Received-SPF: pass (google.com: domain of linux-kernel+bounces-130548-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
From: Sugar Zhang <sugar.zhang@rock-chips.com>
To: heiko@sntech.de,
	vkoul@kernel.org
Cc: linux-rockchip@lists.infradead.org,
	Sugar Zhang <sugar.zhang@rock-chips.com>,
	dmaengine@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH v2 2/2] dmaengine: pl330: Add support for audio interleaved transfer
Date: Wed,  3 Apr 2024 21:18:23 +0800
Message-Id: <20240403211810.v2.2.I27a3219694d4d2a43fcfb35e1b069340dc94a5f0@changeid>
In-Reply-To: <1712150304-60832-1-git-send-email-sugar.zhang@rock-chips.com>
References: <1712150304-60832-1-git-send-email-sugar.zhang@rock-chips.com>
Precedence: bulk

This patch add support for interleaved transfer which used
for interleaved audio or 2d video data transfer.

for audio situation, we add 'nump' for number of period frames.

e.g. combine 2 stream into a union one by 2D dma.

DAI0: 16CH
+-------------------------------------------------------------+
| Frame-1 | Frame-2 | Frame-3 | Frame-4 | ...... Frame-'numf' |
+-------------------------------------------------------------+

DAI1: 16CH
+-------------------------------------------------------------+
| Frame-1 | Frame-2 | Frame-3 | Frame-4 | ...... Frame-'numf' |
+-------------------------------------------------------------+

DAI0 + DAI1: 32CH

+-------------------------------------------------------------+
| DAI0-F1 | DAI1-F1 | DAI0-F2 | DAI1-F2 | ......              |
+-------------------------------------------------------------+
|      Frame-1      |      Frame-2      | ...... Frame-'numf' |

For audio situation, we have buffer_size and period_size,
the 'numf' is the buffer_size. so, we need another one for
period_size, e.g. 'nump'.

| Frame-1 | ~ | Frame-'nump' | ~ | Frame-'nump+1' | ~ |  Frame-'numf' |
|

As the above shown:

each DAI0 transfer 1 Frame, should skip a gap size (DAI1-F1)
each DAI1 transfer 1 Frame, should skip a gap size (DAI0-F1)

So, the interleaved template describe as follows:

DAI0:

struct dma_interleaved_template *xt;

xt->sgl[0].size = DAI0-F1;
xt->sgl[0].icg =  DAI1-F1;
xt->nump = nump; //the period_size in frames
xt->numf = numf; //the buffer_size in frames

DAI1:

struct dma_interleaved_template *xt;

xt->sgl[0].size = DAI1-F1;
xt->sgl[0].icg =  DAI0-F1;
xt->nump = nump; //the period_size in frames
xt->numf = numf; //the buffer_size in frames

Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
---

(no changes since v1)

 drivers/dma/pl330.c | 169 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 163 insertions(+), 6 deletions(-)

diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 5f6d7f1..449ed6b 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -239,6 +239,7 @@ enum pl330_byteswap {
 
 #define BYTE_TO_BURST(b, ccr)	((b) / BRST_SIZE(ccr) / BRST_LEN(ccr))
 #define BURST_TO_BYTE(c, ccr)	((c) * BRST_SIZE(ccr) * BRST_LEN(ccr))
+#define BYTE_MOD_BURST_LEN(b, ccr)	(((b) / BRST_SIZE(ccr)) % BRST_LEN(ccr))
 
 /*
  * With 256 bytes, we can do more than 2.5MB and 5MB xfers per req
@@ -545,6 +546,8 @@ struct dma_pl330_desc {
 	unsigned peri:5;
 	/* Hook to attach to DMAC's list of reqs with due callback */
 	struct list_head rqd;
+	/* interleaved size */
+	struct data_chunk sgl;
 };
 
 struct _xfer_spec {
@@ -577,6 +580,22 @@ static inline u32 get_revision(u32 periph_id)
 	return (periph_id >> PERIPH_REV_SHIFT) & PERIPH_REV_MASK;
 }
 
