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

[alsa-devel] [PATCH] drm/i915: Enable VLV audio chicken bit for LPE audio
by Takashi Iwai 03 Feb '17

03 Feb '17

The audio chicken bit (register offset 0x62f38) seems required to make DP audio working on some machines. At least, on Dell Wyse 3040, I failed to get the audio unless this bit is set once. Strangely, the bit seems necessary only once, and it persists after that, even some power-off cycles. The register is supposedly write-only, so it's no evidence whether the bit keeps effect persistently. But, judging from the experiment, it looks enough to set it up once at the device initialization. The patch is basically a cut from the original patch by Pierre-Louis Bossart. Cc: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- drivers/gpu/drm/i915/i915_reg.h | 3 +++ drivers/gpu/drm/i915/intel_lpe_audio.c | 7 +++++++ 2 files changed, 10 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 4e24ba0cdbe8..4f15a3dc6d98 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -2062,6 +2062,9 @@ enum skl_disp_power_wells { #define I915_HDMI_LPE_AUDIO_SIZE 0x1000 /* DisplayPort Audio w/ LPE */ +#define VLV_AUD_CHICKEN_BIT_REG _MMIO(VLV_DISPLAY_BASE + 0x62F38) +#define VLV_CHICKEN_BIT_DBG_ENABLE (1 << 0) + #define _VLV_AUD_PORT_EN_B_DBG (VLV_DISPLAY_BASE + 0x62F20) #define _VLV_AUD_PORT_EN_C_DBG (VLV_DISPLAY_BASE + 0x62F30) #define _VLV_AUD_PORT_EN_D_DBG (VLV_DISPLAY_BASE + 0x62F34) diff --git a/drivers/gpu/drm/i915/intel_lpe_audio.c b/drivers/gpu/drm/i915/intel_lpe_audio.c index d3ffe0012692..3ba2799e93bd 100644 --- a/drivers/gpu/drm/i915/intel_lpe_audio.c +++ b/drivers/gpu/drm/i915/intel_lpe_audio.c @@ -248,6 +248,13 @@ static int lpe_audio_setup(struct drm_i915_private *dev_priv) goto err_free_irq; } + /* enable chicken bit; at least this is required for Dell Wyse 3040 + * with DP outputs (but only sometimes by some reason!) + */ + I915_WRITE(VLV_AUD_CHICKEN_BIT_REG, + I915_READ(VLV_AUD_CHICKEN_BIT_REG) | + VLV_CHICKEN_BIT_DBG_ENABLE); + return 0; err_free_irq: irq_free_desc(dev_priv->lpe_audio.irq); -- 2.11.0

2 3

[alsa-devel] [PATCH v2 00/10] Add sun8i A33 audio driver
by Mylène Josserand 03 Feb '17

03 Feb '17

Hello everyone, This a V2 of my Allwinner A33 (sun8i) audio codec driver. Tested on linux-next-20170105 tag with one remaining patch from Chen-Yu's series: http://lkml.iu.edu/hypermail/linux/kernel/1611.3/01256.html Patch to apply before this serie: https://patchwork.kernel.org/patch/9423999/ Changes since V1: - Remove the analog codec driver as a better version has been committed by Chen-Yu Tsai and is already merged. - Remove the audio-card as simple-card can be used - The DMA maxburst is set to 8 in the sun4i-i2s instead of adding the maxburst of 4 in Sun6i dma engine. - Create a new compatible for sun4i-i2s to handle the reset line. - Fix various problems in sun8i-codec driver according to V1's reviews - Add the pm_runtime hooks in sun8i-codec driver to prepare/ unprepare clocks. - Update the DTS according to Chen-Yu's analog codec driver. - Rename sun8i-codec's clocks to "bus" and "mod" - The first "delay" issue from V1 is fixed by using a delay when enabling the headphone amplifier to let the amplifier being up. Patch 1 increases the DMA maxburst value of sun4i-i2s to 8. See Chen-Yu's commit log for more details: https://patchwork.kernel.org/patch/9411805/ Patch 2 adds the CLK_SET_RATE_PARENT flag to ccu sun8i-a33 clocks. Patches 3 and 4 add a new compatible "allwinner,sun6i-a31-i2s" to handle the reset line for sun4i-i2s driver. It uses a quirk to use a version with or without reset lines. Patch 5 adds the sun8i codec driver which represents the digital part of the A33 codec. It supports only playback features. Path 6 fixes the previous issue of a "first time delay" in V1 (see cover letter). Do not hesitate if you have comments on this patch. Patches 7 adds the dt-bindings documentation for new audio driver added in this serie (sun8i-codec). Patch 8 adds the cpu DAI, codec and audio nodes to sun8i-a33 device tree. Patches 9 and 10 enable the audio on Parrot and Sinlinx's boards. The DAI for this A33 codec is the same than for A20: "sun4i-i2s". Currently, all the drivers handle only the playback feature. The other ones (such as capture) and all other interfaces except headphone are not supported. I will send a patch to handle the capture with microphones in next few weeks. Examples of amixer commands: amixer set 'Headphone' 75% amixer set 'Headphone' on amixer set 'DAC' on amixer set 'Right DAC Mixer RSlot 0' on amixer set 'Left DAC Mixer LSlot 0' on It was tested on Parrot and Sinlinx boards. Let me know if you have any comments on this serie. Thank you in advance, Best regards, Mylène Josserand (10): ASoC: sun4i-i2s: Increase DMA max burst to 8 clk: ccu-sun8i-a33: Add CLK_SET_RATE_PARENT to ac-dig dt-bindings: sound: Add new reset compatible for sun4i-i2s ASoC: sun4i-i2s: Add quirks to handle new compatible for reset ASoC: Add sun8i digital audio codec ASoC: sun8i-codec-analog: Add amplifier event to fix first delay dt-bindings: sound: Add sun8i audio documentation ARM: dts: sun8i: Add audio codec, dai and card for A33 ARM: dts: sun8i: parrot: Enable audio nodes ARM: dts: sun8i: sinlinx: Enable audio nodes .../devicetree/bindings/sound/sun4i-i2s.txt | 23 + .../devicetree/bindings/sound/sun8i-codec.txt | 76 +++ arch/arm/boot/dts/sun8i-a33-sinlinx-sina33.dts | 8 + arch/arm/boot/dts/sun8i-a33.dtsi | 47 ++ arch/arm/boot/dts/sun8i-r16-parrot.dts | 8 + drivers/clk/sunxi-ng/ccu-sun8i-a33.c | 2 +- sound/soc/sunxi/Kconfig | 11 + sound/soc/sunxi/Makefile | 1 + sound/soc/sunxi/sun4i-i2s.c | 61 ++- sound/soc/sunxi/sun8i-codec-analog.c | 30 +- sound/soc/sunxi/sun8i-codec.c | 531 +++++++++++++++++++++ 11 files changed, 791 insertions(+), 7 deletions(-) create mode 100644 Documentation/devicetree/bindings/sound/sun8i-codec.txt create mode 100644 sound/soc/sunxi/sun8i-codec.c -- 2.11.0

