[alsa-devel] [PATCH v3 0/5] Beaglebone-Black HDMI audio
Changes since v2: - Change compatible property from "ti,gpio-clock" to "ti,gpio-gate-clock" - Some minor cleanups
The code has a functional dependency to: http://www.mail-archive.com/linux-omap@vger.kernel.org/msg109264.html
Without the above patch the audio card does not probe.
The code has been rebased on top of Linux 3.17-rc5. The patches bellow, the above dependency, and couple of commits to add BBB HDMI audio support to omap2plus_defconfig can be pulled from:
https://github.com/jsarha/linux.git linux-master-bbb-hdmi-audio
Cheers, Jyri
Jyri Sarha (5): clk: ti: add "ti,gpio-gate-clock" controlled clock drm/tilcdc: Add I2S HDMI audio config for tda998x ASoC: davinci-evm: HDMI audio support for TDA998x trough McASP I2S bus ASoC: davinci: HDMI audio build for AM33XX and TDA998x ARM: dts: am335x-boneblack: Add HDMI audio support
.../bindings/clock/ti/gpio-gate-clock.txt | 21 ++ .../bindings/sound/davinci-evm-audio.txt | 6 +- arch/arm/boot/dts/am335x-boneblack.dts | 52 +++++ drivers/clk/ti/Makefile | 2 +- drivers/clk/ti/gpio.c | 202 ++++++++++++++++++++ drivers/gpu/drm/tilcdc/tilcdc_slave.c | 24 ++- sound/soc/davinci/Kconfig | 12 ++ sound/soc/davinci/davinci-evm.c | 82 +++++++- 8 files changed, 395 insertions(+), 6 deletions(-) create mode 100644 Documentation/devicetree/bindings/clock/ti/gpio-gate-clock.txt create mode 100644 drivers/clk/ti/gpio.c
The added ti,gpio-gate-clock is a basic clock that can be enabled and disabled trough a gpio output. The DT binding document for the clock is also added. For EPROBE_DEFER handling the registering of the clock has to be delayed until of_clk_get() call time.
Signed-off-by: Jyri Sarha jsarha@ti.com --- .../bindings/clock/ti/gpio-gate-clock.txt | 21 ++ drivers/clk/ti/Makefile | 2 +- drivers/clk/ti/gpio.c | 202 ++++++++++++++++++++ 3 files changed, 224 insertions(+), 1 deletion(-) create mode 100644 Documentation/devicetree/bindings/clock/ti/gpio-gate-clock.txt create mode 100644 drivers/clk/ti/gpio.c
diff --git a/Documentation/devicetree/bindings/clock/ti/gpio-gate-clock.txt b/Documentation/devicetree/bindings/clock/ti/gpio-gate-clock.txt new file mode 100644 index 0000000..6156f82 --- /dev/null +++ b/Documentation/devicetree/bindings/clock/ti/gpio-gate-clock.txt @@ -0,0 +1,21 @@ +Binding for simple gpio controlled clock. + +This binding uses the common clock binding[1]. + +[1] Documentation/devicetree/bindings/clock/clock-bindings.txt + +Required properties: +- compatible : shall be "ti,gpio-gate-clock". +- #clock-cells : from common clock binding; shall be set to 0. +- enable-gpios : GPIO reference for enabling and disabling the clock. + +Optional properties: +- clocks: Maximum of one parent clock is supported. + +Example: + clock { + compatible = "ti,gpio-gate-clock"; + clocks = <&parentclk>; + #clock-cells = <0>; + enable-gpios = <&gpio 1 GPIO_ACTIVE_HIGH>; + }; diff --git a/drivers/clk/ti/Makefile b/drivers/clk/ti/Makefile index ed4d0aa..88074d3 100644 --- a/drivers/clk/ti/Makefile +++ b/drivers/clk/ti/Makefile @@ -1,7 +1,7 @@ ifneq ($(CONFIG_OF),) obj-y += clk.o autoidle.o clockdomain.o clk-common = dpll.o composite.o divider.o gate.o \ - fixed-factor.o mux.o apll.o + fixed-factor.o mux.o apll.o gpio.o obj-$(CONFIG_SOC_AM33XX) += $(clk-common) clk-33xx.o obj-$(CONFIG_ARCH_OMAP2) += $(clk-common) interface.o clk-2xxx.o obj-$(CONFIG_ARCH_OMAP3) += $(clk-common) interface.o clk-3xxx.o diff --git a/drivers/clk/ti/gpio.c b/drivers/clk/ti/gpio.c new file mode 100644 index 0000000..b973801 --- /dev/null +++ b/drivers/clk/ti/gpio.c @@ -0,0 +1,202 @@ +/* + * Copyright (C) 2012 - 2014 Texas Instruments Incorporated - http://www.ti.com + * Author: Jyri Sarha jsarha@ti.com + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Gpio controlled clock implementation + */ + +#include <linux/clk-provider.