March 2009 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] ASoC updates for 2.6.30
by Mark Brown 06 Apr '09

06 Apr '09

The following changes since commit 1f2186951e02f2a5bcda9459f63136918932385a: Takashi Iwai (1): ALSA: Fix wrong pointer to dev_err() in arm/pxa2xx-ac97-lib.c are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound-2.6.git for-2.6.30 Philipp Zabel (2): ASoC: pxa-ssp: Use 16-bit DMA for magician stereo ASoC: Add Magician machine support sound/soc/pxa/Kconfig | 10 + sound/soc/pxa/Makefile | 2 + sound/soc/pxa/magician.c | 560 ++++++++++++++++++++++++++++++++++++++++++++++ sound/soc/pxa/pxa-ssp.c | 12 +- 4 files changed, 581 insertions(+), 3 deletions(-) create mode 100644 sound/soc/pxa/magician.c

3 3

[alsa-devel] problem with speaker-test
by Sedji Gaouaou 02 Apr '09

02 Apr '09

Hi, I am trying to use speaker-test on my sam9g20 board, with the Wolfson wm8731, but I can't have any tone... Instead it returns this error: # speaker-test -c 2 -t1 speaker-test 1.0.14 Playback device is default Stream parameters are 48000Hz, S16_LE, 2 channels Using 16 octaves of pink noise Rate set to 48000Hz (requested 48000Hz) Buffer size range from 16 to 1073745880 Period size range from 0 to 0 Using max buffer size 1073745880 Periods = 4 was set period_size = 0 was set buffer_size = 8192 Unable to set sw params for playback: Invalid argument Setting of swparams failed: Invalid argument So I tried: # speaker-test -D"name" -c 2 -t1, but I don't know exactly what should I pass as "name". I did: aplay -L to list all the devices, but I don't see anything... Is anybody got an idea here? Regards, Sedji

2 2

[alsa-devel] [patch] tlv320aic3x: disable ADC/DAC while changing clock
by Daniel Glöckner 02 Apr '09

