Merge branch 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma

Pull slave-dma updates from Vinod Koul:
 - new driver for BCM2835 used in R-pi
 - new driver for MOXA ART
 - dma_get_any_slave_channel API for DT based systems
 - minor fixes and updates spread acrooss driver

[ The fsl-ssi dual fifo mode support addition clashed badly with the
  other changes to fsl-ssi that came in through the sound merge.  I did
  a very rough cut at fixing up the conflict, but Nicolin Chen (author
  of both sides) will need to verify and check things ]

* 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma: (36 commits)
  dmaengine: mmp_pdma: fix mismerge
  dma: pl08x: Export pl08x_filter_id
  acpi-dma: align documentation with kernel-doc format
  dma: fix vchan_cookie_complete() debug print
  DMA: dmatest: extend the "device" module parameter to 32 characters
  drivers/dma: fix error return code
  dma: omap: Set debug level to debugging messages
  dmaengine: fix kernel-doc style typos for few comments
  dma: tegra: add support for Tegra148/124
  dma: dw: use %pad instead of casting dma_addr_t
  dma: dw: join split up messages
  dma: dw: fix style of multiline comment
  dmaengine: k3dma: fix sparse warnings
  dma: pl330: Use dma_get_slave_channel() in the of xlate callback
  dma: pl330: Differentiate between submitted and issued descriptors
  dmaengine: sirf: Add device_slave_caps interface
  DMA: Freescale: change BWC from 256 bytes to 1024 bytes
  dmaengine: Add MOXA ART DMA engine driver
  dmaengine: Add DMA_PRIVATE to BCM2835 driver
  dma: imx-sdma: Assign a default script number for ROM firmware cases
  ...

Documentation/devicetree/bindings/dma/bcm2835-dma.txt

+* BCM2835 DMA controller
+
+The BCM2835 DMA controller has 16 channels in total.
+Only the lower 13 channels have an associated IRQ.
+Some arbitrary channels are used by the firmware
+(1,3,6,7 in the current firmware version).
+The channels 0,2 and 3 have special functionality
+and should not be used by the driver.
+
+Required properties:
+- compatible: Should be "brcm,bcm2835-dma".
+- reg: Should contain DMA registers location and length.
+- interrupts: Should contain the DMA interrupts associated
+		to the DMA channels in ascending order.
+- #dma-cells: Must be <1>, the cell in the dmas property of the
+		client device represents the DREQ number.
+- brcm,dma-channel-mask: Bit mask representing the channels
+			 not used by the firmware in ascending order,
+			 i.e. first channel corresponds to LSB.
+
+Example:
+
+dma: dma@7e007000 {
+	compatible = "brcm,bcm2835-dma";
+	reg = <0x7e007000 0xf00>;
+	interrupts = <1 16>,
+		     <1 17>,
+		     <1 18>,
+		     <1 19>,
+		     <1 20>,
+		     <1 21>,
+		     <1 22>,
+		     <1 23>,
+		     <1 24>,
+		     <1 25>,
+		     <1 26>,
+		     <1 27>,
+		     <1 28>;
+
+	#dma-cells = <1>;
+	brcm,dma-channel-mask = <0x7f35>;
+};
+
+DMA clients connected to the BCM2835 DMA controller must use the format
+described in the dma.txt file, using a two-cell specifier for each channel.
+
+Example:
+
+bcm2835_i2s: i2s@7e203000 {
+	compatible = "brcm,bcm2835-i2s";
+	reg = <	0x7e203000 0x20>,
+	      < 0x7e101098 0x02>;
+
+	dmas = <&dma 2>,
+	       <&dma 3>;
+	dma-names = "tx", "rx";
+};

Documentation/devicetree/bindings/dma/fsl-imx-sdma.txt

@@ -42,6 +42,7 @@ The full ID of peripheral types can be found below.
 	19	IPU Memory
 	20	ASRC
 	21	ESAI
+	22	SSI Dual FIFO	(needs firmware ver >= 2)
 
 The third cell specifies the transfer priority as below.
 

Documentation/devicetree/bindings/dma/moxa,moxart-dma.txt

+MOXA ART DMA Controller
+
+See dma.txt first
+
+Required properties:
+
+- compatible :	Must be "moxa,moxart-dma"
+- reg :		Should contain registers location and length
+- interrupts :	Should contain an interrupt-specifier for the sole
+		interrupt generated by the device
+- #dma-cells :	Should be 1, a single cell holding a line request number
+
+Example:
+
+	dma: dma@90500000 {
+		compatible = "moxa,moxart-dma";
+		reg = <0x90500080 0x40>;
+		interrupts = <24 0>;
+		#dma-cells = <1>;
+	};
+
+
+Clients:
+
+DMA clients connected to the MOXA ART DMA controller must use the format
+described in the dma.txt file, using a two-cell specifier for each channel:
+a phandle plus one integer cells.
+The two cells in order are:
+
+1. A phandle pointing to the DMA controller.
+2. Peripheral identifier for the hardware handshaking interface.
+
+Example:
+Use specific request line passing from dma
+For example, MMC request line is 5
+
+	sdhci: sdhci@98e00000 {
+		compatible = "moxa,moxart-sdhci";
+		reg = <0x98e00000 0x5C>;
+		interrupts = <5 0>;
+		clocks = <&clk_apb>;
+		dmas =  <&dma 5>,
+			<&dma 5>;
+		dma-names = "tx", "rx";
+	};

drivers/dma/Kconfig

@@ -306,6 +306,12 @@ config DMA_OMAP
 	select DMA_ENGINE
 	select DMA_VIRTUAL_CHANNELS
 
+config DMA_BCM2835
+	tristate "BCM2835 DMA engine support"
+	depends on (ARCH_BCM2835 || MACH_BCM2708)
+	select DMA_ENGINE
+	select DMA_VIRTUAL_CHANNELS
+
 config TI_CPPI41
 	tristate "AM33xx CPPI41 DMA support"
 	depends on ARCH_OMAP
@@ -336,6 +342,14 @@ config K3_DMA
 	  Support the DMA engine for Hisilicon K3 platform
 	  devices.
 
