linux-renesas-soc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v4 0/3] Add RZ/G2L Sound support
@ 2021-08-06 10:29 Biju Das
  2021-08-06 10:29 ` [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver Biju Das
                   ` (2 more replies)
  0 siblings, 3 replies; 11+ messages in thread
From: Biju Das @ 2021-08-06 10:29 UTC (permalink / raw)
  To: Jaroslav Kysela, Takashi Iwai, Philipp Zabel, Rob Herring
  Cc: Biju Das, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, Krzysztof Kozlowski, Laurent Pinchart,
	Lad Prabhakar, alsa-devel, Geert Uytterhoeven, Chris Paterson,
	Biju Das, linux-renesas-soc

This patch series aims to add ASoC support on RZ/G2L SoC's.

It is based on the work done by Chris Brandt for RZ/A ASoC driver.

v3->v4:
 * Updated the subject line as per style for the subsystem.
 * Removed select SND_SIMPLE_CARD from Kconfig
 * Added C++ comments for copyright and driver description.
 * Moved validation of channels in hw_params
 * removed asm issue reported by bot as well as Mark
 * replaced master/slave macros with provider/consumer macros
 * Improved locking and added more null pointer checks.
v2->v3:
 * Fixed the dependency on KCONFIG
 * Merged the binding patch with dma feature added
 * Updated dt binding example with encoded #dma-cells value.
 * Improved Error handling in probe function
 * Removed the passing legacy channel configuration parameters from
   dmaengine_slave_config function
 * started using dma_request_chan instead of deprecated
   dma_request_slave_channel
 * Removed SoC dtsi and config patches from this series. Will send it later.
v1->v2:
 * Rebased to latest rc kernel

Biju Das (3):
  ASoC: sh: Add RZ/G2L SSIF-2 driver
  ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel
    configuration parameters
  ASoC: sh: rz-ssi: Add SSI DMAC support

 .../bindings/sound/renesas,rz-ssi.yaml        |    4 +-
 sound/soc/sh/Kconfig                          |    6 +
 sound/soc/sh/Makefile                         |    4 +
 sound/soc/sh/rz-ssi.c                         | 1069 +++++++++++++++++
 4 files changed, 1081 insertions(+), 2 deletions(-)
 create mode 100644 sound/soc/sh/rz-ssi.c

-- 
2.17.1


^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver
  2021-08-06 10:29 [PATCH v4 0/3] Add RZ/G2L Sound support Biju Das
@ 2021-08-06 10:29 ` Biju Das
  2021-08-06 11:27   ` Mark Brown
  2021-08-06 10:29 ` [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters Biju Das
  2021-08-06 10:29 ` [PATCH v4 3/3] ASoC: sh: rz-ssi: Add SSI DMAC support Biju Das
  2 siblings, 1 reply; 11+ messages in thread
From: Biju Das @ 2021-08-06 10:29 UTC (permalink / raw)
  To: Jaroslav Kysela, Takashi Iwai, Philipp Zabel
  Cc: Biju Das, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, Laurent Pinchart, Randy Dunlap,
	Krzysztof Kozlowski, alsa-devel, Geert Uytterhoeven,
	Chris Paterson, Biju Das, Prabhakar Mahadev Lad,
	linux-renesas-soc

Add serial sound interface(SSIF-2) driver support for
RZ/G2L SoC.

Based on the work done by Chris Brandt for RZ/A SSI driver.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
---
v3->v4:
 * Updated the subject line as per style for the subsystem.
 * Removed select SND_SIMPLE_CARD from Kconfig
 * Added C++ comments for copyright and driver description.
 * Moved validation of channels in hw_params
 * removed asm issue reported by bot as well as Mark
 * replaced master/slave macros with provider/consumer macros
 * Improved locking and added more null pointer checks.
v2->v3:
 * Fixed the dependency on KCONFIG
 * Improved Error handling in probe function
v1->v2:
 * No change
---
 sound/soc/sh/Kconfig  |   6 +
 sound/soc/sh/Makefile |   4 +
 sound/soc/sh/rz-ssi.c | 877 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 887 insertions(+)
 create mode 100644 sound/soc/sh/rz-ssi.c

diff --git a/sound/soc/sh/Kconfig b/sound/soc/sh/Kconfig
index 346c806ba390..ae46f187cc2a 100644
--- a/sound/soc/sh/Kconfig
+++ b/sound/soc/sh/Kconfig
@@ -45,6 +45,12 @@ config SND_SOC_RCAR
 	help
 	  This option enables R-Car SRU/SCU/SSIU/SSI sound support
 
+config SND_SOC_RZ
+	tristate "RZ/G2L series SSIF-2 support"
+	depends on ARCH_R9A07G044 || COMPILE_TEST
+	help
+	  This option enables RZ/G2L SSIF-2 sound support.
+
 ##
 ## Boards
 ##
diff --git a/sound/soc/sh/Makefile b/sound/soc/sh/Makefile
index 51bd7c81671c..f6fd79948f6a 100644
--- a/sound/soc/sh/Makefile
+++ b/sound/soc/sh/Makefile
@@ -22,3 +22,7 @@ snd-soc-migor-objs		:= migor.o
 