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/gpio.h> +#include <linux/of_gpio.h> +#include <linux/err.h> +#include <linux/device.h> + +struct clk_gpio { + struct clk_hw hw; + struct gpio_desc *gpiod; +}; + +/** + * DOC: basic gpio controlled clock which can be enabled and disabled + * with gpio output + * Traits of this clock: + * prepare - clk_(un)prepare only ensures parent is (un)prepared + * enable - clk_enable and clk_disable are functional & control gpio + * rate - inherits rate from parent. No clk_set_rate support + * parent - fixed parent. No clk_set_parent support + */ + +#define to_clk_gpio(_hw) container_of(_hw, struct clk_gpio, hw) + +static int clk_gpio_enable(struct clk_hw *hw) +{ + struct clk_gpio *clk = to_clk_gpio(hw); + + gpiod_set_value(clk->gpiod, 1); + + return 0; +} + +static void clk_gpio_disable(struct clk_hw *hw) +{ + struct clk_gpio *clk = to_clk_gpio(hw); + + gpiod_set_value(clk->gpiod, 0); +} + +static int clk_gpio_is_enabled(struct clk_hw *hw) +{ + struct clk_gpio *clk = to_clk_gpio(hw); + + return gpiod_get_value(clk->gpiod); +} + +static const struct clk_ops clk_gpio_ops = { + .enable = clk_gpio_enable, + .disable = clk_gpio_disable, + .is_enabled = clk_gpio_is_enabled, +}; + +/** + * clk_register_gpio - register a gpip clock with the clock framework + * @dev: device that is registering this clock + * @name: name of this clock + * @parent_name: name of this clock's parent + * @gpiod: gpio descriptor to control this clock + */ +static struct clk *clk_register_gpio(struct device *dev, const char *name, + const char *parent_name, struct gpio_desc *gpiod, + unsigned long flags) +{ + struct clk_gpio *clk_gpio = NULL; + struct clk *clk = ERR_PTR(-EINVAL); + struct clk_init_data init = { NULL }; + unsigned long gpio_flags; + int err; + + if (gpiod_is_active_low(gpiod)) + gpio_flags = GPIOF_OUT_INIT_HIGH; + else + gpio_flags = GPIOF_OUT_INIT_LOW; + + if (dev) + err = devm_gpio_request_one(dev, desc_to_gpio(gpiod), + gpio_flags, name); + else + err = gpio_request_one(desc_to_gpio(gpiod), gpio_flags, name); + + if (err) { + pr_err("%s: %s: Error requesting clock control gpio %u\n", + __func__, name, desc_to_gpio(gpiod)); + return ERR_PTR(err); + } + + if (dev) + clk_gpio = devm_kzalloc(dev, sizeof(struct clk_gpio), + GFP_KERNEL); + else + clk_gpio = kzalloc(sizeof(struct clk_gpio), GFP_KERNEL); + + if (!clk_gpio) + goto clk_register_gpio_err; + + init.name = name; + init.ops = &clk_gpio_ops; + init.flags = flags | CLK_IS_BASIC; + init.parent_names = (parent_name ? &parent_name : NULL); + init.num_parents = (parent_name ? 1 : 0); + + clk_gpio->gpiod = gpiod; + clk_gpio->hw.init = &init; + + clk = clk_register(dev, &clk_gpio->hw); + + if (!IS_ERR(clk)) + return clk; + + if (!dev) + kfree(clk_gpio); + +clk_register_gpio_err: + gpiod_put(gpiod); + + return clk; +} + +/** + * The clk_register_gpio has to be delayed, because the EPROBE_DEFER + * can not be handled properly at of_clk_init() call time. + */ + +struct clk_gpio_delayed_register_data { + struct device_node *node; + struct mutex lock; + struct clk *clk; +}; + +static +struct clk *of_clk_gpio_delayed_register_get(struct of_phandle_args *clkspec, + void *_data) +{ + struct clk_gpio_delayed_register_data *data = _data; + struct clk *clk; + const char *clk_name = data->node->name; + const char *parent_name; + struct gpio_desc *gpiod; + int gpio; + + mutex_lock(&data->lock); + + if (data->clk) { + mutex_unlock(&data->lock); + return data->clk; + } + + gpio = of_get_named_gpio_flags(data->node, "enable-gpios", 0, NULL); + if (gpio < 0) { + mutex_unlock(&data->lock); + if (gpio != -EPROBE_DEFER) + pr_err("%s: %s: Can't get 'enable-gpios' DT property\n", + __func__, clk_name); + return ERR_PTR(gpio); + } + gpiod = gpio_to_desc(gpio); + + parent_name = of_clk_get_parent_name(data->node, 0); + + clk = clk_register_gpio(NULL, clk_name, parent_name, gpiod, 0); + if (IS_ERR(clk)) { + mutex_unlock(&data->lock); + return clk; + } + + data->clk = clk; + mutex_unlock(&data->lock); + + return clk; +} + +/** + * of_gpio_clk_setup() - Setup function for gpio controlled clock + */ +static void __init of_gpio_clk_setup(struct device_node *node) +{ + struct clk_gpio_delayed_register_data *data; + + data = kzalloc(sizeof(struct clk_gpio_delayed_register_data), + GFP_KERNEL); + if (!data) + return; + + data->node = node; + mutex_init(&data->lock); + + of_clk_add_provider(node, of_clk_gpio_delayed_register_get, data); +} +CLK_OF_DECLARE(gpio_clk, "ti,gpio-gate-clock", of_gpio_clk_setup);
On 16/09/14 23:40, Jyri Sarha wrote:
The added ti,gpio-gate-clock is a basic clock that can be enabled and disabled trough a gpio output. The DT binding document for the clock is also added. For EPROBE_DEFER handling the registering of the clock has to be delayed until of_clk_get() call time.