+config MOXART_DMA
+	tristate "MOXART DMA support"
+	depends on ARCH_MOXART
+	select DMA_ENGINE
+	select DMA_VIRTUAL_CHANNELS
+	help
+	  Enable support for the MOXA ART SoC DMA controller.
+
 config DMA_ENGINE
 	bool
 

drivers/dma/Makefile

@@ -38,7 +38,9 @@ obj-$(CONFIG_EP93XX_DMA) += ep93xx_dma.o
 obj-$(CONFIG_DMA_SA11X0) += sa11x0-dma.o
 obj-$(CONFIG_MMP_TDMA) += mmp_tdma.o
 obj-$(CONFIG_DMA_OMAP) += omap-dma.o
+obj-$(CONFIG_DMA_BCM2835) += bcm2835-dma.o
 obj-$(CONFIG_MMP_PDMA) += mmp_pdma.o
 obj-$(CONFIG_DMA_JZ4740) += dma-jz4740.o
 obj-$(CONFIG_TI_CPPI41) += cppi41.o
 obj-$(CONFIG_K3_DMA) += k3dma.o
+obj-$(CONFIG_MOXART_DMA) += moxart-dma.o

drivers/dma/acpi-dma.c

@@ -30,11 +30,12 @@ static DEFINE_MUTEX(acpi_dma_lock);
  * @adev:	ACPI device to match with
  * @adma:	struct acpi_dma of the given DMA controller
  *
- * Returns 1 on success, 0 when no information is available, or appropriate
- * errno value on error.
- *
  * In order to match a device from DSDT table to the corresponding CSRT device
  * we use MMIO address and IRQ.
+ *
+ * Return:
+ * 1 on success, 0 when no information is available, or appropriate errno value
+ * on error.
  */
 static int acpi_dma_parse_resource_group(const struct acpi_csrt_group *grp,
 		struct acpi_device *adev, struct acpi_dma *adma)
@@ -101,7 +102,6 @@ static int acpi_dma_parse_resource_group(const struct acpi_csrt_group *grp,
  *
  * We are using this table to get the request line range of the specific DMA
  * controller to be used later.
- *
  */
 static void acpi_dma_parse_csrt(struct acpi_device *adev, struct acpi_dma *adma)
 {
@@ -141,10 +141,11 @@ static void acpi_dma_parse_csrt(struct acpi_device *adev, struct acpi_dma *adma)
  * @data		pointer to controller specific data to be used by
  *			translation function
  *
- * Returns 0 on success or appropriate errno value on error.
- *
  * Allocated memory should be freed with appropriate acpi_dma_controller_free()
  * call.
+ *
+ * Return:
+ * 0 on success or appropriate errno value on error.
  */
 int acpi_dma_controller_register(struct device *dev,
 		struct dma_chan *(*acpi_dma_xlate)
@@ -188,6 +189,9 @@ EXPORT_SYMBOL_GPL(acpi_dma_controller_register);
  * @dev:	struct device of DMA controller
  *
  * Memory allocated by acpi_dma_controller_register() is freed here.
+ *
+ * Return:
+ * 0 on success or appropriate errno value on error.
  */
 int acpi_dma_controller_free(struct device *dev)
 {
@@ -225,6 +229,9 @@ static void devm_acpi_dma_release(struct device *dev, void *res)
  * Managed acpi_dma_controller_register(). DMA controller registered by this
  * function are automatically freed on driver detach. See
  * acpi_dma_controller_register() for more information.
+ *
+ * Return:
+ * 0 on success or appropriate errno value on error.
  */
 int devm_acpi_dma_controller_register(struct device *dev,
 		struct dma_chan *(*acpi_dma_xlate)
@@ -267,8 +274,6 @@ EXPORT_SYMBOL_GPL(devm_acpi_dma_controller_free);
  * @adma:	struct acpi_dma of DMA controller
  * @dma_spec:	dma specifier to update
  *
- * Returns 0, if no information is avaiable, -1 on mismatch, and 1 otherwise.
- *
  * Accordingly to ACPI 5.0 Specification Table 6-170 "Fixed DMA Resource
  * Descriptor":
  *	DMA Request Line bits is a platform-relative number uniquely
@@ -276,6 +281,9 @@ EXPORT_SYMBOL_GPL(devm_acpi_dma_controller_free);
  *	mapping is done in a controller-specific OS driver.
  * That's why we can safely adjust slave_id when the appropriate controller is
  * found.
+ *
+ * Return:
+ * 0, if no information is avaiable, -1 on mismatch, and 1 otherwise.
  */
 static int acpi_dma_update_dma_spec(struct acpi_dma *adma,
 		struct acpi_dma_spec *dma_spec)
@@ -334,7 +342,8 @@ static int acpi_dma_parse_fixed_dma(struct acpi_resource *res, void *data)
  * @dev:	struct device to get DMA request from
  * @index:	index of FixedDMA descriptor for @dev
  *
- * Returns pointer to appropriate dma channel on success or NULL on error.
+ * Return:
+ * Pointer to appropriate dma channel on success or NULL on error.
  */
 struct dma_chan *acpi_dma_request_slave_chan_by_index(struct device *dev,
 		size_t index)
@@ -403,7 +412,8 @@ EXPORT_SYMBOL_GPL(acpi_dma_request_slave_chan_by_index);
  * translate the names "tx" and "rx" here based on the most common case where
  * the first FixedDMA descriptor is TX and second is RX.
  *
- * Returns pointer to appropriate dma channel on success or NULL on error.
+ * Return:
+ * Pointer to appropriate dma channel on success or NULL on error.
  */
 struct dma_chan *acpi_dma_request_slave_chan_by_name(struct device *dev,
 		const char *name)
@@ -427,8 +437,10 @@ EXPORT_SYMBOL_GPL(acpi_dma_request_slave_chan_by_name);
  * @adma: pointer to ACPI DMA controller data
  *
  * A simple translation function for ACPI based devices. Passes &struct
- * dma_spec to the DMA controller driver provided filter function. Returns
- * pointer to the channel if found or %NULL otherwise.
+ * dma_spec to the DMA controller driver provided filter function.
+ *
+ * Return:
+ * Pointer to the channel if found or %NULL otherwise.
  */
 struct dma_chan *acpi_dma_simple_xlate(struct acpi_dma_spec *dma_spec,
 		struct acpi_dma *adma)

drivers/dma/amba-pl08x.c

@@ -83,6 +83,7 @@
 #include <linux/dmaengine.h>
 #include <linux/dmapool.h>
 #include <linux/dma-mapping.h>
+#include <linux/export.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
@@ -1771,6 +1772,7 @@ bool pl08x_filter_id(struct dma_chan *chan, void *chan_id)
 
 	return false;
 }
+EXPORT_SYMBOL_GPL(pl08x_filter_id);
 
 /*
  * Just check that the device is there and active
@@ -2167,7 +2169,7 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
 	/* Register slave channels */
 	ret = pl08x_dma_init_virtual_channels(pl08x, &pl08x->slave,
 			pl08x->pd->num_slave_channels, true);
-	if (ret <= 0) {
+	if (ret < 0) {
 		dev_warn(&pl08x->adev->dev,
 			"%s failed to enumerate slave channels - %d\n",
 				__func__, ret);

drivers/dma/cppi41.c

@@ -972,8 +972,10 @@ static int cppi41_dma_probe(struct platform_device *pdev)
 		goto err_chans;
 
 	irq = irq_of_parse_and_map(dev->of_node, 0);
-	if (!irq)
+	if (!irq) {
+		ret = -EINVAL;
 		goto err_irq;
+	}
 
 	cppi_writel(USBSS_IRQ_PD_COMP, cdd->usbss_mem + USBSS_IRQ_ENABLER);
 

drivers/dma/dmatest.c

@@ -31,7 +31,7 @@ module_param_string(channel, test_channel, sizeof(test_channel),
 		S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(channel, "Bus ID of the channel to test (default: any)");
 
