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

[alsa-devel] [PATCH] Make snd_BUG_ON() always evaluate and return the conditional expression.
by Christine Spang 06 Mar '13

06 Mar '13

Having snd_BUG_ON() only evaluate its conditional when CONFIG_SND_DEBUG is set leads to frequent bugs, since other similar macros in the kernel have different behavior. Let's make snd_BUG_ON() act like those macros so it will stop being accidentally misused. Signed-off-by: Christine Spang <christine.spang(a)oracle.com> --- Documentation/DocBook/writing-an-alsa-driver.tmpl | 12 +++++------- include/sound/core.h | 24 ++++++++--------------- 2 files changed, 13 insertions(+), 23 deletions(-) diff --git a/Documentation/DocBook/writing-an-alsa-driver.tmpl b/Documentation/DocBook/writing-an-alsa-driver.tmpl index bd6fee2..06741e9 100644 --- a/Documentation/DocBook/writing-an-alsa-driver.tmpl +++ b/Documentation/DocBook/writing-an-alsa-driver.tmpl @@ -6164,14 +6164,12 @@ struct _snd_pcm_runtime { <para> The macro takes an conditional expression to evaluate. - When <constant>CONFIG_SND_DEBUG</constant>, is set, the - expression is actually evaluated. If it's non-zero, it shows - the warning message such as + When <constant>CONFIG_SND_DEBUG</constant>, is set, if the + expression is non-zero, it shows the warning message such as <computeroutput>BUG? (xxx)</computeroutput> - normally followed by stack trace. It returns the evaluated - value. - When no <constant>CONFIG_SND_DEBUG</constant> is set, this - macro always returns zero. + normally followed by stack trace. + + In both cases it returns the evaluated value. </para> </section> diff --git a/include/sound/core.h b/include/sound/core.h index 7cede2d..a63680b 100644 --- a/include/sound/core.h +++ b/include/sound/core.h @@ -379,18 +379,10 @@ void __snd_printk(unsigned int level, const char *file, int line, * snd_BUG_ON - debugging check macro * @cond: condition to evaluate * - * When CONFIG_SND_DEBUG is set, this macro evaluates the given condition, - * and call WARN() and returns the value if it's non-zero. - * - * When CONFIG_SND_DEBUG is not set, this just returns zero, and the given - * condition is ignored. - * - * NOTE: the argument won't be evaluated at all when CONFIG_SND_DEBUG=n. - * Thus, don't put any statement that influences on the code behavior, - * such as pre/post increment, to the argument of this macro. - * If you want to evaluate and give a warning, use standard WARN_ON(). + * Has the same behavior as WARN_ON when CONFIG_SND_DEBUG is set, + * otherwise just evaluates the conditional and returns the value. */ -#define snd_BUG_ON(cond) WARN((cond), "BUG? (%s)\n", __stringify(cond)) +#define snd_BUG_ON(cond) WARN_ON((cond)) #else /* !CONFIG_SND_DEBUG */ @@ -400,11 +392,11 @@ __printf(2, 3) static inline void _snd_printd(int level, const char *format, ...) {} #define snd_BUG() do { } while (0) -static inline int __snd_bug_on(int cond) -{ - return 0; -} -#define snd_BUG_ON(cond) __snd_bug_on(0 && (cond)) /* always false */ + +#define snd_BUG_ON(condition) ({ \ + int __ret_warn_on = !!(condition); \ + unlikely(__ret_warn_on); \ +}) #endif /* CONFIG_SND_DEBUG */ -- 1.8.2.rc0

3 5

Re: [alsa-devel] [PATCH] ASoC: samsung: let device core setup the default pin configuration
by Mark Brown 06 Mar '13

06 Mar '13

On Wed, Mar 06, 2013 at 12:42:09PM +0100, Heiko Stübner wrote: > What is going away is the deprecated non-pinctrl way to setup the pins on dt > platforms. OK, can we have a changelog improvement please? I didn't read the patch properly because the changelog said it would break my primary development system.