3 24

[alsa-devel] Applied "ASoC: sun4i-i2s: Add quirks to handle a31 compatible" to the asoc tree
by Mark Brown 03 Feb '17

03 Feb '17

The patch ASoC: sun4i-i2s: Add quirks to handle a31 compatible has been applied to the asoc tree at git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted. You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed. If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced. Please add any relevant lists and maintainers to the CCs when replying to this mail. Thanks, Mark >From 2ad6f30de7087515a0bc2a718fca6681a57739a0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Myl=C3=A8ne=20Josserand?= <mylene.josserand(a)free-electrons.com> Date: Thu, 2 Feb 2017 10:24:16 +0100 Subject: [PATCH] ASoC: sun4i-i2s: Add quirks to handle a31 compatible MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Some SoCs have a reset line that must be asserted/deasserted. This patch adds a quirk to handle the new compatible "allwinner,sun6i-a31-i2s" which will deassert the reset line on probe function and assert it on remove's one. This new compatible is useful in case of A33 codec driver, for example. Signed-off-by: Mylène Josserand <mylene.josserand(a)free-electrons.com> Acked-by: Maxime Ripard <maxime.ripard(a)free-electrons.com> Signed-off-by: Mark Brown <broonie(a)kernel.org> --- sound/soc/sunxi/sun4i-i2s.c | 57 +++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 55 insertions(+), 2 deletions(-) diff --git a/sound/soc/sunxi/sun4i-i2s.c b/sound/soc/sunxi/sun4i-i2s.c index f24d19526603..268f2bf691b3 100644 --- a/sound/soc/sunxi/sun4i-i2s.c +++ b/sound/soc/sunxi/sun4i-i2s.c @@ -14,9 +14,11 @@ #include <linux/clk.h> #include <linux/dmaengine.h> #include <linux/module.h> +#include <linux/of_device.h> #include <linux/platform_device.h> #include <linux/pm_runtime.h> #include <linux/regmap.h> +#include <linux/reset.h> #include <sound/dmaengine_pcm.h> #include <sound/pcm_params.h> @@ -92,6 +94,7 @@ struct sun4i_i2s { struct clk *bus_clk; struct clk *mod_clk; struct regmap *regmap; + struct reset_control *rst; unsigned int mclk_freq; @@ -651,9 +654,22 @@ static int sun4i_i2s_runtime_suspend(struct device *dev) return 0; } +struct sun4i_i2s_quirks { + bool has_reset; +}; + +static const struct sun4i_i2s_quirks sun4i_a10_i2s_quirks = { + .has_reset = false, +}; + +static const struct sun4i_i2s_quirks sun6i_a31_i2s_quirks = { + .has_reset = true, +}; + static int sun4i_i2s_probe(struct platform_device *pdev) { struct sun4i_i2s *i2s; + const struct sun4i_i2s_quirks *quirks; struct resource *res; void __iomem *regs; int irq, ret; @@ -674,6 +690,12 @@ static int sun4i_i2s_probe(struct platform_device *pdev) return irq; } + quirks = of_device_get_match_data(&pdev->dev); + if (!quirks) { + dev_err(&pdev->dev, "Failed to determine the quirks to use\n"); + return -ENODEV; + } + i2s->bus_clk = devm_clk_get(&pdev->dev, "apb"); if (IS_ERR(i2s->bus_clk)) { dev_err(&pdev->dev, "Can't get our bus clock\n"); @@ -692,7 +714,24 @@ static int sun4i_i2s_probe(struct platform_device *pdev) dev_err(&pdev->dev, "Can't get our mod clock\n"); return PTR_ERR(i2s->mod_clk); } - + + if (quirks->has_reset) { + i2s->rst = devm_reset_control_get(&pdev->dev, NULL); + if (IS_ERR(i2s->rst)) { + dev_err(&pdev->dev, "Failed to get reset control\n"); + return PTR_ERR(i2s->rst); + } + } + + if (!IS_ERR(i2s->rst)) { + ret = reset_control_deassert(i2s->rst); + if (ret) { + dev_err(&pdev->dev, + "Failed to deassert the reset control\n"); + return -EINVAL; + } + } + i2s->playback_dma_data.addr = res->start + SUN4I_I2S_FIFO_TX_REG; i2s->playback_dma_data.maxburst = 4; @@ -727,23 +766,37 @@ static int sun4i_i2s_probe(struct platform_device *pdev) sun4i_i2s_runtime_suspend(&pdev->dev); err_pm_disable: pm_runtime_disable(&pdev->dev); + if (!IS_ERR(i2s->rst)) + reset_control_assert(i2s->rst); return ret; } static int sun4i_i2s_remove(struct platform_device *pdev) { + struct sun4i_i2s *i2s = dev_get_drvdata(&pdev->dev); + snd_dmaengine_pcm_unregister(&pdev->dev); pm_runtime_disable(&pdev->dev); if (!pm_runtime_status_suspended(&pdev->dev)) sun4i_i2s_runtime_suspend(&pdev->dev); + if (!IS_ERR(i2s->rst)) + reset_control_assert(i2s->rst); + return 0; } static const struct of_device_id sun4i_i2s_match[] = { - { .compatible = "allwinner,sun4i-a10-i2s", }, + { + .compatible = "allwinner,sun4i-a10-i2s", + .data = &sun4i_a10_i2s_quirks, + }, + { + .compatible = "allwinner,sun6i-a31-i2s", + .data = &sun6i_a31_i2s_quirks, + }, {} }; MODULE_DEVICE_TABLE(of, sun4i_i2s_match); -- 2.11.0

