March 2013 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH 1/2] ASoC: compress - add support for metadata apis
by Vinod Koul 26 Mar '13

26 Mar '13

From: Jeeja KP <jeeja.kp(a)intel.com> Compress core added metadata apis in 9727b4, so add same in ASoC Signed-off-by: Jeeja KP <jeeja.kp(a)intel.com> Signed-off-by: Vinod Koul <vinod.koul(a)intel.com> --- Rebased on asoc-next sound/soc/soc-compress.c | 27 +++++++++++++++++++++++++++ 1 files changed, 27 insertions(+), 0 deletions(-) diff --git a/sound/soc/soc-compress.c b/sound/soc/soc-compress.c index b5b3db7..f9b2197 100644 --- a/sound/soc/soc-compress.c +++ b/sound/soc/soc-compress.c @@ -322,11 +322,38 @@ static int soc_compr_copy(struct snd_compr_stream *cstream, return ret; } +static int sst_compr_set_metadata(struct snd_compr_stream *cstream, + struct snd_compr_metadata *metadata) +{ + struct snd_soc_pcm_runtime *rtd = cstream->private_data; + struct snd_soc_platform *platform = rtd->platform; + int ret = 0; + + if (platform->driver->compr_ops && platform->driver->compr_ops->set_metadata) + ret = platform->driver->compr_ops->set_metadata(cstream, metadata); + + return ret; +} + +static int sst_compr_get_metadata(struct snd_compr_stream *cstream, + struct snd_compr_metadata *metadata) +{ + struct snd_soc_pcm_runtime *rtd = cstream->private_data; + struct snd_soc_platform *platform = rtd->platform; + int ret = 0; + + if (platform->driver->compr_ops && platform->driver->compr_ops->get_metadata) + ret = platform->driver->compr_ops->get_metadata(cstream, metadata); + + return ret; +} /* ASoC Compress operations */ static struct snd_compr_ops soc_compr_ops = { .open = soc_compr_open, .free = soc_compr_free, .set_params = soc_compr_set_params, + .set_metadata = sst_compr_set_metadata, + .get_metadata = sst_compr_get_metadata, .get_params = soc_compr_get_params, .trigger = soc_compr_trigger, .pointer = soc_compr_pointer, -- 1.7.0.4

2 2

[alsa-devel] [PATCH 00/14] ASoC: Cleanup dmaengine PCM users
by Lars-Peter Clausen 26 Mar '13