2 1

Re: [alsa-devel] ALSA: usb: Work around CM6631 sample
by Demian Martin 06 Mar '13

06 Mar '13

Date: Tue, 5 Mar 2013 23:03:03 +0100 From: Torstein Hegge <hegge(a)resisty.net> To: alsa-devel(a)alsa-project.org Subject: [alsa-devel] [PATCH v3] ALSA: usb: Work around CM6631 sample rate change bug Message-ID: <20130305220303.GD20417(a)pvv.ntnu.no> Content-Type: text/plain; charset=utf-8 The C-Media CM6631 USB-to-S/PDIF receiver doesn't respond to changes in sample rate while the interface is active. Reset the interface after setting the sampling frequency on sample rate changes, to ensure that the sample rate set by snd_usb_init_sample_rate() is used. Otherwise, the device will try to use the sample rate of the previous stream, causing distorted sound on sample rate changes. The reset is performed for all current C-Media UAC V2.0 devices, including CM6610, CM6620 and CM6631, but has only been tested with CM6631. Signed-off-by: Torstein Hegge <hegge(a)resisty.net> I met with the CMedia team late last year. They told me that the chips are different and they were showing me a CM6631a that has some key differences from the CM6631. The two are pin compatible. I got the impression that the driver requirements are different. I asked about Linux support and what the problem was. First, on UAC2 Their software wizard dug deep into the problem with their CM6631 and ALSA. He concluded that there is a sequence problem between setting sample rate and sending the "play' command. Something about UAC1 method not being the same as UAC2. He indicated that the ALSA driver handles this differently from the OSX and their Windows drivers. I may well have this description confused. In any case he has a revised firmware for the new CM6631A chip that addresses this in ALSA. I asked them to contribute the change but they seemed to get distracted with other projects. Needless to say for them the volume on these chips has been small. I just requested that they contribute to this effort again. I don't know if it will happen. Demian

2 1

[alsa-devel] [PATCH 4/4] ASoC: arizona: Optimise FLL loop gains
by Mark Brown 06 Mar '13

06 Mar '13

For optimal performance the FLL loop gain should be adjusted depending on the frequency of the input clock for the loop. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/codecs/arizona.c | 43 +++++++++++++++++++++++++++++++++++++++---- 1 file changed, 39 insertions(+), 4 deletions(-) diff --git a/sound/soc/codecs/arizona.c b/sound/soc/codecs/arizona.c index 41c116f..29dd641 100644 --- a/sound/soc/codecs/arizona.c +++ b/sound/soc/codecs/arizona.c @@ -1175,6 +1175,16 @@ static struct { { 1000000, 13500000, 0, 1 }, }; +static struct { + unsigned int min; + unsigned int max; + u16 gain; +} fll_gains[] = { + { 0, 256000, 0 }, + { 256000, 1000000, 2 }, + { 1000000, 13500000, 4 }, +}; + struct arizona_fll_cfg { int n; int theta; @@ -1182,6 +1192,7 @@ struct arizona_fll_cfg { int refdiv; int outdiv; int fratio; + int gain; }; static int arizona_calc_fll(struct arizona_fll *fll, @@ -1241,6 +1252,18 @@ static int arizona_calc_fll(struct arizona_fll *fll, return -EINVAL; } + for (i = 0; i < ARRAY_SIZE(fll_gains); i++) { + if (fll_gains[i].min <= Fref && Fref <= fll_gains[i].max) { + cfg->gain = fll_gains[i].gain; + break; + } + } + if (i == ARRAY_SIZE(fll_gains)) { + arizona_fll_err(fll, "Unable to find gain for Fref=%uHz\n", + Fref); + return -EINVAL; + } + cfg->n = target / (ratio * Fref); if (target % (ratio * Fref)) { @@ -1268,13 +1291,15 @@ static int arizona_calc_fll(struct arizona_fll *fll, cfg->n, cfg->theta, cfg->lambda); arizona_fll_dbg(fll, "FRATIO=%x(%d) OUTDIV=%x REFCLK_DIV=%x\n", cfg->fratio, cfg->fratio, cfg->outdiv, cfg->refdiv); + arizona_fll_dbg(fll, "GAIN=%d\n", cfg->gain); return 0; } static void arizona_apply_fll(struct arizona *arizona, unsigned int base, - struct arizona_fll_cfg *cfg, int source) + struct arizona_fll_cfg *cfg, int source, + bool sync) { regmap_update_bits(arizona->regmap, base + 3, ARIZONA_FLL1_THETA_MASK, cfg->theta); @@ -1289,6 +1314,15 @@ static void arizona_apply_fll(struct arizona *arizona, unsigned int base, cfg->refdiv << ARIZONA_FLL1_CLK_REF_DIV_SHIFT | source << ARIZONA_FLL1_CLK_REF_SRC_SHIFT); + if (sync) + regmap_update_bits(arizona->regmap, base + 0x7, + ARIZONA_FLL1_GAIN_MASK, + cfg->gain << ARIZONA_FLL1_GAIN_SHIFT); + else + regmap_update_bits(arizona->regmap, base + 0x9, + ARIZONA_FLL1_GAIN_MASK, + cfg->gain << ARIZONA_FLL1_GAIN_SHIFT); + regmap_update_bits(arizona->regmap, base + 2, ARIZONA_FLL1_CTRL_UPD | ARIZONA_FLL1_N_MASK, ARIZONA_FLL1_CTRL_UPD | cfg->n); @@ -1326,17 +1360,18 @@ static void arizona_enable_fll(struct arizona_fll *fll, ARIZONA_FLL1_OUTDIV_MASK, ref->outdiv << ARIZONA_FLL1_OUTDIV_SHIFT); - arizona_apply_fll(arizona, fll->base, ref, fll->ref_src); + arizona_apply_fll(arizona, fll->base, ref, fll->ref_src, + false); if (fll->sync_src >= 0) arizona_apply_fll(arizona, fll->base + 0x10, sync, - fll->sync_src); + fll->sync_src, true); } else if (fll->sync_src >= 0) { regmap_update_bits(arizona->regmap, fll->base + 5, ARIZONA_FLL1_OUTDIV_MASK, sync->outdiv << ARIZONA_FLL1_OUTDIV_SHIFT); arizona_apply_fll(arizona, fll->base, sync, - fll->sync_src); + fll->sync_src, false); } else { arizona_fll_err(fll, "No clocks provided\n"); return; -- 1.7.10.4

