November 2017 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH] ASoC: rsnd: dma.c: spin lock is no longer needed in IRQ handler
by Kuninori Morimoto 29 Nov '17

29 Nov '17

From: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> DMA handler had needed to calculate pointer before, but it doesn't need now. Thus, we can remove unnecessary spin lock from DMAC handler. Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> --- sound/soc/sh/rcar/dma.c | 18 ------------------ 1 file changed, 18 deletions(-) diff --git a/sound/soc/sh/rcar/dma.c b/sound/soc/sh/rcar/dma.c index 4d750bdf..41de234 100644 --- a/sound/soc/sh/rcar/dma.c +++ b/sound/soc/sh/rcar/dma.c @@ -71,25 +71,7 @@ struct rsnd_dma_ctrl { static void __rsnd_dmaen_complete(struct rsnd_mod *mod, struct rsnd_dai_stream *io) { - struct rsnd_priv *priv = rsnd_mod_to_priv(mod); - bool elapsed = false; - unsigned long flags; - - /* - * Renesas sound Gen1 needs 1 DMAC, - * Gen2 needs 2 DMAC. - * In Gen2 case, it are Audio-DMAC, and Audio-DMAC-peri-peri. - * But, Audio-DMAC-peri-peri doesn't have interrupt, - * and this driver is assuming that here. - */ - spin_lock_irqsave(&priv->lock, flags); - if (rsnd_io_is_working(io)) - elapsed = true; - - spin_unlock_irqrestore(&priv->lock, flags); - - if (elapsed) rsnd_dai_period_elapsed(io); } -- 1.9.1

2 1

[alsa-devel] [PATCH 1/2] ASoC: nau8825: disable crosstalk by default
by John Hsu 29 Nov '17