 obj-$(CONFIG_SND_SH7760_AC97)	+= snd-soc-sh7760-ac97.o
 obj-$(CONFIG_SND_SIU_MIGOR)	+= snd-soc-migor.o
+
+# RZ/G2L
+snd-soc-rz-ssi-objs		:= rz-ssi.o
+obj-$(CONFIG_SND_SOC_RZ)	+= snd-soc-rz-ssi.o
diff --git a/sound/soc/sh/rz-ssi.c b/sound/soc/sh/rz-ssi.c
new file mode 100644
index 000000000000..0809423a76fa
--- /dev/null
+++ b/sound/soc/sh/rz-ssi.c
@@ -0,0 +1,877 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Renesas RZ/G2L ASoC Serial Sound Interface (SSIF-2) Driver
+//
+// Copyright (C) 2021 Renesas Electronics Corp.
+// Copyright (C) 2019 Chris Brandt.
+//
+
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/reset.h>
+#include <sound/soc.h>
+
+/* REGISTER OFFSET */
+#define SSICR			0x000
+#define SSISR			0x004
+#define SSIFCR			0x010
+#define SSIFSR			0x014
+#define SSIFTDR			0x018
+#define SSIFRDR			0x01c
+#define SSIOFR			0x020
+#define SSISCR			0x024
+
+/* SSI REGISTER BITS */
+#define SSICR_DWL(x)		(((x) & 0x7) << 19)
+#define SSICR_SWL(x)		(((x) & 0x7) << 16)
+#define SSICR_MST		BIT(14)
+#define SSICR_CKDV(x)		(((x) & 0xf) << 4)
+
+#define SSICR_CKS		BIT(30)
+#define SSICR_TUIEN		BIT(29)
+#define SSICR_TOIEN		BIT(28)
+#define SSICR_RUIEN		BIT(27)
+#define SSICR_ROIEN		BIT(26)
+#define SSICR_MST		BIT(14)
+#define SSICR_BCKP		BIT(13)
+#define SSICR_LRCKP		BIT(12)
+#define SSICR_CKDV(x)		(((x) & 0xf) << 4)
+#define SSICR_TEN		BIT(1)
+#define SSICR_REN		BIT(0)
+
+#define SSISR_TUIRQ		BIT(29)
+#define SSISR_TOIRQ		BIT(28)
+#define SSISR_RUIRQ		BIT(27)
+#define SSISR_ROIRQ		BIT(26)
+#define SSISR_IIRQ		BIT(25)
+
+#define SSIFCR_AUCKE		BIT(31)
+#define SSIFCR_SSIRST		BIT(16)
+#define SSIFCR_TIE		BIT(3)
+#define SSIFCR_RIE		BIT(2)
+#define SSIFCR_TFRST		BIT(1)
+#define SSIFCR_RFRST		BIT(0)
+
+#define SSIFSR_TDC_MASK		0x3f
+#define SSIFSR_TDC_SHIFT	24
+#define SSIFSR_RDC_MASK		0x3f
+#define SSIFSR_RDC_SHIFT	8
+
+#define SSIFSR_TDC(x)		(((x) & 0x1f) << 24)
+#define SSIFSR_TDE		BIT(16)
+#define SSIFSR_RDC(x)		(((x) & 0x1f) << 8)
+#define SSIFSR_RDF		BIT(0)
+
+#define SSIOFR_LRCONT		BIT(8)
+
+#define SSISCR_TDES(x)		(((x) & 0x1f) << 8)
+#define SSISCR_RDFS(x)		(((x) & 0x1f) << 0)
+
+/* Pre allocated buffers sizes */
+#define PREALLOC_BUFFER		(SZ_32K)
+#define PREALLOC_BUFFER_MAX	(SZ_32K)
+
+#define SSI_RATES		SNDRV_PCM_RATE_8000_48000 /* 8k-44.1kHz */
+#define SSI_FMTS		SNDRV_PCM_FMTBIT_S16_LE
+#define SSI_CHAN_MIN		2
+#define SSI_CHAN_MAX		2
+#define SSI_FIFO_DEPTH		32
+
+struct rz_ssi_priv;
+
+struct rz_ssi_stream {
+	struct rz_ssi_priv *priv;
+	struct snd_pcm_substream *substream;
+	int fifo_sample_size;	/* sample capacity of SSI FIFO */
+	int period_counter;	/* for keeping track of periods transferred */
+	int sample_width;
+	int buffer_pos;		/* current frame position in the buffer */
+	int running;		/* 0=stopped, 1=running */
+
+	int uerr_num;
+	int oerr_num;
+
+	int (*transfer)(struct rz_ssi_priv *ssi, struct rz_ssi_stream *strm);
+};
+
+struct rz_ssi_priv {
+	void __iomem *base;
+	struct platform_device *pdev;
+	struct reset_control *rstc;
+	struct device *dev;
+	struct clk *sfr_clk;
+	struct clk *clk;
+
+	int irq_int;
+	int irq_tx;
+	int irq_rx;
+
+	spinlock_t lock;
+
+	/*
+	 * The SSI supports full-duplex transmission and reception.
+	 * However, if an error occurs, channel reset (both transmission
+	 * and reception reset) is required.
+	 * So it is better to use as half-duplex (playing and recording
+	 * should be done on separate channels).
+	 */
+	struct rz_ssi_stream playback;
+	struct rz_ssi_stream capture;
+
+	/* clock */
+	unsigned long audio_mck;
+	unsigned long audio_clk_1;
+	unsigned long audio_clk_2;
+
+	bool lrckp_fsync_fall;	/* LR clock polarity (SSICR.LRCKP) */
+	bool bckp_rise;	/* Bit clock polarity (SSICR.BCKP) */
+};
+
+static void rz_ssi_reg_writel(struct rz_ssi_priv *priv, uint reg, u32 data)
+{
+	writel(data, (priv->base + reg));
+}
+
+static u32 rz_ssi_reg_readl(struct rz_ssi_priv *priv, uint reg)
+{
+	return readl(priv->base + reg);
+}
+
+static void rz_ssi_reg_mask_setl(struct rz_ssi_priv __iomem *priv, uint reg,
+				 u32 bclr, u32 bset)
+{
+	u32 val;
+
+	val = readl(priv->base + reg);
+	val = (val & ~bclr) | bset;
+	writel(val, (priv->base + reg));
+}
+
+static inline struct snd_soc_dai *
+rz_ssi_get_dai(struct snd_pcm_substream *substream)
+{
+	struct snd_soc_pcm_runtime *rtd = asoc_substream_to_rtd(substream);
+
+	return asoc_rtd_to_cpu(rtd, 0);
+}
+
+static inline bool rz_ssi_stream_is_play(struct rz_ssi_priv *ssi,
+					 struct snd_pcm_substream *substream)
+{
+	return substream->stream == SNDRV_PCM_STREAM_PLAYBACK;
+}
+
+static inline struct rz_ssi_stream *
+rz_ssi_stream_get(struct rz_ssi_priv *ssi, struct snd_pcm_substream *substream)
+{
+	struct rz_ssi_stream *stream = &ssi->playback;
+
+	if (substream->stream != SNDRV_PCM_STREAM_PLAYBACK)
+		stream = &ssi->capture;
+
+	return stream;
+}
+
+static int rz_ssi_stream_is_valid(struct rz_ssi_priv *ssi,
+				  struct rz_ssi_stream *strm)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&ssi->lock, flags);
+	ret = !!(strm->substream && strm->substream->runtime);
+	spin_unlock_irqrestore(&ssi->lock, flags);
+
+	return ret;
+}
+
+static int rz_ssi_stream_init(struct rz_ssi_priv *ssi,
+			      struct rz_ssi_stream *strm,
+			      struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+
+	if (runtime->sample_bits != 16) {
+		dev_err(ssi->dev, "Unsupported sample width: %d\n",
+			runtime->sample_bits);
+		return -EINVAL;
+	}
+
+	if (runtime->frame_bits != 32) {
+		dev_err(ssi->dev, "Unsupported frame width: %d\n",
+			runtime->sample_bits);
+		return -EINVAL;
+	}
+
+	strm->substream = substream;
+	strm->sample_width = samples_to_bytes(runtime, 1);
+	strm->period_counter = 0;
+	strm->buffer_pos = 0;
+
+	strm->oerr_num = 0;
+	strm->uerr_num = 0;
+	strm->running = 0;
+
+	/* fifo init */
+	strm->fifo_sample_size = SSI_FIFO_DEPTH;
+
+	return 0;
+}
+
+static void rz_ssi_stream_quit(struct rz_ssi_priv *ssi,
+			       struct rz_ssi_stream *strm)
+{
+	struct snd_soc_dai *dai = rz_ssi_get_dai(strm->substream);
+
+	strm->substream = NULL;
+
+	if (strm->oerr_num > 0)
+		dev_info(dai->dev, "overrun = %d\n", strm->oerr_num);
+
+	if (strm->uerr_num > 0)
+		dev_info(dai->dev, "underrun = %d\n", strm->uerr_num);
+}
+
+static int rz_ssi_clk_setup(struct rz_ssi_priv *ssi, unsigned int rate,
+			    unsigned int channels)
+{
+	static s8 ckdv[16] = { 1,  2,  4,  8, 16, 32, 64, 128,
+			       6, 12, 24, 48, 96, -1, -1, -1 };
+	unsigned int channel_bits = 32;	/* System Word Length */
+	unsigned long bclk_rate = rate * channels * channel_bits;
+	unsigned int div;
+	unsigned int i;
+	u32 ssicr = 0;
+	u32 clk_ckdv;
+
+	/* Clear AUCKE so we can set MST */
+	rz_ssi_reg_writel(ssi, SSIFCR, 0);
+
+	/* Continue to output LRCK pin even when idle */
+	rz_ssi_reg_writel(ssi, SSIOFR, SSIOFR_LRCONT);
+	if (ssi->audio_clk_1 && ssi->audio_clk_2) {
+		if (ssi->audio_clk_1 % bclk_rate)
+			ssi->audio_mck = ssi->audio_clk_2;
+		else
+			ssi->audio_mck = ssi->audio_clk_1;
+	}
+
+	/* Clock setting */
+	ssicr |= SSICR_MST;
+	if (ssi->audio_mck == ssi->audio_clk_1)
+		ssicr |= SSICR_CKS;
+	if (ssi->bckp_rise)
+		ssicr |= SSICR_BCKP;
+	if (ssi->lrckp_fsync_fall)
+		ssicr |= SSICR_LRCKP;
+
+	/* Determine the clock divider */
+	clk_ckdv = 0;
+	div = ssi->audio_mck / bclk_rate;
+	/* try to find an match */
+	for (i = 0; i < ARRAY_SIZE(ckdv); i++) {
+		if (ckdv[i] == div) {
+			clk_ckdv = i;
+			break;
+		}
+	}
+
+	if (i == ARRAY_SIZE(ckdv)) {
+		dev_err(ssi->dev, "Rate not divisible by audio clock source\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * DWL: Data Word Length = 16 bits
+	 * SWL: System Word Length = 32 bits
+	 */
+	ssicr |= SSICR_CKDV(clk_ckdv);
+	ssicr |= SSICR_DWL(1) | SSICR_SWL(3);
+	rz_ssi_reg_writel(ssi, SSICR, ssicr);
+	rz_ssi_reg_writel(ssi, SSIFCR,
+			  (SSIFCR_AUCKE | SSIFCR_TFRST | SSIFCR_RFRST));
+
+	return 0;
+}
+
+static int rz_ssi_start_stop(struct rz_ssi_priv *ssi,
+			     struct rz_ssi_stream *strm,
+			     int start)
+{
+	struct snd_pcm_substream *substream = strm->substream;
+	u32 ssicr, ssifcr;
+	int timeout;
+
+	if (start) {
