[alsa-devel] FSL SSI 32 bit transfer (WAS Re: [RFC PATCH] ASoC: fsl-ssi: Add support for CBM_CFS 16/24 bit)
Hi all
I have a quick question. Do you know if 32 bit transfer on dma is suppose to work on IMX6? 32LE sample 2 channels
Michael
On Thu, Jul 31, 2014 at 7:09 AM, Michael Trimarchi michael@amarulasolutions.com wrote:
Hi
Il 29/lug/2014 08:18 "Markus Pargmann" mpa@pengutronix.de ha scritto:
Hi,
On Mon, Jul 28, 2014 at 02:46:38PM +0200, Michael Trimarchi wrote:
Signed-off-by: Michael Trimarchi michael@amarulasolutions.com
A commit description what you are doing in this patch and how your setup looks like would be good.
sound/soc/fsl/fsl_ssi.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-)
diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 9bfef55..743e50d 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -724,10 +724,19 @@ static int fsl_ssi_hw_params(struct snd_pcm_substream *substream, regmap_update_bits(regs, CCSR_SSI_SRCCR, CCSR_SSI_SxCCR_WL_MASK, wl);
if (!fsl_ssi_is_ac97(ssi_private))
if (!fsl_ssi_is_ac97(ssi_private)) {
u8 i2smode;
if (ssi_private->i2s_mode == (CCSR_SSI_SCR_I2S_MODE_NORMAL
|
CCSR_SSI_SCR_NET) && sample_size >= 24)
This condition seems a bit fragile using the i2s_mode variable.
I don't know about this bitclock master mode, why is this condition necessary, what is the difference with sample_size >= 24?
I will do in better way and cleanup. Problem is how data is packed in 32 bit sample. So 24/32 should be master but 16/32 seems a different story. It packs like 16/16 bit.
I will take a look during the weekend
Michael
Regards,
Markus
i2smode = CCSR_SSI_SCR_I2S_MODE_MASTER;
else
i2smode = ssi_private->i2s_mode;
regmap_update_bits(regs, CCSR_SSI_SCR, CCSR_SSI_SCR_NET |
CCSR_SSI_SCR_I2S_MODE_MASK,
channels == 1 ? 0 :
ssi_private->i2s_mode);
channels == 1 ? 0 : i2smode);
} return 0;
} @@ -789,6 +798,15 @@ static int _fsl_ssi_set_dai_fmt(struct fsl_ssi_private *ssi_private, CCSR_SSI_SxCCR_DC_MASK, CCSR_SSI_SxCCR_DC(2)); break;
case SND_SOC_DAIFMT_CBM_CFS:
ssi_private->i2s_mode |=
CCSR_SSI_SCR_I2S_MODE_NORMAL;
regmap_update_bits(regs, CCSR_SSI_STCCR,
CCSR_SSI_SxCCR_DC_MASK,
CCSR_SSI_SxCCR_DC(2));
regmap_update_bits(regs, CCSR_SSI_SRCCR,
CCSR_SSI_SxCCR_DC_MASK,
CCSR_SSI_SxCCR_DC(2));
break; case SND_SOC_DAIFMT_CBM_CFM: ssi_private->i2s_mode |=
CCSR_SSI_SCR_I2S_MODE_SLAVE; break; @@ -853,6 +871,11 @@ static int _fsl_ssi_set_dai_fmt(struct fsl_ssi_private *ssi_private, case SND_SOC_DAIFMT_CBM_CFM: scr &= ~CCSR_SSI_SCR_SYS_CLK_EN; break;
case SND_SOC_DAIFMT_CBM_CFS:
strcr &= ~CCSR_SSI_STCR_TXDIR;
strcr |= CCSR_SSI_STCR_TFDIR;
scr &= ~CCSR_SSI_SCR_SYS_CLK_EN;
break; default: return -EINVAL; }
-- 1.8.1.2
-- Pengutronix e.K. | | Industrial Linux Solutions |http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone:+49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555|
Hi
On Sat, Aug 2, 2014 at 12:39 PM, Michael Trimarchi michael@amarulasolutions.com wrote:
Hi all
I have a quick question. Do you know if 32 bit transfer on dma is suppose to work on IMX6? 32LE sample 2 channels
I have already verified and the shift register is 24 bit
Michael
Michael
On Thu, Jul 31, 2014 at 7:09 AM, Michael Trimarchi michael@amarulasolutions.com wrote:
Hi
Il 29/lug/2014 08:18 "Markus Pargmann" mpa@pengutronix.de ha scritto:
Hi,
On Mon, Jul 28, 2014 at 02:46:38PM +0200, Michael Trimarchi wrote:
Signed-off-by: Michael Trimarchi michael@amarulasolutions.com
A commit description what you are doing in this patch and how your setup looks like would be good.
