| Message ID | 1407164887-22703-1-git-send-email-fabio.falzoi84@gmail.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | cf4f7fc3e7336e2e946880890e60ed36178889ea |
| Headers | show |
On 08/04/2014 10:08 AM, Fabio Falzoi wrote: > Add SND_SOC_DAIFMT_CBM_CFS support for Freescale architecture. > Successfully tested on i.MX 6Quad Wandboard and UDOO boards connected to > the pcm1792a codec. > In CBM_CFS mode, when using a sample size of 16 bits, we cannot use > CCSR_SSI_SCR_I2S_MODE_MASTER since we get a frame sync every 16 bits. > > Signed-off-by: Michael Trimarchi<michael@amarulasolutions.com> > Signed-off-by: Fabio Falzoi<fabio.falzoi84@gmail.com> > Tested-by: Angelo Adamo<adamo.a60@gmail.com> Acked-by: Timur Tabi <timur@tabi.org>
On Mon, Aug 4, 2014 at 12:08 PM, Fabio Falzoi <fabio.falzoi84@gmail.com> wrote: > Add SND_SOC_DAIFMT_CBM_CFS support for Freescale architecture. > Successfully tested on i.MX 6Quad Wandboard and UDOO boards connected to > the pcm1792a codec. Great, do you plan to submit the audio support into imx6q-udoo.dts? Thanks
Hi Fabio On Mon, Aug 4, 2014 at 5:27 PM, Fabio Estevam <festevam@gmail.com> wrote: > On Mon, Aug 4, 2014 at 12:08 PM, Fabio Falzoi <fabio.falzoi84@gmail.com> wrote: >> Add SND_SOC_DAIFMT_CBM_CFS support for Freescale architecture. >> Successfully tested on i.MX 6Quad Wandboard and UDOO boards connected to >> the pcm1792a codec. > > Great, do you plan to submit the audio support into imx6q-udoo.dts? > > Thanks udoo is not based on pcm1792a but on ac97 codec. I have seen somewhere the driver I can cleanup and let it go. We have solded the pcm1792a on wandboard and udoo ;) and on pandaboard es :D. Michael
On Mon, Aug 4, 2014 at 12:29 PM, Michael Trimarchi <michael@amarulasolutions.com> wrote: > udoo is not based on pcm1792a but on ac97 codec. I have seen somewhere > the driver I can cleanup and let it go. We have solded the pcm1792a on wandboard > and udoo ;) and on pandaboard es :D. Ah, good point :-) Yes, Udoo has the vt1613 codec (https://github.com/UDOOboard/Kernel_Unico/blob/master/sound/soc/codecs/vt1613.c ), which has not been upstreamed yet. Thanks
On Mon, Aug 04, 2014 at 05:08:07PM +0200, Fabio Falzoi wrote: > Add SND_SOC_DAIFMT_CBM_CFS support for Freescale architecture. > Successfully tested on i.MX 6Quad Wandboard and UDOO boards connected to > the pcm1792a codec. > In CBM_CFS mode, when using a sample size of 16 bits, we cannot use > CCSR_SSI_SCR_I2S_MODE_MASTER since we get a frame sync every 16 bits. Applied, thanks.
Hi Mark On Thu, Aug 7, 2014 at 7:45 PM, Mark Brown <broonie@kernel.org> wrote: > On Mon, Aug 04, 2014 at 05:08:07PM +0200, Fabio Falzoi wrote: >> Add SND_SOC_DAIFMT_CBM_CFS support for Freescale architecture. >> Successfully tested on i.MX 6Quad Wandboard and UDOO boards connected to >> the pcm1792a codec. >> In CBM_CFS mode, when using a sample size of 16 bits, we cannot use >> CCSR_SSI_SCR_I2S_MODE_MASTER since we get a frame sync every 16 bits. > > Applied, thanks. where do you apply it (I mean branch)? Michael
On Fri, Sep 12, 2014 at 2:37 PM, Michael Trimarchi
<michael@amarulasolutions.com> wrote:
> where do you apply it (I mean branch)?
https://git.kernel.org/cgit/linux/kernel/git/broonie/sound.git/commit/sound/soc/fsl/fsl_ssi.c?h=for-next&id=cf4f7fc3e7336e2e946880890e60ed36178889ea
It also shows in linux-next.
Hi Fabio On Fri, Sep 12, 2014 at 8:07 PM, Fabio Estevam <festevam@gmail.com> wrote: > On Fri, Sep 12, 2014 at 2:37 PM, Michael Trimarchi > <michael@amarulasolutions.com> wrote: > >> where do you apply it (I mean branch)? > > https://git.kernel.org/cgit/linux/kernel/git/broonie/sound.git/commit/sound/soc/fsl/fsl_ssi.c?h=for-next&id=cf4f7fc3e7336e2e946880890e60ed36178889ea > > It also shows in linux-next. Thank you Michael
diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 87eb577..2fc3e66 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -259,6 +259,11 @@ static bool fsl_ssi_is_i2s_master(struct fsl_ssi_private *ssi_private) SND_SOC_DAIFMT_CBS_CFS; } +static bool fsl_ssi_is_i2s_cbm_cfs(struct fsl_ssi_private *ssi_private) +{ + return (ssi_private->dai_fmt & SND_SOC_DAIFMT_MASTER_MASK) == + SND_SOC_DAIFMT_CBM_CFS; +} /** * fsl_ssi_isr: SSI interrupt handler * @@ -705,6 +710,23 @@ static int fsl_ssi_hw_params(struct snd_pcm_substream *substream, } } + if (!fsl_ssi_is_ac97(ssi_private)) { + u8 i2smode; + /* + * Switch to normal net mode in order to have a frame sync + * signal every 32 bits instead of 16 bits + */ + if (fsl_ssi_is_i2s_cbm_cfs(ssi_private) && sample_size == 16) + i2smode = CCSR_SSI_SCR_I2S_MODE_NORMAL | + CCSR_SSI_SCR_NET; + 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 : i2smode); + } + /* * FIXME: The documentation says that SxCCR[WL] should not be * modified while the SSI is enabled. The only time this can @@ -724,11 +746,6 @@ 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)) - regmap_update_bits(regs, CCSR_SSI_SCR, - CCSR_SSI_SCR_NET | CCSR_SSI_SCR_I2S_MODE_MASK, - channels == 1 ? 0 : ssi_private->i2s_mode); - return 0; } @@ -780,6 +797,7 @@ static int _fsl_ssi_set_dai_fmt(struct fsl_ssi_private *ssi_private, switch (fmt & SND_SOC_DAIFMT_FORMAT_MASK) { case SND_SOC_DAIFMT_I2S: switch (fmt & SND_SOC_DAIFMT_MASTER_MASK) { + case SND_SOC_DAIFMT_CBM_CFS: case SND_SOC_DAIFMT_CBS_CFS: ssi_private->i2s_mode |= CCSR_SSI_SCR_I2S_MODE_MASTER; regmap_update_bits(regs, CCSR_SSI_STCCR, @@ -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; }