Signed-off-by: Jyri Sarha jsarha@ti.com
.../bindings/clock/ti/gpio-gate-clock.txt | 21 ++ drivers/clk/ti/Makefile | 2 +- drivers/clk/ti/gpio.c | 202 ++++++++++++++++++++ 3 files changed, 224 insertions(+), 1 deletion(-) create mode 100644 Documentation/devicetree/bindings/clock/ti/gpio-gate-clock.txt create mode 100644 drivers/clk/ti/gpio.c
Why is this a TI clock? Sounds like a generic one to me.
In any case, this should go through Mike.
Tomi
The configuration is needed for HDMI audio. The "swap" and "mirr" parameters have to be correctly set in the configuration in order to have proper colors in the HDMI picture.
Signed-off-by: Jyri Sarha jsarha@ti.com --- drivers/gpu/drm/tilcdc/tilcdc_slave.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_slave.c b/drivers/gpu/drm/tilcdc/tilcdc_slave.c index 3775fd4..5dd8d78 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_slave.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_slave.c @@ -19,6 +19,7 @@ #include <linux/pinctrl/pinmux.h> #include <linux/pinctrl/consumer.h> #include <drm/drm_encoder_slave.h> +#include <drm/i2c/tda998x.h>
#include "tilcdc_drv.h"
@@ -111,8 +112,29 @@ static const struct drm_encoder_helper_funcs slave_encoder_helper_funcs = { .restore = drm_i2c_encoder_restore, };
+static struct tda998x_encoder_params tda998x_pdata = { + .swap_b = 0x3, + .mirr_b = 0x0, + .swap_a = 0x2, + .mirr_a = 0x0, + .swap_d = 0x1, + .mirr_d = 0x0, + .swap_c = 0x0, + .mirr_c = 0x0, + .swap_f = 0x5, + .mirr_f = 0x0, + .swap_e = 0x4, + .mirr_e = 0x0, + .audio_cfg = 0x3, /* I2S mode */ + .audio_clk_cfg = 1, /* select clock pin */ + .audio_frame[1] = 1, /* channels - 1 */ + .audio_format = AFMT_I2S, + .audio_sample_rate = 48000, +}; + static const struct i2c_board_info info = { - I2C_BOARD_INFO("tda998x", 0x70) + I2C_BOARD_INFO("tda998x", 0x70), + .platform_data = &tda998x_pdata, };
static struct drm_encoder *slave_encoder_create(struct drm_device *dev,
Add machine driver support for BeagleBone-Black HDMI audio. BBB has NXP TDA998X HDMI transmitter connected to McASP port in I2S mode. The 44100 Hz sample-rate and it's multiples can not be accurately produced on BBB. The only supported sample format is SNDRV_PCM_FORMAT_S32_LE. The 8 least significant bits are ignored.
Signed-off-by: Jyri Sarha jsarha@ti.com --- .../bindings/sound/davinci-evm-audio.txt | 6 +- sound/soc/davinci/davinci-evm.c | 82 +++++++++++++++++++- 2 files changed, 84 insertions(+), 4 deletions(-)
diff --git a/Documentation/devicetree/bindings/sound/davinci-evm-audio.txt b/Documentation/devicetree/bindings/sound/davinci-evm-audio.txt index 963e100..7560146 100644 --- a/Documentation/devicetree/bindings/sound/davinci-evm-audio.txt +++ b/Documentation/devicetree/bindings/sound/davinci-evm-audio.txt @@ -1,7 +1,9 @@ -* Texas Instruments SoC audio setups with TLV320AIC3X Codec +* TI SoC audio using McASP to connect to TLV320AIC3X or HDMI endcoder
Required properties: -- compatible : "ti,da830-evm-audio" : forDM365/DA8xx/OMAPL1x/AM33xx +- compatible: should contain one of: + * "ti,da830-evm-audio" for DM365/DA8xx/OMAPL1x/AM33xx with TLV320AIC3X + * "ti,beaglebone-black-audio" for Beaglebone-black HDMI audio - ti,model : The user-visible name of this sound complex. - ti,audio-codec : The phandle of the TLV320AIC3x audio codec - ti,mcasp-controller : The phandle of the McASP controller diff --git a/sound/soc/davinci/davinci-evm.c b/sound/soc/davinci/davinci-evm.c index a50010e..3d0acba2a 100644 --- a/sound/soc/davinci/davinci-evm.c +++ b/sound/soc/davinci/davinci-evm.c @@ -21,6 +21,7 @@ #include <sound/core.h> #include <sound/pcm.h> #include <sound/soc.h> +#include <sound/pcm_params.h>
#include <asm/dma.h> #include <asm/mach-types.h> @@ -83,12 +84,46 @@ static int evm_hw_params(struct snd_pcm_substream *substream, return 0; }
