December 2011 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH 0/5] snd-usb endpoint rework, take 5
by Daniel Mack 22 Dec '11

22 Dec '11

I actually wanted to wait for more input, but as the flaws reported so far were rather minor, I addressed to following details in this new round as mentioned by Takashi and Clemens: - removed the faulty "(!u || u->urb)" check - guard the result of min() with the spinlock inside next_packet_size() - removed the tasklet and made the handler a function that is called in-place Thanks again for the reviews, and keep them coming :) Daniel Daniel Mack (5): ALSA: snd-usb: add snd_usb_audio-wide mutex ALSA: snd-usb: implement new endpoint streaming model ALSA: snd-usb: switch over to new endpoint streaming logic ALSA: snd-usb: remove old streaming logic ALSA: snd-usb: add support for implicit feedback sound/usb/card.c | 10 +- sound/usb/card.h | 62 +++ sound/usb/endpoint.c | 1454 +++++++++++++++++++++++++------------------------ sound/usb/endpoint.h | 32 +- sound/usb/pcm.c | 437 +++++++++++++--- sound/usb/stream.c | 31 +- sound/usb/usbaudio.h | 2 + 7 files changed, 1224 insertions(+), 804 deletions(-) -- 1.7.5.4

2 11

Re: [alsa-devel] [GIT PULL] ARM: amba: Enable module alias autogeneration for AMBA drivers
by Russell King - ARM Linux 22 Dec '11

22 Dec '11

On Thu, Dec 22, 2011 at 11:21:25AM +0000, Dave Martin wrote: > On Mon, Dec 19, 2011 at 09:39:42PM +0000, Russell King - ARM Linux wrote: > > I have pulled it, I just haven't merged it in anywhere yet. > > OK, great -- thanks. > > I wasn't too familiar with the procedure, so I just wanted > to be sure that I hadn't done things wrong. FYI, I'll be pushing out an update to my tree Friday evening.

1 0

[alsa-devel] [PATCH v2 0/9] OMAP4: ASoC: Support for PandaBoard family
by Peter Ujfalusi 22 Dec '11

22 Dec '11

Hello, the following series will add ASoC support for PandaBoards. PandaBoards have different audio routings compared to SDP4430/Blaze boards, but the differences not that big to justify a new ASoC machine driver. This series is based on the first three patch from the v1 sereis (those, which has been applied by Liam). Changes since v1: As suggested by Mark the platform data, machine driver has been converted to process feature flags instead of Board IDs. - The ASoC machine driver registers all DAPM widgets. Based on the received configuration it will disable the unused paths - PCM for dmic is only created, if the board has digital mic conencted - Jack functionality will be only used on boards supporting it - Sound card name for SDP4430 has net been changed Regards, Peter --- Peter Ujfalusi (9): include: platform_data: Platform data header for OMAP4 ASoC audio OMAP4: 4430sdp: Register platform device for OMAP4 audio ASoC: omap-abe-twl6040: Convert to platform deriver ASoC: twl6040: Convert MICBIAS to SUPPLY widget ASoC: omap-abe-twl6040: Add complete DAPM routing ASoC: omap-abe-twl6040: DAI link selection based on platform data ASoC: omap-abe-twl6040: Configure card according to platform data OMAP4: omap4panda: Enable audio support ASoC: Kconfig: OMAP4: Enable support for PandaBoards arch/arm/mach-omap2/board-4430sdp.c | 27 ++++ arch/arm/mach-omap2/board-omap4panda.c | 56 ++++++++- include/linux/platform_data/omap-abe-twl6040.h | 48 +++++++ sound/soc/codecs/twl6040.c | 16 +- sound/soc/omap/Kconfig | 5 +- sound/soc/omap/omap-abe-twl6040.c | 178 ++++++++++++++++-------- 6 files changed, 260 insertions(+), 70 deletions(-) create mode 100644 include/linux/platform_data/omap-abe-twl6040.h -- 1.7.8.1

3 17

[alsa-devel] [PATCH v2 0/9] OMAP4: ASoC: Support for PandaBoard family
by Peter Ujfalusi 22 Dec '11