26 Mar '13

Hi, This series cleans up a few of the current users of the dmaengine pcm lib in order to make the introduction of a new more generic PCM dmaengine driver more straight forward. The series consists mainly of two types of changes. One is removing all unnecessary wrappers of snd_soc_dmaengine_close(). For some reason quite a few drivers started providing their own pcm_close callback, which does nothing but calls snd_soc_dmaengine_close, although snd_soc_dmaengine_close can be used directly as the pcm_close callback if nothing else needs to be done. The other is getting rid of snd_dmaengine_pcm_{get,set}_data(). These two functions were initially introduced to deal with a few oddball dmaengine drivers which require that data is passed to them via the private_data field of the channel and we had to keep the data allocated until the channel had been released. A few other driver started using these functions although there is no need to do so. E.g. some drivers just call snd_dmaengine_pcm_set_data(), but never attempt to read the data again using snd_dmaengine_pcm_get_data(). For the drivers, for which snd_dmaengine_pcm_{set,get}_data() was originally introduced, the responsibility of keeping the dma config allocated is shifted from the pcm driver to the i2s drivers. Which in turn makes the code much more straight forward since for most i2s drivers the dma config is either static or setup once at driver probe time. The series has only been compile tested, so please test these patches, especially the non trivial ones for ep93xx, mmp, imx and mxs. - Lars Lars-Peter Clausen (14): ASoC: ux500_pcm: Remove duplicated SNDRV_PCM_HW_PARAM_PERIODS constraint ASoC: spear_pcm: No need to wrap snd_dmaengine_pcm_close() ASoC: omap-pcm: No need to wrap snd_dmaengine_pcm_close() ASoC: tegra_pcm: No need to wrap snd_dmaengine_pcm_close() ASoC: ux500_pcm: No need to wrap snd_dmaengine_pcm_close() ASoC: atmel-pcm-dma: No need to wrap snd_dmaengine_pcm_close() ASoC: ux500_pcm: No need to use snd_dmaengine_pcm_set_data() ASoC: speaer_pcm: No need to use snd_dmaengine_pcm_set_data() ASoC: atmel-pcm-dma: Do not use snd_dmaengine_pcm_{set,get}_data() ASoC: ep93xx: Use ep93xx_dma_params instead of ep93xx_pcm_dma_params ASoC: mmp-pcm: Allocate dma filter parameters on the stack ASoC: imx-pcm: Embed the imx_dma_data struct in the dma_params struct ASoC: mxs: Embed the mxs_dma_data struct in the mxs_pcm_dma_params struct ASoC: dmaengine-pcm: Remove snd_dmaengine_pcm_{set,get}_data include/sound/dmaengine_pcm.h | 3 --- sound/soc/atmel/atmel-pcm-dma.c | 23 +++++++--------------- sound/soc/cirrus/edb93xx.c | 1 - sound/soc/cirrus/ep93xx-ac97.c | 9 +++++---- sound/soc/cirrus/ep93xx-i2s.c | 12 ++++++------ sound/soc/cirrus/ep93xx-pcm.c | 37 +++-------------------------------- sound/soc/cirrus/ep93xx-pcm.h | 20 ------------------- sound/soc/cirrus/simone.c | 2 -- sound/soc/cirrus/snappercl15.c | 1 - sound/soc/fsl/fsl_ssi.c | 17 ++++++++-------- sound/soc/fsl/imx-pcm-dma.c | 35 ++------------------------------- sound/soc/fsl/imx-pcm.h | 17 ++++++++++++++-- sound/soc/fsl/imx-ssi.c | 12 ++++++++---- sound/soc/mxs/mxs-pcm.c | 43 +++++------------------------------------ sound/soc/mxs/mxs-pcm.h | 4 +++- sound/soc/mxs/mxs-saif.c | 6 +++--- sound/soc/omap/omap-pcm.c | 8 +------- sound/soc/pxa/mmp-pcm.c | 33 +++++-------------------------- sound/soc/soc-dmaengine-pcm.c | 29 --------------------------- sound/soc/spear/spear_pcm.c | 18 ++--------------- sound/soc/tegra/tegra_pcm.c | 8 +------- sound/soc/ux500/ux500_pcm.c | 25 +----------------------- 22 files changed, 76 insertions(+), 287 deletions(-) delete mode 100644 sound/soc/cirrus/ep93xx-pcm.h -- 1.8.0

9 30

[alsa-devel] [PATCH] ASoC: omap: Call omap_mcbsp_set_threshold() from mcbsp hw_params
by Lars-Peter Clausen 26 Mar '13