+/* If changing sample format the tda998x configuration (REG_CTS_N) needs + to be changed. */ +#define TDA998X_SAMPLE_FORMAT SNDRV_PCM_FORMAT_S32_LE +static int evm_tda998x_startup(struct snd_pcm_substream *substream) +{ + struct snd_pcm_runtime *runtime = substream->runtime; + struct snd_mask *fmt = constrs_mask(&runtime->hw_constraints, + SNDRV_PCM_HW_PARAM_FORMAT); + snd_mask_none(fmt); + snd_mask_set(fmt, TDA998X_SAMPLE_FORMAT); + + return evm_startup(substream); +} + +static int evm_tda998x_hw_params(struct snd_pcm_substream *substream, + struct snd_pcm_hw_params *params) +{ + struct snd_soc_pcm_runtime *rtd = substream->private_data; + struct snd_soc_dai *cpu_dai = rtd->cpu_dai; + struct snd_soc_card *soc_card = rtd->card; + struct snd_soc_card_drvdata_davinci *drvdata = + snd_soc_card_get_drvdata(soc_card); + + return snd_soc_dai_set_sysclk(cpu_dai, 0, drvdata->sysclk, + SND_SOC_CLOCK_IN); +} + static struct snd_soc_ops evm_ops = { .startup = evm_startup, .shutdown = evm_shutdown, .hw_params = evm_hw_params, };
+ +static struct snd_soc_ops evm_tda998x_ops = { + .startup = evm_tda998x_startup, + .shutdown = evm_shutdown, + .hw_params = evm_tda998x_hw_params, +}; + /* davinci-evm machine dapm widgets */ static const struct snd_soc_dapm_widget aic3x_dapm_widgets[] = { SND_SOC_DAPM_HP("Headphone Jack", NULL), @@ -149,6 +184,35 @@ static int evm_aic3x_init(struct snd_soc_pcm_runtime *rtd) return 0; }
+static const struct snd_soc_dapm_widget tda998x_dapm_widgets[] = { + SND_SOC_DAPM_OUTPUT("HDMI Out"), +}; + +static int evm_tda998x_init(struct snd_soc_pcm_runtime *rtd) +{ + struct snd_soc_dai *cpu_dai = rtd->cpu_dai; + struct snd_soc_dapm_context *dapm = &rtd->codec->dapm; + struct snd_soc_card *soc_card = rtd->card; + int ret; + + ret = snd_soc_dai_set_clkdiv(cpu_dai, 0, 1); + if (ret < 0) + return ret; + + snd_soc_dapm_new_controls(dapm, tda998x_dapm_widgets, + ARRAY_SIZE(tda998x_dapm_widgets)); + + ret = snd_soc_of_parse_audio_routing(soc_card, "ti,audio-routing"); + + /* not connected */ + snd_soc_dapm_disable_pin(dapm, "RX"); + + /* always connected */ + snd_soc_dapm_enable_pin(dapm, "HDMI Out"); + + return 0; +} + /* davinci-evm digital audio interface glue - connects codec <--> CPU */ static struct snd_soc_dai_link dm6446_evm_dai = { .name = "TLV320AIC3X", @@ -334,7 +398,7 @@ static struct snd_soc_card da850_snd_soc_card = { #if defined(CONFIG_OF)
/* - * The struct is used as place holder. It will be completely + * The structs are used as place holders. They will be completely * filled with data from dt node. */ static struct snd_soc_dai_link evm_dai_tlv320aic3x = { @@ -347,10 +411,24 @@ static struct snd_soc_dai_link evm_dai_tlv320aic3x = { SND_SOC_DAIFMT_IB_NF, };
+static struct snd_soc_dai_link evm_dai_tda998x_hdmi = { + .name = "NXP TDA998x HDMI Chip", + .stream_name = "HDMI", + .codec_dai_name = "hdmi-hifi", + .ops = &evm_tda998x_ops, + .init = evm_tda998x_init, + .dai_fmt = (SND_SOC_DAIFMT_CBS_CFS | SND_SOC_DAIFMT_I2S | + SND_SOC_DAIFMT_IB_NF), +}; + static const struct of_device_id davinci_evm_dt_ids[] = { { .compatible = "ti,da830-evm-audio", - .data = (void *) &evm_dai_tlv320aic3x, + .data = &evm_dai_tlv320aic3x, + }, + { + .compatible = "ti,beaglebone-black-audio", + .data = &evm_dai_tda998x_hdmi, }, { /* sentinel */ } };
Adds configuration option for HDMI audio support for AM33XX based boards with NXP TDA998x HDMI transmitter. The audio is connected to NXP TDA998x trough McASP running in i2s mode.
Signed-off-by: Jyri Sarha jsarha@ti.com --- sound/soc/davinci/Kconfig | 12 ++++++++++++ 1 file changed, 12 insertions(+)
diff --git a/sound/soc/davinci/Kconfig b/sound/soc/davinci/Kconfig index d69510c..0c8f143 100644 --- a/sound/soc/davinci/Kconfig +++ b/sound/soc/davinci/Kconfig @@ -43,6 +43,18 @@ config SND_AM33XX_SOC_EVM AM335X-EVMSK, and BeagelBone with AudioCape boards have this setup.
+config SND_AM335X_SOC_NXPTDA_EVM + tristate "HDMI Audio for the AM33XX chip based boards with TDA998x" + depends on SND_EDMA_SOC && SOC_AM33XX && DRM_I2C_NXP_TDA998X + select SND_SOC_HDMI_CODEC + select SND_DAVINCI_SOC_MCASP + select SND_DAVINCI_SOC_GENERIC_EVM + help + Say Y or M if you want to add support for HDMI SoC audio on + AM33XX boards with NXP TDA998x HDMI transmitter. For example + BeagleBoneBack. The audio is connected to NXP TDA998x trough + McASP running in i2s mode. + config SND_DAVINCI_SOC_EVM tristate "SoC Audio support for DaVinci DM6446, DM355 or DM365 EVM" depends on SND_DAVINCI_SOC && I2C
On Tue, Sep 16, 2014 at 11:40:17PM +0300, Jyri Sarha wrote:
Adds configuration option for HDMI audio support for AM33XX based boards with NXP TDA998x HDMI transmitter. The audio is connected to NXP TDA998x trough McASP running in i2s mode.