+static inline u32 _emit_ADDH(unsigned dry_run, u8 buf[],
+		enum pl330_dst da, u16 val)
+{
+	if (dry_run)
+		return SZ_DMAADDH;
+
+	buf[0] = CMD_DMAADDH;
+	buf[0] |= (da << 1);
+	*((__le16 *)&buf[1]) = cpu_to_le16(val);
+
+	PL330_DBGCMD_DUMP(SZ_DMAADDH, "\tDMAADDH %s %u\n",
+		da == 1 ? "DA" : "SA", val);
+
+	return SZ_DMAADDH;
+}
+
 static inline u32 _emit_END(unsigned dry_run, u8 buf[])
 {
 	if (dry_run)
@@ -1190,7 +1209,7 @@ static inline int _ldst_peripheral(struct pl330_dmac *pl330,
 				 const struct _xfer_spec *pxs, int cyc,
 				 enum pl330_cond cond)
 {
-	int off = 0;
+	int off = 0, i = 0, burstn = 1;
 
 	/*
 	 * do FLUSHP at beginning to clear any stale dma requests before the
@@ -1198,12 +1217,36 @@ static inline int _ldst_peripheral(struct pl330_dmac *pl330,
 	 */
 	if (!(pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP))
 		off += _emit_FLUSHP(dry_run, &buf[off], pxs->desc->peri);
+
+	if (pxs->desc->sgl.size) {
+		WARN_ON(BYTE_MOD_BURST_LEN(pxs->desc->sgl.size, pxs->ccr));
+		burstn = BYTE_TO_BURST(pxs->desc->sgl.size, pxs->ccr);
+	}
+
 	while (cyc--) {
-		off += _emit_WFP(dry_run, &buf[off], cond, pxs->desc->peri);
-		off += _emit_load(dry_run, &buf[off], cond, pxs->desc->rqtype,
-			pxs->desc->peri);
-		off += _emit_store(dry_run, &buf[off], cond, pxs->desc->rqtype,
-			pxs->desc->peri);
+		for (i = 0; i < burstn; i++) {
+			off += _emit_WFP(dry_run, &buf[off], cond, pxs->desc->peri);
+			off += _emit_load(dry_run, &buf[off], cond, pxs->desc->rqtype,
+				pxs->desc->peri);
+			off += _emit_store(dry_run, &buf[off], cond, pxs->desc->rqtype,
+				pxs->desc->peri);
+		}
+
+		switch (pxs->desc->rqtype) {
+		case DMA_DEV_TO_MEM:
+			if (pxs->desc->sgl.dst_icg)
+				off += _emit_ADDH(dry_run, &buf[off], DST,
+						  pxs->desc->sgl.dst_icg);
+			break;
+		case DMA_MEM_TO_DEV:
+			if (pxs->desc->sgl.src_icg)
+				off += _emit_ADDH(dry_run, &buf[off], SRC,
+						  pxs->desc->sgl.src_icg);
+			break;
+		default:
+			WARN_ON(1);
+			break;
+		}
 	}
 
 	return off;
@@ -1382,6 +1425,9 @@ static inline int _setup_loops(struct pl330_dmac *pl330,
 		BRST_SIZE(ccr);
 	int off = 0;
 
+	if (pxs->desc->sgl.size)
+		bursts = x->bytes / pxs->desc->sgl.size;
+
 	while (bursts) {
 		c = bursts;
 		off += _loop(pl330, dry_run, &buf[off], &c, pxs);
@@ -2624,6 +2670,9 @@ static struct dma_pl330_desc *pl330_get_desc(struct dma_pl330_chan *pch)
 
 	desc->peri = peri_id ? pch->chan.chan_id : 0;
 	desc->rqcfg.pcfg = &pch->dmac->pcfg;
+	desc->sgl.size = 0;
+	desc->sgl.src_icg = 0;
+	desc->sgl.dst_icg = 0;
 
 	dma_async_tx_descriptor_init(&desc->txd, &pch->chan);
 
@@ -2774,6 +2823,110 @@ static struct dma_async_tx_descriptor *pl330_prep_dma_cyclic(
 	return &desc->txd;
 }
 