+		bool is_play = rz_ssi_stream_is_play(ssi, substream);
+
+		ssicr = rz_ssi_reg_readl(ssi, SSICR);
+		ssifcr = rz_ssi_reg_readl(ssi, SSIFCR) & ~0xF;
+
+		/* FIFO interrupt thresholds */
+		rz_ssi_reg_writel(ssi, SSISCR,
+				  SSISCR_TDES(strm->fifo_sample_size / 2 - 1) |
+				  SSISCR_RDFS(0));
+
+		/* enable IRQ */
+		if (is_play) {
+			ssicr |= SSICR_TUIEN | SSICR_TOIEN;
+			ssifcr |= SSIFCR_TIE | SSIFCR_RFRST;
+		} else {
+			ssicr |= SSICR_RUIEN | SSICR_ROIEN;
+			ssifcr |= SSIFCR_RIE | SSIFCR_TFRST;
+		}
+
+		rz_ssi_reg_writel(ssi, SSICR, ssicr);
+		rz_ssi_reg_writel(ssi, SSIFCR, ssifcr);
+
+		/* Clear all error flags */
+		rz_ssi_reg_mask_setl(ssi, SSISR,
+				     (SSISR_TOIRQ | SSISR_TUIRQ | SSISR_ROIRQ |
+				      SSISR_RUIRQ), 0);
+
+		strm->running = 1;
+		ssicr |= is_play ? SSICR_TEN : SSICR_REN;
+		rz_ssi_reg_writel(ssi, SSICR, ssicr);
+
+	} else {
+		strm->running = 0;
+
+		/* Disable TX/RX */
+		rz_ssi_reg_mask_setl(ssi, SSICR, SSICR_TEN | SSICR_REN, 0);
+
+		/* Disable irqs */
+		rz_ssi_reg_mask_setl(ssi, SSICR, SSICR_TUIEN | SSICR_TOIEN |
+				     SSICR_RUIEN | SSICR_ROIEN, 0);
+		rz_ssi_reg_mask_setl(ssi, SSIFCR, SSIFCR_TIE | SSIFCR_RIE, 0);
+
+		/* Clear all error flags */
+		rz_ssi_reg_mask_setl(ssi, SSISR,
+				     (SSISR_TOIRQ | SSISR_TUIRQ | SSISR_ROIRQ |
+				      SSISR_RUIRQ), 0);
+
+		/* Wait for idle */
+		timeout = 100;
+		while (--timeout) {
+			if (rz_ssi_reg_readl(ssi, SSISR) | SSISR_IIRQ)
+				break;
+			udelay(1);
+		}
+
+		if (!timeout)
+			dev_info(ssi->dev, "timeout waiting for SSI idle\n");
+
+		/* Hold FIFOs in reset */
+		rz_ssi_reg_mask_setl(ssi, SSIFCR, 0,
+				     SSIFCR_TFRST | SSIFCR_RFRST);
+	}
+
+	return 0;
+}
+
+static void rz_ssi_pointer_update(struct rz_ssi_stream *strm, int frames)
+{
+	struct snd_pcm_substream *substream = strm->substream;
+	struct snd_pcm_runtime *runtime;
+	int current_period;
+
+	if (!strm->running || !substream || !substream->runtime)
+		return;
+
+	runtime = substream->runtime;
+	strm->buffer_pos += frames;
+	WARN_ON(strm->buffer_pos > runtime->buffer_size);
+
+	/* ring buffer */
+	if (strm->buffer_pos == runtime->buffer_size)
+		strm->buffer_pos = 0;
+
+	current_period = strm->buffer_pos / runtime->period_size;
+	if (strm->period_counter != current_period) {
+		snd_pcm_period_elapsed(strm->substream);
+		strm->period_counter = current_period;
+	}
+}
+
+static int rz_ssi_pio_recv(struct rz_ssi_priv *ssi, struct rz_ssi_stream *strm)
+{
+	struct snd_pcm_substream *substream = strm->substream;
+	struct snd_pcm_runtime *runtime;
+	u16 *buf;
+	int fifo_samples;
+	int frames_left;
+	int samples = 0;
+	int i;
+
+	if (!rz_ssi_stream_is_valid(ssi, strm))
+		return -EINVAL;
+
+	runtime = substream->runtime;
+	/* frames left in this period */
+	frames_left = runtime->period_size - (strm->buffer_pos %
+					      runtime->period_size);
+	if (frames_left == 0)
+		frames_left = runtime->period_size;
+
+	/* Samples in RX FIFO */
+	fifo_samples = (rz_ssi_reg_readl(ssi, SSIFSR) >>
+			SSIFSR_RDC_SHIFT) & SSIFSR_RDC_MASK;
+
+	/* Only read full frames at a time */
+	while (frames_left && (fifo_samples >= runtime->channels)) {
+		samples += runtime->channels;
+		fifo_samples -= runtime->channels;
+		frames_left--;
+	}
+
+	/* not enough samples yet */
+	if (samples == 0)
+		return 0;
+
+	/* calculate new buffer index */
+	buf = (u16 *)(runtime->dma_area);
+	buf += strm->buffer_pos * runtime->channels;
+
+	/* Note, only supports 16-bit samples */
+	for (i = 0; i < samples; i++)
+		*buf++ = (u16)(rz_ssi_reg_readl(ssi, SSIFRDR) >> 16);
+
+	rz_ssi_reg_mask_setl(ssi, SSIFSR, SSIFSR_RDF, 0);
+	rz_ssi_pointer_update(strm, samples / runtime->channels);
+
+	/*
+	 * If we finished this period, but there are more samples in
+	 * the RX FIFO, call this function again
+	 */
+	if (frames_left == 0 && fifo_samples >= runtime->channels)
+		rz_ssi_pio_recv(ssi, strm);
+
+	return 0;
+}
+
+static int rz_ssi_pio_send(struct rz_ssi_priv *ssi, struct rz_ssi_stream *strm)
+{
+	struct snd_pcm_substream *substream = strm->substream;
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	int sample_space;
+	int samples = 0;
+	int frames_left;
+	int i;
+	u32 ssifsr;
+	u16 *buf;
+
+	if (!rz_ssi_stream_is_valid(ssi, strm))
+		return -EINVAL;
+
+	/* frames left in this period */
+	frames_left = runtime->period_size - (strm->buffer_pos %
+					      runtime->period_size);
+	if (frames_left == 0)
+		frames_left = runtime->period_size;
+
+	sample_space = strm->fifo_sample_size;
+	ssifsr = rz_ssi_reg_readl(ssi, SSIFSR);
+	sample_space -= (ssifsr >> SSIFSR_TDC_SHIFT) & SSIFSR_TDC_MASK;
+
+	/* Only add full frames at a time */
+	while (frames_left && (sample_space >= runtime->channels)) {
+		samples += runtime->channels;
+		sample_space -= runtime->channels;
+		frames_left--;
+	}
+
+	/* no space to send anything right now */
+	if (samples == 0)
+		return 0;
+
+	/* calculate new buffer index */
+	buf = (u16 *)(runtime->dma_area);
+	buf += strm->buffer_pos * runtime->channels;
+
+	/* Note, only supports 16-bit samples */
+	for (i = 0; i < samples; i++)
+		rz_ssi_reg_writel(ssi, SSIFTDR, ((u32)(*buf++) << 16));
+
+	rz_ssi_reg_mask_setl(ssi, SSIFSR, SSIFSR_TDE, 0);
+	rz_ssi_pointer_update(strm, samples / runtime->channels);
+
+	return 0;
+}
+
+static irqreturn_t rz_ssi_interrupt(int irq, void *data)
+{
+	struct rz_ssi_stream *strm = NULL;
+	struct rz_ssi_priv *ssi = data;
+	u32 ssisr = rz_ssi_reg_readl(ssi, SSISR);
+
+	if (ssi->playback.substream)
+		strm = &ssi->playback;
+	else if (ssi->capture.substream)
+		strm = &ssi->capture;
+	else
+		return IRQ_HANDLED; /* Left over TX/RX interrupt */
+
+	if (irq == ssi->irq_int) { /* error or idle */
+		if (ssisr & SSISR_TUIRQ)
+			strm->uerr_num++;
+		if (ssisr & SSISR_TOIRQ)
+			strm->oerr_num++;
+		if (ssisr & SSISR_RUIRQ)
+			strm->uerr_num++;
+		if (ssisr & SSISR_ROIRQ)
+			strm->oerr_num++;
+
+		if (ssisr & (SSISR_TUIRQ | SSISR_TOIRQ | SSISR_RUIRQ |
+			     SSISR_ROIRQ)) {
+			/* Error handling */
+			/* You must reset (stop/restart) after each interrupt */
+			rz_ssi_start_stop(ssi, strm, 0);
+
+			/* Clear all flags */
+			rz_ssi_reg_mask_setl(ssi, SSISR, SSISR_TOIRQ |
+					     SSISR_TUIRQ | SSISR_ROIRQ |
+					     SSISR_RUIRQ, 0);
+
+			/* Add/remove more data */
+			strm->transfer(ssi, strm);
+
+			/* Resume */
+			rz_ssi_start_stop(ssi, strm, 1);
+		}
+	}
+
+	if (!strm->running)
+		return IRQ_HANDLED;
+
+	/* tx data empty */
+	if (irq == ssi->irq_tx)
+		strm->transfer(ssi, &ssi->playback);
+
+	/* rx data full */
+	if (irq == ssi->irq_rx) {
+		strm->transfer(ssi, &ssi->capture);
+		rz_ssi_reg_mask_setl(ssi, SSIFSR, SSIFSR_RDF, 0);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int rz_ssi_dai_trigger(struct snd_pcm_substream *substream, int cmd,
+			      struct snd_soc_dai *dai)
+{
+	struct rz_ssi_priv *ssi = snd_soc_dai_get_drvdata(dai);
+	struct rz_ssi_stream *strm = rz_ssi_stream_get(ssi, substream);
+	int ret = 0;
+	unsigned long flags;
+
+	switch (cmd) {
+	case SNDRV_PCM_TRIGGER_START:
+		/* Soft Reset */
+		rz_ssi_reg_mask_setl(ssi, SSIFCR, 0, SSIFCR_SSIRST);
+		rz_ssi_reg_mask_setl(ssi, SSIFCR, SSIFCR_SSIRST, 0);
+		udelay(5);
+
+		spin_lock_irqsave(&ssi->lock, flags);
+		ret = rz_ssi_stream_init(ssi, strm, substream);
+		spin_unlock_irqrestore(&ssi->lock, flags);
+		if (ret)
+			goto done;
+
+		ret = strm->transfer(ssi, strm);
+		if (ret)
+			goto done;
+
+		ret = rz_ssi_start_stop(ssi, strm, 1);
+		break;
+	case SNDRV_PCM_TRIGGER_STOP:
+		rz_ssi_start_stop(ssi, strm, 0);
+		spin_lock_irqsave(&ssi->lock, flags);
+		rz_ssi_stream_quit(ssi, strm);
+		spin_unlock_irqrestore(&ssi->lock, flags);
+		break;
+	}
+
+done:
+	return ret;
+}
+
+static int rz_ssi_dai_set_fmt(struct snd_soc_dai *dai, unsigned int fmt)
+{
+	struct rz_ssi_priv *ssi = snd_soc_dai_get_drvdata(dai);
+
+	switch (fmt & SND_SOC_DAIFMT_CLOCK_PROVIDER_MASK) {
+	case SND_SOC_DAIFMT_CBC_CFC:
+		break;
+	default:
+		dev_err(ssi->dev, "Codec should be clk and frame consumer\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * set clock polarity
+	 *
