. But since those patches will anyway go through different trees, it is a bit inevitable. And it is not dramatically important, since audio support on Armada 370 DB is already broken since the split of the CS42L51 driver. Thanks, Thomas Thomas Petazzoni (6): ASoC: cs42l51: re-add of_match_table pointer ASoC: cs42l51: make driver user-selectable ASoC: kirkwood: get rid of armada-370-db driver ARM: mvebu: remove conflicting muxing on Armada 370 DB ARM: mvebu: use simple-card DT binding for audio on Armada 370 DB ARM: mvebu: update mvebu_v7_defconfig for Armada 370 audio arch/arm/boot/dts/armada-370-db.dts | 80 +++++++++++-------- arch/arm/boot/dts/armada-370.dtsi | 1 + arch/arm/configs/mvebu_v7_defconfig | 4 +- sound/soc/codecs/Kconfig | 2 +- sound/soc/codecs/cs42l51-i2c.c | 7 ++ sound/soc/codecs/cs42l51.c | 5 -- sound/soc/kirkwood/Kconfig | 13 +--- sound/soc/kirkwood/Makefile | 4 - sound/soc/kirkwood/armada-370-db.c | 148 ------------------------------------ 9 files changed, 63 insertions(+), 201 deletions(-) delete mode 100644 sound/soc/kirkwood/armada-370-db.c -- 2.0.0

5 17

[alsa-devel] [PATCH 0/8] ASoC: rsnd: add Volume Ramp support
by Kuninori Morimoto 06 Nov '14

06 Nov '14

Hi Mark These patches add Volume Ramp support on Renesas R-Car sound driver. Kuninori Morimoto (8): ASoC: rsnd: tidyup debug information when read/write ASoC: rsnd: tidyup RSND_DVC_VOLUME_NUM to RSND_DVC_CHANNELS ASoC: rsnd: add struct rsnd_dvc_cfg and control DVC settings ASoC: rsnd: control DVC_DVUCR under rsnd_dvc_volume_update() ASoC: rsnd: move DVC_DVUER settings under rsnd_dvc_volume_update() ASoC: rsnd: enable multiple DVC valume settings ASoC: rsnd: enable single DVC valume settings ASoC: rsnd: Add Volume Ramp support sound/soc/sh/rcar/dvc.c | 155 ++++++++++++++++++++++++++++++++++------------ sound/soc/sh/rcar/gen.c | 30 ++++++--- sound/soc/sh/rcar/rsnd.h | 6 ++ 3 files changed, 145 insertions(+), 46 deletions(-) Best regards --- Kuninori Morimoto

2 24

[alsa-devel] [PATCH] ASoC: Intel: mrfld - remove non static definition
by Vinod Koul 06 Nov '14

06 Nov '14

sst_save_shim64() is defined as static in code but header is non static. Since this is not used other than file where defined remove non static definition Reported-by: kbuild test robot <fengguang.wu(a)intel.com> Signed-off-by: Vinod Koul <vinod.koul(a)intel.com> --- sound/soc/intel/sst/sst.h | 2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/sound/soc/intel/sst/sst.h b/sound/soc/intel/sst/sst.h index 3ee555e..2dcbf47 100644 --- a/sound/soc/intel/sst/sst.h +++ b/sound/soc/intel/sst/sst.h @@ -508,8 +508,6 @@ int sst_prepare_and_post_msg(struct intel_sst_drv *sst, size_t mbox_data_len, const void *mbox_data, void **data, bool large, bool fill_dsp, bool sync, bool response); -void sst_save_shim64(struct intel_sst_drv *ctx, void __iomem *shim, - struct sst_shim_regs64 *shim_regs); void sst_process_pending_msg(struct work_struct *work); int sst_assign_pvt_id(struct intel_sst_drv *sst_drv_ctx); void sst_init_stream(struct stream_info *stream, -- 1.7.0.4

2 1

[alsa-devel] [PATCH 1/3] ASoC: ssm4567: Add DAC high-pass-filter control
by Lars-Peter Clausen 06 Nov '14

06 Nov '14

Add a switch which can be used to enable/disable the DAC high-pass-filter. Signed-off-by: Lars-Peter Clausen <lars(a)metafoo.de> --- sound/soc/codecs/ssm4567.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/soc/codecs/ssm4567.c b/sound/soc/codecs/ssm4567.c index 4b5c17f..e1e33d8 100644 --- a/sound/soc/codecs/ssm4567.c +++ b/sound/soc/codecs/ssm4567.c @@ -145,6 +145,8 @@ static const struct snd_kcontrol_new ssm4567_snd_controls[] = { SOC_SINGLE_TLV("Master Playback Volume", SSM4567_REG_DAC_VOLUME, 0, 0xff, 1, ssm4567_vol_tlv), SOC_SINGLE("DAC Low Power Mode Switch", SSM4567_REG_DAC_CTRL, 4, 1, 0), + SOC_SINGLE("DAC High Pass Filter Switch", SSM4567_REG_DAC_CTRL, + 5, 1, 0), }; static const struct snd_soc_dapm_widget ssm4567_dapm_widgets[] = { -- 1.8.0