1 0

[alsa-devel] [PATCH 3/4] ASoC: arizona: Increase FLL synchroniser bandwidth for high frequencies
by Mark Brown 06 Mar '13

06 Mar '13

If we are using a high freqency SYNCCLK then increasing the bandwidth of the synchroniser improves performance. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/codecs/arizona.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/sound/soc/codecs/arizona.c b/sound/soc/codecs/arizona.c index 81da6ef..41c116f 100644 --- a/sound/soc/codecs/arizona.c +++ b/sound/soc/codecs/arizona.c @@ -1342,6 +1342,17 @@ static void arizona_enable_fll(struct arizona_fll *fll, return; } + /* + * Increase the bandwidth if we're not using a low frequency + * sync source. + */ + if (fll->sync_src >= 0 && fll->sync_freq > 100000) + regmap_update_bits(arizona->regmap, fll->base + 0x17, + ARIZONA_FLL1_SYNC_BW, 0); + else + regmap_update_bits(arizona->regmap, fll->base + 0x17, + ARIZONA_FLL1_SYNC_BW, ARIZONA_FLL1_SYNC_BW); + if (!arizona_is_enabled_fll(fll)) pm_runtime_get(arizona->dev); -- 1.7.10.4

1 0

[alsa-devel] [PATCH 2/4] mfd: wm5102: Map in additional FLL control registers
by Mark Brown 06 Mar '13