22 Dec '11

Hello, the following series will add ASoC support for PandaBoards. PandaBoards have different audio routings compared to SDP4430/Blaze boards, but the differences not that big to justify a new ASoC machine driver. This series is based on the first three patch from the v1 sereis (those, which has been applied by Liam). Changes since v1: As suggested by Mark the platform data, machine driver has been converted to process feature flags instead of Board IDs. - The ASoC machine driver registers all DAPM widgets. Based on the received configuration it will disable the unused paths - PCM for dmic is only created, if the board has digital mic conencted - Jack functionality will be only used on boards supporting it - Sound card name for SDP4430 has net been changed Regards, Peter --- Peter Ujfalusi (9): include: platform_data: Platform data header for OMAP4 ASoC audio OMAP4: 4430sdp: Register platform device for OMAP4 audio ASoC: omap-abe-twl6040: Convert to platform deriver ASoC: twl6040: Convert MICBIAS to SUPPLY widget ASoC: omap-abe-twl6040: Add complete DAPM routing ASoC: omap-abe-twl6040: DAI link selection based on platform data ASoC: omap-abe-twl6040: Configure card according to platform data OMAP4: omap4panda: Enable audio support ASoC: Kconfig: OMAP4: Enable support for PandaBoards arch/arm/mach-omap2/board-4430sdp.c | 27 ++++ arch/arm/mach-omap2/board-omap4panda.c | 56 ++++++++- include/linux/platform_data/omap-abe-twl6040.h | 48 +++++++ sound/soc/codecs/twl6040.c | 16 +- sound/soc/omap/Kconfig | 5 +- sound/soc/omap/omap-abe-twl6040.c | 178 ++++++++++++++++-------- 6 files changed, 260 insertions(+), 70 deletions(-) create mode 100644 include/linux/platform_data/omap-abe-twl6040.h -- 1.7.8.1

4 20

[alsa-devel] Audio BoF @ ELC 2012 US
by Patrick Lai 22 Dec '11

22 Dec '11

Hi, Is there going to be Audio BOF session at ELC 2012 US. Liam/Mark: Are you guys going to attend ELC? Thanks Patrick -- Sent by an employee of the Qualcomm Innovation Center, Inc. The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

3 2

[alsa-devel] [PATCH] ASoC: Declare soc_new_pcm() properly
by Mark Brown 22 Dec '11

22 Dec '11

Ensure that everything is seeing the same declaration by moving it to a header file rather than putting the declaration in soc-core.c Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- include/sound/soc.h | 1 + sound/soc/soc-core.c | 2 -- 2 files changed, 1 insertions(+), 2 deletions(-) diff --git a/include/sound/soc.h b/include/sound/soc.h index e5414ed..312dfd3 100644 --- a/include/sound/soc.h +++ b/include/sound/soc.h @@ -323,6 +323,7 @@ int snd_soc_platform_read(struct snd_soc_platform *platform, unsigned int reg); int snd_soc_platform_write(struct snd_soc_platform *platform, unsigned int reg, unsigned int val); +int soc_new_pcm(struct snd_soc_pcm_runtime *rtd, int num); /* Utility functions to get clock rates from various things */ int snd_soc_calc_frame_size(int sample_size, int channels, int tdm_slots); diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index d3fcdc1..d19d0d1 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -59,8 +59,6 @@ static LIST_HEAD(dai_list); static LIST_HEAD(platform_list); static LIST_HEAD(codec_list); -int soc_new_pcm(struct snd_soc_pcm_runtime *rtd, int num); - /* * This is a timeout to do a DAPM powerdown after a stream is closed(). * It can be used to eliminate pops between different playback streams, e.g. -- 1.7.7.3

2 1

[alsa-devel] [PATCH] ALSA: hda/jack - Also add jack kctls for Conexant codecs
by David Henningsson 22 Dec '11

22 Dec '11

Signed-off-by: David Henningsson <david.henningsson(a)canonical.com> --- Hi Takashi, A patch for your jack branch - it seems you forgot to add kctls for conexant? Also it would be nice to see this merged, soon it might be too late for 3.3? sound/pci/hda/patch_conexant.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index ae9c028..99942e6 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -4325,6 +4325,7 @@ static int cx_auto_build_input_controls(struct hda_codec *codec) static int cx_auto_build_controls(struct hda_codec *codec) { + struct conexant_spec *spec = codec->spec; int err; err = cx_auto_build_output_controls(codec); @@ -4333,6 +4334,9 @@ static int cx_auto_build_controls(struct hda_codec *codec) err = cx_auto_build_input_controls(codec); if (err < 0) return err; + err = snd_hda_jack_add_kctls(codec, &spec->autocfg); + if (err < 0) + return err; return conexant_build_controls(codec); } -- 1.7.5.4

3 10

[alsa-devel] [PATCH] ASoC: fsl: Add .owner to struct snd_soc_card
by Axel Lin 22 Dec '11