-static char test_device[20];
+static char test_device[32];
 module_param_string(device, test_device, sizeof(test_device),
 		S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(device, "Bus ID of the DMA Engine to test (default: any)");
@@ -89,7 +89,7 @@ MODULE_PARM_DESC(verbose, "Enable \"success\" result messages (default: off)");
 struct dmatest_params {
 	unsigned int	buf_size;
 	char		channel[20];
-	char		device[20];
+	char		device[32];
 	unsigned int	threads_per_chan;
 	unsigned int	max_channels;
 	unsigned int	iterations;

drivers/dma/dw/core.c

@@ -218,8 +218,10 @@ static inline void dwc_do_single_block(struct dw_dma_chan *dwc,
 	struct dw_dma	*dw = to_dw_dma(dwc->chan.device);
 	u32		ctllo;
 
-	/* Software emulation of LLP mode relies on interrupts to continue
-	 * multi block transfer. */
+	/*
+	 * Software emulation of LLP mode relies on interrupts to continue
+	 * multi block transfer.
+	 */
 	ctllo = desc->lli.ctllo | DWC_CTLL_INT_EN;
 
 	channel_writel(dwc, SAR, desc->lli.sar);
@@ -253,8 +255,7 @@ static void dwc_dostart(struct dw_dma_chan *dwc, struct dw_desc *first)
 						&dwc->flags);
 		if (was_soft_llp) {
 			dev_err(chan2dev(&dwc->chan),
-				"BUG: Attempted to start new LLP transfer "
-				"inside ongoing one\n");
+				"BUG: Attempted to start new LLP transfer inside ongoing one\n");
 			return;
 		}
 
@@ -420,8 +421,7 @@ static void dwc_scan_descriptors(struct dw_dma *dw, struct dw_dma_chan *dwc)
 		return;
 	}
 
-	dev_vdbg(chan2dev(&dwc->chan), "%s: llp=0x%llx\n", __func__,
-			(unsigned long long)llp);
+	dev_vdbg(chan2dev(&dwc->chan), "%s: llp=%pad\n", __func__, &llp);
 
 	list_for_each_entry_safe(desc, _desc, &dwc->active_list, desc_node) {
 		/* Initial residue value */
@@ -567,8 +567,8 @@ static void dwc_handle_cyclic(struct dw_dma *dw, struct dw_dma_chan *dwc,
 			unlikely(status_xfer & dwc->mask)) {
 		int i;
 
-		dev_err(chan2dev(&dwc->chan), "cyclic DMA unexpected %s "
-				"interrupt, stopping DMA transfer\n",
+		dev_err(chan2dev(&dwc->chan),
+			"cyclic DMA unexpected %s interrupt, stopping DMA transfer\n",
 			status_xfer ? "xfer" : "error");
 
 		spin_lock_irqsave(&dwc->lock, flags);
@@ -711,9 +711,8 @@ dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 	u32			ctllo;
 
 	dev_vdbg(chan2dev(chan),
-			"%s: d0x%llx s0x%llx l0x%zx f0x%lx\n", __func__,
-			(unsigned long long)dest, (unsigned long long)src,
-			len, flags);
+			"%s: d%pad s%pad l0x%zx f0x%lx\n", __func__,
+			&dest, &src, len, flags);
 
 	if (unlikely(!len)) {
 		dev_dbg(chan2dev(chan), "%s: length is zero!\n", __func__);
@@ -1401,9 +1400,9 @@ struct dw_cyclic_desc *dw_dma_cyclic_prep(struct dma_chan *chan,
 	/* Let's make a cyclic list */
 	last->lli.llp = cdesc->desc[0]->txd.phys;
 
-	dev_dbg(chan2dev(&dwc->chan), "cyclic prepared buf 0x%llx len %zu "
-			"period %zu periods %d\n", (unsigned long long)buf_addr,
-			buf_len, period_len, periods);
+	dev_dbg(chan2dev(&dwc->chan),
+			"cyclic prepared buf %pad len %zu period %zu periods %d\n",
+			&buf_addr, buf_len, period_len, periods);
 
 	cdesc->periods = periods;
 	dwc->cdesc = cdesc;
@@ -1603,9 +1602,11 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 			dev_dbg(chip->dev, "DWC_PARAMS[%d]: 0x%08x\n", i,
 					   dwc_params);
 
-			/* Decode maximum block size for given channel. The
+			/*
+			 * Decode maximum block size for given channel. The
 			 * stored 4 bit value represents blocks from 0x00 for 3
-			 * up to 0x0a for 4095. */
+			 * up to 0x0a for 4095.
+			 */
 			dwc->block_size =
 				(4 << ((max_blk_size >> 4 * i) & 0xf)) - 1;
 			dwc->nollp =

drivers/dma/edma.c

@@ -699,7 +699,7 @@ static int edma_alloc_chan_resources(struct dma_chan *chan)
 	echan->alloced = true;
 	echan->slot[0] = echan->ch_num;
 
-	dev_info(dev, "allocated channel for %u:%u\n",
+	dev_dbg(dev, "allocated channel for %u:%u\n",
 		EDMA_CTLR(echan->ch_num), EDMA_CHAN_SLOT(echan->ch_num));
 
 	return 0;
@@ -736,7 +736,7 @@ static void edma_free_chan_resources(struct dma_chan *chan)
 		echan->alloced = false;
 	}
 
-	dev_info(dev, "freeing channel for %u\n", echan->ch_num);
+	dev_dbg(dev, "freeing channel for %u\n", echan->ch_num);
 }
 
 /* Send pending descriptor to hardware */

drivers/dma/fsldma.h

@@ -41,7 +41,7 @@
  * channel is allowed to transfer before the DMA engine pauses
  * the current channel and switches to the next channel
  */
-#define FSL_DMA_MR_BWC         0x08000000
+#define FSL_DMA_MR_BWC         0x0A000000
 
