linux-kernel - Re: [PATCH v2 6/6] ASoC: atmel_ssc_dai: Enable shared FSYNC source in frame-slave mode

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20190908133929.GA32003@qmqm.qmqm.pl>
Date:   Sun, 8 Sep 2019 15:39:30 +0200
From:   mirq-linux@...e.qmqm.pl
To:     Codrin.Ciubotariu@...rochip.com
Cc:     alsa-devel@...a-project.org, devicetree@...r.kernel.org,
        alexandre.belloni@...tlin.com, arnd@...db.de, 3chas3@...il.com,
        gregkh@...uxfoundation.org, perex@...ex.cz, lgirdwood@...il.com,
        Ludovic.Desroches@...rochip.com, broonie@...nel.org,
        mark.rutland@....com, Nicolas.Ferre@...rochip.com,
        robh-dt@...nel.org, tiwai@...e.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 6/6] ASoC: atmel_ssc_dai: Enable shared FSYNC source
 in frame-slave mode

On Mon, Aug 26, 2019 at 03:05:06PM +0000, Codrin.Ciubotariu@...rochip.com wrote:
> On 24.08.2019 23:26, Michał Mirosław wrote:
> > SSC driver allows only synchronous TX and RX. In slave mode for BCLK
> > it uses only one of TK or RK pin, but for LRCLK it configured separate
> > inputs from TF and RF pins. Allow configuration with common FS signal.
> > 
> > Signed-off-by: Michał Mirosław <mirq-linux@...e.qmqm.pl>
> > 
> > ---
> >   v2: use alternate DT binding
> >       split DT and drivers/misc changes
> > 
> > ---
> >   sound/soc/atmel/atmel_ssc_dai.c | 26 ++++++++++++++++++++++----
> >   1 file changed, 22 insertions(+), 4 deletions(-)
> > 
> > diff --git a/sound/soc/atmel/atmel_ssc_dai.c b/sound/soc/atmel/atmel_ssc_dai.c
> > index 48e9eef34c0f..035d4da58f2b 100644
> > --- a/sound/soc/atmel/atmel_ssc_dai.c
> > +++ b/sound/soc/atmel/atmel_ssc_dai.c
> > @@ -605,14 +605,32 @@ static int atmel_ssc_hw_params(struct snd_pcm_substream *substream,
> >   		return -EINVAL;
> >   	}
> >   
> > -	if (!atmel_ssc_cfs(ssc_p)) {
> > +	if (atmel_ssc_cfs(ssc_p)) {
> > +		/*
> > +		 * SSC provides LRCLK
> > +		 *
> > +		 * Both TF and RF are generated, so use them directly.
> > +		 */
> > +		rcmr |=	  SSC_BF(RCMR_START, fs_edge);
> > +		tcmr |=	  SSC_BF(TCMR_START, fs_edge);
> 
> Hmm, how would this work if capture and playback start/run at the same time?

Same as it did before this patch: as there is only one bi-directional link
between SSC and codec, whichever stream starts first defines the rate.

> > +	} else {
> >   		fslen = fslen_ext = 0;
> >   		rcmr_period = tcmr_period = 0;
> >   		fs_osync = SSC_FSOS_NONE;
> > -	}
> >   
> > -	rcmr |=	  SSC_BF(RCMR_START, fs_edge);
> > -	tcmr |=	  SSC_BF(TCMR_START, fs_edge);
> > +		if (ssc->lrclk_from_tf_pin) {
> > +			rcmr |=	  SSC_BF(RCMR_START, SSC_START_TX_RX);
> > +			tcmr |=	  SSC_BF(TCMR_START, fs_edge);
> > +		} else if (ssc->lrclk_from_rf_pin) {
> > +			/* assume RF is to be used when RK is used as BCLK input */
> 
> This comment is not longer true...

Removed for next version.

> 
> > +			/* Note: won't work correctly on SAMA5D2 due to errata */
> > +			rcmr |=	  SSC_BF(RCMR_START, fs_edge);
> > +			tcmr |=	  SSC_BF(TCMR_START, SSC_START_TX_RX);
> > +		} else {
> > +			rcmr |=	  SSC_BF(RCMR_START, fs_edge);
> > +			tcmr |=	  SSC_BF(TCMR_START, fs_edge);
> > +		}
> > +	}
> >   
> >   	if (atmel_ssc_cbs(ssc_p)) {
> >   		/*
> > 
> 
> Thanks and best regards,
> Codrin

Best Regards,
Michał Mirosław