+	 * "normal" BCLK = Signal is available at rising edge of BCLK
+	 * "normal" FSYNC = (I2S) Left ch starts with falling FSYNC edge
+	 */
+	switch (fmt & SND_SOC_DAIFMT_INV_MASK) {
+	case SND_SOC_DAIFMT_NB_NF:
+		ssi->bckp_rise = false;
+		ssi->lrckp_fsync_fall = false;
+		break;
+	case SND_SOC_DAIFMT_NB_IF:
+		ssi->bckp_rise = false;
+		ssi->lrckp_fsync_fall = true;
+		break;
+	case SND_SOC_DAIFMT_IB_NF:
+		ssi->bckp_rise = true;
+		ssi->lrckp_fsync_fall = false;
+		break;
+	case SND_SOC_DAIFMT_IB_IF:
+		ssi->bckp_rise = true;
+		ssi->lrckp_fsync_fall = true;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* only i2s support */
+	switch (fmt & SND_SOC_DAIFMT_FORMAT_MASK) {
+	case SND_SOC_DAIFMT_I2S:
+		break;
+	default:
+		dev_err(ssi->dev, "Only I2S mode is supported.\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int rz_ssi_dai_hw_params(struct snd_pcm_substream *substream,
+				struct snd_pcm_hw_params *params,
+				struct snd_soc_dai *dai)
+{
+	struct rz_ssi_priv *ssi = snd_soc_dai_get_drvdata(dai);
+	int channels = params_channels(params);
+
+	if (channels != 2) {
+		dev_err(ssi->dev, "Number of channels not matched\n");
+		return -EINVAL;
+	}
+
+	return rz_ssi_clk_setup(ssi, params_rate(params),
+				params_channels(params));
+}
+
+static const struct snd_soc_dai_ops rz_ssi_dai_ops = {
+	.trigger	= rz_ssi_dai_trigger,
+	.set_fmt	= rz_ssi_dai_set_fmt,
+	.hw_params	= rz_ssi_dai_hw_params,
+};
+
+static const struct snd_pcm_hardware rz_ssi_pcm_hardware = {
+	.info			= SNDRV_PCM_INFO_INTERLEAVED	|
+				  SNDRV_PCM_INFO_MMAP		|
+				  SNDRV_PCM_INFO_MMAP_VALID,
+	.buffer_bytes_max	= PREALLOC_BUFFER,
+	.period_bytes_min	= 32,
+	.period_bytes_max	= 8192,
+	.channels_min		= SSI_CHAN_MIN,
+	.channels_max		= SSI_CHAN_MAX,
+	.periods_min		= 1,
+	.periods_max		= 32,
+	.fifo_size		= 32 * 2,
+};
+
+static int rz_ssi_pcm_open(struct snd_soc_component *component,
+			   struct snd_pcm_substream *substream)
+{
+	snd_soc_set_runtime_hwparams(substream, &rz_ssi_pcm_hardware);
+
+	return snd_pcm_hw_constraint_integer(substream->runtime,
+					    SNDRV_PCM_HW_PARAM_PERIODS);
+}
+
+static snd_pcm_uframes_t rz_ssi_pcm_pointer(struct snd_soc_component *component,
+					    struct snd_pcm_substream *substream)
+{
+	struct snd_soc_dai *dai = rz_ssi_get_dai(substream);
+	struct rz_ssi_priv *ssi = snd_soc_dai_get_drvdata(dai);
+	struct rz_ssi_stream *strm = rz_ssi_stream_get(ssi, substream);
+
+	return strm->buffer_pos;
+}
+
+static int rz_ssi_pcm_new(struct snd_soc_component *component,
+			  struct snd_soc_pcm_runtime *rtd)
+{
+	snd_pcm_set_managed_buffer_all(rtd->pcm, SNDRV_DMA_TYPE_DEV,
+				       rtd->card->snd_card->dev,
+				       PREALLOC_BUFFER, PREALLOC_BUFFER_MAX);
+	return 0;
+}
+
+static struct snd_soc_dai_driver rz_ssi_soc_dai[] = {
+	{
+		.name			= "rz-ssi-dai",
+		.playback = {
+			.rates		= SSI_RATES,
+			.formats	= SSI_FMTS,
+			.channels_min	= SSI_CHAN_MIN,
+			.channels_max	= SSI_CHAN_MAX,
+		},
+		.capture = {
+			.rates		= SSI_RATES,
+			.formats	= SSI_FMTS,
+			.channels_min	= SSI_CHAN_MIN,
+			.channels_max	= SSI_CHAN_MAX,
+		},
+		.ops = &rz_ssi_dai_ops,
+	},
+};
+
+static const struct snd_soc_component_driver rz_ssi_soc_component = {
+	.name		= "rz-ssi",
+	.open		= rz_ssi_pcm_open,
+	.pointer	= rz_ssi_pcm_pointer,
+	.pcm_construct	= rz_ssi_pcm_new,
+};
+
+static int rz_ssi_probe(struct platform_device *pdev)
+{
+	struct rz_ssi_priv *ssi;
+	struct clk *audio_clk;
+	int ret;
+
+	ssi = devm_kzalloc(&pdev->dev, sizeof(*ssi), GFP_KERNEL);
+	if (!ssi)
+		return -ENOMEM;
+
+	ssi->pdev = pdev;
+	ssi->dev = &pdev->dev;
+	ssi->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(ssi->base))
+		return PTR_ERR(ssi->base);
+
+	ssi->clk = devm_clk_get(&pdev->dev, "ssi");
+	if (IS_ERR(ssi->clk))
+		return PTR_ERR(ssi->clk);
+
+	ssi->sfr_clk = devm_clk_get(&pdev->dev, "ssi_sfr");
+	if (IS_ERR(ssi->sfr_clk))
+		return PTR_ERR(ssi->sfr_clk);
+
+	audio_clk = devm_clk_get(&pdev->dev, "audio_clk1");
+	if (IS_ERR(audio_clk))
+		return dev_err_probe(&pdev->dev, PTR_ERR(audio_clk),
+				     "no audio clk1");
+
+	ssi->audio_clk_1 = clk_get_rate(audio_clk);
+	audio_clk = devm_clk_get(&pdev->dev, "audio_clk2");
+	if (IS_ERR(audio_clk))
+		return dev_err_probe(&pdev->dev, PTR_ERR(audio_clk),
+				     "no audio clk2");
+
+	ssi->audio_clk_2 = clk_get_rate(audio_clk);
+	if (!(ssi->audio_clk_1 || ssi->audio_clk_2))
+		return dev_err_probe(&pdev->dev, -EINVAL,
+				     "no audio clk1 or audio clk2");
+
+	ssi->audio_mck = ssi->audio_clk_1 ? ssi->audio_clk_1 : ssi->audio_clk_2;
+
+	ssi->playback.transfer = rz_ssi_pio_send;
+	ssi->capture.transfer = rz_ssi_pio_recv;
+	ssi->playback.priv = ssi;
+	ssi->capture.priv = ssi;
+
+	/* Error Interrupt */
+	ssi->irq_int = platform_get_irq_byname(pdev, "int_req");
+	if (ssi->irq_int < 0)
+		return dev_err_probe(&pdev->dev, -ENODEV,
+				     "Unable to get SSI int_req IRQ\n");
+
+	ret = devm_request_irq(&pdev->dev, ssi->irq_int, &rz_ssi_interrupt,
+			       0, dev_name(&pdev->dev), ssi);
+	if (ret < 0)
+		return dev_err_probe(&pdev->dev, ret,
+				     "irq request error (int_req)\n");
+
+	/* Tx and Rx interrupts (pio only) */
+	ssi->irq_tx = platform_get_irq_byname(pdev, "dma_tx");
+	if (ssi->irq_tx < 0)
+		return dev_err_probe(&pdev->dev, -ENODEV,
+				     "Unable to get SSI dma_tx IRQ\n");
+
+	ret = devm_request_irq(&pdev->dev, ssi->irq_tx, &rz_ssi_interrupt, 0,
+			       dev_name(&pdev->dev), ssi);
+	if (ret < 0)
+		return dev_err_probe(&pdev->dev, ret,
+				     "irq request error (dma_tx)\n");
+
+	ssi->irq_rx = platform_get_irq_byname(pdev, "dma_rx");
+	if (ssi->irq_rx < 0)
+		return dev_err_probe(&pdev->dev, -ENODEV,
+				     "Unable to get SSI dma_rx IRQ\n");
+
+	ret = devm_request_irq(&pdev->dev, ssi->irq_rx, &rz_ssi_interrupt, 0,
+			       dev_name(&pdev->dev), ssi);
+	if (ret < 0)
+		return dev_err_probe(&pdev->dev, ret,
+				     "irq request error (dma_rx)\n");
+
+	ssi->rstc = devm_reset_control_get_exclusive(&pdev->dev, NULL);
+	if (IS_ERR(ssi->rstc))
+		return PTR_ERR(ssi->rstc);
+
+	reset_control_deassert(ssi->rstc);
+	pm_runtime_enable(&pdev->dev);
+	pm_runtime_resume_and_get(&pdev->dev);
+
+	spin_lock_init(&ssi->lock);
+	dev_set_drvdata(&pdev->dev, ssi);
+	ret = devm_snd_soc_register_component(&pdev->dev, &rz_ssi_soc_component,
+					      rz_ssi_soc_dai,
+					      ARRAY_SIZE(rz_ssi_soc_dai));
+	if (ret < 0) {
+		pm_runtime_put(ssi->dev);
+		pm_runtime_disable(ssi->dev);
+		reset_control_assert(ssi->rstc);
+		dev_err(&pdev->dev, "failed to register snd component\n");
+	}
+
+	return ret;
+}
+
+static int rz_ssi_remove(struct platform_device *pdev)
+{
+	struct rz_ssi_priv *ssi = dev_get_drvdata(&pdev->dev);
+
+	pm_runtime_put(ssi->dev);
+	pm_runtime_disable(ssi->dev);
+	reset_control_assert(ssi->rstc);
+
+	return 0;
+}
+
+static const struct of_device_id rz_ssi_of_match[] = {
+	{ .compatible = "renesas,rz-ssi", },
+	{/* Sentinel */},
+};
+MODULE_DEVICE_TABLE(of, rz_ssi_of_match);
+
+static struct platform_driver rz_ssi_driver = {
+	.driver	= {
+		.name	= "rz-ssi-pcm-audio",
+		.of_match_table = rz_ssi_of_match,
+	},
+	.probe		= rz_ssi_probe,
+	.remove		= rz_ssi_remove,
+};
+
+module_platform_driver(rz_ssi_driver);
+
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("Renesas RZ/G2L ASoC Serial Sound Interface Driver");
+MODULE_AUTHOR("Biju Das <biju.das.jz@bp.renesas.com>");
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters
  2021-08-06 10:29 [PATCH v4 0/3] Add RZ/G2L Sound support Biju Das
  2021-08-06 10:29 ` [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver Biju Das
@ 2021-08-06 10:29 ` Biju Das
  2021-08-09 12:57   ` Geert Uytterhoeven
  2021-08-13 20:13   ` Rob Herring
  2021-08-06 10:29 ` [PATCH v4 3/3] ASoC: sh: rz-ssi: Add SSI DMAC support Biju Das
  2 siblings, 2 replies; 11+ messages in thread
From: Biju Das @ 2021-08-06 10:29 UTC (permalink / raw)
  To: Rob Herring
  Cc: Biju Das, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, alsa-devel, devicetree, Geert Uytterhoeven,
	Chris Paterson, Biju Das, Prabhakar Mahadev Lad,
	linux-renesas-soc