1 0

[alsa-devel] [PATCH] ASoC: rsnd: fixup reset timing of sync convert_rate
by Kuninori Morimoto 03 Feb '17

03 Feb '17

From: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> Sync convert rate settings should be availabled *after* Playing. Thus, src->sync should be reset first of init function. Otherwise, it will set remaining settings when it start playing. This patch fixes it. Thanks to Yokoyama-san Reported-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx(a)renesas.com> Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx(a)renesas.com> --- >> Mark This is bug fix for Linus tree sound/soc/sh/rcar/src.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sound/soc/sh/rcar/src.c b/sound/soc/sh/rcar/src.c index 3a8f65b..42db48d 100644 --- a/sound/soc/sh/rcar/src.c +++ b/sound/soc/sh/rcar/src.c @@ -390,6 +390,9 @@ static int rsnd_src_init(struct rsnd_mod *mod, { struct rsnd_src *src = rsnd_mod_to_src(mod); + /* reset sync convert_rate */ + src->sync.val = 0; + rsnd_mod_power_on(mod); rsnd_src_activation(mod); @@ -398,9 +401,6 @@ static int rsnd_src_init(struct rsnd_mod *mod, rsnd_src_status_clear(mod); - /* reset sync convert_rate */ - src->sync.val = 0; - return 0; } -- 2.7.4

2 1

Re: [alsa-devel] [RFC] [ALSA][CONTROL] Added 2 ioctls for reading/writing values of multiple controls in one go (at once)
by Takashi Iwai 03 Feb '17

03 Feb '17

On Thu, 02 Feb 2017 04:45:48 +0100, Takashi Sakamoto wrote: > > Hi, > > On Feb 2 2017 12:14, Satendra Singh Thakur wrote: > > From: satendra singh thakur <satendra.t(a)samsung.com> > > > > -Added 2 ioctls in alsa driver's control interface > > -Added an ioctl to read values of multiple elements at once > > -Added an ioctl to write values of multiple elements at once > > -In the absence of above ioctls user needs to call N ioctls to > > read/write value of N elements which requires N context switches > > -Proposed ioctls will allow accessing N elements' values in a single > > context switch > > -Above mentioned ioctl will be useful for alsa utils such as amixer > > which reads all controls of given sound card > > > > Signed-off-by: Satendra Singh Thakur <satendra.t(a)samsung.com> > > --- > > include/uapi/sound/asound.h | 8 +++++ > > sound/core/control.c | 69 +++++++++++++++++++++++++++++++++++++++++++ > > 2 files changed, 77 insertions(+) > > I'm _strongly_ interested in your two patches, because it has a > potentiality to purge ASoC abuse of TLV feature, which was introduced > in 2014 with a bad reviewing process. I don't think it can be a replacement for the extended TLV usages. The proposed API is nothing but a loop of ctl elem read/write, and I'm not sure whether it worth to introduce the new ioctls just for that. Takashi