2 3

[alsa-devel] [PATCH] ALSA: usb-audio: Trigger PCM XRUN at XRUN
by Takashi Iwai 06 Nov '14

06 Nov '14

The usb-audio driver detects XRUN at its complete callback, but the actual code to trigger PCM XRUN is commented out because it caused deadlock in the past. This patch revives the PCM trigger properly. It resulted in more than just enabling snd_pcm_stop(), but it had to deduce the PCM substream with proper NULL checks and holds the stream lock around the call. Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/usb/endpoint.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/sound/usb/endpoint.c b/sound/usb/endpoint.c index 167d0c1643e1..a4679913b0aa 100644 --- a/sound/usb/endpoint.c +++ b/sound/usb/endpoint.c @@ -348,6 +348,8 @@ static void snd_complete_urb(struct urb *urb) { struct snd_urb_ctx *ctx = urb->context; struct snd_usb_endpoint *ep = ctx->ep; + struct snd_pcm_substream *substream; + unsigned long flags; int err; if (unlikely(urb->status == -ENOENT || /* unlinked */ @@ -364,8 +366,6 @@ static void snd_complete_urb(struct urb *urb) goto exit_clear; if (snd_usb_endpoint_implicit_feedback_sink(ep)) { - unsigned long flags; - spin_lock_irqsave(&ep->lock, flags); list_add_tail(&ctx->ready_list, &ep->ready_playback_urbs); spin_unlock_irqrestore(&ep->lock, flags); @@ -389,7 +389,12 @@ static void snd_complete_urb(struct urb *urb) return; usb_audio_err(ep->chip, "cannot submit urb (err = %d)\n", err); - //snd_pcm_stop(substream, SNDRV_PCM_STATE_XRUN); + if (ep->data_subs && ep->data_subs->pcm_substream) { + substream = ep->data_subs->pcm_substream; + snd_pcm_stream_lock_irqsave(substream, flags); + snd_pcm_stop(substream, SNDRV_PCM_STATE_XRUN); + snd_pcm_stream_unlock_irqrestore(substream, flags); + } exit_clear: clear_bit(ctx->index, &ep->active_mask); -- 2.1.3

1 0

[alsa-devel] [PATCH] ASoC:rt5645:Add JD function support
by bardliao＠realtek.com 06 Nov '14

06 Nov '14

From: Bard Liao <bardliao(a)realtek.com> rt5645 codec support jack detection function. The patch will set related registers if JD function is used. Signed-off-by: Bard Liao <bardliao(a)realtek.com> --- include/sound/rt5645.h | 3 +++ sound/soc/codecs/rt5645.c | 20 ++++++++++++++++++++ sound/soc/codecs/rt5645.h | 5 +++++ 3 files changed, 28 insertions(+) diff --git a/include/sound/rt5645.h b/include/sound/rt5645.h index a535271..937f421 100644 --- a/include/sound/rt5645.h +++ b/include/sound/rt5645.h @@ -23,6 +23,9 @@ struct rt5645_platform_data { unsigned int hp_det_gpio; bool gpio_hp_det_active_high; + + /* true if codec's jd function is used */ + bool en_jd_func; }; #endif diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 0e92e63..1c099ee 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -2205,6 +2205,13 @@ static int rt5645_probe(struct snd_soc_codec *codec) snd_soc_update_bits(codec, RT5645_CHARGE_PUMP, 0x0300, 0x0200); + /* for JD function */ + if (rt5645->pdata.en_jd_func) { + snd_soc_dapm_force_enable_pin(&codec->dapm, "JD Power"); + snd_soc_dapm_force_enable_pin(&codec->dapm, "LDO2"); + snd_soc_dapm_sync(&codec->dapm); + } + return 0; } @@ -2438,6 +2445,19 @@ static int rt5645_i2c_probe(struct i2c_client *i2c, } + if (rt5645->pdata.en_jd_func) { + regmap_update_bits(rt5645->regmap, RT5645_GEN_CTRL3, + RT5645_IRQ_CLK_GATE_CTRL | RT5645_MICINDET_MANU, + RT5645_IRQ_CLK_GATE_CTRL | RT5645_MICINDET_MANU); + regmap_update_bits(rt5645->regmap, RT5645_IN1_CTRL1, + RT5645_CBJ_BST1_EN, RT5645_CBJ_BST1_EN); + regmap_update_bits(rt5645->regmap, RT5645_JD_CTRL3, + RT5645_JD_CBJ_EN | RT5645_JD_CBJ_POL, + RT5645_JD_CBJ_EN | RT5645_JD_CBJ_POL); + regmap_update_bits(rt5645->regmap, RT5645_MICBIAS, + RT5645_IRQ_CLK_INT, RT5645_IRQ_CLK_INT); + } + if (rt5645->i2c->irq) { ret = request_threaded_irq(rt5645->i2c->irq, NULL, rt5645_irq, IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING diff --git a/sound/soc/codecs/rt5645.h b/sound/soc/codecs/rt5645.h index 5ec2520..82f681b 100644 --- a/sound/soc/codecs/rt5645.h +++ b/sound/soc/codecs/rt5645.h @@ -1348,6 +1348,8 @@ #define RT5645_PWR_CLK25M_SFT 4 #define RT5645_PWR_CLK25M_PD (0x0 << 4) #define RT5645_PWR_CLK25M_PU (0x1 << 4) +#define RT5645_IRQ_CLK_MCLK (0x0 << 3) +#define RT5645_IRQ_CLK_INT (0x1 << 3) /* VAD Control 4 (0x9d) */ #define RT5645_VAD_SEL_MASK (0x3 << 8) @@ -2116,6 +2118,9 @@ enum { #define RT5645_RXDP2_SEL_ADC (0x1 << 3) #define RT5645_RXDP2_SEL_SFT (3) +/* General Control3 (0xfc) */ +#define RT5645_IRQ_CLK_GATE_CTRL (0x1 << 11) +#define RT5645_MICINDET_MANU (0x1 << 7) /* Vendor ID (0xfd) */ #define RT5645_VER_C 0x2 -- 1.8.1.1.439.g50a6b54