The DMAC driver on RZ/G2L expects the slave channel configuration
to be passed in dmas property.
This patch updates the example node to include the encoded slave
channel configuration.

Fixes: 5df6dfbb6de815ba3a("ASoC: dt-bindings: sound: renesas,rz-ssi: Document DMA support")
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
---
Note:-
 This patch is based on [1]
 [1]:- https://patchwork.kernel.org/project/linux-renesas-soc/patch/20210806095322.2326-2-biju.das.jz@bp.renesas.com/

v3->v4:
 * Updated bindings as the DMAC driver on RZ/G2L expects the
   slave channel configuration to be passed in dmas property.
v2->v3:
 * Merged the binding patch with dmas added
 * Updated dt binding example with encoded #dma-cells value.
v1->v2:
 * Rebased on 5.14-rc2.
---
 Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml b/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
index 471937cb8d05..9923e7b8264f 100644
--- a/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
+++ b/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
@@ -93,8 +93,8 @@ examples:
             clock-names = "ssi", "ssi_sfr", "audio_clk1", "audio_clk2";
             power-domains = <&cpg>;
             resets = <&cpg R9A07G044_SSI0_RST_M2_REG>;
-            dmas = <&dmac 0x255>,
-                   <&dmac 0x256>;
+            dmas = <&dmac 0x2655>,
+                   <&dmac 0x2656>;
             dma-names = "tx", "rx";
             #sound-dai-cells = <0>;
     };
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v4 3/3] ASoC: sh: rz-ssi: Add SSI DMAC support
  2021-08-06 10:29 [PATCH v4 0/3] Add RZ/G2L Sound support Biju Das
  2021-08-06 10:29 ` [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver Biju Das
  2021-08-06 10:29 ` [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters Biju Das
@ 2021-08-06 10:29 ` Biju Das
  2 siblings, 0 replies; 11+ messages in thread
From: Biju Das @ 2021-08-06 10:29 UTC (permalink / raw)
  To: Jaroslav Kysela, Takashi Iwai, Philipp Zabel
  Cc: Biju Das, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, Laurent Pinchart, Randy Dunlap,
	Krzysztof Kozlowski, alsa-devel, Geert Uytterhoeven,
	Chris Paterson, Biju Das, Prabhakar Mahadev Lad,
	linux-renesas-soc

Add SSI DMAC support to RZ/G2L SoC.

Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
---
Note:-
 This patch has run time dependency on [1]
 [1] https://patchwork.kernel.org/project/linux-renesas-soc/patch/20210806095322.2326-4-biju.das.jz@bp.renesas.com/

v3->v4:
 * Updated the subject line as per style for the subsystem.
 * started using dmaengine_terminate_async instead of 
   dmaengine_terminate_sync as it is called from atomic context.
v2->v3:
 * Removed the passing legacy channel configuration parameters from dmaengine_slave_config function
 * started using dma_request_chan instead of deprecated dma_request_slave_channel
v1->v2:
 * No change
---
 sound/soc/sh/rz-ssi.c | 256 ++++++++++++++++++++++++++++++++++++------
 1 file changed, 224 insertions(+), 32 deletions(-)

diff --git a/sound/soc/sh/rz-ssi.c b/sound/soc/sh/rz-ssi.c
index 0809423a76fa..d1cde9d9207e 100644
--- a/sound/soc/sh/rz-ssi.c
+++ b/sound/soc/sh/rz-ssi.c
@@ -7,6 +7,7 @@
 //
 
 #include <linux/clk.h>
+#include <linux/dmaengine.h>
 #include <linux/io.h>
 #include <linux/module.h>
 #include <linux/of_device.h>
@@ -86,6 +87,7 @@ struct rz_ssi_stream {
 	struct rz_ssi_priv *priv;
 	struct snd_pcm_substream *substream;
 	int fifo_sample_size;	/* sample capacity of SSI FIFO */
+	int dma_buffer_pos;	/* The address for the next DMA descriptor */
 	int period_counter;	/* for keeping track of periods transferred */
 	int sample_width;
 	int buffer_pos;		/* current frame position in the buffer */
@@ -94,6 +96,8 @@ struct rz_ssi_stream {
 	int uerr_num;
 	int oerr_num;
 
+	struct dma_chan *dma_ch;
+
 	int (*transfer)(struct rz_ssi_priv *ssi, struct rz_ssi_stream *strm);
 };
 
@@ -105,6 +109,7 @@ struct rz_ssi_priv {
 	struct clk *sfr_clk;
 	struct clk *clk;
 
+	phys_addr_t phys;
 	int irq_int;
 	int irq_tx;
 	int irq_rx;
@@ -128,8 +133,11 @@ struct rz_ssi_priv {
 
 	bool lrckp_fsync_fall;	/* LR clock polarity (SSICR.LRCKP) */
 	bool bckp_rise;	/* Bit clock polarity (SSICR.BCKP) */
+	bool dma_rt;
 };
 
+static void rz_ssi_dma_complete(void *data);
+
 static void rz_ssi_reg_writel(struct rz_ssi_priv *priv, uint reg, u32 data)
 {
 	writel(data, (priv->base + reg));
@@ -175,6 +183,11 @@ rz_ssi_stream_get(struct rz_ssi_priv *ssi, struct snd_pcm_substream *substream)
 	return stream;
 }
 
+static inline bool rz_ssi_is_dma_enabled(struct rz_ssi_priv *ssi)
+{
+	return (ssi->playback.dma_ch || ssi->capture.dma_ch);
+}
+
 static int rz_ssi_stream_is_valid(struct rz_ssi_priv *ssi,
 				  struct rz_ssi_stream *strm)
 {
@@ -208,6 +221,7 @@ static int rz_ssi_stream_init(struct rz_ssi_priv *ssi,
 
 	strm->substream = substream;
 	strm->sample_width = samples_to_bytes(runtime, 1);
+	strm->dma_buffer_pos = 0;
 	strm->period_counter = 0;
 	strm->buffer_pos = 0;
 
@@ -312,9 +326,12 @@ static int rz_ssi_start_stop(struct rz_ssi_priv *ssi,
 		ssifcr = rz_ssi_reg_readl(ssi, SSIFCR) & ~0xF;
 
 		/* FIFO interrupt thresholds */
-		rz_ssi_reg_writel(ssi, SSISCR,
-				  SSISCR_TDES(strm->fifo_sample_size / 2 - 1) |
-				  SSISCR_RDFS(0));
+		if (rz_ssi_is_dma_enabled(ssi))
+			rz_ssi_reg_writel(ssi, SSISCR, 0);
+		else
+			rz_ssi_reg_writel(ssi, SSISCR,
+					  SSISCR_TDES(strm->fifo_sample_size / 2 - 1) |
+					  SSISCR_RDFS(0));
 
 		/* enable IRQ */
 		if (is_play) {
@@ -343,6 +360,10 @@ static int rz_ssi_start_stop(struct rz_ssi_priv *ssi,
 		/* Disable TX/RX */
 		rz_ssi_reg_mask_setl(ssi, SSICR, SSICR_TEN | SSICR_REN, 0);
 