1 0

[alsa-devel] [RFC] [ALSA][CONTROL] Added 2 ioctls for reading/writing values of multiple controls in one go (at once)
by Satendra Singh Thakur 03 Feb '17

03 Feb '17

From: satendra singh thakur <satendra.t(a)samsung.com> -Added 2 ioctls in alsa driver's control interface -Added an ioctl to read values of multiple elements at once -Added an ioctl to write values of multiple elements at once -In the absence of above ioctls user needs to call N ioctls to read/write value of N elements which requires N context switches -Proposed ioctls will allow accessing N elements' values in a single context switch -Above mentioned ioctl will be useful for alsa utils such as amixer which reads all controls of given sound card Signed-off-by: Satendra Singh Thakur <satendra.t(a)samsung.com> --- include/uapi/sound/asound.h | 8 +++++ sound/core/control.c | 69 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 77 insertions(+) diff --git a/include/uapi/sound/asound.h b/include/uapi/sound/asound.h index be353a7..a0dcee7 100644 --- a/include/uapi/sound/asound.h +++ b/include/uapi/sound/asound.h @@ -948,6 +948,11 @@ struct snd_ctl_tlv { unsigned int length; /* in bytes aligned to 4 */ unsigned int tlv[0]; /* first TLV */ }; +/*Struct to read/write multiple element values */ +struct snd_ctl_elem_values { + unsigned int num_vals;/* Number of values*/ + struct snd_ctl_elem_value *pvals;/* Pointer to the array of values */ +}; #define SNDRV_CTL_IOCTL_PVERSION _IOR('U', 0x00, int) #define SNDRV_CTL_IOCTL_CARD_INFO _IOR('U', 0x01, struct snd_ctl_card_info) @@ -974,6 +979,9 @@ struct snd_ctl_tlv { #define SNDRV_CTL_IOCTL_RAWMIDI_PREFER_SUBDEVICE _IOW('U', 0x42, int) #define SNDRV_CTL_IOCTL_POWER _IOWR('U', 0xd0, int) #define SNDRV_CTL_IOCTL_POWER_STATE _IOR('U', 0xd1, int) +/*Multipe controls' values read and write*/ +#define SNDRV_CTL_IOCTL_ELEMS_READ _IOWR('U', 0xe0, struct snd_ctl_elem_values) +#define SNDRV_CTL_IOCTL_ELEMS_WRITE _IOWR('U', 0xe1, struct snd_ctl_elem_values) /* * Read interface. diff --git a/sound/core/control.c b/sound/core/control.c index fb096cb..d60dc98 100644 --- a/sound/core/control.c +++ b/sound/core/control.c @@ -942,6 +942,7 @@ static int snd_ctl_elem_read_user(struct snd_card *card, return result; } + static int snd_ctl_elem_write(struct snd_card *card, struct snd_ctl_file *file, struct snd_ctl_elem_value *control) { @@ -1000,6 +1001,70 @@ static int snd_ctl_elem_write_user(struct snd_ctl_file *file, return result; } +/** + * snd_ctl_elems_rw_user - Read/Write values of more than one element, + * one by one + * @card: the card to which element belongs to + * @pucontrols: user-space pointer to struct snd_ctl_elem_values + * @write_flag: this flag distinguises write or read type request + * + * This function reads/writes the value of controls with the given IDs + * of the same card + * Return: On full/partial success, It returns number of successful + * controls read/written. + * On failure, it returns appropriate error + */ +static int snd_ctl_elems_rw_user(struct snd_ctl_file *file, + struct snd_ctl_elem_values __user *pucontrols, + bool write_flag) +{ + struct snd_ctl_elem_values controls; + struct snd_ctl_elem_value control; + struct snd_ctl_elem_value __user *puvals; + struct snd_card *card = file->card; + int result; + int vals; + int controls_count = 0; + + if (copy_from_user(&controls, pucontrols, sizeof(controls))) + return -EFAULT; + if (!controls.num_vals || controls.num_vals > card->controls_count) + return -EINVAL; + /*assign user-space pointer**/ + puvals = (struct snd_ctl_elem_value __user *) controls.pvals; + for (vals = 0; vals < controls.num_vals; vals++) { + if (copy_from_user(&control, puvals + vals, sizeof(control))) + return -EFAULT; + snd_power_lock(card); + result = snd_power_wait(card, SNDRV_CTL_POWER_D0); + if (result >= 0) { + if (write_flag == true) + result = snd_ctl_elem_write(card, file, + &control); + else + result = snd_ctl_elem_read(card, &control); + } + snd_power_unlock(card); + if (result < 0) { + /**If control failed to set/get + * inform user by sending back -1 in reserved field + * so that one can try again for failed elements + */ + control.reserved[0] = (unsigned char) -1; + pr_err("ALSA: control: snd_ctl_elem_write/read failed\ + for control name = %s\n", control.id.name); + } else { + controls_count++; + } + if (copy_to_user(puvals + vals, &control, sizeof(control))) + return -EFAULT; + } + pr_debug("ALSA: control: Num values successfully read/written %u\n",\ + controls_count); + /**Return successful control count to user**/ + return controls_count; +} + static int snd_ctl_elem_lock(struct snd_ctl_file *file, struct snd_ctl_elem_id __user *_id) { @@ -1515,8 +1580,12 @@ static long snd_ctl_ioctl(struct file *file, unsigned int cmd, unsigned long arg return snd_ctl_elem_info_user(ctl, argp); case SNDRV_CTL_IOCTL_ELEM_READ: return snd_ctl_elem_read_user(card, argp); + case SNDRV_CTL_IOCTL_ELEMS_READ: + return snd_ctl_elems_rw_user(ctl, argp, false); case SNDRV_CTL_IOCTL_ELEM_WRITE: return snd_ctl_elem_write_user(ctl, argp); + case SNDRV_CTL_IOCTL_ELEMS_WRITE: + return snd_ctl_elems_rw_user(ctl, argp, true); case SNDRV_CTL_IOCTL_ELEM_LOCK: return snd_ctl_elem_lock(ctl, argp); case SNDRV_CTL_IOCTL_ELEM_UNLOCK: -- 1.7.9.5