 /* Special MR definition for MPC8349 */
 #define FSL_DMA_MR_EOTIE	0x00000080

drivers/dma/imx-sdma.c

@@ -323,6 +323,7 @@ struct sdma_engine {
 	struct clk			*clk_ipg;
 	struct clk			*clk_ahb;
 	spinlock_t			channel_0_lock;
+	u32				script_number;
 	struct sdma_script_start_addrs	*script_addrs;
 	const struct sdma_driver_data	*drvdata;
 };
@@ -724,6 +725,10 @@ static void sdma_get_pc(struct sdma_channel *sdmac,
 		per_2_emi = sdma->script_addrs->app_2_mcu_addr;
 		emi_2_per = sdma->script_addrs->mcu_2_app_addr;
 		break;
+	case IMX_DMATYPE_SSI_DUAL:
+		per_2_emi = sdma->script_addrs->ssish_2_mcu_addr;
+		emi_2_per = sdma->script_addrs->mcu_2_ssish_addr;
+		break;
 	case IMX_DMATYPE_SSI_SP:
 	case IMX_DMATYPE_MMC:
 	case IMX_DMATYPE_SDHC:
@@ -1238,6 +1243,7 @@ static void sdma_issue_pending(struct dma_chan *chan)
 }
 
 #define SDMA_SCRIPT_ADDRS_ARRAY_SIZE_V1	34
+#define SDMA_SCRIPT_ADDRS_ARRAY_SIZE_V2	38
 
 static void sdma_add_scripts(struct sdma_engine *sdma,
 		const struct sdma_script_start_addrs *addr)
@@ -1246,7 +1252,11 @@ static void sdma_add_scripts(struct sdma_engine *sdma,
 	s32 *saddr_arr = (u32 *)sdma->script_addrs;
 	int i;
 
-	for (i = 0; i < SDMA_SCRIPT_ADDRS_ARRAY_SIZE_V1; i++)
+	/* use the default firmware in ROM if missing external firmware */
+	if (!sdma->script_number)
+		sdma->script_number = SDMA_SCRIPT_ADDRS_ARRAY_SIZE_V1;
+
+	for (i = 0; i < sdma->script_number; i++)
 		if (addr_arr[i] > 0)
 			saddr_arr[i] = addr_arr[i];
 }
@@ -1272,6 +1282,17 @@ static void sdma_load_firmware(const struct firmware *fw, void *context)
 		goto err_firmware;
 	if (header->ram_code_start + header->ram_code_size > fw->size)
 		goto err_firmware;
+	switch (header->version_major) {
+		case 1:
+			sdma->script_number = SDMA_SCRIPT_ADDRS_ARRAY_SIZE_V1;
+			break;
+		case 2:
+			sdma->script_number = SDMA_SCRIPT_ADDRS_ARRAY_SIZE_V2;
+			break;
+		default:
+			dev_err(sdma->dev, "unknown firmware version\n");
+			goto err_firmware;
+	}
 
 	addr = (void *)header + header->script_addrs_start;
 	ram_code = (void *)header + header->ram_code_start;

drivers/dma/k3dma.c

@@ -477,7 +477,7 @@ static struct dma_async_tx_descriptor *k3_dma_prep_slave_sg(
 	dma_addr_t addr, src = 0, dst = 0;
 	int num = sglen, i;
 
-	if (sgl == 0)
+	if (sgl == NULL)
 		return NULL;
 
 	for_each_sg(sgl, sg, sglen, i) {
@@ -817,7 +817,7 @@ static int k3_dma_resume(struct device *dev)
 	return 0;
 }
 
-SIMPLE_DEV_PM_OPS(k3_dma_pmops, k3_dma_suspend, k3_dma_resume);
+static SIMPLE_DEV_PM_OPS(k3_dma_pmops, k3_dma_suspend, k3_dma_resume);
 
 static struct platform_driver k3_pdma_driver = {
 	.driver		= {

drivers/dma/mmp_tdma.c

@@ -121,11 +121,13 @@ struct mmp_tdma_chan {
 	int				idx;
 	enum mmp_tdma_type		type;
 	int				irq;
-	unsigned long			reg_base;
+	void __iomem			*reg_base;
 
 	size_t				buf_len;
 	size_t				period_len;
 	size_t				pos;
+
+	struct gen_pool			*pool;
 };
 
 #define TDMA_CHANNEL_NUM 2
@@ -182,7 +184,7 @@ static void mmp_tdma_pause_chan(struct mmp_tdma_chan *tdmac)
 
 static int mmp_tdma_config_chan(struct mmp_tdma_chan *tdmac)
 {
-	unsigned int tdcr;
+	unsigned int tdcr = 0;
 
 	mmp_tdma_disable_chan(tdmac);
 
@@ -324,7 +326,7 @@ static void mmp_tdma_free_descriptor(struct mmp_tdma_chan *tdmac)
 	struct gen_pool *gpool;
 	int size = tdmac->desc_num * sizeof(struct mmp_tdma_desc);
 
-	gpool = sram_get_gpool("asram");
+	gpool = tdmac->pool;
 	if (tdmac->desc_arr)
 		gen_pool_free(gpool, (unsigned long)tdmac->desc_arr,
 				size);
@@ -374,7 +376,7 @@ struct mmp_tdma_desc *mmp_tdma_alloc_descriptor(struct mmp_tdma_chan *tdmac)
 	struct gen_pool *gpool;
 	int size = tdmac->desc_num * sizeof(struct mmp_tdma_desc);
 
-	gpool = sram_get_gpool("asram");
+	gpool = tdmac->pool;
 	if (!gpool)
 		return NULL;
 
@@ -505,7 +507,8 @@ static int mmp_tdma_remove(struct platform_device *pdev)
 }
 
 static int mmp_tdma_chan_init(struct mmp_tdma_device *tdev,
-						int idx, int irq, int type)
+					int idx, int irq,
+					int type, struct gen_pool *pool)
 {
 	struct mmp_tdma_chan *tdmac;
 
@@ -526,7 +529,8 @@ static int mmp_tdma_chan_init(struct mmp_tdma_device *tdev,
 	tdmac->chan.device = &tdev->device;
 	tdmac->idx	   = idx;
 	tdmac->type	   = type;
-	tdmac->reg_base	   = (unsigned long)tdev->base + idx * 4;
+	tdmac->reg_base	   = tdev->base + idx * 4;
+	tdmac->pool	   = pool;
 	tdmac->status = DMA_COMPLETE;
 	tdev->tdmac[tdmac->idx] = tdmac;
 	tasklet_init(&tdmac->tasklet, dma_do_tasklet, (unsigned long)tdmac);
@@ -553,6 +557,7 @@ static int mmp_tdma_probe(struct platform_device *pdev)
 	int i, ret;
 	int irq = 0, irq_num = 0;
 	int chan_num = TDMA_CHANNEL_NUM;
+	struct gen_pool *pool;
 
 	of_id = of_match_device(mmp_tdma_dt_ids, &pdev->dev);
 	if (of_id)
@@ -579,6 +584,15 @@ static int mmp_tdma_probe(struct platform_device *pdev)
 
 	INIT_LIST_HEAD(&tdev->device.channels);
 