+		/* Cancel all remaining DMA transactions */
+		if (rz_ssi_is_dma_enabled(ssi))
+			dmaengine_terminate_async(strm->dma_ch);
+
 		/* Disable irqs */
 		rz_ssi_reg_mask_setl(ssi, SSICR, SSICR_TUIEN | SSICR_TOIEN |
 				     SSICR_RUIEN | SSICR_ROIEN, 0);
@@ -559,12 +580,143 @@ static irqreturn_t rz_ssi_interrupt(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
+static int rz_ssi_dma_slave_config(struct rz_ssi_priv *ssi,
+				   struct dma_chan *dma_ch, bool is_play)
+{
+	struct dma_slave_config cfg;
+
+	memset(&cfg, 0, sizeof(cfg));
+
+	cfg.direction = is_play ? DMA_MEM_TO_DEV : DMA_DEV_TO_MEM;
+	cfg.dst_addr = ssi->phys + SSIFTDR;
+	cfg.src_addr = ssi->phys + SSIFRDR;
+	cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+	cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+
+	return dmaengine_slave_config(dma_ch, &cfg);
+}
+
+static int rz_ssi_dma_transfer(struct rz_ssi_priv *ssi,
+			       struct rz_ssi_stream *strm)
+{
+	struct snd_pcm_substream *substream = strm->substream;
+	struct dma_async_tx_descriptor *desc;
+	struct snd_pcm_runtime *runtime;
+	enum dma_transfer_direction dir;
+	u32 dma_paddr, dma_size;
+	int amount;
+
+	if (!rz_ssi_stream_is_valid(ssi, strm))
+		return -EINVAL;
+
+	runtime = substream->runtime;
+	if (runtime->status->state == SNDRV_PCM_STATE_DRAINING)
+		/*
+		 * Stream is ending, so do not queue up any more DMA
+		 * transfers otherwise we play partial sound clips
+		 * because we can't shut off the DMA quick enough.
+		 */
+		return 0;
+
+	dir = rz_ssi_stream_is_play(ssi, substream) ? DMA_MEM_TO_DEV : DMA_DEV_TO_MEM;
+
+	/* Always transfer 1 period */
+	amount = runtime->period_size;
+
+	/* DMA physical address and size */
+	dma_paddr = runtime->dma_addr + frames_to_bytes(runtime,
+							strm->dma_buffer_pos);
+	dma_size = frames_to_bytes(runtime, amount);
+	desc = dmaengine_prep_slave_single(strm->dma_ch, dma_paddr, dma_size,
+					   dir,
+					   DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!desc) {
+		dev_err(ssi->dev, "dmaengine_prep_slave_single() fail\n");
+		return -ENOMEM;
+	}
+
+	desc->callback = rz_ssi_dma_complete;
+	desc->callback_param = strm;
+
+	if (dmaengine_submit(desc) < 0) {
+		dev_err(ssi->dev, "dmaengine_submit() fail\n");
+		return -EIO;
+	}
+
+	/* Update DMA pointer */
+	strm->dma_buffer_pos += amount;
+	if (strm->dma_buffer_pos >= runtime->buffer_size)
+		strm->dma_buffer_pos = 0;
+
+	/* Start DMA */
+	dma_async_issue_pending(strm->dma_ch);
+
+	return 0;
+}
+
+static void rz_ssi_dma_complete(void *data)
+{
+	struct rz_ssi_stream *strm = (struct rz_ssi_stream *)data;
+
+	if (!strm->running || !strm->substream || !strm->substream->runtime)
+		return;
+
+	/* Note that next DMA transaction has probably already started */
+	rz_ssi_pointer_update(strm, strm->substream->runtime->period_size);
+
+	/* Queue up another DMA transaction */
+	rz_ssi_dma_transfer(strm->priv, strm);
+}
+
+static void rz_ssi_release_dma_channels(struct rz_ssi_priv *ssi)
+{
+	if (ssi->playback.dma_ch) {
+		dma_release_channel(ssi->playback.dma_ch);
+		ssi->playback.dma_ch = NULL;
+		if (ssi->dma_rt)
+			ssi->dma_rt = false;
+	}
+
+	if (ssi->capture.dma_ch) {
+		dma_release_channel(ssi->capture.dma_ch);
+		ssi->capture.dma_ch = NULL;
+	}
+}
+
+static int rz_ssi_dma_request(struct rz_ssi_priv *ssi, struct device *dev)
+{
+	ssi->playback.dma_ch = dma_request_chan(dev, "tx");
+	ssi->capture.dma_ch = dma_request_chan(dev, "rx");
+	if (!ssi->playback.dma_ch && !ssi->capture.dma_ch) {
+		ssi->playback.dma_ch = dma_request_chan(dev, "rt");
+		if (!ssi->playback.dma_ch)
+			goto no_dma;
+
+		ssi->dma_rt = true;
+	}
+
+	if (ssi->playback.dma_ch &&
+	    (rz_ssi_dma_slave_config(ssi, ssi->playback.dma_ch, true) < 0))
+		goto no_dma;
+
+	if (ssi->capture.dma_ch &&
+	    (rz_ssi_dma_slave_config(ssi, ssi->capture.dma_ch, false) < 0))
+		goto no_dma;
+
+	return 0;
+
+no_dma:
+	rz_ssi_release_dma_channels(ssi);
+
+	return -ENODEV;
+}
+
 static int rz_ssi_dai_trigger(struct snd_pcm_substream *substream, int cmd,
 			      struct snd_soc_dai *dai)
 {
 	struct rz_ssi_priv *ssi = snd_soc_dai_get_drvdata(dai);
 	struct rz_ssi_stream *strm = rz_ssi_stream_get(ssi, substream);
-	int ret = 0;
+	int ret = 0, i, num_transfer = 1;
 	unsigned long flags;
 
 	switch (cmd) {
@@ -580,9 +732,29 @@ static int rz_ssi_dai_trigger(struct snd_pcm_substream *substream, int cmd,
 		if (ret)
 			goto done;
 
-		ret = strm->transfer(ssi, strm);
-		if (ret)
-			goto done;
+		if (ssi->dma_rt) {
+			bool is_playback;
+
+			is_playback = rz_ssi_stream_is_play(ssi, substream);
+			ret = rz_ssi_dma_slave_config(ssi, ssi->playback.dma_ch,
+						      is_playback);
+			/* Fallback to pio */
+			if (ret < 0) {
+				ssi->playback.transfer = rz_ssi_pio_send;
+				ssi->capture.transfer = rz_ssi_pio_recv;
+				rz_ssi_release_dma_channels(ssi);
+			}
+		}
+
+		/* For DMA, queue up multiple DMA descriptors */
+		if (rz_ssi_is_dma_enabled(ssi))
+			num_transfer = 4;
+
+		for (i = 0; i < num_transfer; i++) {
+			ret = strm->transfer(ssi, strm);
+			if (ret)
+				goto done;
+		}
 
 		ret = rz_ssi_start_stop(ssi, strm, 1);
 		break;
@@ -743,6 +915,7 @@ static int rz_ssi_probe(struct platform_device *pdev)
 {
 	struct rz_ssi_priv *ssi;
 	struct clk *audio_clk;
+	struct resource *res;
 	int ret;
 
 	ssi = devm_kzalloc(&pdev->dev, sizeof(*ssi), GFP_KERNEL);
@@ -751,10 +924,11 @@ static int rz_ssi_probe(struct platform_device *pdev)
 
 	ssi->pdev = pdev;
 	ssi->dev = &pdev->dev;
-	ssi->base = devm_platform_ioremap_resource(pdev, 0);
+	ssi->base = devm_platform_get_and_ioremap_resource(pdev, 0, &res);
 	if (IS_ERR(ssi->base))
 		return PTR_ERR(ssi->base);
 
+	ssi->phys = res->start;
 	ssi->clk = devm_clk_get(&pdev->dev, "ssi");
 	if (IS_ERR(ssi->clk))
 		return PTR_ERR(ssi->clk);
@@ -781,8 +955,18 @@ static int rz_ssi_probe(struct platform_device *pdev)
 
 	ssi->audio_mck = ssi->audio_clk_1 ? ssi->audio_clk_1 : ssi->audio_clk_2;
 
-	ssi->playback.transfer = rz_ssi_pio_send;
-	ssi->capture.transfer = rz_ssi_pio_recv;
+	/* Detect DMA support */
+	ret = rz_ssi_dma_request(ssi, &pdev->dev);
+	if (ret < 0) {
+		dev_warn(&pdev->dev, "DMA not available, using PIO\n");
+		ssi->playback.transfer = rz_ssi_pio_send;
+		ssi->capture.transfer = rz_ssi_pio_recv;
+	} else {
+		dev_info(&pdev->dev, "DMA enabled");
+		ssi->playback.transfer = rz_ssi_dma_transfer;
+		ssi->capture.transfer = rz_ssi_dma_transfer;
+	}
+
 	ssi->playback.priv = ssi;
 	ssi->capture.priv = ssi;
 
@@ -798,28 +982,32 @@ static int rz_ssi_probe(struct platform_device *pdev)
 		return dev_err_probe(&pdev->dev, ret,
 				     "irq request error (int_req)\n");
 
-	/* Tx and Rx interrupts (pio only) */
-	ssi->irq_tx = platform_get_irq_byname(pdev, "dma_tx");
-	if (ssi->irq_tx < 0)
-		return dev_err_probe(&pdev->dev, -ENODEV,
-				     "Unable to get SSI dma_tx IRQ\n");
-
-	ret = devm_request_irq(&pdev->dev, ssi->irq_tx, &rz_ssi_interrupt, 0,
-			       dev_name(&pdev->dev), ssi);
-	if (ret < 0)
-		return dev_err_probe(&pdev->dev, ret,
-				     "irq request error (dma_tx)\n");
-
-	ssi->irq_rx = platform_get_irq_byname(pdev, "dma_rx");
-	if (ssi->irq_rx < 0)
-		return dev_err_probe(&pdev->dev, -ENODEV,
-				     "Unable to get SSI dma_rx IRQ\n");
-
-	ret = devm_request_irq(&pdev->dev, ssi->irq_rx, &rz_ssi_interrupt, 0,
-			       dev_name(&pdev->dev), ssi);
-	if (ret < 0)
-		return dev_err_probe(&pdev->dev, ret,
-				     "irq request error (dma_rx)\n");
+	if (!rz_ssi_is_dma_enabled(ssi)) {
+		/* Tx and Rx interrupts (pio only) */
+		ssi->irq_tx = platform_get_irq_byname(pdev, "dma_tx");
+		if (ssi->irq_tx < 0)
+			return dev_err_probe(&pdev->dev, -ENODEV,
+					     "Unable to get SSI dma_tx IRQ\n");
+
+		ret = devm_request_irq(&pdev->dev, ssi->irq_tx,
+				       &rz_ssi_interrupt, 0,
+				       dev_name(&pdev->dev), ssi);
+		if (ret < 0)
+			return dev_err_probe(&pdev->dev, ret,
+					     "irq request error (dma_tx)\n");
+
+		ssi->irq_rx = platform_get_irq_byname(pdev, "dma_rx");
+		if (ssi->irq_rx < 0)
+			return dev_err_probe(&pdev->dev, -ENODEV,
+					     "Unable to get SSI dma_rx IRQ\n");
+
+		ret = devm_request_irq(&pdev->dev, ssi->irq_rx,
+				       &rz_ssi_interrupt, 0,
+				       dev_name(&pdev->dev), ssi);
+		if (ret < 0)
+			return dev_err_probe(&pdev->dev, ret,
+					     "irq request error (dma_rx)\n");
+	}
 
 	ssi->rstc = devm_reset_control_get_exclusive(&pdev->dev, NULL);
 	if (IS_ERR(ssi->rstc))
@@ -835,6 +1023,8 @@ static int rz_ssi_probe(struct platform_device *pdev)
 					      rz_ssi_soc_dai,
 					      ARRAY_SIZE(rz_ssi_soc_dai));
 	if (ret < 0) {
+		rz_ssi_release_dma_channels(ssi);
+
 		pm_runtime_put(ssi->dev);
 		pm_runtime_disable(ssi->dev);
 		reset_control_assert(ssi->rstc);
@@ -848,6 +1038,8 @@ static int rz_ssi_remove(struct platform_device *pdev)
 {
 	struct rz_ssi_priv *ssi = dev_get_drvdata(&pdev->dev);
 