06 Mar '13

Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- drivers/mfd/wm5102-tables.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/mfd/wm5102-tables.c b/drivers/mfd/wm5102-tables.c index f3c2e9a..768fd42 100644 --- a/drivers/mfd/wm5102-tables.c +++ b/drivers/mfd/wm5102-tables.c @@ -291,12 +291,14 @@ static const struct reg_default wm5102_reg_default[] = { { 0x00000176, 0x0000 }, /* R374 - FLL1 Control 6 */ { 0x00000177, 0x0181 }, /* R375 - FLL1 Loop Filter Test 1 */ { 0x00000178, 0x0000 }, /* R376 - FLL1 NCO Test 0 */ + { 0x00000179, 0x0000 }, /* R377 - FLL1 Control 7 */ { 0x00000181, 0x0000 }, /* R385 - FLL1 Synchroniser 1 */ { 0x00000182, 0x0000 }, /* R386 - FLL1 Synchroniser 2 */ { 0x00000183, 0x0000 }, /* R387 - FLL1 Synchroniser 3 */ { 0x00000184, 0x0000 }, /* R388 - FLL1 Synchroniser 4 */ { 0x00000185, 0x0000 }, /* R389 - FLL1 Synchroniser 5 */ { 0x00000186, 0x0000 }, /* R390 - FLL1 Synchroniser 6 */ + { 0x00000187, 0x0001 }, /* R391 - FLL1 Synchroniser 7 */ { 0x00000189, 0x0000 }, /* R393 - FLL1 Spread Spectrum */ { 0x0000018A, 0x0004 }, /* R394 - FLL1 GPIO Clock */ { 0x00000191, 0x0000 }, /* R401 - FLL2 Control 1 */ @@ -307,12 +309,14 @@ static const struct reg_default wm5102_reg_default[] = { { 0x00000196, 0x0000 }, /* R406 - FLL2 Control 6 */ { 0x00000197, 0x0000 }, /* R407 - FLL2 Loop Filter Test 1 */ { 0x00000198, 0x0000 }, /* R408 - FLL2 NCO Test 0 */ + { 0x00000199, 0x0000 }, /* R409 - FLL2 Control 7 */ { 0x000001A1, 0x0000 }, /* R417 - FLL2 Synchroniser 1 */ { 0x000001A2, 0x0000 }, /* R418 - FLL2 Synchroniser 2 */ { 0x000001A3, 0x0000 }, /* R419 - FLL2 Synchroniser 3 */ { 0x000001A4, 0x0000 }, /* R420 - FLL2 Synchroniser 4 */ { 0x000001A5, 0x0000 }, /* R421 - FLL2 Synchroniser 5 */ { 0x000001A6, 0x0000 }, /* R422 - FLL2 Synchroniser 6 */ + { 0x000001A7, 0x0001 }, /* R423 - FLL2 Synchroniser 7 */ { 0x000001A9, 0x0000 }, /* R425 - FLL2 Spread Spectrum */ { 0x000001AA, 0x0004 }, /* R426 - FLL2 GPIO Clock */ { 0x00000200, 0x0006 }, /* R512 - Mic Charge Pump 1 */ @@ -1056,12 +1060,14 @@ static bool wm5102_readable_register(struct device *dev, unsigned int reg) case ARIZONA_FLL1_CONTROL_6: case ARIZONA_FLL1_LOOP_FILTER_TEST_1: case ARIZONA_FLL1_NCO_TEST_0: + case ARIZONA_FLL1_CONTROL_7: case ARIZONA_FLL1_SYNCHRONISER_1: case ARIZONA_FLL1_SYNCHRONISER_2: case ARIZONA_FLL1_SYNCHRONISER_3: case ARIZONA_FLL1_SYNCHRONISER_4: case ARIZONA_FLL1_SYNCHRONISER_5: case ARIZONA_FLL1_SYNCHRONISER_6: + case ARIZONA_FLL1_SYNCHRONISER_7: case ARIZONA_FLL1_SPREAD_SPECTRUM: case ARIZONA_FLL1_GPIO_CLOCK: case ARIZONA_FLL2_CONTROL_1: @@ -1072,12 +1078,14 @@ static bool wm5102_readable_register(struct device *dev, unsigned int reg) case ARIZONA_FLL2_CONTROL_6: case ARIZONA_FLL2_LOOP_FILTER_TEST_1: case ARIZONA_FLL2_NCO_TEST_0: + case ARIZONA_FLL2_CONTROL_7: case ARIZONA_FLL2_SYNCHRONISER_1: case ARIZONA_FLL2_SYNCHRONISER_2: case ARIZONA_FLL2_SYNCHRONISER_3: case ARIZONA_FLL2_SYNCHRONISER_4: case ARIZONA_FLL2_SYNCHRONISER_5: case ARIZONA_FLL2_SYNCHRONISER_6: + case ARIZONA_FLL2_SYNCHRONISER_7: case ARIZONA_FLL2_SPREAD_SPECTRUM: case ARIZONA_FLL2_GPIO_CLOCK: case ARIZONA_MIC_CHARGE_PUMP_1: -- 1.7.10.4

1 0

[alsa-devel] [PATCH] ASoC: wm5102: Apply a SYSCLK patch for later revs
by Mark Brown 06 Mar '13

06 Mar '13

Evaluation has identified some performance improvements to the device. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/codecs/wm5102.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/sound/soc/codecs/wm5102.c b/sound/soc/codecs/wm5102.c index deb1097..cef288c 100644 --- a/sound/soc/codecs/wm5102.c +++ b/sound/soc/codecs/wm5102.c @@ -573,6 +573,13 @@ static const struct reg_default wm5102_sysclk_reva_patch[] = { { 0x025e, 0x0112 }, }; +static const struct reg_default wm5102_sysclk_revb_patch[] = { + { 0x3081, 0x08FE }, + { 0x3083, 0x00ED }, + { 0x30C1, 0x08FE }, + { 0x30C3, 0x00ED }, +}; + static int wm5102_sysclk_ev(struct snd_soc_dapm_widget *w, struct snd_kcontrol *kcontrol, int event) { @@ -587,6 +594,10 @@ static int wm5102_sysclk_ev(struct snd_soc_dapm_widget *w, patch = wm5102_sysclk_reva_patch; patch_size = ARRAY_SIZE(wm5102_sysclk_reva_patch); break; + default: + patch = wm5102_sysclk_revb_patch; + patch_size = ARRAY_SIZE(wm5102_sysclk_revb_patch); + break; } switch (event) { -- 1.7.10.4