2 2

[alsa-devel] [PATCH 00/53] Intel LPE audio driver cleanups
by Takashi Iwai 03 Feb '17

03 Feb '17

Hi, this is a fairly largish patchset to fix and clean up Intel LPE audio driver. Most of the patches are to clean up and refactoring of the codes. As you can see in the diffstat below, quite lots of codes (over 1600 lines) are reduced. Now the driver structure is flattened to a single layer, and it looks much like a normal driver. Besides the refactoring, there are a few major fixes: e.g. the fix for the PCM disconnection that confuses PA,. the fix for the possible sleep-in-atomic at hotplug, the fix for racy chmap accesses. As new features, the driver supports the runtime PM properly now, it sets the CA bit for DP, and it exposes the ELD via ALSA control element. The current code is available in topic/intel-lpe-audio-cleanup branch, and it's based on the DP-audio branch, topic/intel-lpe-audio-dp that I'm going to merge to for-next soon. Takashi === Takashi Iwai (53): ALSA: x86: Don't set PCM state to DISCONNECTED ALSA: x86: Remove v1 ops and structs ALSA: x86: Drop indirect calls of had_ops ALSA: x86: Replace indirect register ops with direct calls ALSA: x86: Replace indirect query_ops with direct calls ALSA: x86: Drop snd_intel_had_interface indirect calls ALSA: x86: Pass snd_intelhad object to helpers ALSA: x86: Handle the error from hdmi_audio_probe() properly ALSA: x86: Drop useless mutex at probe ALSA: x86: Call event callback directly ALSA: x86: Fix possible stale interrupt calls ALSA: x86: Drop unused mid_hdmi_audio_is_busy() ALSA: x86: Drop the global platform device reference ALSA: x86: Drop global hlpe_state ALSA: x86: Drop global ELD copy ALSA: x86: Move the global underrun_count to struct snd_intelhad ALSA: x86: Drop unused hw_silence field ALSA: x86: Move dma_mask debug print into intel_hdmi_lpe_audio.c ALSA: x86: Embed snd_intelhad into snd_card ALSA: x86: Drop superfluous CHT PCI ID check ALSA: x86: Check platform_data earlier ALSA: x86: Call snd_card_register() at the end ALSA: x86: Drop unused hdmi_audio_query() ALSA: x86: Flatten two abstraction layers ALSA: x86: Fold intel_hdmi_audio_if.c into main file ALSA: x86: Replace pr_xxx() with dev_xxx() ALSA: x86: Fix for CONFIG_PM=n ALSA: x86: Remove indirect call of snd_pcm_period_elapsed() ALSA: x86: Drop unused fields from snd_intelhad struct ALSA: x86: Drop superfluous PCM private_free ALSA: x86: Fix sleep-in-atomic via i915 notification ALSA: x86: Remove superfluous check at resume ALSA: x86: Drop had_get_hwstate() ALSA: x86: Tidy up codes ALSA: x86: Remove _v[12] suffices ALSA: x86: Constfy tables ALSA: x86: Remove superfluous irqsave flags ALSA: x86: Fix racy access to chmap ALSA: x86: Drop flag_underrun field ALSA: x86: Drop superfluous state field ALSA: x86: Drop redundant had_stream_pvt ALSA: x86: Drop unused fields from pcm_stream_info ALSA: x86: Properly manage PCM substream lifetype ALSA: x86: Implement runtime PM ALSA: x86: Move stream status into pcm_stream_info ALSA: x86: Use the standard ELD bytes definitions ALSA: x86: Reduce redundant register field names ALSA: x86: Clean up unused defines and inclusions ALSA: x86: Create ELD control element ALSA: x86: Set CA bits for DisplayPort too ALSA: x86: Simplify comments ALSA: x86: Yet more tidy-up and clean-ups ALSA: x86: Rename drv_status to connected sound/x86/Makefile | 4 +- sound/x86/intel_hdmi_audio.c | 1901 +++++++++++++++++++------------------- sound/x86/intel_hdmi_audio.h | 117 +-- sound/x86/intel_hdmi_audio_if.c | 548 ----------- sound/x86/intel_hdmi_lpe_audio.c | 665 ------------- sound/x86/intel_hdmi_lpe_audio.h | 470 ++-------- 6 files changed, 1026 insertions(+), 2679 deletions(-) delete mode 100644 sound/x86/intel_hdmi_audio_if.c delete mode 100644 sound/x86/intel_hdmi_lpe_audio.c -- 2.11.0