So, Jean-Francois is also trying to do things with the TDA998x - what's the story with that, is this joined up at all?
On 09/17/2014 10:41 PM, Mark Brown wrote:
On Tue, Sep 16, 2014 at 11:40:17PM +0300, Jyri Sarha wrote:
Adds configuration option for HDMI audio support for AM33XX based boards with NXP TDA998x HDMI transmitter. The audio is connected to NXP TDA998x trough McASP running in i2s mode.
So, Jean-Francois is also trying to do things with the TDA998x - what's the story with that, is this joined up at all?
Not really. This basic functionality does not touch tda998x at all on the fly, but just sets i2s configuation in the beginning and bangs the bits trough McASP. But as long as the old platform data for tda998x still works after Jean-Francois' patches I should be safe.
The problem with the Jean-Francois' patches is the DT support. The BBB HDMI video is implemented trough tilcdc-slave mechanism without a DT node for the tda encoder, which renders Jean-Francois' approach unusable for BBB.
The whole tilcdc slave approach may not be the most elegant way to use the tda driver, but it does not look like it is going to change any time soon.
Best regards, Jyri
ps. I have been thinking on something similar to Jean-Francois' patch for SiI9022 which I have been working on lately. Also I have been wondering if it would be possible to come up with a generic ASoC codec component driver or library that could be used with any HDMI encoder to produce the ASoC codec component. Unfortunately I am in too early stage to produce anything more concrete.
On Thu, 18 Sep 2014 00:13:09 +0300 Jyri Sarha jsarha@ti.com wrote:
So, Jean-Francois is also trying to do things with the TDA998x - what's the story with that, is this joined up at all?
Not really. This basic functionality does not touch tda998x at all on the fly, but just sets i2s configuation in the beginning and bangs the bits trough McASP. But as long as the old platform data for tda998x still works after Jean-Francois' patches I should be safe.
It should. And you may test it.
The problem with the Jean-Francois' patches is the DT support. The BBB HDMI video is implemented trough tilcdc-slave mechanism without a DT node for the tda encoder, which renders Jean-Francois' approach unusable for BBB.
I used the TDA998x with DT support for a long time with a hack by using parts of drm_i2c_encoder_init(). Now, I think that this function could be used as it is just setting no platform_data in info.
The whole tilcdc slave approach may not be the most elegant way to use the tda driver, but it does not look like it is going to change any time soon.
Right: I proposed a patch for that and it was rejected.
Best regards, Jyri
ps. I have been thinking on something similar to Jean-Francois' patch for SiI9022 which I have been working on lately. Also I have been wondering if it would be possible to come up with a generic ASoC codec component driver or library that could be used with any HDMI encoder to produce the ASoC codec component. Unfortunately I am in too early stage to produce anything more concrete.
Here are some thoughts about this topic.
The video and audio worlds don't know about each other. The only solution I found is to let the encoder create the codec as a child device. Then, the codec knows from which encoder it depends. This could have been done using Russell's components, but the codec should have been declared in the DT. This is not useful.
The codec interacts with the encoder in 2 ways: - it uses the HDMI parameters retrieved by the encoder, - it gives the audio source type to the encoder. I used exported functions for that, but, for a generic codec, theses functions could be given through the codec platform_data.
The codec declares the DAI(s) prior to know the encoder. The DAI table must be in the codec because of the snd_soc_dai_ops. For a generic codec, this DAI table could be built dynamically from information (name, id) also given through the codec platform_data.
On 09/18/2014 11:25 AM, Jean-Francois Moine wrote:
On Thu, 18 Sep 2014 00:13:09 +0300 Jyri Sarha jsarha@ti.com wrote:
So, Jean-Francois is also trying to do things with the TDA998x - what's the story with that, is this joined up at all?
Not really. This basic functionality does not touch tda998x at all on the fly, but just sets i2s configuation in the beginning and bangs the bits trough McASP. But as long as the old platform data for tda998x still works after Jean-Francois' patches I should be safe.
It should. And you may test it.
I tested by BBB HDMI audio patches with "[PATCH v5] ASoC: tda998x: Add a codec to the HDMI transmitter", and it works otherwise, but the default for CTS_N_K has changed from 3 to 2. I just needed to change the supported format from S32_LE to S24_LE to compensate the change.
The problem with the Jean-Francois' patches is the DT support. The BBB HDMI video is implemented trough tilcdc-slave mechanism without a DT node for the tda encoder, which renders Jean-Francois' approach unusable for BBB.
I used the TDA998x with DT support for a long time with a hack by using parts of drm_i2c_encoder_init(). Now, I think that this function could be used as it is just setting no platform_data in info.
The whole tilcdc slave approach may not be the most elegant way to use the tda driver, but it does not look like it is going to change any time soon.
Right: I proposed a patch for that and it was rejected.
The tilcdc is not actively maintained by anybody ATM and that is why it is so hard to make any bigger changes to it. I hope we can change this at some point, but for now I am just trying to get a simple HDMI audio working on BBB with minimal changes.
Best regards, Jyri
ps. I have been thinking on something similar to Jean-Francois' patch for SiI9022 which I have been working on lately. Also I have been wondering if it would be possible to come up with a generic ASoC codec component driver or library that could be used with any HDMI encoder to produce the ASoC codec component. Unfortunately I am in too early stage to produce anything more concrete.