1 0

Re: [alsa-devel] [PATCH v3 09/12] ASoC: dmaengine_pcm: add snd_dmaengine_generic_pcm_open()
by Shawn Guo 06 Mar '13

06 Mar '13

On Wed, Mar 06, 2013 at 12:28:18AM +0100, Marek Vasut wrote: > Hi Shawn, > > > With generic DMA device tree binding and helper function > > dma_request_slave_channel() in place, dmaengine_pcm should support > > that in requesting DMA channel for users that support generic DMA > > device tree binding. > > > > Add a new API snd_dmaengine_generic_pcm_open() as the variant of > > snd_dmaengine_pcm_open(). It takes DMA client struct device pointer > > and slave channel name as arguments and calls generic DMA helper > > dma_request_slave_channel() to request DMA channel from dmaengine. > > > > Signed-off-by: Shawn Guo <shawn.guo(a)linaro.org> > > Cc: Mark Brown <broonie(a)opensource.wolfsonmicro.com> > > Cc: alsa-devel(a)alsa-project.org > > --- > > Changes since v2: > > - Drop 'const' from argument 'name' to fix the warning below. > > > [...] > > v3 changelog missing ;-) What did you adjust here please? The "Changes since v2" means v3 changelog. Shawn

1 0

[alsa-devel] [PATCH] ASoC: cs42l73: If Internal MCLK is >= 6.4MHz, then set SCLK to 64*Fs.
by Paul Handrigan 06 Mar '13

06 Mar '13

Signed-off-by: Paul Handrigan <Paul.Handrigan(a)cirrus.com> --- sound/soc/codecs/cs42l73.c | 6 +++++- 1 files changed, 5 insertions(+), 1 deletions(-) diff --git a/sound/soc/codecs/cs42l73.c b/sound/soc/codecs/cs42l73.c index 6361dab..3b20c86 100644 --- a/sound/soc/codecs/cs42l73.c +++ b/sound/soc/codecs/cs42l73.c @@ -1180,7 +1180,11 @@ static int cs42l73_pcm_hw_params(struct snd_pcm_substream *substream, priv->config[id].mmcc &= 0xC0; priv->config[id].mmcc |= cs42l73_mclk_coeffs[mclk_coeff].mmcc; priv->config[id].spc &= 0xFC; - priv->config[id].spc |= MCK_SCLK_MCLK; + /* Use SCLK=64*Fs if internal MCLK >= 6.4MHz */ + if (priv->mclk >= 6400000) + priv->config[id].spc |= MCK_SCLK_64FS; + else + priv->config[id].spc |= MCK_SCLK_MCLK; } else { /* CS42L73 Slave */ priv->config[id].spc &= 0xFC; -- 1.7.0.4

3 2

[alsa-devel] [PATCH] ASoC: omap3pandora: Fix compilation error
by Peter Ujfalusi 06 Mar '13

06 Mar '13

Fixes: sound/soc/omap/omap3pandora.c: In function ‘omap3pandora_dac_event’: sound/soc/omap/omap3pandora.c:92:19: error: request for member ‘dev’ in something not a structure or union make[3]: *** [sound/soc/omap/omap3pandora.o] Error 1 Which is introduced by: dd194b4 ASoC: omap: Check regulator enable for DAC on Pandora Signed-off-by: Peter Ujfalusi <peter.ujfalusi(a)ti.com> --- sound/soc/omap/omap3pandora.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/omap/omap3pandora.c b/sound/soc/omap/omap3pandora.c index 10ced9d..9e46e1d 100644 --- a/sound/soc/omap/omap3pandora.c +++ b/sound/soc/omap/omap3pandora.c @@ -89,7 +89,7 @@ static int omap3pandora_dac_event(struct snd_soc_dapm_widget *w, if (SND_SOC_DAPM_EVENT_ON(event)) { ret = regulator_enable(omap3pandora_dac_reg); if (ret) { - dev_err(w->dapm.dev, "Failed to power DAC: %d\n", ret); + dev_err(w->dapm->dev, "Failed to power DAC: %d\n", ret); return ret; } mdelay(1); -- 1.8.1.5

2 1