29 Nov '17

The driver makes the crosstalk funciton disabled by default which can simplify the codec function. The platform may not need this funciton and reduce the potential risk. Therefore, We change the property "nuvoton,crosstalk-bypass" to "nuvoton,crosstalk-enable". The crosstalk measurement is enabled if the property is set. Otherwise, it is disabled. Besides, add more condition in the entry point of the crosstalk sequence to disable the function completely. Signed-off-by: John Hsu <KCHSU0(a)nuvoton.com> --- .../devicetree/bindings/sound/nau8825.txt | 4 ++-- sound/soc/codecs/nau8825.c | 23 ++++++++++++---------- sound/soc/codecs/nau8825.h | 2 +- 3 files changed, 16 insertions(+), 13 deletions(-) diff --git a/Documentation/devicetree/bindings/sound/nau8825.txt b/Documentation/devicetree/bindings/sound/nau8825.txt index 2f5e973..d16d968 100644 --- a/Documentation/devicetree/bindings/sound/nau8825.txt +++ b/Documentation/devicetree/bindings/sound/nau8825.txt @@ -69,7 +69,7 @@ Optional properties: - nuvoton,jack-insert-debounce: number from 0 to 7 that sets debounce time to 2^(n+2) ms - nuvoton,jack-eject-debounce: number from 0 to 7 that sets debounce time to 2^(n+2) ms - - nuvoton,crosstalk-bypass: make crosstalk function bypass if set. + - nuvoton,crosstalk-enable: make crosstalk function enable if set. - clocks: list of phandle and clock specifier pairs according to common clock bindings for the clocks described in clock-names @@ -98,7 +98,7 @@ Example: nuvoton,short-key-debounce = <2>; nuvoton,jack-insert-debounce = <7>; nuvoton,jack-eject-debounce = <7>; - nuvoton,crosstalk-bypass; + nuvoton,crosstalk-enable; clock-names = "mclk"; clocks = <&tegra_car TEGRA210_CLK_CLK_OUT_2>; diff --git a/sound/soc/codecs/nau8825.c b/sound/soc/codecs/nau8825.c index 714ce17..d3c1a02 100644 --- a/sound/soc/codecs/nau8825.c +++ b/sound/soc/codecs/nau8825.c @@ -815,11 +815,12 @@ static void nau8825_xtalk_work(struct work_struct *work) static void nau8825_xtalk_cancel(struct nau8825 *nau8825) { - /* If the xtalk_protect is true, that means the process is still - * on going. The driver forces to cancel the cross talk task and + /* If the crosstalk is eanbled and the process is on going, + * the driver forces to cancel the crosstalk task and * restores the configuration to original status. */ - if (nau8825->xtalk_protect) { + if (nau8825->xtalk_enable && nau8825->xtalk_state != + NAU8825_XTALK_DONE) { cancel_work_sync(&nau8825->xtalk_work); nau8825_xtalk_clean(nau8825); } @@ -1686,7 +1687,7 @@ static irqreturn_t nau8825_interrupt(int irq, void *data) } else if (active_irq & NAU8825_HEADSET_COMPLETION_IRQ) { if (nau8825_is_jack_inserted(regmap)) { event |= nau8825_jack_insert(nau8825); - if (!nau8825->xtalk_bypass && !nau8825->high_imped) { + if (nau8825->xtalk_enable && !nau8825->high_imped) { /* Apply the cross talk suppression in the * headset without high impedance. */ @@ -1732,8 +1733,10 @@ static irqreturn_t nau8825_interrupt(int irq, void *data) nau8825->xtalk_event_mask = event_mask; } } else if (active_irq & NAU8825_IMPEDANCE_MEAS_IRQ) { - schedule_work(&nau8825->xtalk_work); - clear_irq = NAU8825_IMPEDANCE_MEAS_IRQ; + if (nau8825->xtalk_enable) { + schedule_work(&nau8825->xtalk_work); + clear_irq = NAU8825_IMPEDANCE_MEAS_IRQ; + } } else if ((active_irq & NAU8825_JACK_INSERTION_IRQ_MASK) == NAU8825_JACK_INSERTION_DETECTED) { /* One more step to check GPIO status directly. Thus, the @@ -2440,8 +2443,8 @@ static void nau8825_print_device_properties(struct nau8825 *nau8825) nau8825->jack_insert_debounce); dev_dbg(dev, "jack-eject-debounce: %d\n", nau8825->jack_eject_debounce); - dev_dbg(dev, "crosstalk-bypass: %d\n", - nau8825->xtalk_bypass); + dev_dbg(dev, "crosstalk-enable: %d\n", + nau8825->xtalk_enable); } static int nau8825_read_device_properties(struct device *dev, @@ -2506,8 +2509,8 @@ static int nau8825_read_device_properties(struct device *dev, &nau8825->jack_eject_debounce); if (ret) nau8825->jack_eject_debounce = 0; - nau8825->xtalk_bypass = device_property_read_bool(dev, - "nuvoton,crosstalk-bypass"); + nau8825->xtalk_enable = device_property_read_bool(dev, + "nuvoton,crosstalk-enable"); nau8825->mclk = devm_clk_get(dev, "mclk"); if (PTR_ERR(nau8825->mclk) == -EPROBE_DEFER) { diff --git a/sound/soc/codecs/nau8825.h b/sound/soc/codecs/nau8825.h index 8aee5c86..199d6ea 100644 --- a/sound/soc/codecs/nau8825.h +++ b/sound/soc/codecs/nau8825.h @@ -476,7 +476,7 @@ struct nau8825 { int xtalk_event_mask; bool xtalk_protect; int imp_rms[NAU8825_XTALK_IMM]; - int xtalk_bypass; + int xtalk_enable; }; int nau8825_enable_jack_detect(struct snd_soc_codec *codec, -- 2.6.4

3 7

[alsa-devel] [PATCH v2 2/2] ASoC: fsl_ssi: call _fsl_ssi_set_dai_fmt() just once in AC'97 mode
by Maciej S. Szmigiero 29 Nov '17

29 Nov '17

In AC'97 mode we configure and start SSI RX / TX on probe path via a call to _fsl_ssi_set_dai_fmt() function. We don't need to call this function again later and in fact don't want to do it since this function temporarily sets STCR, SRCR and SCR to some intermediate values. Signed-off-by: Maciej S. Szmigiero <mail(a)maciej.szmigiero.name> --- Changes from v1: The SACCST setup code was split out into a separate commit. sound/soc/fsl/fsl_ssi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 375aaaf6080d..70df6bf832df 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -1116,6 +1116,9 @@ static int fsl_ssi_set_dai_fmt(struct snd_soc_dai *cpu_dai, unsigned int fmt) { struct fsl_ssi_private *ssi_private = snd_soc_dai_get_drvdata(cpu_dai); + if (fsl_ssi_is_ac97(ssi_private)) + return 0; + return _fsl_ssi_set_dai_fmt(cpu_dai->dev, ssi_private, fmt); }

3 2

[alsa-devel] [PATCH v2 1/2] ASoC: fsl_ssi: only enable proper channel slots in AC'97 mode
by Maciej S. Szmigiero 29 Nov '17

