August 2010 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH] ASoC/multi-component: fix au1x platform
by Manuel Lauss 19 Aug '10

19 Aug '10

Hi Liam, Attached is a patch which fixes the Alchemy (au1x) audio platform in your "topic/multi-component" branch. Tested on my Db1200. One minor issue remains: with AC97, the following warning is generated, courtesy of the generic ac97 codec: asoc: ac97-hifi <-> au1xpsc-ac97.1 mapping ok kobject (8fdc59b0): tried to init an initialized object, something is seriously wrong. Call Trace: [<804959c4>] dump_stack+0x8/0x34 [<802a30e4>] kobject_init+0x50/0xcc [<802ec27c>] device_initialize+0x2c/0x70 [<802ecb64>] device_register+0x14/0x28 [<8039e058>] snd_soc_instantiate_cards+0xa00/0xb10 [<8039e27c>] soc_probe+0x114/0x154 [<802ef0fc>] driver_probe_device+0xe4/0x1a0 [<802ee404>] bus_for_each_drv+0x60/0xb0 [<802ef370>] device_attach+0x74/0xa8 [<802ee1e8>] bus_probe_device+0x30/0x54 [<802ec9a0>] device_add+0x384/0x534 [<802f09a4>] platform_device_add+0x15c/0x1c8 [<805ccb14>] db1200_audio_load+0x70/0x9c [<801004fc>] do_one_initcall+0xfc/0x1e0 [<805ba32c>] kernel_init+0xc8/0x168 [<80105904>] kernel_thread_helper+0x10/0x18 ALSA device list: #0: DB1200_AC97 Best regards, Manuel Lauss

2 11

[alsa-devel] [PATCH v7 0/1] A codec driver for the TI WL1273 FM Radio.
by Matti J. Aaltonen 19 Aug '10

19 Aug '10

Hello. And thanks for the comments to patch sets v5 and v6. New in this version is that I've removed the global codec pointer and also removed the codec pointer from the private data. The probe and remove functions have also been revised accordingly. Thanks, Matti Matti J. Aaltonen (1): ASoC: TI WL1273 FM Radio Codec. sound/soc/codecs/wl1273.c | 533 +++++++++++++++++++++++++++++++++++++++++++++ sound/soc/codecs/wl1273.h | 101 +++++++++ 2 files changed, 634 insertions(+), 0 deletions(-) create mode 100644 sound/soc/codecs/wl1273.c create mode 100644 sound/soc/codecs/wl1273.h

3 3

[alsa-devel] [PATCH] ASoC: Multi-component: Use a more adequate name for the CX20442 codec DAI
by Janusz Krzysztofik 19 Aug '10

19 Aug '10

In the process of unification of codec DAI names while implementing multi-component, the CX20442 codec DAI has been renamed to "cx20442-hifi". This new name seems not adequate for a 8kHz voice codec. Use a better name, "cx20442-voice", as suggested by Liam Girdwood. Created against git://git.kernel.org/pub/scm/linux/kernel/git/lrg/asoc-2.6.git commit 5dcba5d6741d4533e0ef696507f93f2a4c738efb dated 12 Aug 2010. Signed-off-by: Janusz Krzysztofik <jkrzyszt(a)tis.icnet.pl> --- --- git/sound/soc/codecs/cx20442.c.orig 2010-08-19 14:48:27.000000000 +0200 +++ git/sound/soc/codecs/cx20442.c 2010-08-19 15:01:55.000000000 +0200 @@ -318,7 +318,7 @@ EXPORT_SYMBOL_GPL(v253_ops); */ static struct snd_soc_dai_driver cx20442_dai = { - .name = "cx20442-hifi", + .name = "cx20442-voice", .playback = { .stream_name = "Playback", .channels_min = 1, --- git/sound/soc/omap/ams-delta.c.orig 2010-08-19 14:48:28.000000000 +0200 +++ git/sound/soc/omap/ams-delta.c 2010-08-19 15:02:05.000000000 +0200 @@ -584,7 +584,7 @@ static struct snd_soc_dai_link ams_delta .name = "CX20442", .stream_name = "CX20442", .cpu_dai_name ="omap-mcbsp-dai.0", - .codec_dai_name = "cx20442-hifi", + .codec_dai_name = "cx20442-voice", .init = ams_delta_cx20442_init, .platform_name = "omap-pcm-audio", .codec_name = "cx20442-codec",