3 59

[alsa-devel] [PATCH v4 2/2] ASoc: Add devictree bindings for CS35L35 CODEC
by Brian Austin 02 Feb '17

02 Feb '17

Add devicetree bindings documentatin for Cirrus Logic CS35L35 Boosted Amplifier Signed-off-by: Brian Austin <brian.austin(a)cirrus.com> --- .../devicetree/bindings/sound/cs35l35.txt | 171 +++++++++++++++++++++ 1 file changed, 171 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/cs35l35.txt diff --git a/Documentation/devicetree/bindings/sound/cs35l35.txt b/Documentation/devicetree/bindings/sound/cs35l35.txt new file mode 100644 index 0000000..8a93b1b --- /dev/null +++ b/Documentation/devicetree/bindings/sound/cs35l35.txt @@ -0,0 +1,171 @@ +CS35L35 Speaker Amplifier + +Required properties: + + - compatible : "cirrus,cs35l35" + + - reg : the I2C address of the device for I2C + + - VA-supply, VP-supply : power supplies for the device, + as covered in + Documentation/devicetree/bindings/regulator/regulator.txt. + + - interrupt-parent : Specifies the phandle of the interrupt controller to + which the IRQs from CS35L35 are delivered to. + - interrupts : IRQ line info CS35L35. + (See Documentation/devicetree/bindings/interrupt-controller/interrupts.txt + for further information relating to interrupt properties) + +Optional properties: + - reset-gpios : gpio used to reset the amplifier + + - cirrus,stereo-config : Boolean to determine if there are 2 AMPs for a + Stereo configuration + + - cirrus,audio-channel : Set Location of Audio Signal on Serial Port + 0 = Data Packet received on Left I2S Channel + 1 = Data Packet received on Right I2S Channel + + - cirrus,advisory-channel : Set Location of Advisory Signal on Serial Port + 0 = Data Packet received on Left I2S Channel + 1 = Data Packet received on Right I2S Channel + + - cirrus,shared-boost : Boolean to enable ClassH tracking of Advisory Signal + if 2 Devices share Boost BST_CTL + + - cirrus,sp-drv-strength : Value for setting the Serial Port drive strength + Table 3-10 of the datasheet lists drive-strength specifications + 0 = 1x (Default) + 1 = .5x + + - cirrus,bst-pdn-fet-on : Boolean to determine if the Boost PDN control + powers down with a rectification FET On or Off. If VSPK is supplied + externally then FET is off. + + - cirrus,boost-ctl : Boost Converter control word. + Step Size: 100 + 0x00 = (Default) VP + Range is 0x01...0x41 + + - cirrus,boost-ipk : Boost-converter peak current limit. + Configures the peak current by monitoring the current through the boost FET. + Step size: 112 + 0x07 = (Default) 2.46 Amps + Range is 0x00...0x19 + + - cirrus,amp-gain-zc : Boolean to determine if to use Amplifier gain-change + zero-cross + +Optional H/G Algorithm sub-node: + + The cs35l35 node can have a single "cirrus,classh-internal-algo" sub-node + that will disable automatic control of the internal H/G Algorithm. + + It is strongly recommended that the Datasheet be referenced when adjusting + or using these Class H Algorithm controls over the internal Algorithm. + Serious damage can occur to the Device and surrounding components. + + - cirrus,classh-internal-algo : Sub-node for the Internal Class H Algorithm + See Section 4.3 Internal Class H Algorithm in the Datasheet. + If not used, the device manages the ClassH Algorithm internally. + +Optional properties for the "cirrus,classh-internal-algo" Sub-node + + Section 7.29 Class H Control + - cirrus,classh-bst-overide : Boolean + - cirrus,classh-bst-max-limit + - cirrus,classh-mem-depth + + Section 7.30 Class H Headroom Control + - cirrus,classh-headroom + + Section 7.31 Class H Release Rate + - cirrus,classh-release-rate + + Section 7.32 Class H Weak FET Drive Control + - cirrus,classh-wk-fet-disable + - cirrus,classh-wk-fet-delay + - cirrus,classh-wk-fet-thld + + Section 7.34 Class H VP Control + - cirrus,classh-vpch-auto + - cirrus,classh-vpch-rate + - cirrus,classh-vpch-man + +Optional Monitor Signal Format sub-node: + + The cs35l35 node can have a single "cirrus,monitor-signal-format" sub-node + for adjusting the Depth, Location and Frame of the Monitoring Signals + for Algorithms. + + See Sections 4.8.2 through 4.8.4 Serial-Port Control in the Datasheet + + -cirrus,monitor-signal-format : Sub-node for the Monitor Signaling Formating + on the I2S Port. Each of the 3 8 bit values in the array contain the settings + for depth, location, and frame. + + If not used, the defaults for the 6 monitor signals is used. + + Sections 7.44 - 7.53 lists values for the depth, location, and frame + for each monitoring signal. + + - cirrus,imon : 3 8 bit values to set the depth, location, and frame + of the IMON monitor signal. + + - cirrus,vmon : 3 8 bit values to set the depth, location, and frame + of the VMON monitor signal. + + - cirrus,vpmon : 3 8 bit values to set the depth, location, and frame + of the VPMON monitor signal. + + - cirrus,vbstmon : 3 8 bit values to set the depth, location, and frame + of the VBSTMON monitor signal + + - cirrus,vpbrstat : 3 8 bit values to set the depth, location, and frame + of the VPBRSTAT monitor signal + + - cirrus,zerofill : 3 8 bit values to set the depth, location, and frame\ + of the ZEROFILL packet in the monitor signal + +Example: + +cs35l35: cs35l35@20 { + compatible = "cirrus,cs35l35"; + reg = <0x20>; + VA-supply = <&dummy_vreg>; + VP-supply = <&dummy_vreg>; + reset-gpios = <&axi_gpio 54 0>; + interrupt-parent = <&gpio8>; + interrupts = <3 IRQ_TYPE_LEVEL_LOW>; + cirrus,boost-ctl = <0x41>; + + cirrus,stereo-config; + cirrus,audio-channel = <0x00>; + cirrus,advisory-channel = <0x01>; + cirrus,shared-boost; + + cirrus,classh-internal-algo { + cirrus,classh-bst-overide; + cirrus,classh-bst-max-limit = <0x01>; + cirrus,classh-mem-depth = <0x01>; + cirrus,classh-release-rate = <0x08>; + cirrus,classh-headroom-millivolt = <0x0B>; + cirrus,classh-wk-fet-disable = <0x01>; + cirrus,classh-wk-fet-delay = <0x04>; + cirrus,classh-wk-fet-thld = <0x01>; + cirrus,classh-vpch-auto = <0x01>; + cirrus,classh-vpch-rate = <0x02>; + cirrus,classh-vpch-man = <0x05>; + }; + + /* Depth, Location, Frame */ + cirrus,monitor-signal-format { + cirrus,imon = /bits/ 8 <0x03 0x00 0x01>; + cirrus,vmon = /bits/ 8 <0x03 0x00 0x00>; + cirrus,vpmon = /bits/ 8 <0x03 0x04 0x00>; + cirrus,vbstmon = /bits/ 8 <0x03 0x04 0x01>; + cirrus,vpbrstat = /bits/ 8 <0x00 0x04 0x00>; + cirrus,zerofill = /bits/ 8 <0x00 0x00 0x00>; + }; + +}; -- 1.9.1