02 Apr '09

While using a tlv320aic3101 as a clock master, it often happened that the clock lines got stuck while changing the sampling rate. In a discussion with TI technical support it was suggested to shut down the ADC and DAC while there are changes being made to the clock. Following that rule resolves the problem, but merely writing the registers to switch the ADC/DAC off is not enough. One needs to poll the power status registers to wait until they have completely shut down. The aforementioned rule implies that we can not disable the PLL in stand-by bias level. Signed-off-by: Daniel Glöckner <dg(a)emlix.com> --- sound/soc/codecs/tlv320aic3x.c | 89 ++++++++++++++++++++++++++++------------ sound/soc/codecs/tlv320aic3x.h | 8 ++++ 2 files changed, 71 insertions(+), 26 deletions(-) diff --git a/sound/soc/codecs/tlv320aic3x.c b/sound/soc/codecs/tlv320aic3x.c index aea0cb7..4c44022 100644 --- a/sound/soc/codecs/tlv320aic3x.c +++ b/sound/soc/codecs/tlv320aic3x.c @@ -740,6 +740,57 @@ static int aic3x_add_widgets(struct snd_soc_codec *codec) return 0; } +static u8 aic3x_power_codec(struct snd_soc_codec *codec, u8 new) +{ + u8 val, old; + + val = aic3x_read_reg_cache(codec, LINE1L_2_LADC_CTRL); + old = (val & LADC_PWR_ON) ? 1 : 0; + if ((old ^ new) & 1) { + val ^= LADC_PWR_ON; + aic3x_write(codec, LINE1L_2_LADC_CTRL, val); + } + + val = aic3x_read_reg_cache(codec, LINE1R_2_RADC_CTRL); + old += (val & RADC_PWR_ON) ? 2 : 0; + if ((old ^ new) & 2) { + val ^= RADC_PWR_ON; + aic3x_write(codec, LINE1R_2_RADC_CTRL, val); + } + + val = aic3x_read_reg_cache(codec, DAC_PWR); + old += (val & LDAC_PWR_ON) ? 4 : 0; + old += (val & RDAC_PWR_ON) ? 8 : 0; + if ((old ^ new) & 4) + val ^= LDAC_PWR_ON; + if ((old ^ new) & 8) + val ^= RDAC_PWR_ON; + if ((old ^ new) & 12) + aic3x_write(codec, DAC_PWR, val); + + return old; +} + +static u8 aic3x_power_off_wait(struct snd_soc_codec *codec) +{ + int timeout = 1000; + u8 old, data; + old = aic3x_power_codec(codec, 0); + + do { + aic3x_read(codec, ADC_FLAGS, &data); + } while ((data & (LADC_PWR_STATUS | RADC_PWR_STATUS)) && --timeout > 0); + + do { + aic3x_read(codec, MODULE_PWR_STATUS, &data); + } while ((data & (LDAC_PWR_STATUS | RDAC_PWR_STATUS)) && --timeout > 0); + + if (timeout < 0) + printk(KERN_WARNING "aic3x: timeout waiting for ADC/DAC to shut down"); + + return old; +} + static int aic3x_hw_params(struct snd_pcm_substream *substream, struct snd_pcm_hw_params *params, struct snd_soc_dai *dai) @@ -749,9 +800,11 @@ static int aic3x_hw_params(struct snd_pcm_substream *substream, struct snd_soc_codec *codec = socdev->codec; struct aic3x_priv *aic3x = codec->private_data; int codec_clk = 0, bypass_pll = 0, fsref, last_clk = 0; - u8 data, r, p, pll_q, pll_p = 1, pll_r = 1, pll_j = 1; + u8 data, r, p, pll_q, pll_p = 1, pll_r = 1, pll_j = 1, state_saved; u16 pll_d = 1; + state_saved = aic3x_power_off_wait(codec); + /* select data word length */ data = aic3x_read_reg_cache(codec, AIC3X_ASD_INTF_CTRLB) & (~(0x3 << 4)); @@ -805,8 +858,10 @@ static int aic3x_hw_params(struct snd_pcm_substream *substream, data |= (data << 4); aic3x_write(codec, AIC3X_SAMPLE_RATE_SEL_REG, data); - if (bypass_pll) + if (bypass_pll) { + aic3x_power_codec(codec, state_saved); return 0; + } /* Use PLL * find an apropriate setup for j, d, r and p by iterating over @@ -847,6 +902,7 @@ static int aic3x_hw_params(struct snd_pcm_substream *substream, if (last_clk == 0) { printk(KERN_ERR "%s(): unable to setup PLL\n", __func__); + aic3x_power_codec(codec, state_saved); return -EINVAL; } @@ -857,7 +913,11 @@ static int aic3x_hw_params(struct snd_pcm_substream *substream, aic3x_write(codec, AIC3X_PLL_PROGC_REG, (pll_d >> 6) << PLLD_MSB_SHIFT); aic3x_write(codec, AIC3X_PLL_PROGD_REG, (pll_d & 0x3F) << PLLD_LSB_SHIFT); + data = aic3x_read_reg_cache(codec, AIC3X_PLL_PROGA_REG); + aic3x_write(codec, AIC3X_PLL_PROGA_REG, data | (pll_p << PLLP_SHIFT) + | PLL_ENABLE); + aic3x_power_codec(codec, state_saved); return 0; } @@ -951,37 +1011,14 @@ static int aic3x_set_bias_level(struct snd_soc_codec *codec, switch (level) { case SND_SOC_BIAS_ON: - /* all power is driven by DAPM system */ - if (aic3x->master) { - /* enable pll */ - reg = aic3x_read_reg_cache(codec, AIC3X_PLL_PROGA_REG); - aic3x_write(codec, AIC3X_PLL_PROGA_REG, - reg | PLL_ENABLE); - } break; case SND_SOC_BIAS_PREPARE: break; case SND_SOC_BIAS_STANDBY: - /* - * all power is driven by DAPM system, - * so output power is safe if bypass was set - */ - if (aic3x->master) { - /* disable pll */ - reg = aic3x_read_reg_cache(codec, AIC3X_PLL_PROGA_REG); - aic3x_write(codec, AIC3X_PLL_PROGA_REG, - reg & ~PLL_ENABLE); - } break; case SND_SOC_BIAS_OFF: /* force all power off */ - reg = aic3x_read_reg_cache(codec, LINE1L_2_LADC_CTRL); - aic3x_write(codec, LINE1L_2_LADC_CTRL, reg & ~LADC_PWR_ON); - reg = aic3x_read_reg_cache(codec, LINE1R_2_RADC_CTRL); - aic3x_write(codec, LINE1R_2_RADC_CTRL, reg & ~RADC_PWR_ON); - - reg = aic3x_read_reg_cache(codec, DAC_PWR); - aic3x_write(codec, DAC_PWR, reg & ~(LDAC_PWR_ON | RDAC_PWR_ON)); + aic3x_power_off_wait(codec); reg = aic3x_read_reg_cache(codec, HPLOUT_CTRL); aic3x_write(codec, HPLOUT_CTRL, reg & ~HPLOUT_PWR_ON); diff --git a/sound/soc/codecs/tlv320aic3x.h b/sound/soc/codecs/tlv320aic3x.h index ac827e5..0b4fedf 100644 --- a/sound/soc/codecs/tlv320aic3x.h +++ b/sound/soc/codecs/tlv320aic3x.h @@ -69,6 +69,8 @@ #define RAGC_CTRL_B 30 #define RAGC_CTRL_C 31 +/* ADC Flag register */ +#define ADC_FLAGS 34 /* DAC Power and Left High Power Output control registers */ #define DAC_PWR 37 #define HPLCOM_CFG 37 @@ -126,6 +128,8 @@ #define DACR1_2_LLOPM_VOL 85 #define LLOPM_CTRL 86 #define RLOPM_CTRL 93 +/* Module Power status register */ +#define MODULE_PWR_STATUS 94 /* GPIO/IRQ registers */ #define AIC3X_STICKY_IRQ_FLAGS_REG 96 #define AIC3X_RT_IRQ_FLAGS_REG 97 @@ -191,6 +195,10 @@ #define MONOLOPM_PWR_ON 0x01 #define LLOPM_PWR_ON 0x01 #define RLOPM_PWR_ON 0x01 +#define LADC_PWR_STATUS 0x40 +#define RADC_PWR_STATUS 0x04 +#define LDAC_PWR_STATUS 0x80 +#define RDAC_PWR_STATUS 0x40 #define INVERT_VOL(val) (0x7f - val) -- 1.6.2.107.ge47ee