+	rz_ssi_release_dma_channels(ssi);
+
 	pm_runtime_put(ssi->dev);
 	pm_runtime_disable(ssi->dev);
 	reset_control_assert(ssi->rstc);
-- 
2.17.1


^ permalink raw reply related	[flat|nested] 11+ messages in thread

* Re: [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver
  2021-08-06 10:29 ` [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver Biju Das
@ 2021-08-06 11:27   ` Mark Brown
  2021-08-06 12:28     ` Biju Das
  0 siblings, 1 reply; 11+ messages in thread
From: Mark Brown @ 2021-08-06 11:27 UTC (permalink / raw)
  To: Biju Das
  Cc: Jaroslav Kysela, Takashi Iwai, Philipp Zabel, Liam Girdwood,
	Vinod Koul, Chris Brandt, Kuninori Morimoto, Laurent Pinchart,
	Randy Dunlap, Krzysztof Kozlowski, alsa-devel,
	Geert Uytterhoeven, Chris Paterson, Biju Das,
	Prabhakar Mahadev Lad, linux-renesas-soc

[-- Attachment #1: Type: text/plain, Size: 2084 bytes --]

On Fri, Aug 06, 2021 at 11:29:28AM +0100, Biju Das wrote:

> +static int rz_ssi_stream_init(struct rz_ssi_priv *ssi,
> +			      struct rz_ssi_stream *strm,
> +			      struct snd_pcm_substream *substream)
> +{
> +	struct snd_pcm_runtime *runtime = substream->runtime;
> +
> +	if (runtime->sample_bits != 16) {
> +		dev_err(ssi->dev, "Unsupported sample width: %d\n",
> +			runtime->sample_bits);
> +		return -EINVAL;
> +	}
> +
> +	if (runtime->frame_bits != 32) {
> +		dev_err(ssi->dev, "Unsupported frame width: %d\n",
> +			runtime->sample_bits);
> +		return -EINVAL;
> +	}

You should be doing *all* this validation at the time things are
configured, not during trigger.

> +static int rz_ssi_start_stop(struct rz_ssi_priv *ssi,
> +			     struct rz_ssi_stream *strm,
> +			     int start)
> +{
> +	struct snd_pcm_substream *substream = strm->substream;
> +	u32 ssicr, ssifcr;
> +	int timeout;
> +
> +	if (start) {
> +		bool is_play = rz_ssi_stream_is_play(ssi, substream);

...

> +	} else {
> +		strm->running = 0;

...

> +	}
> +
> +	return 0;
> +}

This is two functions merged into one with zero shared code, just make
them two separate functions and then people don't need to guess if true
is start or stop.

> +	switch (cmd) {
> +	case SNDRV_PCM_TRIGGER_START:
> +		/* Soft Reset */
> +		rz_ssi_reg_mask_setl(ssi, SSIFCR, 0, SSIFCR_SSIRST);
> +		rz_ssi_reg_mask_setl(ssi, SSIFCR, SSIFCR_SSIRST, 0);
> +		udelay(5);
> +
> +		spin_lock_irqsave(&ssi->lock, flags);
> +		ret = rz_ssi_stream_init(ssi, strm, substream);
> +		spin_unlock_irqrestore(&ssi->lock, flags);

It's not clear what this lock is intended to accomplish.  It's only used
in trigger like this and in _stream_is_valid().  In trigger() we're
already in atomic context so don't have to worry about simultaneous
calls to trigger() while in _stream_is_valid() we're just checking if
there's a runtime present but since the lock is taken and held within
the function the status could change before we've even returned to the
caller.  The two uses don't seem joined up with each other and neither
seems to do much.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

^ permalink raw reply	[flat|nested] 11+ messages in thread

* RE: [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver
  2021-08-06 11:27   ` Mark Brown
@ 2021-08-06 12:28     ` Biju Das
  0 siblings, 0 replies; 11+ messages in thread
From: Biju Das @ 2021-08-06 12:28 UTC (permalink / raw)
  To: Mark Brown
  Cc: Jaroslav Kysela, Takashi Iwai, Philipp Zabel, Liam Girdwood,
	Vinod Koul, Chris Brandt, Kuninori Morimoto, Laurent Pinchart,
	Randy Dunlap, Krzysztof Kozlowski, alsa-devel,
	Geert Uytterhoeven, Chris Paterson, Biju Das,
	Prabhakar Mahadev Lad, linux-renesas-soc

Hi Mark,

Thanks for the feedback.

> Subject: Re: [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver
> 
> On Fri, Aug 06, 2021 at 11:29:28AM +0100, Biju Das wrote:
> 
> > +static int rz_ssi_stream_init(struct rz_ssi_priv *ssi,
> > +			      struct rz_ssi_stream *strm,
> > +			      struct snd_pcm_substream *substream) {
> > +	struct snd_pcm_runtime *runtime = substream->runtime;
> > +
> > +	if (runtime->sample_bits != 16) {
> > +		dev_err(ssi->dev, "Unsupported sample width: %d\n",
> > +			runtime->sample_bits);
> > +		return -EINVAL;
> > +	}
> > +
> > +	if (runtime->frame_bits != 32) {
> > +		dev_err(ssi->dev, "Unsupported frame width: %d\n",
> > +			runtime->sample_bits);
> > +		return -EINVAL;
> > +	}
> 
> You should be doing *all* this validation at the time things are
> configured, not during trigger.

OK, Will move sample_bits validation to hw_params.

From core/pcm_native.c, 

frame_bits = channels * sample_bits;

since we are validating both channels and sample_bits,
there is no need to validate frame_bits.

So I am dropping frame_bits validation on next version.

> 
> > +static int rz_ssi_start_stop(struct rz_ssi_priv *ssi,
> > +			     struct rz_ssi_stream *strm,
> > +			     int start)
> > +{
> > +	struct snd_pcm_substream *substream = strm->substream;
> > +	u32 ssicr, ssifcr;
> > +	int timeout;
> > +
> > +	if (start) {
> > +		bool is_play = rz_ssi_stream_is_play(ssi, substream);
> 
> ...
> 
> > +	} else {
> > +		strm->running = 0;
> 
> ...
> 
> > +	}
> > +
> > +	return 0;
> > +}
> 
> This is two functions merged into one with zero shared code, just make
> them two separate functions and then people don't need to guess if true is
> start or stop.

OK. Will split this into 2 separate function.

> 
> > +	switch (cmd) {
> > +	case SNDRV_PCM_TRIGGER_START:
> > +		/* Soft Reset */
> > +		rz_ssi_reg_mask_setl(ssi, SSIFCR, 0, SSIFCR_SSIRST);
> > +		rz_ssi_reg_mask_setl(ssi, SSIFCR, SSIFCR_SSIRST, 0);
> > +		udelay(5);
> > +
> > +		spin_lock_irqsave(&ssi->lock, flags);
> > +		ret = rz_ssi_stream_init(ssi, strm, substream);
> > +		spin_unlock_irqrestore(&ssi->lock, flags);
> 
> It's not clear what this lock is intended to accomplish.  It's only used
> in trigger like this and in _stream_is_valid().  In trigger() we're
> already in atomic context so don't have to worry about simultaneous calls
> to trigger() while in _stream_is_valid() we're just checking if there's a
> runtime present but since the lock is taken and held within the function
> the status could change before we've even returned to the caller.  The two
> uses don't seem joined up with each other and neither seems to do much.

Thanks for  pointing out. After re-checking, locking is not needed at this
place. It is required only for _stream_is_valid() and _stream_quit().

Cheers,
Biju

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters
  2021-08-06 10:29 ` [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters Biju Das
@ 2021-08-09 12:57   ` Geert Uytterhoeven
  2021-08-12  7:19     ` Biju Das
  2021-08-13 20:12     ` Rob Herring
  2021-08-13 20:13   ` Rob Herring
  1 sibling, 2 replies; 11+ messages in thread
From: Geert Uytterhoeven @ 2021-08-09 12:57 UTC (permalink / raw)
  To: Biju Das
  Cc: Rob Herring, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, ALSA Development Mailing List,
	open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS,
	Geert Uytterhoeven, Chris Paterson, Biju Das,
	Prabhakar Mahadev Lad, Linux-Renesas

Hi Biju,

On Fri, Aug 6, 2021 at 12:29 PM Biju Das <biju.das.jz@bp.renesas.com> wrote:
> The DMAC driver on RZ/G2L expects the slave channel configuration
> to be passed in dmas property.
> This patch updates the example node to include the encoded slave
> channel configuration.
>
> Fixes: 5df6dfbb6de815ba3a("ASoC: dt-bindings: sound: renesas,rz-ssi: Document DMA support")
> Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
> Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>

> v3->v4:
>  * Updated bindings as the DMAC driver on RZ/G2L expects the
>    slave channel configuration to be passed in dmas property.

Thanks for the update!

> --- a/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> +++ b/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> @@ -93,8 +93,8 @@ examples:
>              clock-names = "ssi", "ssi_sfr", "audio_clk1", "audio_clk2";
>              power-domains = <&cpg>;
>              resets = <&cpg R9A07G044_SSI0_RST_M2_REG>;
> -            dmas = <&dmac 0x255>,
> -                   <&dmac 0x256>;
> +            dmas = <&dmac 0x2655>,
> +                   <&dmac 0x2656>;
>              dma-names = "tx", "rx";
>              #sound-dai-cells = <0>;
>      };

I think it would be good to describe the expected format in the description
for the dmas property, so the DTS writer knows what the numerical
values in the example really mean.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

^ permalink raw reply	[flat|nested] 11+ messages in thread

* RE: [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters
  2021-08-09 12:57   ` Geert Uytterhoeven
@ 2021-08-12  7:19     ` Biju Das
  2021-08-13 20:12     ` Rob Herring
  1 sibling, 0 replies; 11+ messages in thread
From: Biju Das @ 2021-08-12  7:19 UTC (permalink / raw)
  To: Geert Uytterhoeven
  Cc: Rob Herring, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, ALSA Development Mailing List,
	open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS,
	Geert Uytterhoeven, Chris Paterson, Biju Das,
	Prabhakar Mahadev Lad, Linux-Renesas

Hi Geert,

Thanks for the update.