Here are some thoughts about this topic.
The video and audio worlds don't know about each other. The only solution I found is to let the encoder create the codec as a child device. Then, the codec knows from which encoder it depends. This could have been done using Russell's components, but the codec should have been declared in the DT. This is not useful.
There is anothere option of putting the ASoC codec code into an ASoC library module and registering the ASoC codec component under the HDMI encoder device. However, this approach does not allow building the ASoC as modules if the HDMI encoder driver is built in. So a separate platform device for the codec component is probably better.
The codec interacts with the encoder in 2 ways:
- it uses the HDMI parameters retrieved by the encoder,
- it gives the audio source type to the encoder.
I used exported functions for that, but, for a generic codec, theses functions could be given through the codec platform_data.
The codec declares the DAI(s) prior to know the encoder. The DAI table must be in the codec because of the snd_soc_dai_ops. For a generic codec, this DAI table could be built dynamically from information (name, id) also given through the codec platform_data.
This is pretty much what I have been thinking too. The API between the generic ASoC HDMI codec could use already existing the linux headers derived HDMI standard. The ASoC side code should build the hdmi audio infoframe and configure the stream header bits etc, so basically do all the generic stuff coming from HDMI standard. Writing the bits to the appropriate registers in the encoder chip would be left to the HDMI encoder driver.
Best regards, Jyri
Adds mcasp0_pins, clk_mcasp0_fixed, clk_mcasp0, mcasp0, hdmi_audio, and sound nodes.
Signed-off-by: Jyri Sarha jsarha@ti.com --- arch/arm/boot/dts/am335x-boneblack.dts | 52 ++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+)
diff --git a/arch/arm/boot/dts/am335x-boneblack.dts b/arch/arm/boot/dts/am335x-boneblack.dts index 305975d..1487f7c 100644 --- a/arch/arm/boot/dts/am335x-boneblack.dts +++ b/arch/arm/boot/dts/am335x-boneblack.dts @@ -59,12 +59,35 @@ 0x1b0 0x03 /* xdma_event_intr0, OMAP_MUX_MODE3 | AM33XX_PIN_OUTPUT */ >; }; + + mcasp0_pins: mcasp0_pins { + pinctrl-single,pins = < + 0x1ac (PIN_INPUT_PULLUP | MUX_MODE0) /* mcasp0_ahclkx.mcasp0_ahclkx */ + 0x19c (PIN_OUTPUT_PULLDOWN | MUX_MODE2) /* mcasp0_ahclkr.mcasp0_axr2 */ + 0x194 (PIN_OUTPUT_PULLUP | MUX_MODE0) /* mcasp0_fsx.mcasp0_fsx */ + 0x190 (PIN_OUTPUT_PULLDOWN | MUX_MODE0) /* mcasp0_aclkx.mcasp0_aclkx */ + 0x06c (PIN_OUTPUT_PULLDOWN | MUX_MODE7) /* gpmc_a11.GPIO1_27 */ + >; + }; };
&lcdc { status = "okay"; };
+&mcasp0 { + pinctrl-names = "default"; + pinctrl-0 = <&mcasp0_pins>; + status = "okay"; + op-mode = <0>; /* MCASP_IIS_MODE */ + tdm-slots = <2>; + serial-dir = < /* 0: INACTIVE, 1: TX, 2: RX */ + 0 0 1 0 + >; + tx-num-evt = <1>; + rx-num-evt = <1>; +}; + / { hdmi { compatible = "ti,tilcdc,slave"; @@ -74,4 +97,33 @@ pinctrl-1 = <&nxp_hdmi_bonelt_off_pins>; status = "okay"; }; + + clk_mcasp0_fixed: clk_mcasp0_fixed { + #clock-cells = <0>; + compatible = "fixed-clock"; + clock-frequency = <24576000>; + }; + + clk_mcasp0: clk_mcasp0 { + #clock-cells = <0>; + compatible = "ti,gpio-gate-clock"; + clocks = <&clk_mcasp0_fixed>; + enable-gpios = <&gpio1 27 0>; /* BeagleBone Black Clk enable on GPIO1_27 */ + }; + + hdmi_audio: hdmi_audio@0 { + compatible = "linux,hdmi-audio"; + status = "okay"; + }; + + sound { + compatible = "ti,beaglebone-black-audio"; + ti,model = "TI BeagleBone Black"; + ti,audio-codec = <&hdmi_audio>; + ti,mcasp-controller = <&mcasp0>; + ti,audio-routing = + "HDMI Out", "TX"; + clocks = <&clk_mcasp0>; + clock-names = "mclk"; + }; };
This patch is invalid now that Mike Turquette took the generic clk-gpio-gate in. The other patches in the series are still Ok.
I'll mail a new dts patch shortly.
Best regards, Jyri
On 09/16/2014 11:40 PM, Jyri Sarha wrote:
Adds mcasp0_pins, clk_mcasp0_fixed, clk_mcasp0, mcasp0, hdmi_audio, and sound nodes.