26 Mar '13

The omap PCM driver provides a set_threshold callback which gets called by the PCM driver when either playback or capture is started. The only DAI driver which sets this callback is the mcbsp driver. This patch removes the callback from the PCM driver and moves the invocation of the omap_mcbsp_set_threshold() function to the mcbsp hw_params callback since this is the only place where the threshold size can change. Doing so allows us to use the default dmaengine PCM trigger callback in the omap PCM driver. Signed-off-by: Lars-Peter Clausen <lars(a)metafoo.de> --- Only compile tested, but I don't see why it should not work Signed-off-by: Lars-Peter Clausen <lars(a)metafoo.de> --- sound/soc/omap/omap-mcbsp.c | 12 +++++------- sound/soc/omap/omap-pcm.c | 33 +-------------------------------- sound/soc/omap/omap-pcm.h | 1 - 3 files changed, 6 insertions(+), 40 deletions(-) diff --git a/sound/soc/omap/omap-mcbsp.c b/sound/soc/omap/omap-mcbsp.c index 8d2defd..406fc87 100644 --- a/sound/soc/omap/omap-mcbsp.c +++ b/sound/soc/omap/omap-mcbsp.c @@ -62,24 +62,22 @@ enum { * Stream DMA parameters. DMA request line and port address are set runtime * since they are different between OMAP1 and later OMAPs */ -static void omap_mcbsp_set_threshold(struct snd_pcm_substream *substream) +static void omap_mcbsp_set_threshold(struct snd_pcm_substream *substream, + unsigned int packet_size) { struct snd_soc_pcm_runtime *rtd = substream->private_data; struct snd_soc_dai *cpu_dai = rtd->cpu_dai; struct omap_mcbsp *mcbsp = snd_soc_dai_get_drvdata(cpu_dai); - struct omap_pcm_dma_data *dma_data; int words; - dma_data = snd_soc_dai_get_dma_data(rtd->cpu_dai, substream); - /* * Configure McBSP threshold based on either: * packet_size, when the sDMA is in packet mode, or based on the * period size in THRESHOLD mode, otherwise use McBSP threshold = 1 * for mono streams. */ - if (dma_data->packet_size) - words = dma_data->packet_size; + if (packet_size) + words = packet_size; else words = 1; @@ -245,7 +243,6 @@ static int omap_mcbsp_dai_hw_params(struct snd_pcm_substream *substream, return -EINVAL; } if (mcbsp->pdata->buffer_size) { - dma_data->set_threshold = omap_mcbsp_set_threshold; if (mcbsp->dma_op_mode == MCBSP_DMA_MODE_THRESHOLD) { int period_words, max_thrsh; int divider = 0; @@ -276,6 +273,7 @@ static int omap_mcbsp_dai_hw_params(struct snd_pcm_substream *substream, /* Use packet mode for non mono streams */ pkt_size = channels; } + omap_mcbsp_set_threshold(substream, pkt_size); } dma_data->packet_size = pkt_size; diff --git a/sound/soc/omap/omap-pcm.c b/sound/soc/omap/omap-pcm.c index c722c2e..1626fb7 100644 --- a/sound/soc/omap/omap-pcm.c +++ b/sound/soc/omap/omap-pcm.c @@ -129,37 +129,6 @@ static int omap_pcm_hw_free(struct snd_pcm_substream *substream) return 0; } -static int omap_pcm_trigger(struct snd_pcm_substream *substream, int cmd) -{ - struct snd_soc_pcm_runtime *rtd = substream->private_data; - struct omap_pcm_dma_data *dma_data; - int ret = 0; - - dma_data = snd_soc_dai_get_dma_data(rtd->cpu_dai, substream); - - switch (cmd) { - case SNDRV_PCM_TRIGGER_START: - case SNDRV_PCM_TRIGGER_RESUME: - case SNDRV_PCM_TRIGGER_PAUSE_RELEASE: - /* Configure McBSP internal buffer usage */ - if (dma_data->set_threshold) - dma_data->set_threshold(substream); - break; - - case SNDRV_PCM_TRIGGER_STOP: - case SNDRV_PCM_TRIGGER_SUSPEND: - case SNDRV_PCM_TRIGGER_PAUSE_PUSH: - break; - default: - ret = -EINVAL; - } - - if (ret == 0) - ret = snd_dmaengine_pcm_trigger(substream, cmd); - - return ret; -} - static snd_pcm_uframes_t omap_pcm_pointer(struct snd_pcm_substream *substream) { snd_pcm_uframes_t offset; @@ -208,7 +177,7 @@ static struct snd_pcm_ops omap_pcm_ops = { .ioctl = snd_pcm_lib_ioctl, .hw_params = omap_pcm_hw_params, .hw_free = omap_pcm_hw_free, - .trigger = omap_pcm_trigger, + .trigger = snd_dmaengine_pcm_trigger, .pointer = omap_pcm_pointer, .mmap = omap_pcm_mmap, }; diff --git a/sound/soc/omap/omap-pcm.h b/sound/soc/omap/omap-pcm.h index cabe74c..39e6e45 100644 --- a/sound/soc/omap/omap-pcm.h +++ b/sound/soc/omap/omap-pcm.h @@ -31,7 +31,6 @@ struct omap_pcm_dma_data { char *name; /* stream identifier */ int dma_req; /* DMA request line */ unsigned long port_addr; /* transmit/receive register */ - void (*set_threshold)(struct snd_pcm_substream *substream); int data_type; /* 8, 16, 32 (bits) or 0 to let omap-pcm * to decide the sDMA data type */ int packet_size; /* packet size only in PACKET mode */ -- 1.8.0

3 2

Re: [alsa-devel] [PATCH] ASoC: core: Fix to check return value of snd_soc_update_bits_locked()
by Mark Brown 26 Mar '13

26 Mar '13

On Tue, Mar 26, 2013 at 02:41:05PM +0900, Joonyoung Shim wrote: > It can be 0 or 1 return value of snd_soc_update_bits_locked() when it is > success. So just check return value is negative. Applied, thanks.