22 Dec '11

Missed .owner of struct snd_soc_card will prevent the module from being removed from underneath its users. Reported-by: Lothar Waßmann <LW(a)KARO-electronics.de> Signed-off-by: Axel Lin <axel.lin(a)gmail.com> --- sound/soc/fsl/efika-audio-fabric.c | 14 +++++++------- sound/soc/fsl/pcm030-audio-fabric.c | 14 +++++++------- 2 files changed, 14 insertions(+), 14 deletions(-) diff --git a/sound/soc/fsl/efika-audio-fabric.c b/sound/soc/fsl/efika-audio-fabric.c index 108b5d8..b2acd329 100644 --- a/sound/soc/fsl/efika-audio-fabric.c +++ b/sound/soc/fsl/efika-audio-fabric.c @@ -31,8 +31,6 @@ #define DRV_NAME "efika-audio-fabric" -static struct snd_soc_card card; - static struct snd_soc_dai_link efika_fabric_dai[] = { { .name = "AC97", @@ -52,6 +50,13 @@ static struct snd_soc_dai_link efika_fabric_dai[] = { }, }; +static struct snd_soc_card card = { + .name = "Efika", + .owner = THIS_MODULE, + .dai_link = efika_fabric_dai, + .num_links = ARRAY_SIZE(efika_fabric_dai), +}; + static __init int efika_fabric_init(void) { struct platform_device *pdev; @@ -60,11 +65,6 @@ static __init int efika_fabric_init(void) if (!of_machine_is_compatible("bplan,efika")) return -ENODEV; - card.name = "Efika"; - card.dai_link = efika_fabric_dai; - card.num_links = ARRAY_SIZE(efika_fabric_dai); - - pdev = platform_device_alloc("soc-audio", 1); if (!pdev) { pr_err("efika_fabric_init: platform_device_alloc() failed\n"); diff --git a/sound/soc/fsl/pcm030-audio-fabric.c b/sound/soc/fsl/pcm030-audio-fabric.c index ba4d85e..b3af55d 100644 --- a/sound/soc/fsl/pcm030-audio-fabric.c +++ b/sound/soc/fsl/pcm030-audio-fabric.c @@ -31,8 +31,6 @@ #define DRV_NAME "pcm030-audio-fabric" -static struct snd_soc_card card; - static struct snd_soc_dai_link pcm030_fabric_dai[] = { { .name = "AC97", @@ -52,6 +50,13 @@ static struct snd_soc_dai_link pcm030_fabric_dai[] = { }, }; +static struct snd_soc_card card = { + .name = "pcm030", + .owner = THIS_MODULE, + .dai_link = pcm030_fabric_dai, + .num_links = ARRAY_SIZE(pcm030_fabric_dai), +}; + static __init int pcm030_fabric_init(void) { struct platform_device *pdev; @@ -60,11 +65,6 @@ static __init int pcm030_fabric_init(void) if (!of_machine_is_compatible("phytec,pcm030")) return -ENODEV; - - card.name = "pcm030"; - card.dai_link = pcm030_fabric_dai; - card.num_links = ARRAY_SIZE(pcm030_fabric_dai); - pdev = platform_device_alloc("soc-audio", 1); if (!pdev) { pr_err("pcm030_fabric_init: platform_device_alloc() failed\n"); -- 1.7.5.4

2 1

Re: [alsa-devel] ASoC updates for 3.2
by Mark Brown 22 Dec '11

22 Dec '11

On Thu, Dec 22, 2011 at 08:38:39PM +0800, Shawn Guo wrote: > On 22 December 2011 20:02, Mark Brown > > If the API change is only in -next this shouldn't have been applied to > > the ASoC tree at all. This needs go along with the change that adds the > > new API otherwise you break bisection. Things like this *really* need > > to be made clear when posting patches. > I did mention that the patch series addresses the issue seen on -next > tree in the cover letter. And I replied your query as below. You shouldn't rely on cover letters getting noticed, they're too easy to disassociate from the patch. > > With patches like this you really need to supply more context - knowing > > which release the commit you're talking about appears in is useful for > > exammple, > This patch only applies to -next, as the offending patch only sits on > -next. Yeah, that was after I'd applied the patch. TBH given that this is an incompatible API change the original patch relly ought to be updating all users, there shouldn't be separate patches for the drivers.

1 0

Re: [alsa-devel] ASoC updates for 3.2
by Mark Brown 22 Dec '11

22 Dec '11

On Thu, Dec 22, 2011 at 07:21:44PM +0800, Shawn Guo wrote: > On 22 December 2011 19:04, Mark Brown > > Shawn Guo (1): > > ASoC: mxs: correct 'direction' of device_prep_dma_cyclic > If I remember correctly, the API change only sits on -next. You may > not want to push this patch for 3.2, I guess? Oh, ffs. git describe shows the change as being based off 3.1 so it would be expected to appear in the 3.2 merge window but looking at Linus' tree it doesn't seem to be there. I'll revert the change. If the API change is only in -next this shouldn't have been applied to the ASoC tree at all. This needs go along with the change that adds the new API otherwise you break bisection. Things like this *really* need to be made clear when posting patches. I'll revert the change, you should get anyone else who's applied these changes to do so as well.

2 1