+static struct dma_async_tx_descriptor *pl330_prep_interleaved_dma(
+	struct dma_chan *chan, struct dma_interleaved_template *xt,
+	unsigned long flags)
+{
+	struct dma_pl330_desc *desc = NULL, *first = NULL;
+	struct dma_pl330_chan *pch = to_pchan(chan);
+	struct pl330_dmac *pl330 = pch->dmac;
+	unsigned int i;
+	dma_addr_t dst;
+	dma_addr_t src;
+	size_t size, src_icg, dst_icg, period_bytes, buffer_bytes, full_period_bytes;
+	size_t nump = 0, numf = 0;
+
+	if (!xt->numf || !xt->sgl[0].size || xt->frame_size != 1)
+		return NULL;
+	nump = xt->nump;
+	numf = xt->numf;
+	size = xt->sgl[0].size;
+	period_bytes = size * nump;
+	buffer_bytes = size * numf;
+
+	if (flags & DMA_PREP_REPEAT && (!nump || (numf % nump)))
+		return NULL;
+
+	src_icg = dmaengine_get_src_icg(xt, &xt->sgl[0]);
+	dst_icg = dmaengine_get_dst_icg(xt, &xt->sgl[0]);
+
+	pl330_config_write(chan, &pch->slave_config, xt->dir);
+
+	if (!pl330_prep_slave_fifo(pch, xt->dir))
+		return NULL;
+
+	for (i = 0; i < numf / nump; i++) {
+		desc = pl330_get_desc(pch);
+		if (!desc) {
+			unsigned long iflags;
+
+			dev_err(pch->dmac->ddma.dev, "%s:%d Unable to fetch desc\n",
+				__func__, __LINE__);
+
+			if (!first)
+				return NULL;
+
+			spin_lock_irqsave(&pl330->pool_lock, iflags);
+
+			while (!list_empty(&first->node)) {
+				desc = list_entry(first->node.next,
+						struct dma_pl330_desc, node);
+				list_move_tail(&desc->node, &pl330->desc_pool);
+			}
+
+			list_move_tail(&first->node, &pl330->desc_pool);
+
+			spin_unlock_irqrestore(&pl330->pool_lock, iflags);
+
+			return NULL;
+		}
+
+		switch (xt->dir) {
+		case DMA_MEM_TO_DEV:
+			desc->rqcfg.src_inc = 1;
+			desc->rqcfg.dst_inc = 0;
+			src = xt->src_start + period_bytes * i;
+			dst = pch->fifo_dma;
+			full_period_bytes = (size + src_icg) * nump;
+			break;
+		case DMA_DEV_TO_MEM:
+			desc->rqcfg.src_inc = 0;
+			desc->rqcfg.dst_inc = 1;
+			src = pch->fifo_dma;
+			dst = xt->dst_start + period_bytes * i;
+			full_period_bytes = (size + dst_icg) * nump;
+			break;
+		default:
+			break;
+		}
+
+		desc->rqtype = xt->dir;
+		desc->rqcfg.brst_size = pch->burst_sz;
+		desc->rqcfg.brst_len = pch->burst_len;
+		desc->bytes_requested = full_period_bytes;
+		desc->sgl.size = size;
+		desc->sgl.src_icg = src_icg;
+		desc->sgl.dst_icg = dst_icg;
+		fill_px(&desc->px, dst, src, period_bytes);
+
+		if (!first)
+			first = desc;
+		else
+			list_add_tail(&desc->node, &first->node);
+	}
+
+	if (!desc)
+		return NULL;
+
+	if (flags & DMA_PREP_REPEAT)
+		pch->cyclic = true;
+
+	dev_dbg(chan->device->dev, "size: %zu, src_icg: %zu, dst_icg: %zu, nump: %zu, numf: %zu\n",
+		size, src_icg, dst_icg, nump, numf);
+
+	return &desc->txd;
+}
+
 static struct dma_async_tx_descriptor *
 pl330_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dst,
 		dma_addr_t src, size_t len, unsigned long flags)
@@ -3129,12 +3282,16 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 		dma_cap_set(DMA_SLAVE, pd->cap_mask);
 		dma_cap_set(DMA_CYCLIC, pd->cap_mask);
 		dma_cap_set(DMA_PRIVATE, pd->cap_mask);
+		dma_cap_set(DMA_INTERLEAVE, pd->cap_mask);
+		dma_cap_set(DMA_REPEAT, pd->cap_mask);
+		dma_cap_set(DMA_LOAD_EOT, pd->cap_mask);
 	}
 
 	pd->device_alloc_chan_resources = pl330_alloc_chan_resources;
 	pd->device_free_chan_resources = pl330_free_chan_resources;
 	pd->device_prep_dma_memcpy = pl330_prep_dma_memcpy;
 	pd->device_prep_dma_cyclic = pl330_prep_dma_cyclic;
+	pd->device_prep_interleaved_dma = pl330_prep_interleaved_dma;
 	pd->device_tx_status = pl330_tx_status;
 	pd->device_prep_slave_sg = pl330_prep_slave_sg;
 	pd->device_config = pl330_config;
-- 
2.7.4