3 14

[alsa-devel] [PATCH 0/7] snd-atmel-ac97c: misc fixes and error reporting
by Hans-Christian Egtvedt 02 Apr '09

02 Apr '09

This patch series contains six minor fixes and one enhancment of the snd-atmel-ac97c driver. It is based on the for-next branch in sound-2.6 git repo. The enhancment is enabling interrupts for error reporting, this might be useful if you have sound troubles (glitches) on a stressed system. -- Best regards, Hans-Christian Egtvedt

2 10

[alsa-devel] [PATCH] Add configuration for HP EliteBook 2530p
by Matthew Garrett 01 Apr '09

01 Apr '09

The HP EliteBook 2530p needs the AD1884A_LAPTOP codec setup. Signed-off-by: Matthew Garrett <mjg(a)redhat.com> diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c index 5bb48ee..945b029 100644 --- a/sound/pci/hda/patch_analog.c +++ b/sound/pci/hda/patch_analog.c @@ -3924,6 +3924,7 @@ static struct snd_pci_quirk ad1884a_cfg_tbl[] = { SND_PCI_QUIRK(0x103c, 0x3056, "HP", AD1884A_MOBILE), SND_PCI_QUIRK_MASK(0x103c, 0xfff0, 0x3070, "HP", AD1884A_MOBILE), SND_PCI_QUIRK_MASK(0x103c, 0xfff0, 0x30e0, "HP laptop", AD1884A_LAPTOP), + SND_PCI_QUIRK_MASK(0x103c, 0xfff0, 0x30e1, "HP laptop", AD1884A_LAPTOP), SND_PCI_QUIRK_MASK(0x103c, 0xff00, 0x3600, "HP laptop", AD1884A_LAPTOP), SND_PCI_QUIRK(0x17aa, 0x20ac, "Thinkpad X300", AD1884A_THINKPAD), {} -- Matthew Garrett | mjg59(a)srcf.ucam.org

1 1

[alsa-devel] Bug 2552 typo in ice1724.c
by Alan Horstmann 31 Mar '09

31 Mar '09

Hi, I noticed this trivial typo report on bugtrack, -no 2552 and thought it may as well be fixed, and the bug closed. Patch inline & attached. Not sure if it needs a sign-off? Alan ice1724.c-name-typo-fix.diff Fix trivial name string typo as reported in bug 2552 Signed-off-by: Alan Horstmann <gineera(a)aspect135.co.uk> --- alsa-kernel/pci/ice1712/ice1724-orig.c 2008-06-27 09:11:58.000000000 +0100 +++ alsa-kernel/pci/ice1712/ice1724.c 2009-01-05 11:11:33.000000000 +0000 @@ -1239,7 +1239,7 @@ if (ice->force_pdma4 || ice->force_rdma1) name = "ICE1724 Secondary"; else - name = "IEC1724 IEC958"; + name = "ICE1724 IEC958"; err = snd_pcm_new(ice->card, name, device, play, capt, &pcm); if (err < 0) return err;