1 0

Re: [alsa-devel] [PATCH -next] ASoC: arizona: remove duplicated include from arizona.c
by Mark Brown 26 Mar '13

26 Mar '13

On Tue, Mar 26, 2013 at 02:47:08PM +0800, Wei Yongjun wrote: > From: Wei Yongjun <yongjun_wei(a)trendmicro.com.cn> > > Remove duplicated include. Applied, thanks.

1 0

[alsa-devel] ALSA regression Macbook Air 5, 2: No sound from the loudspeaker
by Daniel Wagner 26 Mar '13

26 Mar '13

Hi, I experience a regression on my Macbook Air 2012 model with the 3.9-rc kernels (v3.9-rc2-284-gaea8b5d) The symptoms are that there is just no sound from the loudspeakers. The headphone output works fine. I was able to pin point it down to one commit 1077a024 "ALSA: hda - Use generic parser for Cirrus codec" If I revert this one and all came back to work. By comparing the alsa-info.txt output I figured that there is something wrong with the GPIO settings on node 0x01: This is the excerpt with the previous patch mentioned reverted: Codec: Cirrus Logic CS4206 Address: 0 AFG Function Id: 0x1 (unsol 0) Vendor Id: 0x10134206 Subsystem Id: 0x106b5600 Revision Id: 0x100302 No Modem Function Group found Default PCM: rates [0x7f0]: 32000 44100 48000 88200 96000 176400 192000 bits [0x1e]: 16 20 24 32 formats [0x3]: PCM FLOAT Default Amp-In caps: N/A Default Amp-Out caps: N/A State of AFG node 0x01: Power states: D0 D3 CLKSTOP EPSS Power: setting=D0, actual=D0 GPIO: io=4, o=0, i=0, unsolicited=0, wake=0 IO[0]: enable=0, dir=0, wake=0, sticky=0, data=0, unsol=0 IO[1]: enable=1, dir=1, wake=0, sticky=0, data=0, unsol=0 IO[2]: enable=0, dir=0, wake=0, sticky=0, data=0, unsol=0 IO[3]: enable=1, dir=1, wake=0, sticky=0, data=1, unsol=0 and here the output vanilla kernel: Codec: Cirrus Logic CS4206 Address: 0 AFG Function Id: 0x1 (unsol 0) Vendor Id: 0x10134206 Subsystem Id: 0x106b5600 Revision Id: 0x100302 No Modem Function Group found Default PCM: rates [0x7f0]: 32000 44100 48000 88200 96000 176400 192000 bits [0x1e]: 16 20 24 32 formats [0x3]: PCM FLOAT Default Amp-In caps: N/A Default Amp-Out caps: N/A State of AFG node 0x01: Power states: D0 D3 CLKSTOP EPSS Power: setting=D0, actual=D0 GPIO: io=4, o=0, i=0, unsolicited=0, wake=0 IO[0]: enable=0, dir=0, wake=0, sticky=0, data=0, unsol=0 IO[1]: enable=1, dir=1, wake=0, sticky=0, data=0, unsol=0 IO[2]: enable=0, dir=0, wake=0, sticky=0, data=0, unsol=0 IO[3]: enable=1, dir=1, wake=0, sticky=0, data=0, unsol=0 The diff is "IO[3] data=1 vs data=0". I used then the patch function from the awesome hda-analyzer tool to patch this on the vanilla kernel and there suddenly I have output from the speakers. I don't know if someone else has reported this already or if there is already a fix. I was without internet for few days and didn't report this in time. If that is the case, sorry for the noise. The complete logs from hda-analyzer are here: http://www.monom.org/misc/alsa/alsa-info.txt-v3.9-rc2-284-gaea8b5d http://www.monom.org/misc/alsa/alsa-info.txt-revert-1077a024 cheers, daniel

1 1

[alsa-devel] channel swapping issue on OMAP3/TWL4030 is back
by Peter Meerwald 26 Mar '13