+	if (pdev->dev.of_node)
+		pool = of_get_named_gen_pool(pdev->dev.of_node, "asram", 0);
+	else
+		pool = sram_get_gpool("asram");
+	if (!pool) {
+		dev_err(&pdev->dev, "asram pool not available\n");
+		return -ENOMEM;
+	}
+
 	if (irq_num != chan_num) {
 		irq = platform_get_irq(pdev, 0);
 		ret = devm_request_irq(&pdev->dev, irq,
@@ -590,7 +604,7 @@ static int mmp_tdma_probe(struct platform_device *pdev)
 	/* initialize channel parameters */
 	for (i = 0; i < chan_num; i++) {
 		irq = (irq_num != chan_num) ? 0 : platform_get_irq(pdev, i);
-		ret = mmp_tdma_chan_init(tdev, i, irq, type);
+		ret = mmp_tdma_chan_init(tdev, i, irq, type, pool);
 		if (ret)
 			return ret;
 	}

drivers/dma/omap-dma.c

@@ -190,7 +190,7 @@ static int omap_dma_alloc_chan_resources(struct dma_chan *chan)
 {
 	struct omap_chan *c = to_omap_dma_chan(chan);
 
-	dev_info(c->vc.chan.device->dev, "allocating channel for %u\n", c->dma_sig);
+	dev_dbg(c->vc.chan.device->dev, "allocating channel for %u\n", c->dma_sig);
 
 	return omap_request_dma(c->dma_sig, "DMA engine",
 		omap_dma_callback, c, &c->dma_ch);
@@ -203,7 +203,7 @@ static void omap_dma_free_chan_resources(struct dma_chan *chan)
 	vchan_free_chan_resources(&c->vc);
 	omap_free_dma(c->dma_ch);
 
-	dev_info(c->vc.chan.device->dev, "freeing channel for %u\n", c->dma_sig);
+	dev_dbg(c->vc.chan.device->dev, "freeing channel for %u\n", c->dma_sig);
 }
 
 static size_t omap_dma_sg_size(struct omap_sg *sg)

drivers/dma/pl330.c

@@ -543,7 +543,9 @@ struct dma_pl330_chan {
 	/* DMA-Engine Channel */
 	struct dma_chan chan;
 
-	/* List of to be xfered descriptors */
+	/* List of submitted descriptors */
+	struct list_head submitted_list;
+	/* List of issued descriptors */
 	struct list_head work_list;
 	/* List of completed descriptors */
 	struct list_head completed_list;
@@ -578,12 +580,16 @@ struct dma_pl330_dmac {
 	/* DMA-Engine Device */
 	struct dma_device ddma;
 
+	/* Holds info about sg limitations */
+	struct device_dma_parameters dma_parms;
+
 	/* Pool of descriptors available for the DMAC's channels */
 	struct list_head desc_pool;
 	/* To protect desc_pool manipulation */
 	spinlock_t pool_lock;
 
 	/* Peripheral channels connected to this DMAC */
+	unsigned int num_peripherals;
 	struct dma_pl330_chan *peripherals; /* keep at end */
 };
 
@@ -606,11 +612,6 @@ struct dma_pl330_desc {
 	struct dma_pl330_chan *pchan;
 };
 
-struct dma_pl330_filter_args {
-	struct dma_pl330_dmac *pdmac;
-	unsigned int chan_id;
-};
-
 static inline void _callback(struct pl330_req *r, enum pl330_op_err err)
 {
 	if (r && r->xfer_cb)
@@ -2298,16 +2299,6 @@ static void dma_pl330_rqcb(void *token, enum pl330_op_err err)
 	tasklet_schedule(&pch->task);
 }
 
-static bool pl330_dt_filter(struct dma_chan *chan, void *param)
-{
-	struct dma_pl330_filter_args *fargs = param;
-
-	if (chan->device != &fargs->pdmac->ddma)
-		return false;
-
-	return (chan->chan_id == fargs->chan_id);
-}
-
 bool pl330_filter(struct dma_chan *chan, void *param)
 {
 	u8 *peri_id;
@@ -2325,23 +2316,16 @@ static struct dma_chan *of_dma_pl330_xlate(struct of_phandle_args *dma_spec,
 {
 	int count = dma_spec->args_count;
 	struct dma_pl330_dmac *pdmac = ofdma->of_dma_data;
-	struct dma_pl330_filter_args fargs;
-	dma_cap_mask_t cap;
-
-	if (!pdmac)
-		return NULL;
+	unsigned int chan_id;
 
 	if (count != 1)
 		return NULL;
 
-	fargs.pdmac = pdmac;
-	fargs.chan_id = dma_spec->args[0];
-
-	dma_cap_zero(cap);
-	dma_cap_set(DMA_SLAVE, cap);
-	dma_cap_set(DMA_CYCLIC, cap);
+	chan_id = dma_spec->args[0];
+	if (chan_id >= pdmac->num_peripherals)
+		return NULL;
 
-	return dma_request_channel(cap, pl330_dt_filter, &fargs);
+	return dma_get_slave_channel(&pdmac->peripherals[chan_id].chan);
 }
 
 static int pl330_alloc_chan_resources(struct dma_chan *chan)
@@ -2385,6 +2369,11 @@ static int pl330_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd, unsigned
 		pl330_chan_ctrl(pch->pl330_chid, PL330_OP_FLUSH);
 
 		/* Mark all desc done */
+		list_for_each_entry(desc, &pch->submitted_list, node) {
+			desc->status = FREE;
+			dma_cookie_complete(&desc->txd);
+		}
+
 		list_for_each_entry(desc, &pch->work_list , node) {
 			desc->status = FREE;
 			dma_cookie_complete(&desc->txd);
@@ -2395,6 +2384,7 @@ static int pl330_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd, unsigned
 			dma_cookie_complete(&desc->txd);
 		}
 
+		list_splice_tail_init(&pch->submitted_list, &pdmac->desc_pool);
 		list_splice_tail_init(&pch->work_list, &pdmac->desc_pool);
 		list_splice_tail_init(&pch->completed_list, &pdmac->desc_pool);
 		spin_unlock_irqrestore(&pch->lock, flags);
@@ -2453,7 +2443,14 @@ pl330_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 
 static void pl330_issue_pending(struct dma_chan *chan)
 {
-	pl330_tasklet((unsigned long) to_pchan(chan));
+	struct dma_pl330_chan *pch = to_pchan(chan);
+	unsigned long flags;
+
+	spin_lock_irqsave(&pch->lock, flags);
+	list_splice_tail_init(&pch->submitted_list, &pch->work_list);
+	spin_unlock_irqrestore(&pch->lock, flags);
+
+	pl330_tasklet((unsigned long)pch);
 }
 
