Merge tag 'dmaengine-5.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine

Pull dmaengine updates from Vinod Koul:
 "Core:
   - Support out of order dma completion
   - Support for repeating transaction

  New controllers:
   - Support for Actions S700 DMA engine
   - Renesas R8A774E1, r8a7742 controller binding
   - New driver for Xilinx DPDMA controller

  Other:
   - Support of out of order dma completion in idxd driver
   - W=1 warning cleanup of subsystem
   - Updates to ti-k3-dma, dw, idxd drivers"

* tag 'dmaengine-5.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (68 commits)
  dmaengine: dw: Don't include unneeded header to platform data header
  dmaengine: Actions: Add support for S700 DMA engine
  dmaengine: Actions: get rid of bit fields from dma descriptor
  dt-bindings: dmaengine: convert Actions Semi Owl SoCs bindings to yaml
  dmaengine: idxd: add missing invalid flags field to completion
  dmaengine: dw: Initialize max_sg_burst capability
  dmaengine: dw: Introduce max burst length hw config
  dmaengine: dw: Initialize min and max burst DMA device capability
  dmaengine: dw: Set DMA device max segment size parameter
  dmaengine: dw: Take HC_LLP flag into account for noLLP auto-config
  dmaengine: Introduce DMA-device device_caps callback
  dmaengine: Introduce max SG burst capability
  dmaengine: Introduce min burst length capability
  dt-bindings: dma: dw: Add max burst transaction length property
  dt-bindings: dma: dw: Convert DW DMAC to DT binding
  dmaengine: ti: k3-udma: Query throughput level information from hardware
  dmaengine: ti: k3-udma: Use defines for capabilities register parsing
  dmaengine: xilinx: dpdma: Fix kerneldoc warning
  dmaengine: xilinx: dpdma: add missing kernel doc
  dmaengine: xilinx: dpdma: remove comparison of unsigned expression
  ...

1 files changed, 39 insertions, 27 deletions

diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 88b884cbb7c1..2c508ee672b9 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -33,7 +33,8 @@
 #define PL330_MAX_PERI		32
 #define PL330_MAX_BURST         16
 
-#define PL330_QUIRK_BROKEN_NO_FLUSHP BIT(0)
+#define PL330_QUIRK_BROKEN_NO_FLUSHP	BIT(0)
+#define PL330_QUIRK_PERIPH_BURST	BIT(1)
 
 enum pl330_cachectrl {
 	CCTRL0,		/* Noncacheable and nonbufferable */
@@ -284,7 +285,7 @@ struct pl330_config {
 	u32		irq_ns;
 };
 
-/**
+/*
  * Request Configuration.
  * The PL330 core does not modify this and uses the last
  * working configuration if the request doesn't provide any.
@@ -509,6 +510,10 @@ static struct pl330_of_quirks {
 	{
 		.quirk = "arm,pl330-broken-no-flushp",
 		.id = PL330_QUIRK_BROKEN_NO_FLUSHP,
+	},
+	{
+		.quirk = "arm,pl330-periph-burst",
+		.id = PL330_QUIRK_PERIPH_BURST,
 	}
 };
 
@@ -885,6 +890,12 @@ static inline void _execute_DBGINSN(struct pl330_thread *thrd,
 	void __iomem *regs = thrd->dmac->base;
 	u32 val;
 
+	/* If timed out due to halted state-machine */
+	if (_until_dmac_idle(thrd)) {
+		dev_err(thrd->dmac->ddma.dev, "DMAC halted!\n");
+		return;
+	}
+
 	val = (insn[0] << 16) | (insn[1] << 24);
 	if (!as_manager) {
 		val |= (1 << 0);
@@ -895,12 +906,6 @@ static inline void _execute_DBGINSN(struct pl330_thread *thrd,
 	val = le32_to_cpu(*((__le32 *)&insn[2]));
 	writel(val, regs + DBGINST1);
 
-	/* If timed out due to halted state-machine */
-	if (_until_dmac_idle(thrd)) {
-		dev_err(thrd->dmac->ddma.dev, "DMAC halted!\n");
-		return;
-	}
-
 	/* Get going */
 	writel(0, regs + DBGCMD);
 }
@@ -1183,9 +1188,6 @@ static inline int _ldst_peripheral(struct pl330_dmac *pl330,
 {
 	int off = 0;
 