26 Mar '13

Hello, I'm running Linux 3.7 on a beagle-xm and observe very reproducible stereo channel swapping issue on playback of a stereo stream (one channel has a sine, the other is zero); the channel swap occurs on starting the playback with a scope I have observed that the audio data indeed swaps relative to the FSX; so the problem is on the I2S bus I see the issue with the TWL4030 audio codec; McBSP2 is slave, TWL4030 is master on a custom board I see the same issue with a TLV320AIC3104 on McBSP3 (have tried to configured the TLV320AIC3104 as slave and master, same channel swapping issue for both configs) the channel swapping issue has been discussed here, http://comments.gmane.org/gmane.linux.alsa.devel/82178, and patches have been pointed out for 2.6.32 it appears that the swapping issue is back, probably due to the DMA rework in recent kernel? can someone confirm the issue? any suggestions? thanks, regards, p. -- Peter Meerwald +43-664-2444418 (mobile)

4 11

[alsa-devel] [bug] sbawe: the PnP won't stop
by David Flater 26 Mar '13

26 Mar '13

I've got an ISA PnP SB AWE type card in a system with a non-working DMA 5. Without intervention, the Sound Blaster always gets the broken DMA. I have no BIOS options to control the allocation of DMAs, and the PnP configuration done within ALSA ignores /etc/isapnp.conf [*]. DMA 6 works fine if I can get sbawe to stop re-initializing the PnP and just use it. # Use setup from isapnp with dma16 moved to 6. modprobe snd-sbawe isapnp=0 port=0x220 mpu_port=0x330 fm_port=0x388 awe_port=0x620 irq=5 dma8=1 dma16=6 modprobe snd-emu8000-synth Problem is, regardless whether isapnp is 0 or 1, sbawe always PnPs the card and puts it back on broken DMA 5. If I have already got the card correctly configured with /sbin/isapnp beforehand, sbawe leaves the original resources tied up as a ghost card and allocates again (/proc/dma): 1: SoundBlaster - 8bit 2: floppy 3: SoundBlaster - 8bit 4: cascade 5: SoundBlaster - 16bit 6: SoundBlaster - 16bit To get it working I ripped the following lines of code out of kernel 3.8.4, blocking PnP completely. As for a proper nondestructive fix, I'm not sure whether the intent was (A) you were only supposed to get one of the two registrations, or (B) the PnP registration was not supposed to reinit everything if it was already set up. --- sound/isa/sb/sb16.c~ 2013-03-14 14:27:14.000000000 -0400 +++ sound/isa/sb/sb16.c 2013-03-23 20:18:50.137987332 -0400 @@ -674,13 +674,6 @@ #ifdef CONFIG_PNP if (!err) isa_registered = 1; - - err = pnp_register_card_driver(&sb16_pnpc_driver); - if (!err) - pnp_registered = 1; - - if (isa_registered) - err = 0; #endif return err; } [*] So is there a different config file for PnP within ALSA? -- DWF

1 1

Re: [alsa-devel] BUG: soft lockup when recording audio on MX28EVK with ASoC sgtl5000
by Fabio Estevam 26 Mar '13

26 Mar '13

Hi Hector, On Mon, Mar 25, 2013 at 7:52 AM, Hector Palacios <hector.palacios(a)digi.com> wrote: > Hello, > > I just tried recording audio on Freescale's MX28EVK that uses ASoC sgtl5000 > (kernel v3.8) with: > > arecord -M -f cd sound.wav --duration 10 > > and got a scheduler message: > > [ 789.041847] [sched_delayed] sched: RT throttling activated I have just tested arecord on 3.8,4 and I haven't seen the 'soft lockup' bug. However, I see that mxs saif gets busy after the record and any attempt to do an 'aplay' after the first arecord fails. With the attached patch applied I am able to do arecord/aplay sequence several times. I don't have a cable handy here to actually test if the sound is recorded properly or not. Could you please test it against 3.8.4? Thanks, Fabio Estevam

1 1

[alsa-devel] [PATCH 1/2] ASoC: tegra: add Tegra114 support to the AHUB driver
by Stephen Warren 25 Mar '13