29 Nov '17

We need to make sure that only proper channel slots (in SACCST register) are enabled at playback start time since some AC'97 CODECs (like VT1613 on UDOO board) were observed requesting via SLOTREQ spurious ones just after an AC'97 link is started but before the CODEC is configured by its driver. When a bit for some channel slot is set in a SLOTREQ request then SSI sets the relevant bit in SACCST automatically, which then 'sticks' until it is manually unset. The SACCST register is not writable directly, we have to use SACCDIS and SACCEN registers to configure it instead (these aren't normal registers: writing a '1' bit at some position in SACCEN sets the relevant bit in SACCST; SACCDIS operates in a similar way but allows unsetting bits in SACCST). Theoretically, this should be necessary only for the very first playback but since some CODECs are so untrustworthy and extra channel slots enabled mean ruined playback let's play safe here and make sure that no extra slots are enabled in SACCST every time a playback is started. Signed-off-by: Maciej S. Szmigiero <mail(a)maciej.szmigiero.name> --- Changes from v1: Split out this part from "fsl_ssi: call _fsl_ssi_set_dai_fmt() just once in AC'97 mode" commit, describe the problem and its solution better both in the commit message and in the code, move the SACCST setup code into a separate function and call it from TX config instead of doing it from trigger handler function. sound/soc/fsl/fsl_ssi.c | 52 +++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 46 insertions(+), 6 deletions(-) diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 48bb850a34d9..375aaaf6080d 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -574,8 +574,54 @@ static void fsl_ssi_rx_config(struct fsl_ssi_private *ssi_private, bool enable) fsl_ssi_config(ssi_private, enable, &ssi_private->rxtx_reg_val.rx); } +static void fsl_ssi_tx_ac97_saccst_setup(struct fsl_ssi_private *ssi_private) +{ + struct regmap *regs = ssi_private->regs; + + /* no SACC{ST,EN,DIS} regs on imx21-class SSI */ + if (!ssi_private->soc->imx21regs) { + /* + * Note that these below aren't just normal registers. + * They are a way to disable or enable bits in SACCST + * register: + * - writing a '1' bit at some position in SACCEN sets the + * relevant bit in SACCST, + * - writing a '1' bit at some position in SACCDIS unsets + * the relevant bit in SACCST register. + * + * The two writes below first disable all channels slots, + * then enable just slots 3 & 4 ("PCM Playback Left Channel" + * and "PCM Playback Right Channel"). + */ + regmap_write(regs, CCSR_SSI_SACCDIS, 0xff); + regmap_write(regs, CCSR_SSI_SACCEN, 0x300); + } +} + static void fsl_ssi_tx_config(struct fsl_ssi_private *ssi_private, bool enable) { + /* + * Why are we setting up SACCST everytime we are starting a + * playback? + * Some CODECs (like VT1613 CODEC on UDOO board) like to + * (sometimes) set extra bits in their SLOTREQ requests. + * When a bit is set in a SLOTREQ request then SSI sets the + * relevant bit in SACCST automatically (it is enough if a bit was + * set in a SLOTREQ just once, bits in SACCST are 'sticky'). + * If an extra slot gets enabled that's a disaster for playback + * because some of normal left or right channel samples are + * redirected instead to this extra slot. + * + * A workaround implemented in fsl-asoc-card of setting an + * appropriate CODEC register so that slots 3 & 4 (the normal + * stereo playback slots) are used for S/PDIF seems to mostly fix + * this issue on the UDOO board but since this CODEC is so + * untrustworthy let's play safe here and make sure that no extra + * slots are enabled every time a playback is started. + */ + if (enable && fsl_ssi_is_ac97(ssi_private)) + fsl_ssi_tx_ac97_saccst_setup(ssi_private); + fsl_ssi_config(ssi_private, enable, &ssi_private->rxtx_reg_val.tx); } @@ -630,12 +676,6 @@ static void fsl_ssi_setup_ac97(struct fsl_ssi_private *ssi_private) regmap_write(regs, CCSR_SSI_SACNT, CCSR_SSI_SACNT_AC97EN | CCSR_SSI_SACNT_FV); - /* no SACC{ST,EN,DIS} regs on imx21-class SSI */ - if (!ssi_private->soc->imx21regs) { - regmap_write(regs, CCSR_SSI_SACCDIS, 0xff); - regmap_write(regs, CCSR_SSI_SACCEN, 0x300); - } - /* * Enable SSI, Transmit and Receive. AC97 has to communicate with the * codec before a stream is started.