1 1

[PATCH] ASoC: Don't defer resume work for AC97 codecs
by Mark Brown 31 Mar '09

31 Mar '09

AC97 devices may have other drivers hanging off them directly so need to have resumed when the resume function returns meaning that we can't defer the resume - complete it immediately for them. Non-AC97 devices should not have other drivers hanging directly off the ASoC devices. We only really need the deferral for non-AC97 devices - it's there since some I2C buses are very slow and non-AC97 codecs often have large numbers of registers to restore and require delays to bring the codec up cleanly leading to a substantial impact on overall resume time. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> Cc: stable(a)kernel.org --- sound/soc/soc-core.c | 18 ++++++++++++++---- 1 files changed, 14 insertions(+), 4 deletions(-) diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 6e710f7..6c62d4a 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -767,11 +767,21 @@ static int soc_resume(struct platform_device *pdev) { struct snd_soc_device *socdev = platform_get_drvdata(pdev); struct snd_soc_card *card = socdev->card; + struct snd_soc_dai *cpu_dai = card->dai_link[0].cpu_dai; - dev_dbg(socdev->dev, "scheduling resume work\n"); - - if (!schedule_work(&card->deferred_resume_work)) - dev_err(socdev->dev, "resume work item may be lost\n"); + /* AC97 devices might have other drivers hanging off them so + * need to resume immediately. Other drivers don't have that + * problem and may take a substantial amount of time to resume + * due to I/O costs and anti-pop so handle them out of line. + */ + if (cpu_dai->ac97_control) { + dev_dbg(socdev->dev, "Resuming AC97 immediately\n"); + soc_resume_deferred(&card->deferred_resume_work); + } else { + dev_dbg(socdev->dev, "Scheduling resume work\n"); + if (!schedule_work(&card->deferred_resume_work)) + dev_err(socdev->dev, "resume work item may be lost\n"); + } return 0; } -- 1.6.2.1

1 0

[alsa-devel] [PATCH 0/2] ASoC: TWL4030 updates
by Peter Ujfalusi 31 Mar '09

31 Mar '09

Hello, Two small updates for the TWL4030 codec driver: 1. Add support for 96KHz sample rate playback (96KHz capture is not supported) 2. Constrains to protect the already running stream from accidental reconfiguration --- Peter Ujfalusi (2): ASoC: TWL4030: 96KHz playback support ASoC: TWL4030: Add constrains for second stream sound/soc/codecs/twl4030.c | 56 +++++++++++++++++++++++++++++++++++++++++++- 1 files changed, 55 insertions(+), 1 deletions(-)

2 4

[alsa-devel] Non-interleaved multi-channel access, what's required? Oxygen driver...
by olly＠funkster.org 30 Mar '09

30 Mar '09

Hi all, Hope this is the right place for this. I'm working on an app that requires non-interleaved access to 6 output channels, but alas the driver for the card that I'm using (Asus Xonar D1, AV100 / C-Media Oxygen chipset) doesn't support the noninterleaved access type, giving "incorrect access type" in response to this command: if ((err = snd_pcm_hw_params_set_access(playback_handle, hw_params, SND_PCM_ACCESS_RW_NONINTERLEAVED)) < 0) { fprintf(stderr, "cannot set access type (%s)\n", snd_strerror (err)); return false; } In a bout of optimism, I simply added the SNDRV_PCM_INFO_NONINTERLEAVED flag to the .info part of the snd_pcm_hardware struct that's defined for the card in oxygen_pcm.c and recompiled the kernel, not expecting it to really work. To my surprise, it does work! But, only for the "default" card and with two channels. If I try to use any of the multi-channel devices (e.g. front, surround51) then each channel appears to alternate round the buffers for all the channels. So... how do I find out if this is a limitation of the hardware? Can I do anything to the multi-channel part of the driver to make non-interleaved access work correctly? Or is there some clever use of the .asoundrc that can make it do what I need to? Many thanks in advance for any help, -- Oliver Stephenson http://www.funkster.org/ _______________________________________ Sent using funkster.org webmail service

1 0

[alsa-devel] snd_device_name_hint destroys the config
by Henrik Grunell 30 Mar '09

30 Mar '09

Hi, After I've called snd_device_name_hint the configuration seems destroyed. I can no longer access (open) e.g. "default" and "pulse" pcms. As a test, I took the pcm_list function in aplay and put it in my application. The second (and following) call(s) shows that devices are missing compared to the first call. I use Ubuntu 8.10 (1.0.17) and I see the same behavior on Fedora 7 (1.0.14). Is this a known issue? I haven't found anything searching on this. Thanks, Henrik ________________________________ CONFIDENTIALITY NOTE: This e-mail message, including any attachment(s), contains information that may be confidential, protected by the attorney-client or other legal privileges, and/or proprietary non-public information. If you are not an intended recipient of this message or an authorized assistant to an intended recipient, please do not read, copy, use or disclose this communication and notify the sender immediately. It should be noted that any review, retransmission, dissemination or other use of, or taking action or reliance upon, this information by persons or entities other than the intended recipient is prohibited.

1 0