25 Mar '13

From: Stephen Warren <swarren(a)nvidia.com> Tegra114's AHUB shares a design with Tegra30, with the followin changes: * Supports more (10 vs. 4) bi-directional FIFO channels into RAM. * Requires a separate block of registers to support the above. * Supports more attached clients, i.e. new audio multiplexing and de-multiplexing modules. * Is affected by more clocks due to the above. This change fully defines the device tree binding changes required to represent these changes, and minimally extends the driver to support the new hardware, without exposing any of the new FIFO channels. Signed-off-by: Stephen Warren <swarren(a)nvidia.com> --- .../bindings/sound/nvidia,tegra30-ahub.txt | 26 ++++++-- sound/soc/tegra/tegra30_ahub.c | 64 ++++++++++++++------ sound/soc/tegra/tegra30_ahub.h | 16 +++++ 3 files changed, 84 insertions(+), 22 deletions(-) diff --git a/Documentation/devicetree/bindings/sound/nvidia,tegra30-ahub.txt b/Documentation/devicetree/bindings/sound/nvidia,tegra30-ahub.txt index 1ac7b16..0e5c12c 100644 --- a/Documentation/devicetree/bindings/sound/nvidia,tegra30-ahub.txt +++ b/Documentation/devicetree/bindings/sound/nvidia,tegra30-ahub.txt @@ -1,12 +1,22 @@ NVIDIA Tegra30 AHUB (Audio Hub) Required properties: -- compatible : "nvidia,tegra30-ahub" +- compatible : "nvidia,tegra30-ahub", "nvidia,tegra114-ahub", etc. - reg : Should contain the register physical address and length for each of - the AHUB's APBIF registers and the AHUB's own registers. + the AHUB's register blocks. + - Tegra30 requires 2 entries, for the APBIF and AHUB/AUDIO register blocks. + - Tegra114 requires an additional entry, for the APBIF2 register block. - interrupts : Should contain AHUB interrupt -- nvidia,dma-request-selector : The Tegra DMA controller's phandle and - request selector for the first APBIF channel. +- nvidia,dma-request-selector : A list of the DMA channel specifiers. Each + entry contains the Tegra DMA controller's phandle and request selector. + If a single entry is present, the request selectors for the channels are + assumed to be contiguous, and increment from this value. + If multiple values are given, one value must be given per channel. +- clocks : Must contain an entry for each required entry in clock-names. +- clock-names : Must include the following entries: + - Tegra30: Requires d_audio, apbif, i2s0, i2s1, i2s2, i2s3, i2s4, dam0, + dam1, dam2, spdif_in. + - Tegra114: Additionally requires amx, adx. - ranges : The bus address mapping for the configlink register bus. Can be empty since the mapping is 1:1. - #address-cells : For the configlink bus. Should be <1>; @@ -25,7 +35,13 @@ ahub@70080000 { reg = <0x70080000 0x200 0x70080200 0x100>; interrupts = < 0 103 0x04 >; nvidia,dma-request-selector = <&apbdma 1>; - + clocks = <&tegra_car 106>, <&tegra_car 107>, <&tegra_car 30>, + <&tegra_car 11>, <&tegra_car 18>, <&tegra_car 101>, + <&tegra_car 102>, <&tegra_car 108>, <&tegra_car 109>, + <&tegra_car 110>, <&tegra_car 162>; + clock-names = "d_audio", "apbif", "i2s0", "i2s1", "i2s2", + "i2s3", "i2s4", "dam0", "dam1", "dam2", + "spdif_in"; ranges; #address-cells = <1>; #size-cells = <1>; diff --git a/sound/soc/tegra/tegra30_ahub.c b/sound/soc/tegra/tegra30_ahub.c index e5cfb4a..4405c3a 100644 --- a/sound/soc/tegra/tegra30_ahub.c +++ b/sound/soc/tegra/tegra30_ahub.c @@ -287,16 +287,27 @@ int tegra30_ahub_unset_rx_cif_source(enum tegra30_ahub_rxcif rxcif) } EXPORT_SYMBOL_GPL(tegra30_ahub_unset_rx_cif_source); -static const char * const configlink_clocks[] = { - "i2s0", - "i2s1", - "i2s2", - "i2s3", - "i2s4", - "dam0", - "dam1", - "dam2", - "spdif_in", +#define CLK_LIST_MASK_TEGRA30 BIT(0) +#define CLK_LIST_MASK_TEGRA114 BIT(1) + +#define CLK_LIST_MASK_TEGRA30_OR_LATER \ + (CLK_LIST_MASK_TEGRA30 | CLK_LIST_MASK_TEGRA114) + +static const struct { + const char *clk_name; + u32 clk_list_mask; +} configlink_clocks[] = { + { "i2s0", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "i2s1", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "i2s2", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "i2s3", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "i2s4", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "dam0", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "dam1", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "dam2", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "spdif_in", CLK_LIST_MASK_TEGRA30_OR_LATER }, + { "amx", CLK_LIST_MASK_TEGRA114 }, + { "adx", CLK_LIST_MASK_TEGRA114 }, }; #define LAST_REG(name) \ @@ -424,8 +435,24 @@ static const struct regmap_config tegra30_ahub_ahub_regmap_config = { .cache_type = REGCACHE_RBTREE, }; +static struct tegra30_ahub_soc_data soc_data_tegra30 = { + .clk_list_mask = CLK_LIST_MASK_TEGRA30, +}; + +static struct tegra30_ahub_soc_data soc_data_tegra114 = { + .clk_list_mask = CLK_LIST_MASK_TEGRA114, +}; + +static const struct of_device_id tegra30_ahub_of_match[] = { + { .compatible = "nvidia,tegra114-ahub", .data = &soc_data_tegra114 }, + { .compatible = "nvidia,tegra30-ahub", .data = &soc_data_tegra30 }, + {}, +}; + static int tegra30_ahub_probe(struct platform_device *pdev) { + const struct of_device_id *match; + const struct tegra30_ahub_soc_data *soc_data; struct clk *clk; int i; struct resource *res0, *res1, *region; @@ -436,16 +463,24 @@ static int tegra30_ahub_probe(struct platform_device *pdev) if (ahub) return -ENODEV; + match = of_match_device(tegra30_ahub_of_match, &pdev->dev); + if (!match) + return -EINVAL; + soc_data = match->data; + /* * The AHUB hosts a register bus: the "configlink". For this to * operate correctly, all devices on this bus must be out of reset. * Ensure that here. */ for (i = 0; i < ARRAY_SIZE(configlink_clocks); i++) { - clk = clk_get(&pdev->dev, configlink_clocks[i]); + if (!(configlink_clocks[i].clk_list_mask & + soc_data->clk_list_mask)) + continue; + clk = clk_get(&pdev->dev, configlink_clocks[i].clk_name); if (IS_ERR(clk)) { dev_err(&pdev->dev, "Can't get clock %s\n", - configlink_clocks[i]); + configlink_clocks[i].clk_name); ret = PTR_ERR(clk); goto err; } @@ -592,11 +627,6 @@ static int tegra30_ahub_remove(struct platform_device *pdev) return 0; } -static const struct of_device_id tegra30_ahub_of_match[] = { - { .compatible = "nvidia,tegra30-ahub", }, - {}, -}; - static const struct dev_pm_ops tegra30_ahub_pm_ops = { SET_RUNTIME_PM_OPS(tegra30_ahub_runtime_suspend, tegra30_ahub_runtime_resume, NULL) diff --git a/sound/soc/tegra/tegra30_ahub.h b/sound/soc/tegra/tegra30_ahub.h index e690e2e..7189be9 100644 --- a/sound/soc/tegra/tegra30_ahub.h +++ b/sound/soc/tegra/tegra30_ahub.h @@ -468,7 +468,23 @@ extern int tegra30_ahub_set_rx_cif_source(enum tegra30_ahub_rxcif rxcif, enum tegra30_ahub_txcif txcif); extern int tegra30_ahub_unset_rx_cif_source(enum tegra30_ahub_rxcif rxcif); +struct tegra30_ahub_soc_data { + u32 clk_list_mask; + /* + * FIXME: There are many more differences in HW, such as: + * - More APBIF channels. + * - Extra separate chunks of register address space to represent + * the extra APBIF channels. + * - More units connected to the AHUB, so that tegra30_ahub_[rt]xcif + * need expansion, coupled with there being more defined bits in + * the AHUB routing registers. + * However, the driver doesn't support those new features yet, so we + * don't represent them here yet. + */ +}; + struct tegra30_ahub { + const struct tegra30_ahub_soc_data *soc_data; struct device *dev; struct clk *clk_d_audio; struct clk *clk_apbif; -- 1.7.10.4

2 4