> Subject: Re: [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi:
> Update slave dma channel configuration parameters
> 
> Hi Biju,
> 
> On Fri, Aug 6, 2021 at 12:29 PM Biju Das <biju.das.jz@bp.renesas.com>
> wrote:
> > The DMAC driver on RZ/G2L expects the slave channel configuration to
> > be passed in dmas property.
> > This patch updates the example node to include the encoded slave
> > channel configuration.
> >
> > Fixes: 5df6dfbb6de815ba3a("ASoC: dt-bindings: sound: renesas,rz-ssi:
> > Document DMA support")
> > Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
> > Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> 
> > v3->v4:
> >  * Updated bindings as the DMAC driver on RZ/G2L expects the
> >    slave channel configuration to be passed in dmas property.
> 
> Thanks for the update!
> 
> > --- a/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> > +++ b/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> > @@ -93,8 +93,8 @@ examples:
> >              clock-names = "ssi", "ssi_sfr", "audio_clk1", "audio_clk2";
> >              power-domains = <&cpg>;
> >              resets = <&cpg R9A07G044_SSI0_RST_M2_REG>;
> > -            dmas = <&dmac 0x255>,
> > -                   <&dmac 0x256>;
> > +            dmas = <&dmac 0x2655>,
> > +                   <&dmac 0x2656>;
> >              dma-names = "tx", "rx";
> >              #sound-dai-cells = <0>;
> >      };
> 
> I think it would be good to describe the expected format in the
> description for the dmas property, so the DTS writer knows what the
> numerical values in the example really mean.

Agree, will add description for dmas property.

Regards,
Biju

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters
  2021-08-09 12:57   ` Geert Uytterhoeven
  2021-08-12  7:19     ` Biju Das
@ 2021-08-13 20:12     ` Rob Herring
  2021-08-14  9:11       ` Geert Uytterhoeven
  1 sibling, 1 reply; 11+ messages in thread
From: Rob Herring @ 2021-08-13 20:12 UTC (permalink / raw)
  To: Geert Uytterhoeven
  Cc: Biju Das, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, ALSA Development Mailing List,
	open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS,
	Geert Uytterhoeven, Chris Paterson, Biju Das,
	Prabhakar Mahadev Lad, Linux-Renesas

On Mon, Aug 09, 2021 at 02:57:33PM +0200, Geert Uytterhoeven wrote:
> Hi Biju,
> 
> On Fri, Aug 6, 2021 at 12:29 PM Biju Das <biju.das.jz@bp.renesas.com> wrote:
> > The DMAC driver on RZ/G2L expects the slave channel configuration
> > to be passed in dmas property.
> > This patch updates the example node to include the encoded slave
> > channel configuration.
> >
> > Fixes: 5df6dfbb6de815ba3a("ASoC: dt-bindings: sound: renesas,rz-ssi: Document DMA support")
> > Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
> > Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> 
> > v3->v4:
> >  * Updated bindings as the DMAC driver on RZ/G2L expects the
> >    slave channel configuration to be passed in dmas property.
> 
> Thanks for the update!
> 
> > --- a/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> > +++ b/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> > @@ -93,8 +93,8 @@ examples:
> >              clock-names = "ssi", "ssi_sfr", "audio_clk1", "audio_clk2";
> >              power-domains = <&cpg>;
> >              resets = <&cpg R9A07G044_SSI0_RST_M2_REG>;
> > -            dmas = <&dmac 0x255>,
> > -                   <&dmac 0x256>;
> > +            dmas = <&dmac 0x2655>,
> > +                   <&dmac 0x2656>;
> >              dma-names = "tx", "rx";
> >              #sound-dai-cells = <0>;
> >      };
> 
> I think it would be good to describe the expected format in the description
> for the dmas property, so the DTS writer knows what the numerical
> values in the example really mean.

Yes, but that's a property of the DMA controller binding, not this one. 
It could vary from chip to chip.

Rob

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters
  2021-08-06 10:29 ` [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters Biju Das
  2021-08-09 12:57   ` Geert Uytterhoeven
@ 2021-08-13 20:13   ` Rob Herring
  1 sibling, 0 replies; 11+ messages in thread
From: Rob Herring @ 2021-08-13 20:13 UTC (permalink / raw)
  To: Biju Das
  Cc: Geert Uytterhoeven, Rob Herring, Biju Das, alsa-devel,
	Liam Girdwood, linux-renesas-soc, Prabhakar Mahadev Lad,
	devicetree, Kuninori Morimoto, Chris Paterson, Vinod Koul,
	Mark Brown, Chris Brandt

On Fri, 06 Aug 2021 11:29:29 +0100, Biju Das wrote:
> The DMAC driver on RZ/G2L expects the slave channel configuration
> to be passed in dmas property.
> This patch updates the example node to include the encoded slave
> channel configuration.
> 
> Fixes: 5df6dfbb6de815ba3a("ASoC: dt-bindings: sound: renesas,rz-ssi: Document DMA support")
> Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
> Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> ---
> Note:-
>  This patch is based on [1]
>  [1]:- https://patchwork.kernel.org/project/linux-renesas-soc/patch/20210806095322.2326-2-biju.das.jz@bp.renesas.com/
> 
> v3->v4:
>  * Updated bindings as the DMAC driver on RZ/G2L expects the
>    slave channel configuration to be passed in dmas property.
> v2->v3:
>  * Merged the binding patch with dmas added
>  * Updated dt binding example with encoded #dma-cells value.
> v1->v2:
>  * Rebased on 5.14-rc2.
> ---
>  Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 

Acked-by: Rob Herring <robh@kernel.org>

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters
  2021-08-13 20:12     ` Rob Herring
@ 2021-08-14  9:11       ` Geert Uytterhoeven
  0 siblings, 0 replies; 11+ messages in thread
From: Geert Uytterhoeven @ 2021-08-14  9:11 UTC (permalink / raw)
  To: Rob Herring
  Cc: Biju Das, Liam Girdwood, Mark Brown, Vinod Koul, Chris Brandt,
	Kuninori Morimoto, ALSA Development Mailing List,
	open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS,
	Geert Uytterhoeven, Chris Paterson, Biju Das,
	Prabhakar Mahadev Lad, Linux-Renesas

Hi Rob,

On Fri, Aug 13, 2021 at 10:12 PM Rob Herring <robh@kernel.org> wrote:
> On Mon, Aug 09, 2021 at 02:57:33PM +0200, Geert Uytterhoeven wrote:
> > On Fri, Aug 6, 2021 at 12:29 PM Biju Das <biju.das.jz@bp.renesas.com> wrote:
> > > The DMAC driver on RZ/G2L expects the slave channel configuration
> > > to be passed in dmas property.
> > > This patch updates the example node to include the encoded slave
> > > channel configuration.
> > >
> > > Fixes: 5df6dfbb6de815ba3a("ASoC: dt-bindings: sound: renesas,rz-ssi: Document DMA support")
> > > Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
> > > Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> >
> > > v3->v4:
> > >  * Updated bindings as the DMAC driver on RZ/G2L expects the
> > >    slave channel configuration to be passed in dmas property.
> >
> > Thanks for the update!
> >
> > > --- a/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> > > +++ b/Documentation/devicetree/bindings/sound/renesas,rz-ssi.yaml
> > > @@ -93,8 +93,8 @@ examples:
> > >              clock-names = "ssi", "ssi_sfr", "audio_clk1", "audio_clk2";
> > >              power-domains = <&cpg>;
> > >              resets = <&cpg R9A07G044_SSI0_RST_M2_REG>;
> > > -            dmas = <&dmac 0x255>,
> > > -                   <&dmac 0x256>;
> > > +            dmas = <&dmac 0x2655>,
> > > +                   <&dmac 0x2656>;
> > >              dma-names = "tx", "rx";
> > >              #sound-dai-cells = <0>;
> > >      };
> >
> > I think it would be good to describe the expected format in the description
> > for the dmas property, so the DTS writer knows what the numerical
> > values in the example really mean.
>
> Yes, but that's a property of the DMA controller binding, not this one.
> It could vary from chip to chip.

Right, unlike the other Renesas SoCs, RZ/G2L uses the normal DMAC for
audio, not an audio-specific one.

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

^ permalink raw reply	[flat|nested] 11+ messages in thread

end of thread, other threads:[~2021-08-14  9:11 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-08-06 10:29 [PATCH v4 0/3] Add RZ/G2L Sound support Biju Das
2021-08-06 10:29 ` [PATCH v4 1/3] ASoC: sh: Add RZ/G2L SSIF-2 driver Biju Das
2021-08-06 11:27   ` Mark Brown
2021-08-06 12:28     ` Biju Das
2021-08-06 10:29 ` [PATCH v4 2/3] ASoC: dt-bindings: sound: renesas,rz-ssi: Update slave dma channel configuration parameters Biju Das
2021-08-09 12:57   ` Geert Uytterhoeven
2021-08-12  7:19     ` Biju Das
2021-08-13 20:12     ` Rob Herring
2021-08-14  9:11       ` Geert Uytterhoeven
2021-08-13 20:13   ` Rob Herring
2021-08-06 10:29 ` [PATCH v4 3/3] ASoC: sh: rz-ssi: Add SSI DMAC support Biju Das

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).