 /*
@@ -2480,11 +2477,11 @@ static dma_cookie_t pl330_tx_submit(struct dma_async_tx_descriptor *tx)
 
 		dma_cookie_assign(&desc->txd);
 
-		list_move_tail(&desc->node, &pch->work_list);
+		list_move_tail(&desc->node, &pch->submitted_list);
 	}
 
 	cookie = dma_cookie_assign(&last->txd);
-	list_add_tail(&last->node, &pch->work_list);
+	list_add_tail(&last->node, &pch->submitted_list);
 	spin_unlock_irqrestore(&pch->lock, flags);
 
 	return cookie;
@@ -2960,6 +2957,8 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 	else
 		num_chan = max_t(int, pi->pcfg.num_peri, pi->pcfg.num_chan);
 
+	pdmac->num_peripherals = num_chan;
+
 	pdmac->peripherals = kzalloc(num_chan * sizeof(*pch), GFP_KERNEL);
 	if (!pdmac->peripherals) {
 		ret = -ENOMEM;
@@ -2974,6 +2973,7 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 		else
 			pch->chan.private = adev->dev.of_node;
 
+		INIT_LIST_HEAD(&pch->submitted_list);
 		INIT_LIST_HEAD(&pch->work_list);
 		INIT_LIST_HEAD(&pch->completed_list);
 		spin_lock_init(&pch->lock);
@@ -3021,6 +3021,9 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 			"unable to register DMA to the generic DT DMA helpers\n");
 		}
 	}
+
+	adev->dev.dma_parms = &pdmac->dma_parms;
+
 	/*
 	 * This is the limit for transfers with a buswidth of 1, larger
 	 * buswidths will have larger limits.

drivers/dma/ppc4xx/adma.c

@@ -4114,6 +4114,7 @@ static int ppc440spe_adma_probe(struct platform_device *ofdev)
 	regs = ioremap(res.start, resource_size(&res));
 	if (!regs) {
 		dev_err(&ofdev->dev, "failed to ioremap regs!\n");
+		ret = -ENOMEM;
 		goto err_regs_alloc;
 	}
 

drivers/dma/sirf-dma.c

@@ -640,6 +640,25 @@ bool sirfsoc_dma_filter_id(struct dma_chan *chan, void *chan_id)
 }
 EXPORT_SYMBOL(sirfsoc_dma_filter_id);
 
+#define SIRFSOC_DMA_BUSWIDTHS \
+	(BIT(DMA_SLAVE_BUSWIDTH_UNDEFINED) | \
+	BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) | \
+	BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \
+	BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) | \
+	BIT(DMA_SLAVE_BUSWIDTH_8_BYTES))
+
+static int sirfsoc_dma_device_slave_caps(struct dma_chan *dchan,
+	struct dma_slave_caps *caps)
+{
+	caps->src_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+	caps->dstn_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+	caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+	caps->cmd_pause = true;
+	caps->cmd_terminate = true;
+
+	return 0;
+}
+
 static int sirfsoc_dma_probe(struct platform_device *op)
 {
 	struct device_node *dn = op->dev.of_node;
@@ -712,6 +731,7 @@ static int sirfsoc_dma_probe(struct platform_device *op)
 	dma->device_tx_status = sirfsoc_dma_tx_status;
 	dma->device_prep_interleaved_dma = sirfsoc_dma_prep_interleaved;
 	dma->device_prep_dma_cyclic = sirfsoc_dma_prep_cyclic;
+	dma->device_slave_caps = sirfsoc_dma_device_slave_caps;
 
 	INIT_LIST_HEAD(&dma->channels);
 	dma_cap_set(DMA_SLAVE, dma->cap_mask);

drivers/dma/tegra20-apb-dma.c

@@ -100,6 +100,11 @@
 #define TEGRA_APBDMA_APBSEQ_DATA_SWAP		BIT(27)
 #define TEGRA_APBDMA_APBSEQ_WRAP_WORD_1		(1 << 16)
 
+/* Tegra148 specific registers */
+#define TEGRA_APBDMA_CHAN_WCOUNT		0x20
+
+#define TEGRA_APBDMA_CHAN_WORD_TRANSFER		0x24
+
 /*
  * If any burst is in flight and DMA paused then this is the time to complete
  * on-flight burst and update DMA status register.
@@ -109,21 +114,22 @@
 /* Channel base address offset from APBDMA base address */
 #define TEGRA_APBDMA_CHANNEL_BASE_ADD_OFFSET	0x1000
 
-/* DMA channel register space size */
-#define TEGRA_APBDMA_CHANNEL_REGISTER_SIZE	0x20
-
 struct tegra_dma;
 
 /*
  * tegra_dma_chip_data Tegra chip specific DMA data
  * @nr_channels: Number of channels available in the controller.
+ * @channel_reg_size: Channel register size/stride.
  * @max_dma_count: Maximum DMA transfer count supported by DMA controller.
  * @support_channel_pause: Support channel wise pause of dma.
+ * @support_separate_wcount_reg: Support separate word count register.
  */
 struct tegra_dma_chip_data {
 	int nr_channels;
+	int channel_reg_size;
 	int max_dma_count;
 	bool support_channel_pause;
+	bool support_separate_wcount_reg;
 };
 
 /* DMA channel registers */
@@ -133,6 +139,7 @@ struct tegra_dma_channel_regs {
 	unsigned long	apb_ptr;
 	unsigned long	ahb_seq;
 	unsigned long	apb_seq;
+	unsigned long	wcount;
 };
 
 /*
@@ -426,6 +433,8 @@ static void tegra_dma_start(struct tegra_dma_channel *tdc,
 	tdc_write(tdc, TEGRA_APBDMA_CHAN_APBPTR, ch_regs->apb_ptr);
 	tdc_write(tdc, TEGRA_APBDMA_CHAN_AHBSEQ, ch_regs->ahb_seq);
 	tdc_write(tdc, TEGRA_APBDMA_CHAN_AHBPTR, ch_regs->ahb_ptr);
+	if (tdc->tdma->chip_data->support_separate_wcount_reg)
+		tdc_write(tdc, TEGRA_APBDMA_CHAN_WCOUNT, ch_regs->wcount);
 