3 2

[alsa-devel] [PATCH] tlv320aic3x symmetric rates
by Randolph Chung 19 Aug '10

19 Aug '10

The tlv320aic3x codec driver only supports symmetric rates for capture/playback. Set the flag in the DAI accordingly. Signed-off-by: Randolph Chung <tausq(a)parisc-linux.org> diff --git a/sound/soc/codecs/tlv320aic3x.c b/sound/soc/codecs/tlv320aic3x.c index 71a6990..63cad43 100644 --- a/sound/soc/codecs/tlv320aic3x.c +++ b/sound/soc/codecs/tlv320aic3x.c @@ -1116,6 +1116,7 @@ struct snd_soc_dai aic3x_dai = { .rates = AIC3X_RATES, .formats = AIC3X_FORMATS,}, .ops = &aic3x_dai_ops, + .symmetric_rates = 1, }; EXPORT_SYMBOL_GPL(aic3x_dai);

3 5

[alsa-devel] [PATCH v6 0/1] A codec driver for the TI WL1273 FM Radio.
by Matti J. Aaltonen 19 Aug '10

19 Aug '10

Hello. I didn't get any comments to the latest patch set. But now I've had a change to test the codec in multi component environment. I had to change the get_format function because the number of channels couldn't be used for determining the format. Also I added a few audio related defines to the header file. Thank You. Matti J. Aaltonen (1): ASoC: TI WL1273 FM Radio Codec. sound/soc/codecs/wl1273.c | 553 +++++++++++++++++++++++++++++++++++++++++++++ sound/soc/codecs/wl1273.h | 101 ++++++++ 2 files changed, 654 insertions(+), 0 deletions(-) create mode 100644 sound/soc/codecs/wl1273.c create mode 100644 sound/soc/codecs/wl1273.h

3 11

[alsa-devel] [PATCH] Add support for tlv320aic3007 to tlv320aic3x driver
by Randolph Chung 19 Aug '10

19 Aug '10

The attached patch adds support for the tlv320aic3007 codec to the tlv320aic3x driver. The tlv320aic3007 is similar to the aic31, but has an additional class-D speaker amp. The speaker amp control register overlaps with the mono output register of other codecs in this family, so we allow the board driver to pass in a model id to identify the codec being used. randolph

2 3

[alsa-devel] [PATCH] ALSA: intel8x0: Mute External Amplifier by default for ThinkPad X31
by Daniel Chen 19 Aug '10

19 Aug '10

Hi, Please apply the following patch (applicable to all stable linux releases, too) against current sound-2.6 to make audio audible by default for this model ThinkPad. Best, -Dan

2 1

Re: [alsa-devel] Information request - writing a driver for a virtual soundcard
by Daniel Mack 18 Aug '10

18 Aug '10

(please never remove any people from the list of recipients when replying, and please do not top-post) On Wed, Aug 18, 2010 at 09:27:36PM +0200, Olof wrote: > I want to connect my smartphone to my amplifier and then use it as a > virtual soundcard. The actual sound data shall be transmitted over > wireless TCP/IP from my laptop to the smartphone, enabling me to move > around in the flat without wiring. I still don't understand where the actual audio material has its origin, or where it should be sent to, respectively. > Thought a virtual soundcard was a > good idea since it the whole system then would be independent of > application playing sound on the laptop. I intended to do the > compressing & transmission in user space, but perhaps everything can > be done in user space? I wasn't aware of the possibility. Where can I > read more? I think the easiest API to access your existing sound cards and to create virtual sinks and sources is offered by PulseAudio: http://pulseaudio.org/wiki/DeveloperDocumentation In case your distribution uses PulseAudio natively, you souldn't even need to set up anything. HTH, Daniel

2 2

[alsa-devel] [PATCH] [v2] asoc/multi-component: fsl: add support for variable SSI FIFO depth
by Timur Tabi 18 Aug '10