Signed-off-by: Jyri Sarha jsarha@ti.com
arch/arm/boot/dts/am335x-boneblack.dts | 52 ++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+)
diff --git a/arch/arm/boot/dts/am335x-boneblack.dts b/arch/arm/boot/dts/am335x-boneblack.dts index 305975d..1487f7c 100644 --- a/arch/arm/boot/dts/am335x-boneblack.dts +++ b/arch/arm/boot/dts/am335x-boneblack.dts @@ -59,12 +59,35 @@ 0x1b0 0x03 /* xdma_event_intr0, OMAP_MUX_MODE3 | AM33XX_PIN_OUTPUT */ >; };
mcasp0_pins: mcasp0_pins {
pinctrl-single,pins = <
0x1ac (PIN_INPUT_PULLUP | MUX_MODE0) /* mcasp0_ahclkx.mcasp0_ahclkx */
0x19c (PIN_OUTPUT_PULLDOWN | MUX_MODE2) /* mcasp0_ahclkr.mcasp0_axr2 */
0x194 (PIN_OUTPUT_PULLUP | MUX_MODE0) /* mcasp0_fsx.mcasp0_fsx */
0x190 (PIN_OUTPUT_PULLDOWN | MUX_MODE0) /* mcasp0_aclkx.mcasp0_aclkx */
0x06c (PIN_OUTPUT_PULLDOWN | MUX_MODE7) /* gpmc_a11.GPIO1_27 */
>;
}; };
&lcdc { status = "okay"; };
+&mcasp0 {
- pinctrl-names = "default";
- pinctrl-0 = <&mcasp0_pins>;
- status = "okay";
- op-mode = <0>; /* MCASP_IIS_MODE */
- tdm-slots = <2>;
- serial-dir = < /* 0: INACTIVE, 1: TX, 2: RX */
0 0 1 0
>;
- tx-num-evt = <1>;
- rx-num-evt = <1>;
+};
- / { hdmi { compatible = "ti,tilcdc,slave";
@@ -74,4 +97,33 @@ pinctrl-1 = <&nxp_hdmi_bonelt_off_pins>; status = "okay"; };
- clk_mcasp0_fixed: clk_mcasp0_fixed {
#clock-cells = <0>;
compatible = "fixed-clock";
clock-frequency = <24576000>;
- };
- clk_mcasp0: clk_mcasp0 {
#clock-cells = <0>;
compatible = "ti,gpio-gate-clock";
clocks = <&clk_mcasp0_fixed>;
enable-gpios = <&gpio1 27 0>; /* BeagleBone Black Clk enable on GPIO1_27 */
- };
- hdmi_audio: hdmi_audio@0 {
compatible = "linux,hdmi-audio";
status = "okay";
- };
- sound {
compatible = "ti,beaglebone-black-audio";
ti,model = "TI BeagleBone Black";
ti,audio-codec = <&hdmi_audio>;
ti,mcasp-controller = <&mcasp0>;
ti,audio-routing =
"HDMI Out", "TX";
clocks = <&clk_mcasp0>;
clock-names = "mclk";
- }; };
On 17 September 2014 06:40, Jyri Sarha jsarha@ti.com wrote:
Changes since v2:
- Change compatible property from "ti,gpio-clock" to "ti,gpio-gate-clock"
- Some minor cleanups
The code has a functional dependency to: http://www.mail-archive.com/linux-omap@vger.kernel.org/msg109264.html
Without the above patch the audio card does not probe.
The code has been rebased on top of Linux 3.17-rc5. The patches bellow, the above dependency, and couple of commits to add BBB HDMI audio support to omap2plus_defconfig can be pulled from:
https://github.com/jsarha/linux.git linux-master-bbb-hdmi-audio
How do you intend to get this merge, sending patchsets like this without indication to maintainers on a merge strategy is kinda messy.
I'm not sure how maintained tilcdc is.
Dave.
Cheers, Jyri
Jyri Sarha (5): clk: ti: add "ti,gpio-gate-clock" controlled clock drm/tilcdc: Add I2S HDMI audio config for tda998x ASoC: davinci-evm: HDMI audio support for TDA998x trough McASP I2S bus ASoC: davinci: HDMI audio build for AM33XX and TDA998x ARM: dts: am335x-boneblack: Add HDMI audio support
.../bindings/clock/ti/gpio-gate-clock.txt | 21 ++ .../bindings/sound/davinci-evm-audio.txt | 6 +- arch/arm/boot/dts/am335x-boneblack.dts | 52 +++++ drivers/clk/ti/Makefile | 2 +- drivers/clk/ti/gpio.c | 202 ++++++++++++++++++++ drivers/gpu/drm/tilcdc/tilcdc_slave.c | 24 ++- sound/soc/davinci/Kconfig | 12 ++ sound/soc/davinci/davinci-evm.c | 82 +++++++- 8 files changed, 395 insertions(+), 6 deletions(-) create mode 100644 Documentation/devicetree/bindings/clock/ti/gpio-gate-clock.txt create mode 100644 drivers/clk/ti/gpio.c
-- 1.7.9.5
Alsa-devel mailing list Alsa-devel@alsa-project.org http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
On 09/17/2014 04:06 AM, Dave Airlie wrote:
On 17 September 2014 06:40, Jyri Sarha jsarha@ti.com wrote:
Changes since v2:
- Change compatible property from "ti,gpio-clock" to "ti,gpio-gate-clock"
- Some minor cleanups
The code has a functional dependency to: http://www.mail-archive.com/linux-omap@vger.kernel.org/msg109264.html
Without the above patch the audio card does not probe.
The code has been rebased on top of Linux 3.17-rc5. The patches bellow, the above dependency, and couple of commits to add BBB HDMI audio support to omap2plus_defconfig can be pulled from:
https://github.com/jsarha/linux.git linux-master-bbb-hdmi-audio
How do you intend to get this merge, sending patchsets like this without indication to maintainers on a merge strategy is kinda messy.