3 2

[alsa-devel] [PATCH] ASoC: fsl_asrc: Fix typo in a field define
by Nicolin Chen 29 Nov '17

29 Nov '17

ASRFSTi_IAEi has an 11-bit offset as its _SHIFT macro defines. So this patch just fixes that. Reported-by: Laurent Charpentier <laurent.charpentier(a)nxp.com> Signed-off-by: Nicolin Chen <nicoleotsuka(a)gmail.com> --- sound/soc/fsl/fsl_asrc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/fsl/fsl_asrc.h b/sound/soc/fsl/fsl_asrc.h index 0f163ab..52c27a35 100644 --- a/sound/soc/fsl/fsl_asrc.h +++ b/sound/soc/fsl/fsl_asrc.h @@ -260,8 +260,8 @@ #define ASRFSTi_OUTPUT_FIFO_SHIFT 12 #define ASRFSTi_OUTPUT_FIFO_MASK (((1 << ASRFSTi_OUTPUT_FIFO_WIDTH) - 1) << ASRFSTi_OUTPUT_FIFO_SHIFT) #define ASRFSTi_IAEi_SHIFT 11 -#define ASRFSTi_IAEi_MASK (1 << ASRFSTi_OAFi_SHIFT) -#define ASRFSTi_IAEi (1 << ASRFSTi_OAFi_SHIFT) +#define ASRFSTi_IAEi_MASK (1 << ASRFSTi_IAEi_SHIFT) +#define ASRFSTi_IAEi (1 << ASRFSTi_IAEi_SHIFT) #define ASRFSTi_INPUT_FIFO_WIDTH 7 #define ASRFSTi_INPUT_FIFO_SHIFT 0 #define ASRFSTi_INPUT_FIFO_MASK ((1 << ASRFSTi_INPUT_FIFO_WIDTH) - 1) -- 2.1.4

3 2

[alsa-devel] [PATCH] ASoC: soc-utils: add const to dummy_codec
by Kuninori Morimoto 29 Nov '17

29 Nov '17

From: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> Let's align dummy_platform and dummy_codec Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> --- sound/soc/soc-utils.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/soc-utils.c b/sound/soc/soc-utils.c index e30aacb..bcd3da2 100644 --- a/sound/soc/soc-utils.c +++ b/sound/soc/soc-utils.c @@ -288,7 +288,7 @@ static int dummy_dma_open(struct snd_pcm_substream *substream) .ops = &dummy_dma_ops, }; -static struct snd_soc_codec_driver dummy_codec; +static const struct snd_soc_codec_driver dummy_codec; #define STUB_RATES SNDRV_PCM_RATE_8000_192000 #define STUB_FORMATS (SNDRV_PCM_FMTBIT_S8 | \ -- 1.9.1

3 2

[alsa-devel] [PATCH 1/5 v4] ASoC: ep93xx-ac97: Fix platform_get_irq's error checking
by Arvind Yadav 29 Nov '17

29 Nov '17

The platform_get_irq() function returns negative if an error occurs. zero or positive number on success. platform_get_irq() error checking for zero is not correct. Signed-off-by: Arvind Yadav <arvind.yadav.cs(a)gmail.com> --- changes in v2 : irq was unsigned. so changed it to signed. changes in v3 : Add failure case '<= 0' instead of '< 0'. IRQ0 is not valid. changes in v4 : Return -ENODEV insted of irq. sound/soc/cirrus/ep93xx-ac97.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/cirrus/ep93xx-ac97.c b/sound/soc/cirrus/ep93xx-ac97.c index bbf7a92..efeecee 100644 --- a/sound/soc/cirrus/ep93xx-ac97.c +++ b/sound/soc/cirrus/ep93xx-ac97.c @@ -365,7 +365,7 @@ static int ep93xx_ac97_probe(struct platform_device *pdev) { struct ep93xx_ac97_info *info; struct resource *res; - unsigned int irq; + int irq; int ret; info = devm_kzalloc(&pdev->dev, sizeof(*info), GFP_KERNEL); @@ -378,7 +378,7 @@ static int ep93xx_ac97_probe(struct platform_device *pdev) return PTR_ERR(info->regs); irq = platform_get_irq(pdev, 0); - if (!irq) + if (irq <= 0) return -ENODEV; ret = devm_request_irq(&pdev->dev, irq, ep93xx_ac97_interrupt, -- 2.7.4

4 7

[alsa-devel] [PATCH 0/2] ALSA: nm256: Fine-tuning for three function implementations
by SF Markus Elfring 29 Nov '17

29 Nov '17

From: Markus Elfring <elfring(a)users.sourceforge.net> Date: Thu, 16 Nov 2017 18:00:18 +0100 Two update suggestions were taken into account from static source code analysis. Markus Elfring (2): Adjust five function calls together with a variable assignment Use common error handling code in snd_nm256_probe() sound/pci/nm256/nm256.c | 48 +++++++++++++++++++++++++++++------------------- 1 file changed, 29 insertions(+), 19 deletions(-) -- 2.15.0