sound/soc/fsl/fsl_ssi.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-)
diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 9bfef55..743e50d 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -724,10 +724,19 @@ static int fsl_ssi_hw_params(struct snd_pcm_substream *substream, regmap_update_bits(regs, CCSR_SSI_SRCCR, CCSR_SSI_SxCCR_WL_MASK, wl);
if (!fsl_ssi_is_ac97(ssi_private))
if (!fsl_ssi_is_ac97(ssi_private)) {
u8 i2smode;
if (ssi_private->i2s_mode == (CCSR_SSI_SCR_I2S_MODE_NORMAL
|
CCSR_SSI_SCR_NET) && sample_size >= 24)
This condition seems a bit fragile using the i2s_mode variable.
I don't know about this bitclock master mode, why is this condition necessary, what is the difference with sample_size >= 24?
I will do in better way and cleanup. Problem is how data is packed in 32 bit sample. So 24/32 should be master but 16/32 seems a different story. It packs like 16/16 bit.
I will take a look during the weekend
Michael
Regards,
Markus
i2smode = CCSR_SSI_SCR_I2S_MODE_MASTER;
else
i2smode = ssi_private->i2s_mode;
regmap_update_bits(regs, CCSR_SSI_SCR, CCSR_SSI_SCR_NET |
CCSR_SSI_SCR_I2S_MODE_MASK,
channels == 1 ? 0 :
ssi_private->i2s_mode);
channels == 1 ? 0 : i2smode);
} return 0;
} @@ -789,6 +798,15 @@ static int _fsl_ssi_set_dai_fmt(struct fsl_ssi_private *ssi_private, CCSR_SSI_SxCCR_DC_MASK, CCSR_SSI_SxCCR_DC(2)); break;
case SND_SOC_DAIFMT_CBM_CFS:
ssi_private->i2s_mode |=
CCSR_SSI_SCR_I2S_MODE_NORMAL;
regmap_update_bits(regs, CCSR_SSI_STCCR,
CCSR_SSI_SxCCR_DC_MASK,
CCSR_SSI_SxCCR_DC(2));
regmap_update_bits(regs, CCSR_SSI_SRCCR,
CCSR_SSI_SxCCR_DC_MASK,
CCSR_SSI_SxCCR_DC(2));
break; case SND_SOC_DAIFMT_CBM_CFM: ssi_private->i2s_mode |=
CCSR_SSI_SCR_I2S_MODE_SLAVE; break; @@ -853,6 +871,11 @@ static int _fsl_ssi_set_dai_fmt(struct fsl_ssi_private *ssi_private, case SND_SOC_DAIFMT_CBM_CFM: scr &= ~CCSR_SSI_SCR_SYS_CLK_EN; break;
case SND_SOC_DAIFMT_CBM_CFS:
strcr &= ~CCSR_SSI_STCR_TXDIR;
strcr |= CCSR_SSI_STCR_TFDIR;
scr &= ~CCSR_SSI_SCR_SYS_CLK_EN;
break; default: return -EINVAL; }
-- 1.8.1.2
-- Pengutronix e.K. | | Industrial Linux Solutions |http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone:+49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555|
participants (1)
-
Michael Trimarchi