[alsa-devel] [PATCH] ASoC: blackfin: add bf6xx audio dma driver
Bf5xx and bf6xx reused bf5xx-i2s-pcm.c before. Recently tdm driver has been merged into bf5xx-i2s-pcm.c and it's not compatible with bf6xx audio. So create this dma driver and add tdm support later.
Signed-off-by: Scott Jiang scott.jiang.linux@gmail.com --- sound/soc/blackfin/Kconfig | 22 ++-- sound/soc/blackfin/Makefile | 2 + sound/soc/blackfin/bf6xx-pcm.c | 230 ++++++++++++++++++++++++++++++++++++++++ 3 files changed, 248 insertions(+), 6 deletions(-) create mode 100644 sound/soc/blackfin/bf6xx-pcm.c
diff --git a/sound/soc/blackfin/Kconfig b/sound/soc/blackfin/Kconfig index 6410aa2..9d6bd9d 100644 --- a/sound/soc/blackfin/Kconfig +++ b/sound/soc/blackfin/Kconfig @@ -1,8 +1,17 @@ config SND_BF5XX_I2S tristate "SoC I2S Audio for the ADI Blackfin chip" - depends on BLACKFIN - select SND_BF5XX_SOC_SPORT if !BF60x - select SND_BF6XX_SOC_SPORT if BF60x + depends on BLACKFIN && !BF60x + select SND_BF5XX_SOC_SPORT + help + Say Y or M if you want to add support for codecs attached to + the Blackfin SPORT (synchronous serial ports) interface in I2S + mode (supports single stereo In/Out). + You will also need to select the audio interfaces to support below. + +config SND_BF6XX_PCM + tristate "SoC Audio for the ADI Blackfin BF6xx chip" + depends on BF60x + select SND_BF6XX_SOC_SPORT help Say Y or M if you want to add support for codecs attached to the Blackfin SPORT (synchronous serial ports) interface in I2S @@ -11,7 +20,7 @@ config SND_BF5XX_I2S
config SND_BF5XX_SOC_SSM2602 tristate "SoC SSM2602 Audio Codec Add-On Card support" - depends on SND_BF5XX_I2S && SND_SOC_I2C_AND_SPI + depends on (SND_BF5XX_I2S || SND_BF6XX_PCM) && SND_SOC_I2C_AND_SPI select SND_BF5XX_SOC_I2S if !BF60x select SND_BF6XX_SOC_I2S if BF60x select SND_SOC_SSM2602_SPI if SPI_MASTER @@ -45,8 +54,9 @@ config SND_SOC_BFIN_EVAL_ADAU1373
config SND_SOC_BFIN_EVAL_ADAU1X61 tristate "Support for the EVAL-ADAU1X61 board on Blackfin eval boards" - depends on SND_BF5XX_I2S && I2C - select SND_BF5XX_SOC_I2S + depends on (SND_BF5XX_I2S || SND_BF6XX_PCM) && I2C + select SND_BF5XX_SOC_I2S if !BF60x + select SND_BF6XX_SOC_I2S if BF60x select SND_SOC_ADAU1761_I2C help Say Y if you want to add support for the Analog Devices EVAL-ADAU1X61 diff --git a/sound/soc/blackfin/Makefile b/sound/soc/blackfin/Makefile index f21e948..cef39f6 100644 --- a/sound/soc/blackfin/Makefile +++ b/sound/soc/blackfin/Makefile @@ -1,6 +1,7 @@ # Blackfin Platform Support snd-bf5xx-ac97-objs := bf5xx-ac97-pcm.o snd-bf5xx-i2s-objs := bf5xx-i2s-pcm.o +snd-bf6xx-pcm-objs := bf6xx-pcm.o snd-soc-bf5xx-sport-objs := bf5xx-sport.o snd-soc-bf6xx-sport-objs := bf6xx-sport.o snd-soc-bf5xx-ac97-objs := bf5xx-ac97.o @@ -9,6 +10,7 @@ snd-soc-bf6xx-i2s-objs := bf6xx-i2s.o