2 3

[alsa-devel] [PATCH v8] ASoC: omap-mcbsp: Add PM QoS support for McBSP to prevent glitches
by Matt Ranostay 02 Feb '17

02 Feb '17

We can get audio errors if hitting deeper idle states on omaps: [alsa.c:230] error: Fatal problem with alsa output, error -5. [audio.c:614] error: Error in writing audio (Input/output error?)! This seems to happen with off mode idle enabled as power for the whole SoC may get cut off between filling the McBSP fifo using DMA. While active DMA blocks deeper idle states in hardware, McBSP activity does not seem to do so. Basing the QoS latency calculation on the FIFO size, threshold, sample rate, and channels. Based on the original patch by Tony Lindgren Link: https://patchwork.kernel.org/patch/9305867/ Cc: Tony Lindgren <tony(a)atomide.com> Cc: Peter Ujfalusi <peter.ujfalusi(a)ti.com> Signed-off-by: Matt Ranostay <matt(a)ranostay.consulting> --- Changes from v1: * add calculations for latency per number of FIFO locations Changes from v2: * add missing mcbsp.h header change Changes from v3: * base the latency calculations on threshold, buffer size, sample rate, and channels Changes from v4: * using Peter Ujfalusi's suggestions for restoring a higher latency on audio stream completion, or if not applicable remove the QoS request Changes from v5: * clean up latency checking logic * move logic to .prepare and .shutdown to avoid functions that can sleep Changes from v6: * move QoS removal to asoc_mcbsp_remove from omap_mcbsp_cleanup * also remove header include that is unneeded Changes from v7: * fix issue reported by Tony Lindgreen that a player application could close the card after hw_params, and cause an invalid pm_qos_remove_request() sound/soc/omap/mcbsp.h | 3 +++ sound/soc/omap/omap-mcbsp.c | 48 ++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 50 insertions(+), 1 deletion(-) diff --git a/sound/soc/omap/mcbsp.h b/sound/soc/omap/mcbsp.h index 61e93b1c185d..46ae1269a698 100644 --- a/sound/soc/omap/mcbsp.h +++ b/sound/soc/omap/mcbsp.h @@ -323,8 +323,11 @@ struct omap_mcbsp { unsigned int fmt; unsigned int in_freq; + unsigned int latency[2]; int clk_div; int wlen; + + struct pm_qos_request pm_qos_req; }; void omap_mcbsp_config(struct omap_mcbsp *mcbsp, diff --git a/sound/soc/omap/omap-mcbsp.c b/sound/soc/omap/omap-mcbsp.c index d018e966e533..6b40bdbef336 100644 --- a/sound/soc/omap/omap-mcbsp.c +++ b/sound/soc/omap/omap-mcbsp.c @@ -157,6 +157,17 @@ static void omap_mcbsp_dai_shutdown(struct snd_pcm_substream *substream, struct snd_soc_dai *cpu_dai) { struct omap_mcbsp *mcbsp = snd_soc_dai_get_drvdata(cpu_dai); + int tx = (substream->stream == SNDRV_PCM_STREAM_PLAYBACK); + int stream1 = tx ? SNDRV_PCM_STREAM_PLAYBACK : SNDRV_PCM_STREAM_CAPTURE; + int stream2 = tx ? SNDRV_PCM_STREAM_CAPTURE : SNDRV_PCM_STREAM_PLAYBACK; + + if (mcbsp->latency[stream2]) + pm_qos_update_request(&mcbsp->pm_qos_req, + mcbsp->latency[stream2]); + else if (mcbsp->latency[stream1]) + pm_qos_remove_request(&mcbsp->pm_qos_req); + + mcbsp->latency[stream1] = 0; if (!cpu_dai->active) { omap_mcbsp_free(mcbsp); @@ -164,6 +175,28 @@ static void omap_mcbsp_dai_shutdown(struct snd_pcm_substream *substream, } } +static