-	if (pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP)
-		cond = BURST;
-
 	/*
 	 * do FLUSHP at beginning to clear any stale dma requests before the
 	 * first WFP.
@@ -1209,6 +1211,9 @@ static int _bursts(struct pl330_dmac *pl330, unsigned dry_run, u8 buf[],
 	int off = 0;
 	enum pl330_cond cond = BRST_LEN(pxs->ccr) > 1 ? BURST : SINGLE;
 
+	if (pl330->quirks & PL330_QUIRK_PERIPH_BURST)
+		cond = BURST;
+
 	switch (pxs->desc->rqtype) {
 	case DMA_MEM_TO_DEV:
 		/* fall through */
@@ -1231,8 +1236,9 @@ static int _bursts(struct pl330_dmac *pl330, unsigned dry_run, u8 buf[],
 }
 
 /*
- * transfer dregs with single transfers to peripheral, or a reduced size burst
- * for mem-to-mem.
+ * only the unaligned burst transfers have the dregs.
+ * so, still transfer dregs with a reduced size burst
+ * for mem-to-mem, mem-to-dev or dev-to-mem.
  */
 static int _dregs(struct pl330_dmac *pl330, unsigned int dry_run, u8 buf[],
 		const struct _xfer_spec *pxs, int transfer_length)
@@ -1243,22 +1249,31 @@ static int _dregs(struct pl330_dmac *pl330, unsigned int dry_run, u8 buf[],
 	if (transfer_length == 0)
 		return off;
 
+	/*
+	 * dregs_len = (total bytes - BURST_TO_BYTE(bursts, ccr)) /
+	 *             BRST_SIZE(ccr)
+	 * the dregs len must be smaller than burst len,
+	 * so, for higher efficiency, we can modify CCR
+	 * to use a reduced size burst len for the dregs.
+	 */
+	dregs_ccr = pxs->ccr;
+	dregs_ccr &= ~((0xf << CC_SRCBRSTLEN_SHFT) |
+		(0xf << CC_DSTBRSTLEN_SHFT));
+	dregs_ccr |= (((transfer_length - 1) & 0xf) <<
+		CC_SRCBRSTLEN_SHFT);
+	dregs_ccr |= (((transfer_length - 1) & 0xf) <<
+		CC_DSTBRSTLEN_SHFT);
+
 	switch (pxs->desc->rqtype) {
 	case DMA_MEM_TO_DEV:
 		/* fall through */
 	case DMA_DEV_TO_MEM:
-		off += _ldst_peripheral(pl330, dry_run, &buf[off], pxs,
-			transfer_length, SINGLE);
+		off += _emit_MOV(dry_run, &buf[off], CCR, dregs_ccr);
+		off += _ldst_peripheral(pl330, dry_run, &buf[off], pxs, 1,
+					BURST);
 		break;
 
 	case DMA_MEM_TO_MEM:
-		dregs_ccr = pxs->ccr;
-		dregs_ccr &= ~((0xf << CC_SRCBRSTLEN_SHFT) |
-			(0xf << CC_DSTBRSTLEN_SHFT));
-		dregs_ccr |= (((transfer_length - 1) & 0xf) <<
-			CC_SRCBRSTLEN_SHFT);
-		dregs_ccr |= (((transfer_length - 1) & 0xf) <<
-			CC_DSTBRSTLEN_SHFT);
 		off += _emit_MOV(dry_run, &buf[off], CCR, dregs_ccr);
 		off += _ldst_memtomem(dry_run, &buf[off], pxs, 1);
 		break;
@@ -2221,9 +2236,7 @@ static bool pl330_prep_slave_fifo(struct dma_pl330_chan *pch,
 
 static int fixup_burst_len(int max_burst_len, int quirks)
 {
-	if (quirks & PL330_QUIRK_BROKEN_NO_FLUSHP)
-		return 1;
-	else if (max_burst_len > PL330_MAX_BURST)
+	if (max_burst_len > PL330_MAX_BURST)
 		return PL330_MAX_BURST;
 	else if (max_burst_len < 1)
 		return 1;
@@ -3128,8 +3141,7 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 	pd->dst_addr_widths = PL330_DMA_BUSWIDTHS;
 	pd->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
 	pd->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
-	pd->max_burst = ((pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP) ?
-			 1 : PL330_MAX_BURST);
+	pd->max_burst = PL330_MAX_BURST;
 
 	ret = dma_async_device_register(pd);
 	if (ret) {