3 43

[alsa-devel] [PATCH 00/24] ALSA: ctl: refactoring compat layer
by Takashi Sakamoto 29 Nov '17

29 Nov '17

Hi, ALSA control core supports system call compatibility layer because some structures in ALSA control interface includes members of 'long' and 'pointer' types and change own layout according to ABIs. In this point, it's enough for the layer to have handlers for structures on ABIs with 'ILP32' data model. A recent commit of 6236d8bb2afc ('ALSA: ctl: Fix ioctls for X32 ABI') clear that System V ABI for i386 architecture is unique than the other ABIs with 'ILP32' data model. On this ABI, machine type for storage class of 'long long' type has 4 bytes alignment. This is different from the other ABIs. In current implementation of this layer, the same structure is used for i386 ABI and the other ABI with 'ILP32' data model except for x32 ABI. Macro is used to switch between these. But in a view of data model, it's better to define structure for i386 ABI uniquely against the other ABIs including x32 for simplicity. Additionally, this layer includes some points to be improved: - cancel allocation in user space from kernel land - reduce kernel stack usage This patchset is my attempts to improve the layer. For this purpose, this patchset introduces a local framework to describe consistent method to convert and process data for differences of structure layout. Takashi Sakamoto (24): ALSA: ctl: introduce local framework to handle compat ioctl requests ALSA: ctl: add serializer/deserializer of 'elem_list' structure for 32bit ABI compatibility ALSA: ctl: add serializer/deserializer of 'elem_info' structure for 32bit ABI compatibility ALSA: ctl: add serializer/deserializer of 'elem_value' structure for modern 32 bit ABIs compatibility ALSA: ctl: add serializer/deserializer of 'elem_value' structure for i386 ABI compatibility ALSA: ctl: change prototype of local function for ELEM_LIST ioctl ALSA: ctl: add a helper function to allocate for ELEM_LIST request ALSA: ctl: cancel allocation in user space for ELEM_LIST request ALSA: ctl: unify calls of D0-wait function for ELEM_INFO request ALSA: ctl: unify calls of D0-wait function for ELEM_READ request ALSA: ctl: unify calls of D0-wait function for ELEM_WRITE request ALSA: ctl: allocation of elem_info data for ELEM_INFO request ALSA: ctl: change prototype of local function for ELEM_WRITE ioctl ALSA: ctl: change prototype of local function for ELEM_READ ioctl ALSA: ctl: allocation of elem_info data for ELEM_ADD/ELEM_REPLACE requests ALSA: ctl: add replace helper function to allocate own buffer ALSA: ctl: move removal code to replace helper function ALSA: ctl: replacement for compat ELEM_LIST operation for any 32 bit ABI ALSA: ctl: replacement for compat ELEM_INFO operation for any 32 bit ABI ALSA: ctl: replacement for compat ELEM_ADD operation for any 32 bit ABI ALSA: ctl: replacement for compat ELEM_REPLACE operation for any 32 bit ABI ALSA: ctl: replacement for compat ELEM_READ/ELEM_WRITE operation for i386 ABI ALSA: ctl: replacement for compat ELEM_READ/ELEM_WRITE operation for modern 32 bit ABI ALSA: ctl: code cleanup for compat handler sound/core/control.c | 242 ++++++++----- sound/core/control_compat.c | 837 +++++++++++++++++++++++++------------------- 2 files changed, 630 insertions(+), 449 deletions(-) -- 2.14.1

2 30

[alsa-devel] amixer sset volume: Left/Right race
by Eric Rannaud 29 Nov '17

29 Nov '17

In my ~/.xbindkeysrc I have "amixer set Master 1000+" XF86AudioRaiseVolume "amixer set Master 1000-" XF86AudioLowerVolume When I hold the "volume up" key on my keyboard, the left and right channel volumes quickly get out of sync. This is reproducible with this shell loop: for i in $(seq 20); do amixer sset Master 1000+ & done But not with this loop: for i in $(seq 20); do amixer sset Master 1000+; done It appears that xbindkeys invokes multiple amixer concurrently, like the first shell loop, and that amixer/ALSA is somehow setting the channel volumes one at a time, presumably in an unprotected read-modify-update critical section. I'm not familiar with how this works, and where the blame might lay. However, pamixer (https://github.com/cdemoulins/pamixer/blob/master/pulseaudio.cc) uses /usr/include/pulse/pulseaudio.h and pa_context_set_sink_volume_by_index() directly and does not have this problem. for i in $(seq 20); do pamixer --increase 1 & done Thanks. P.S. The "Bug Tracker" link on the left of http://alsa-project.org/main/index.php/Main_Page is broken.

1 0