I'm not sure how maintained tilcdc is.
Well, it is used but AFAIK the people who have been working with it the most have left TI. I think eventually someone at TI needs to take it over, but I do not know anything about that.
I was hoping that because the change to tilcdc is quite minimal it could go in via you. I am sure I could get a reviewed-by and tested-by from from Darren how has bit more experience with tilcdc and maybe from Tomi too if that helps. (Adding Tomi to cc).
The "drm/tilcdc: Add I2S HDMI audio config for tda998x"-patch itself just adds the audio configuration to pda998x pdata and fills the swap, and mirr parameters with default values (they are usually coming in hard coded at the beginning of tda998x_create()).
Best regards, Jyri
On 17/09/14 10:51, Jyri Sarha wrote:
On 09/17/2014 04:06 AM, Dave Airlie wrote:
On 17 September 2014 06:40, Jyri Sarha jsarha@ti.com wrote:
Changes since v2:
- Change compatible property from "ti,gpio-clock" to
"ti,gpio-gate-clock"
- Some minor cleanups
The code has a functional dependency to: http://www.mail-archive.com/linux-omap@vger.kernel.org/msg109264.html
Without the above patch the audio card does not probe.
The code has been rebased on top of Linux 3.17-rc5. The patches bellow, the above dependency, and couple of commits to add BBB HDMI audio support to omap2plus_defconfig can be pulled from:
https://github.com/jsarha/linux.git linux-master-bbb-hdmi-audio
How do you intend to get this merge, sending patchsets like this without indication to maintainers on a merge strategy is kinda messy.
I'm not sure how maintained tilcdc is.
Well, it is used but AFAIK the people who have been working with it the most have left TI. I think eventually someone at TI needs to take it over, but I do not know anything about that.
I was hoping that because the change to tilcdc is quite minimal it could go in via you. I am sure I could get a reviewed-by and tested-by from from Darren how has bit more experience with tilcdc and maybe from Tomi too if that helps. (Adding Tomi to cc).
The "drm/tilcdc: Add I2S HDMI audio config for tda998x"-patch itself just adds the audio configuration to pda998x pdata and fills the swap, and mirr parameters with default values (they are usually coming in hard coded at the beginning of tda998x_create()).
I think Dave's point was that the series touches three different subsystems, and you didn't give any thoughts about how this could be merged.
Must this be merged in one piece, because xxx depends on yyy etc? Can these all be merged separately, via respective trees? If in one piece, you could ask acks from Mr x and Mr y, for the parts they maintain. Etc. And of course, it'd be good to include all the maintainers (At least Mike was missing).
I think the clock patch could be handled totally separate, as it's in no way related to HDMI. The video and audio part may be handled together, if they have dependencies.
Tomi
On 09/19/2014 04:24 PM, Tomi Valkeinen wrote:
On 17/09/14 10:51, Jyri Sarha wrote:
On 09/17/2014 04:06 AM, Dave Airlie wrote:
On 17 September 2014 06:40, Jyri Sarha jsarha@ti.com wrote:
Changes since v2:
- Change compatible property from "ti,gpio-clock" to
"ti,gpio-gate-clock"
- Some minor cleanups
The code has a functional dependency to: http://www.mail-archive.com/linux-omap@vger.kernel.org/msg109264.html
Without the above patch the audio card does not probe.
The code has been rebased on top of Linux 3.17-rc5. The patches bellow, the above dependency, and couple of commits to add BBB HDMI audio support to omap2plus_defconfig can be pulled from:
https://github.com/jsarha/linux.git linux-master-bbb-hdmi-audio
How do you intend to get this merge, sending patchsets like this without indication to maintainers on a merge strategy is kinda messy.
I'm not sure how maintained tilcdc is.
Well, it is used but AFAIK the people who have been working with it the most have left TI. I think eventually someone at TI needs to take it over, but I do not know anything about that.
I was hoping that because the change to tilcdc is quite minimal it could go in via you. I am sure I could get a reviewed-by and tested-by from from Darren how has bit more experience with tilcdc and maybe from Tomi too if that helps. (Adding Tomi to cc).
The "drm/tilcdc: Add I2S HDMI audio config for tda998x"-patch itself just adds the audio configuration to pda998x pdata and fills the swap, and mirr parameters with default values (they are usually coming in hard coded at the beginning of tda998x_create()).
I think Dave's point was that the series touches three different subsystems, and you didn't give any thoughts about how this could be merged.
Must this be merged in one piece, because xxx depends on yyy etc? Can these all be merged separately, via respective trees? If in one piece, you could ask acks from Mr x and Mr y, for the parts they maintain. Etc. And of course, it'd be good to include all the maintainers (At least Mike was missing).
I think the clock patch could be handled totally separate, as it's in no way related to HDMI. The video and audio part may be handled together, if they have dependencies.
There should be no build time breakage no matter what is the merging order. The BBB HDMI audio functionality simply does not work before all the pieces are in.
1. The audio device does not probe if the ti,gpio-gate-clock is misssing. 2. The audio device probes, but there is no sound coming out if the tilcdc_slave change is missing. 3. Without ASoC or DTS changes there is no audio device.
Best regards, Jyri
participants (5)
-
Dave Airlie
-
Jean-Francois Moine
-
Jyri Sarha
-
Mark Brown
-
Tomi Valkeinen