obj-$(CONFIG_SND_BF5XX_AC97) += snd-bf5xx-ac97.o obj-$(CONFIG_SND_BF5XX_I2S) += snd-bf5xx-i2s.o +obj-$(CONFIG_SND_BF6XX_PCM) += snd-bf6xx-pcm.o obj-$(CONFIG_SND_BF5XX_SOC_SPORT) += snd-soc-bf5xx-sport.o obj-$(CONFIG_SND_BF6XX_SOC_SPORT) += snd-soc-bf6xx-sport.o obj-$(CONFIG_SND_BF5XX_SOC_AC97) += snd-soc-bf5xx-ac97.o diff --git a/sound/soc/blackfin/bf6xx-pcm.c b/sound/soc/blackfin/bf6xx-pcm.c new file mode 100644 index 0000000..bad9fbf --- /dev/null +++ b/sound/soc/blackfin/bf6xx-pcm.c @@ -0,0 +1,230 @@ +/* + * bf6xx-pcm.c - Analog Devices BF6XX audio dma driver + * + * Copyright (c) 2014 Analog Devices Inc. + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/device.h> +#include <linux/dma-mapping.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/platform_device.h> +#include <sound/pcm.h> +#include <sound/pcm_params.h> +#include <sound/soc.h> +#include <sound/soc-dai.h> + +#include "bf6xx-sport.h" + +static void bf6xx_dma_irq(void *data) +{ + struct snd_pcm_substream *pcm = data; + snd_pcm_period_elapsed(pcm); +} + +static const struct snd_pcm_hardware bf6xx_pcm_hardware = { + .info = SNDRV_PCM_INFO_INTERLEAVED | + SNDRV_PCM_INFO_MMAP | + SNDRV_PCM_INFO_MMAP_VALID | + SNDRV_PCM_INFO_BLOCK_TRANSFER, + .formats = SNDRV_PCM_FMTBIT_S16_LE | + SNDRV_PCM_FMTBIT_S24_LE | + SNDRV_PCM_FMTBIT_S32_LE, + .period_bytes_min = 32, + .period_bytes_max = 0x10000, + .periods_min = 1, + .periods_max = PAGE_SIZE/32, + .buffer_bytes_max = 0x20000, /* 128 kbytes */ + .fifo_size = 16, +}; + +static int bf6xx_pcm_hw_params(struct snd_pcm_substream *substream, + struct snd_pcm_hw_params *params) +{ + size_t size = params_buffer_bytes(params); + snd_pcm_lib_malloc_pages(substream, size); + + return 0; +} + +static int bf6xx_pcm_hw_free(struct snd_pcm_substream *substream) +{ + snd_pcm_lib_free_pages(substream); + + return 0; +} + +static int bf6xx_pcm_prepare(struct snd_pcm_substream *substream) +{ + struct snd_pcm_runtime *runtime = substream->runtime; + struct sport_device *sport = runtime->private_data; + int period_bytes = frames_to_bytes(runtime, runtime->period_size); + + if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) { + sport_set_tx_callback(sport, bf6xx_dma_irq, substream); + sport_config_tx_dma(sport, runtime->dma_area, + runtime->periods, period_bytes); + } else { + sport_set_rx_callback(sport, bf6xx_dma_irq, substream); + sport_config_rx_dma(sport, runtime->dma_area, + runtime->periods, period_bytes); + } + + return 0; +} + +static int bf6xx_pcm_trigger(struct snd_pcm_substream *substream, int cmd) +{ + struct snd_pcm_runtime *runtime = substream->runtime; + struct sport_device *sport = runtime->private_data; + int ret = 0; + + switch (cmd) { + case SNDRV_PCM_TRIGGER_START: + if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) + sport_tx_start(sport); + else + sport_rx_start(sport); + break; + case SNDRV_PCM_TRIGGER_STOP: + case SNDRV_PCM_TRIGGER_SUSPEND: + case SNDRV_PCM_TRIGGER_PAUSE_PUSH: + if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) + sport_tx_stop(sport); + else + sport_rx_stop(sport); + break; + default: + ret = -EINVAL; + } + + return ret; +} + +static snd_pcm_uframes_t bf6xx_pcm_pointer(struct snd_pcm_substream *substream) +{ + struct snd_pcm_runtime *runtime = substream->runtime; + struct sport_device *sport = runtime->private_data; + unsigned int diff; + snd_pcm_uframes_t frames; + + if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) + diff = sport_curr_offset_tx(sport); + else + diff = sport_curr_offset_rx(sport); + + /* + * TX at least can report one frame beyond the end of the + * buffer if we hit the wraparound case - clamp to within the + * buffer as the ALSA APIs require. + */ + if (diff == snd_pcm_lib_buffer_bytes(substream)) + diff = 0; + + frames = bytes_to_frames(substream->runtime, diff); + + return frames; +} + +static int bf6xx_pcm_open(struct snd_pcm_substream *substream) +{ + struct snd_soc_pcm_runtime *rtd = substream->private_data; + struct snd_soc_dai *cpu_dai = rtd->cpu_dai; + struct sport_device *sport = snd_soc_dai_get_drvdata(cpu_dai); + struct snd_pcm_runtime *runtime = substream->runtime; + struct snd_dma_buffer *buf = &substream->dma_buffer; + int ret; + + snd_soc_set_runtime_hwparams(substream, &bf6xx_pcm_hardware); + + ret = snd_pcm_hw_constraint_integer(runtime, + SNDRV_PCM_HW_PARAM_PERIODS); + if (ret < 0) + return ret; + + if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) + sport->tx_buf = buf->area; + else + sport->rx_buf = buf->area; + + runtime->private_data = sport; + return 0; +} + +static int bf6xx_pcm_mmap(struct snd_pcm_substream *substream, + struct vm_area_struct *vma) +{ + struct snd_pcm_runtime *runtime = substream->runtime; + size_t size = vma->vm_end - vma->vm_start; + vma->vm_start = (unsigned long)runtime->dma_area; + vma->vm_end = vma->vm_start + size; + vma->vm_flags |= VM_SHARED; + + return 0 ; +} + +static struct snd_pcm_ops bf6xx_pcm_ops = { + .open = bf6xx_pcm_open, + .ioctl = snd_pcm_lib_ioctl, + .hw_params = bf6xx_pcm_hw_params, + .hw_free = bf6xx_pcm_hw_free, + .prepare = bf6xx_pcm_prepare, + .trigger = bf6xx_pcm_trigger, + .pointer = bf6xx_pcm_pointer, + .mmap = bf6xx_pcm_mmap, +}; + +static int bf6xx_pcm_new(struct snd_soc_pcm_runtime *rtd) +{ + struct snd_card *card = rtd->card->snd_card; + size_t size = bf6xx_pcm_hardware.buffer_bytes_max; + int ret = 0; + + ret = dma_coerce_mask_and_coherent(card->dev, DMA_BIT_MASK(32)); + if (ret) + return ret; + + return snd_pcm_lib_preallocate_pages_for_all(rtd->pcm, + SNDRV_DMA_TYPE_DEV, card->dev, size, size); +} + +static struct snd_soc_platform_driver bf6xx_soc_platform = { + .ops = &bf6xx_pcm_ops, + .pcm_new = bf6xx_pcm_new, +}; + +static int bfin_soc_platform_probe(struct platform_device *pdev) +{ + return snd_soc_register_platform(&pdev->dev, &bf6xx_soc_platform); +} + +static int bfin_soc_platform_remove(struct platform_device *pdev) +{ + snd_soc_unregister_platform(&pdev->dev); + return 0; +} + +static struct platform_driver bfin_pcm_driver = { + .driver = { + .name = "bfin-i2s-pcm-audio", + .owner = THIS_MODULE, + }, + + .probe = bfin_soc_platform_probe, + .remove = bfin_soc_platform_remove, +}; + +module_platform_driver(bfin_pcm_driver); + +MODULE_DESCRIPTION("Analog Devices BF6XX audio dma driver"); +MODULE_AUTHOR("Scott Jiang Scott.Jiang.Linux@gmail.com"); +MODULE_LICENSE("GPL v2");
On Wed, Sep 24, 2014 at 05:55:20PM +0800, Scott Jiang wrote:
Bf5xx and bf6xx reused bf5xx-i2s-pcm.c before. Recently tdm driver has been merged into bf5xx-i2s-pcm.c and it's not compatible with bf6xx audio. So create this dma driver and add tdm support later.