3 3

[alsa-devel] [PATCH] ALSA: pcm: Update the state properly before notification
by Takashi Iwai 06 Nov '14

06 Nov '14

Some state changes (e.g. snd_pcm_stop()) sets the runtime state after calling snd_timer_notify(). This is basically racy, since the notification may wakes up the user even before the state change. Although the possibility is low, we should set the state before the notifications. Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/core/pcm_native.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index ca224fa2a33a..dfb5031969f8 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -1052,10 +1052,10 @@ static void snd_pcm_post_stop(struct snd_pcm_substream *substream, int state) struct snd_pcm_runtime *runtime = substream->runtime; if (runtime->status->state != state) { snd_pcm_trigger_tstamp(substream); + runtime->status->state = state; if (substream->timer) snd_timer_notify(substream->timer, SNDRV_TIMER_EVENT_MSTOP, &runtime->trigger_tstamp); - runtime->status->state = state; } wake_up(&runtime->sleep); wake_up(&runtime->tsleep); @@ -1204,11 +1204,11 @@ static void snd_pcm_post_suspend(struct snd_pcm_substream *substream, int state) { struct snd_pcm_runtime *runtime = substream->runtime; snd_pcm_trigger_tstamp(substream); + runtime->status->suspended_state = runtime->status->state; + runtime->status->state = SNDRV_PCM_STATE_SUSPENDED; if (substream->timer) snd_timer_notify(substream->timer, SNDRV_TIMER_EVENT_MSUSPEND, &runtime->trigger_tstamp); - runtime->status->suspended_state = runtime->status->state; - runtime->status->state = SNDRV_PCM_STATE_SUSPENDED; wake_up(&runtime->sleep); wake_up(&runtime->tsleep); } @@ -1311,10 +1311,10 @@ static void snd_pcm_post_resume(struct snd_pcm_substream *substream, int state) { struct snd_pcm_runtime *runtime = substream->runtime; snd_pcm_trigger_tstamp(substream); + runtime->status->state = runtime->status->suspended_state; if (substream->timer) snd_timer_notify(substream->timer, SNDRV_TIMER_EVENT_MRESUME, &runtime->trigger_tstamp); - runtime->status->state = runtime->status->suspended_state; } static struct action_ops snd_pcm_action_resume = { -- 2.1.3

1 0

[alsa-devel] [PATCH alsa-lib] pcm: dmix: Don't ignore SND_TIMER_EVENT_MSTOP
by Takashi Iwai 06 Nov '14

06 Nov '14

When a slave PCM gets an error like XRUN, it stops and notifies with SND_TIMER_EVENT_MSTOP event. But the current code filters out this type and eventually hang due to the empty timer queue. The fix is to just add this event type to the filter bit mask. Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- src/pcm/pcm_direct.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/pcm/pcm_direct.c b/src/pcm/pcm_direct.c index 8e37bcba5e19..195fddf06cda 100644 --- a/src/pcm/pcm_direct.c +++ b/src/pcm/pcm_direct.c @@ -1148,6 +1148,7 @@ int snd_pcm_direct_initialize_poll_fd(snd_pcm_direct_t *dmix) dmix->timer_events = (1<<SND_TIMER_EVENT_MSUSPEND) | (1<<SND_TIMER_EVENT_MRESUME) | + (1<<SND_TIMER_EVENT_MSTOP) | (1<<SND_TIMER_EVENT_STOP); /* -- 2.1.3

1 0