18 Aug '10

Add code that programs the DMA and SSI controllers differently based on the FIFO depth of the SSI. The SSI devices on the MPC8610 and the P1022 are identical in every way except one: the transmit and receive FIFO depth. On the MPC8610, the depth is eight. On the P1022, it's fifteen. The device tree nodes for the SSI include a "fsl,fifo-depth" property that specifies the FIFO depth. Signed-off-by: Timur Tabi <timur(a)freescale.com> --- sound/soc/fsl/fsl_dma.c | 67 +++++++++++++++++++++++++++++++++++----------- sound/soc/fsl/fsl_ssi.c | 25 +++++++++++++++-- 2 files changed, 73 insertions(+), 19 deletions(-) diff --git a/sound/soc/fsl/fsl_dma.c b/sound/soc/fsl/fsl_dma.c index 57774cb..dfe1cb9 100644 --- a/sound/soc/fsl/fsl_dma.c +++ b/sound/soc/fsl/fsl_dma.c @@ -60,6 +60,7 @@ struct dma_object { struct snd_soc_platform_driver dai; dma_addr_t ssi_stx_phys; dma_addr_t ssi_srx_phys; + unsigned int ssi_fifo_depth; struct ccsr_dma_channel __iomem *channel; unsigned int irq; bool assigned; @@ -99,6 +100,7 @@ struct fsl_dma_private { unsigned int irq; struct snd_pcm_substream *substream; dma_addr_t ssi_sxx_phys; + unsigned int ssi_fifo_depth; dma_addr_t ld_buf_phys; unsigned int current_link; dma_addr_t dma_buf_phys; @@ -431,6 +433,7 @@ static int fsl_dma_open(struct snd_pcm_substream *substream) else dma_private->ssi_sxx_phys = dma->ssi_srx_phys; + dma_private->ssi_fifo_depth = dma->ssi_fifo_depth; dma_private->dma_channel = dma->channel; dma_private->irq = dma->irq; dma_private->substream = substream; @@ -544,11 +547,11 @@ static int fsl_dma_hw_params(struct snd_pcm_substream *substream, struct device *dev = rtd->platform->dev; /* Number of bits per sample */ - unsigned int sample_size = + unsigned int sample_bits = snd_pcm_format_physical_width(params_format(hw_params)); /* Number of bytes per frame */ - unsigned int frame_size = 2 * (sample_size / 8); + unsigned int sample_bytes = sample_bits / 8; /* Bus address of SSI STX register */ dma_addr_t ssi_sxx_phys = dma_private->ssi_sxx_phys; @@ -588,7 +591,7 @@ static int fsl_dma_hw_params(struct snd_pcm_substream *substream, * that offset here. While we're at it, also tell the DMA controller * how much data to transfer per sample. */ - switch (sample_size) { + switch (sample_bits) { case 8: mr |= CCSR_DMA_MR_DAHTS_1 | CCSR_DMA_MR_SAHTS_1; ssi_sxx_phys += 3; @@ -602,22 +605,42 @@ static int fsl_dma_hw_params(struct snd_pcm_substream *substream, break; default: /* We should never get here */ - dev_err(dev, "unsupported sample size %u\n", sample_size); + dev_err(dev, "unsupported sample size %u\n", sample_bits); return -EINVAL; } /* - * BWC should always be a multiple of the frame size. BWC determines - * how many bytes are sent/received before the DMA controller checks the - * SSI to see if it needs to stop. For playback, the transmit FIFO can - * hold three frames, so we want to send two frames at a time. For - * capture, the receive FIFO is triggered when it contains one frame, so - * we want to receive one frame at a time. + * BWC determines how many bytes are sent/received before the DMA + * controller checks the SSI to see if it needs to stop. BWC should + * always be a multiple of the frame size, so that we always transmit + * whole frames. Each frame occupies two slots in the FIFO. The + * parameter for CCSR_DMA_MR_BWC() is rounded down the next power of two + * (MR[BWC] can only represent even powers of two). + * + * To simplify the process, we set BWC to the largest value that is + * less than or equal to the FIFO watermark. For playback, this ensures + * that we transfer the maximum amount without overrunning the FIFO. + * For capture, this ensures that we transfer the maximum amount without + * underrunning the FIFO. + * + * f = SSI FIFO depth + * w = SSI watermark value (which equals f - 2) + * b = DMA bandwidth count (in bytes) + * s = sample size (in bytes, which equals frame_size * 2) + * + * For playback, we never transmit more than the transmit FIFO + * watermark, otherwise we might write more data than the FIFO can hold. + * The watermark is equal to the FIFO depth minus two. + * + * For capture, two equations must hold: + * w > f - (b / s) + * w >= b / s + * + * So, b > 2 * s, but b must also be <= s * w. To simplify, we set + * b = s * w, which is equal to + * (dma_private->ssi_fifo_depth - 2) * sample_bytes. */ - if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) - mr |= CCSR_DMA_MR_BWC(2 * frame_size); - else - mr |= CCSR_DMA_MR_BWC(frame_size); + mr |= CCSR_DMA_MR_BWC((dma_private->ssi_fifo_depth - 2) * sample_bytes); out_be32(&dma_channel->mr, mr); @@ -871,6 +894,7 @@ static int __devinit fsl_soc_dma_probe(struct of_device *of_dev, struct device_node *np = of_dev->dev.of_node; struct device_node *ssi_np; struct resource res; + const uint32_t *iprop; int ret; /* Find the SSI node that points to us. */ @@ -881,15 +905,17 @@ static int __devinit fsl_soc_dma_probe(struct of_device *of_dev, } ret = of_address_to_resource(ssi_np, 0, &res); - of_node_put(ssi_np); if (ret) { - dev_err(&of_dev->dev, "could not determine device resources\n"); + dev_err(&of_dev->dev, "could not determine resources for %s\n", + ssi_np->full_name); + of_node_put(ssi_np); return ret; } dma = kzalloc(sizeof(*dma) + strlen(np->full_name), GFP_KERNEL); if (!dma) { dev_err(&of_dev->dev, "could not allocate dma object\n"); + of_node_put(ssi_np); return -ENOMEM; } @@ -902,6 +928,15 @@ static int __devinit fsl_soc_dma_probe(struct of_device *of_dev, dma->ssi_stx_phys = res.start + offsetof(struct ccsr_ssi, stx0); dma->ssi_srx_phys = res.start + offsetof(struct ccsr_ssi, srx0); + iprop = of_get_property(ssi_np, "fsl,fifo-depth", NULL); + if (iprop) + dma->ssi_fifo_depth = *iprop; + else + /* Older 8610 DTs didn't have the fifo-depth property */ + dma->ssi_fifo_depth = 8; + + of_node_put(ssi_np); + ret = snd_soc_register_platform(&of_dev->dev, &dma->dai); if (ret) { dev_err(&of_dev->dev, "could not register platform\n"); diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 00e3e62..c377dd2 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -93,6 +93,7 @@ struct fsl_ssi_private { unsigned int playback; unsigned int capture; int asynchronous; + unsigned int fifo_depth; struct snd_soc_dai_driver cpu_dai_drv; struct device_attribute dev_attr; struct platform_device *pdev; @@ -337,11 +338,20 @@ static int fsl_ssi_startup(struct snd_pcm_substream *substream, /* * Set the watermark for transmit FIFI 0 and receive FIFO 0. We - * don't use FIFO 1. Since the SSI only supports stereo, the - * watermark should never be an odd number. + * don't use FIFO 1. We program the transmit water to signal a + * DMA transfer if there are only two (or fewer) elements left + * in the FIFO. Two elements equals one frame (left channel, + * right channel). This value, however, depends on the depth of + * the transmit buffer. + * + * We program the receive FIFO to notify us if at least two + * elements (one frame) have been written to the FIFO. We could + * make this value larger (and maybe we should), but this way + * data will be written to memory as soon as it's available. */ out_be32(&ssi->sfcsr, - CCSR_SSI_SFCSR_TFWM0(6) | CCSR_SSI_SFCSR_RFWM0(2)); + CCSR_SSI_SFCSR_TFWM0(ssi_private->fifo_depth - 2) | + CCSR_SSI_SFCSR_RFWM0(ssi_private->fifo_depth - 2)); /* * We keep the SSI disabled because if we enable it, then the @@ -622,6 +632,7 @@ static int __devinit fsl_ssi_probe(struct of_device *of_dev, struct device_attribute *dev_attr = NULL; struct device_node *np = of_dev->dev.of_node; const char *p, *sprop; + const uint32_t *iprop; struct resource res; char name[64]; @@ -671,6 +682,14 @@ static int __devinit fsl_ssi_probe(struct of_device *of_dev, else ssi_private->cpu_dai_drv.symmetric_rates = 1; + /* Determine the FIFO depth. */ + iprop = of_get_property(np, "fsl,fifo-depth", NULL); + if (iprop) + ssi_private->fifo_depth = *iprop; + else + /* Older 8610 DTs didn't have the fifo-depth property */ + ssi_private->fifo_depth = 8; + /* Initialize the the device_attribute structure */ dev_attr = &ssi_private->dev_attr; dev_attr->attr.name = "statistics"; -- 1.7.0.1

3 3

[alsa-devel] Information request - writing a driver for a virtual soundcard
by Olof 18 Aug '10

18 Aug '10

Hi! Maybe this is the wrong place to ask, but I kinda feel that I'm out of options. My goal is to write a dummy driver which makes the PCM data available as a character device though /dev/whatever. Of course I start with the dummy pcm driver. However I find it difficult to understand how the driver shall poll PCM data from the ALSA framework. I've read the how-to-write-a-driver doc and studied the hda driver but I don't understand how it all fits together with substreams, reading from intermediate buffers, the PCM memory management functions etc. Could someone please help me out with some pointers to additional documentation, a suitable piece of driver code to study or any thing that might help? Kind Regards, Olof

2 1