int omap_mcbsp_dai_prepare(struct snd_pcm_substream *substream, + struct snd_soc_dai *cpu_dai) +{ + struct omap_mcbsp *mcbsp = snd_soc_dai_get_drvdata(cpu_dai); + struct pm_qos_request *pm_qos_req = &mcbsp->pm_qos_req; + int tx = (substream->stream == SNDRV_PCM_STREAM_PLAYBACK); + int stream1 = tx ? SNDRV_PCM_STREAM_PLAYBACK : SNDRV_PCM_STREAM_CAPTURE; + int stream2 = tx ? SNDRV_PCM_STREAM_CAPTURE : SNDRV_PCM_STREAM_PLAYBACK; + int latency = mcbsp->latency[stream2]; + + /* Prevent omap hardware from hitting off between FIFO fills */ + if (!latency || mcbsp->latency[stream1] < latency) + latency = mcbsp->latency[stream1]; + + if (pm_qos_request_active(pm_qos_req)) + pm_qos_update_request(pm_qos_req, latency); + else if (latency) + pm_qos_add_request(pm_qos_req, PM_QOS_CPU_DMA_LATENCY, latency); + + return 0; +} + static int omap_mcbsp_dai_trigger(struct snd_pcm_substream *substream, int cmd, struct snd_soc_dai *cpu_dai) { @@ -226,6 +259,7 @@ static int omap_mcbsp_dai_hw_params(struct snd_pcm_substream *substream, int wlen, channels, wpf; int pkt_size = 0; unsigned int format, div, framesize, master; + unsigned int buffer_size = mcbsp->pdata->buffer_size; dma_data = snd_soc_dai_get_dma_data(cpu_dai, substream); channels = params_channels(params); @@ -240,7 +274,9 @@ static int omap_mcbsp_dai_hw_params(struct snd_pcm_substream *substream, default: return -EINVAL; } - if (mcbsp->pdata->buffer_size) { + if (buffer_size) { + int latency; + if (mcbsp->dma_op_mode == MCBSP_DMA_MODE_THRESHOLD) { int period_words, max_thrsh; int divider = 0; @@ -271,6 +307,12 @@ static int omap_mcbsp_dai_hw_params(struct snd_pcm_substream *substream, /* Use packet mode for non mono streams */ pkt_size = channels; } + + latency = ((((buffer_size - pkt_size) / channels) * 1000) + / (params->rate_num / params->rate_den)); + + mcbsp->latency[substream->stream] = latency; + omap_mcbsp_set_threshold(substream, pkt_size); } @@ -554,6 +596,7 @@ static int omap_mcbsp_dai_set_dai_sysclk(struct snd_soc_dai *cpu_dai, static const struct snd_soc_dai_ops mcbsp_dai_ops = { .startup = omap_mcbsp_dai_startup, .shutdown = omap_mcbsp_dai_shutdown, + .prepare = omap_mcbsp_dai_prepare, .trigger = omap_mcbsp_dai_trigger, .delay = omap_mcbsp_dai_delay, .hw_params = omap_mcbsp_dai_hw_params, @@ -835,6 +878,9 @@ static int asoc_mcbsp_remove(struct platform_device *pdev) if (mcbsp->pdata->ops && mcbsp->pdata->ops->free) mcbsp->pdata->ops->free(mcbsp->id); + if (pm_qos_request_active(&mcbsp->pm_qos_req)) + pm_qos_remove_request(&mcbsp->pm_qos_req); + omap_mcbsp_cleanup(mcbsp); clk_put(mcbsp->fclk); -- 2.10.2

5 11

[alsa-devel] [PATCH 0/2] Baytrail/Cherrytrail audio fixes - take2
by Pierre-Louis Bossart 02 Feb '17

02 Feb '17

Correction to previous series (applies on top for Mark's for-next) - better handling of missing gpiod, continue only on -ENOENT - compiler warning Pierre-Louis Bossart (2): ASoC: rt5645: fix error handling for gpio detection ASoC: cht-bsw-rt5645: fix unused variable compiler warning sound/soc/codecs/rt5645.c | 10 ++++++++-- sound/soc/intel/boards/cht_bsw_rt5645.c | 1 - 2 files changed, 8 insertions(+), 3 deletions(-) -- 2.7.4

3 6