Can you go into a bit more detail on the differences - clearly there's some similarlities since the drivers got merged, can we really not share any code?
2014-09-24 16:26 GMT+08:00 Mark Brown broonie@kernel.org:
On Wed, Sep 24, 2014 at 05:55:20PM +0800, Scott Jiang wrote:
Bf5xx and bf6xx reused bf5xx-i2s-pcm.c before. Recently tdm driver has been merged into bf5xx-i2s-pcm.c and it's not compatible with bf6xx audio. So create this dma driver and add tdm support later.
Can you go into a bit more detail on the differences - clearly there's some similarlities since the drivers got merged, can we really not share any code?
First, sport2 only supports i2s and left justified mode, while sport3 adds i2s packed and right justified mode. Second, in tdm mode for sport2, tx and rx are dependent and have a window granularity of 8 channel. While sport3 doesn't have these limits. What's more, sport3 is not only used on blackfin. That means many patches can only apply to this driver.
Thanks, Scott
On Wed, Sep 24, 2014 at 06:02:40PM +0800, Scott Jiang wrote:
2014-09-24 16:26 GMT+08:00 Mark Brown broonie@kernel.org:
Can you go into a bit more detail on the differences - clearly there's some similarlities since the drivers got merged, can we really not share any code?
First, sport2 only supports i2s and left justified mode, while sport3 adds i2s packed and right justified mode. Second, in tdm mode for sport2, tx and rx are dependent and have a window granularity of 8 channel. While sport3 doesn't have these limits. What's more, sport3 is not only used on blackfin. That means many patches can only apply to this driver.
These sound more like extended functionality and tweaks rather than massive updates which need a completely separate driver - new modes and so on.
2014-09-24 19:42 GMT+08:00 Mark Brown broonie@kernel.org:
On Wed, Sep 24, 2014 at 06:02:40PM +0800, Scott Jiang wrote:
2014-09-24 16:26 GMT+08:00 Mark Brown broonie@kernel.org:
Can you go into a bit more detail on the differences - clearly there's some similarlities since the drivers got merged, can we really not share any code?
First, sport2 only supports i2s and left justified mode, while sport3 adds i2s packed and right justified mode. Second, in tdm mode for sport2, tx and rx are dependent and have a window granularity of 8 channel. While sport3 doesn't have these limits. What's more, sport3 is not only used on blackfin. That means many patches can only apply to this driver.
These sound more like extended functionality and tweaks rather than massive updates which need a completely separate driver - new modes and so on.
Yes, it looks similar in i2s mode. But in tdm mode, because we don't need to copy 2 channels user buffer to 8 channels dma buffer, it might looks very different. If you think this patch is not good, I'd send a new one after I merged tdm driver. By the way, if the channel order of hardware is different from what is expected by alsa, the copy work is left to plugin or driver?
Scott
On Thu, Sep 25, 2014 at 03:24:45PM +0800, Scott Jiang wrote:
2014-09-24 19:42 GMT+08:00 Mark Brown broonie@kernel.org:
These sound more like extended functionality and tweaks rather than massive updates which need a completely separate driver - new modes and so on.
Yes, it looks similar in i2s mode. But in tdm mode, because we don't need to copy 2 channels user buffer to 8 channels dma buffer, it might looks very different. If you think this patch is not good, I'd send a new one after I merged tdm driver.
Well, I'd rather not see the driver split into two separate drivers again if at all possible - ideally just the bits that need splitting. It sounds like you need separate copy operations for the two cases but otherwise things are fairly similar? If the driver ended up being nothing but conditional code that'd be bad but if it's just a few places then it's going to be better to avoid duplicating the entire thing.
By the way, if the channel order of hardware is different from what is expected by alsa, the copy work is left to plugin or driver?
If the format is just not something ALSA understands then the driver ought to do it. If ALSA does understand it but it's just not what most applications want then the plugin should be a better choice.
participants (2)
-
Mark Brown -
Scott Jiang