 	/* Start DMA */
 	tdc_write(tdc, TEGRA_APBDMA_CHAN_CSR,
@@ -465,6 +474,9 @@ static void tegra_dma_configure_for_next(struct tegra_dma_channel *tdc,
 	/* Safe to program new configuration */
 	tdc_write(tdc, TEGRA_APBDMA_CHAN_APBPTR, nsg_req->ch_regs.apb_ptr);
 	tdc_write(tdc, TEGRA_APBDMA_CHAN_AHBPTR, nsg_req->ch_regs.ahb_ptr);
+	if (tdc->tdma->chip_data->support_separate_wcount_reg)
+		tdc_write(tdc, TEGRA_APBDMA_CHAN_WCOUNT,
+						nsg_req->ch_regs.wcount);
 	tdc_write(tdc, TEGRA_APBDMA_CHAN_CSR,
 				nsg_req->ch_regs.csr | TEGRA_APBDMA_CSR_ENB);
 	nsg_req->configured = true;
@@ -718,6 +730,7 @@ static void tegra_dma_terminate_all(struct dma_chan *dc)
 	struct tegra_dma_desc *dma_desc;
 	unsigned long flags;
 	unsigned long status;
+	unsigned long wcount;
 	bool was_busy;
 
 	spin_lock_irqsave(&tdc->lock, flags);
@@ -738,6 +751,10 @@ static void tegra_dma_terminate_all(struct dma_chan *dc)
 		tdc->isr_handler(tdc, true);
 		status = tdc_read(tdc, TEGRA_APBDMA_CHAN_STATUS);
 	}
+	if (tdc->tdma->chip_data->support_separate_wcount_reg)
+		wcount = tdc_read(tdc, TEGRA_APBDMA_CHAN_WORD_TRANSFER);
+	else
+		wcount = status;
 
 	was_busy = tdc->busy;
 	tegra_dma_stop(tdc);
@@ -746,7 +763,7 @@ static void tegra_dma_terminate_all(struct dma_chan *dc)
 		sgreq = list_first_entry(&tdc->pending_sg_req,
 					typeof(*sgreq), node);
 		sgreq->dma_desc->bytes_transferred +=
-				get_current_xferred_count(tdc, sgreq, status);
+				get_current_xferred_count(tdc, sgreq, wcount);
 	}
 	tegra_dma_resume(tdc);
 
@@ -908,6 +925,17 @@ static int get_transfer_param(struct tegra_dma_channel *tdc,
 	return -EINVAL;
 }
 
+static void tegra_dma_prep_wcount(struct tegra_dma_channel *tdc,
+	struct tegra_dma_channel_regs *ch_regs, u32 len)
+{
+	u32 len_field = (len - 4) & 0xFFFC;
+
+	if (tdc->tdma->chip_data->support_separate_wcount_reg)
+		ch_regs->wcount = len_field;
+	else
+		ch_regs->csr |= len_field;
+}
+
 static struct dma_async_tx_descriptor *tegra_dma_prep_slave_sg(
 	struct dma_chan *dc, struct scatterlist *sgl, unsigned int sg_len,
 	enum dma_transfer_direction direction, unsigned long flags,
@@ -991,7 +1019,8 @@ static struct dma_async_tx_descriptor *tegra_dma_prep_slave_sg(
 
 		sg_req->ch_regs.apb_ptr = apb_ptr;
 		sg_req->ch_regs.ahb_ptr = mem;
-		sg_req->ch_regs.csr = csr | ((len - 4) & 0xFFFC);
+		sg_req->ch_regs.csr = csr;
+		tegra_dma_prep_wcount(tdc, &sg_req->ch_regs, len);
 		sg_req->ch_regs.apb_seq = apb_seq;
 		sg_req->ch_regs.ahb_seq = ahb_seq;
 		sg_req->configured = false;
@@ -1120,7 +1149,8 @@ static struct dma_async_tx_descriptor *tegra_dma_prep_dma_cyclic(
 		ahb_seq |= get_burst_size(tdc, burst_size, slave_bw, len);
 		sg_req->ch_regs.apb_ptr = apb_ptr;
 		sg_req->ch_regs.ahb_ptr = mem;
-		sg_req->ch_regs.csr = csr | ((len - 4) & 0xFFFC);
+		sg_req->ch_regs.csr = csr;
+		tegra_dma_prep_wcount(tdc, &sg_req->ch_regs, len);
 		sg_req->ch_regs.apb_seq = apb_seq;
 		sg_req->ch_regs.ahb_seq = ahb_seq;
 		sg_req->configured = false;
@@ -1234,27 +1264,45 @@ static struct dma_chan *tegra_dma_of_xlate(struct of_phandle_args *dma_spec,
 /* Tegra20 specific DMA controller information */
 static const struct tegra_dma_chip_data tegra20_dma_chip_data = {
 	.nr_channels		= 16,
+	.channel_reg_size	= 0x20,
 	.max_dma_count		= 1024UL * 64,
 	.support_channel_pause	= false,
+	.support_separate_wcount_reg = false,
 };
 
 /* Tegra30 specific DMA controller information */
 static const struct tegra_dma_chip_data tegra30_dma_chip_data = {
 	.nr_channels		= 32,
+	.channel_reg_size	= 0x20,
 	.max_dma_count		= 1024UL * 64,
 	.support_channel_pause	= false,
+	.support_separate_wcount_reg = false,
 };
 
 /* Tegra114 specific DMA controller information */
 static const struct tegra_dma_chip_data tegra114_dma_chip_data = {
 	.nr_channels		= 32,
+	.channel_reg_size	= 0x20,
 	.max_dma_count		= 1024UL * 64,
 	.support_channel_pause	= true,
+	.support_separate_wcount_reg = false,
+};
+
+/* Tegra148 specific DMA controller information */
+static const struct tegra_dma_chip_data tegra148_dma_chip_data = {
+	.nr_channels		= 32,
+	.channel_reg_size	= 0x40,
+	.max_dma_count		= 1024UL * 64,
+	.support_channel_pause	= true,
+	.support_separate_wcount_reg = true,
 };
 
 
 static const struct of_device_id tegra_dma_of_match[] = {
 	{
+		.compatible = "nvidia,tegra148-apbdma",
+		.data = &tegra148_dma_chip_data,
+	}, {
 		.compatible = "nvidia,tegra114-apbdma",
 		.data = &tegra114_dma_chip_data,
 	}, {
@@ -1348,7 +1396,7 @@ static int tegra_dma_probe(struct platform_device *pdev)
 		struct tegra_dma_channel *tdc = &tdma->channels[i];
 
 		tdc->chan_base_offset = TEGRA_APBDMA_CHANNEL_BASE_ADD_OFFSET +
-					i * TEGRA_APBDMA_CHANNEL_REGISTER_SIZE;
+					i * cdata->channel_reg_size;
 
 		res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
 		if (!res) {

drivers/dma/virt-dma.h

@@ -84,10 +84,12 @@ static inline bool vchan_issue_pending(struct virt_dma_chan *vc)
 static inline void vchan_cookie_complete(struct virt_dma_desc *vd)
 {
 	struct virt_dma_chan *vc = to_virt_chan(vd->tx.chan);
+	dma_cookie_t cookie;
 
+	cookie = vd->tx.cookie;
 	dma_cookie_complete(&vd->tx);
 	dev_vdbg(vc->chan.device->dev, "txd %p[%x]: marked complete\n",
-		vd, vd->tx.cookie);
+		 vd, cookie);
 	list_add_tail(&vd->node, &vc->desc_completed);
 
 	tasklet_schedule(&vc->task);

include/linux/dmaengine.h

@@ -257,7 +257,7 @@ struct dma_chan_percpu {
  * @dev: class device for sysfs
  * @device_node: used to add this to the device chan list
  * @local: per-cpu pointer to a struct dma_chan_percpu
- * @client-count: how many clients are using this channel
+ * @client_count: how many clients are using this channel
  * @table_count: number of appearances in the mem-to-mem allocation table
  * @private: private data for certain client-channel associations
  */
@@ -279,10 +279,10 @@ struct dma_chan {
 
 /**
  * struct dma_chan_dev - relate sysfs device node to backing channel device
- * @chan - driver channel device
- * @device - sysfs device
- * @dev_id - parent dma_device dev_id
- * @idr_ref - reference count to gate release of dma_device dev_id
+ * @chan: driver channel device
+ * @device: sysfs device
+ * @dev_id: parent dma_device dev_id
+ * @idr_ref: reference count to gate release of dma_device dev_id
  */
 struct dma_chan_dev {
 	struct dma_chan *chan;
@@ -306,9 +306,8 @@ enum dma_slave_buswidth {
 /**
  * struct dma_slave_config - dma slave channel runtime config
  * @direction: whether the data shall go in or out on this slave
- * channel, right now. DMA_TO_DEVICE and DMA_FROM_DEVICE are
- * legal values, DMA_BIDIRECTIONAL is not acceptable since we
- * need to differentiate source and target addresses.
+ * channel, right now. DMA_MEM_TO_DEV and DMA_DEV_TO_MEM are
+ * legal values.
  * @src_addr: this is the physical address where DMA slave data
  * should be read (RX), if the source is memory this argument is
  * ignored.

include/linux/platform_data/dma-imx-sdma.h

@@ -43,6 +43,11 @@ struct sdma_script_start_addrs {
 	s32 dptc_dvfs_addr;
 	s32 utra_addr;
 	s32 ram_code_start_addr;
+	/* End of v1 array */
+	s32 mcu_2_ssish_addr;
+	s32 ssish_2_mcu_addr;
+	s32 hdmi_dma_addr;
+	/* End of v2 array */
 };
 
 /**

include/linux/platform_data/dma-imx.h

@@ -39,6 +39,7 @@ enum sdma_peripheral_type {
 	IMX_DMATYPE_IPU_MEMORY,	/* IPU Memory */
 	IMX_DMATYPE_ASRC,	/* ASRC */
 	IMX_DMATYPE_ESAI,	/* ESAI */
+	IMX_DMATYPE_SSI_DUAL,	/* SSI Dual FIFO */
 };
 
 enum imx_dma_prio {

include/linux/platform_data/dma-mmp_tdma.h

 /*
- *  linux/arch/arm/mach-mmp/include/mach/sram.h
- *
  *  SRAM Memory Management
  *
  *  Copyright (c) 2011 Marvell Semiconductors Inc.
@@ -11,8 +9,8 @@
  *
  */
 
-#ifndef __ASM_ARCH_SRAM_H
-#define __ASM_ARCH_SRAM_H
+#ifndef __DMA_MMP_TDMA_H
+#define __DMA_MMP_TDMA_H
 
 #include <linux/genalloc.h>
 
@@ -32,4 +30,4 @@ struct sram_platdata {
 
 extern struct gen_pool *sram_get_gpool(char *pool_name);
 
-#endif /* __ASM_ARCH_SRAM_H */
+#endif /* __DMA_MMP_TDMA_H */

include/linux/platform_data/dma-mv_xor.h

 /*
- * arch/arm/plat-orion/include/plat/mv_xor.h
- *
  * Marvell XOR platform device data definition file.
  */
 
-#ifndef __PLAT_MV_XOR_H
-#define __PLAT_MV_XOR_H
+#ifndef __DMA_MV_XOR_H
+#define __DMA_MV_XOR_H
 
 #include <linux/dmaengine.h>
 #include <linux/mbus.h>

sound/soc/fsl/fsl_ssi.c

@@ -164,6 +164,7 @@ struct fsl_ssi_private {
 	bool baudclk_locked;
 	bool irq_stats;
 	bool offline_config;
+	bool use_dual_fifo;
 	u8 i2s_mode;
 	spinlock_t baudclk_lock;
 	struct clk *baudclk;
@@ -721,6 +722,12 @@ static int fsl_ssi_setup(struct fsl_ssi_private *ssi_private)
 				CCSR_SSI_SxCCR_DC(2));
 	}
 
+	if (ssi_private->use_dual_fifo) {
+		write_ssi_mask(&ssi->srcr, 0, CCSR_SSI_SRCR_RFEN1);
+		write_ssi_mask(&ssi->stcr, 0, CCSR_SSI_STCR_TFEN1);
+		write_ssi_mask(&ssi->scr, 0, CCSR_SSI_SCR_TCH_EN);
+	}
+
 	return 0;
 }
 
@@ -752,6 +759,15 @@ static int fsl_ssi_startup(struct snd_pcm_substream *substream,
 		spin_unlock_irqrestore(&ssi_private->baudclk_lock, flags);
 	}
 
+	/* When using dual fifo mode, it is safer to ensure an even period
+	 * size. If appearing to an odd number while DMA always starts its
+	 * task from fifo0, fifo1 would be neglected at the end of each
+	 * period. But SSI would still access fifo1 with an invalid data.
+	 */
+	if (ssi_private->use_dual_fifo)
+		snd_pcm_hw_constraint_step(substream->runtime, 0,
+				SNDRV_PCM_HW_PARAM_PERIOD_SIZE, 2);
+
 	return 0;
 }
 
@@ -1370,7 +1386,7 @@ static int fsl_ssi_probe(struct platform_device *pdev)
 
 	if (hw_type == FSL_SSI_MX21 || hw_type == FSL_SSI_MX51 ||
 			hw_type == FSL_SSI_MX35) {
-		u32 dma_events[2];
+		u32 dma_events[2], dmas[4];
 		ssi_private->ssi_on_imx = true;
 
 		ssi_private->clk = devm_clk_get(&pdev->dev, NULL);
@@ -1426,6 +1442,16 @@ static int fsl_ssi_probe(struct platform_device *pdev)
 				goto error_clk;
 			}
 		}
+		/* Should this be merge with the above? */
+		if (!of_property_read_u32_array(pdev->dev.of_node, "dmas", dmas, 4)
+				&& dmas[2] == IMX_DMATYPE_SSI_DUAL) {
+			ssi_private->use_dual_fifo = true;
+			/* When using dual fifo mode, we need to keep watermark
+			 * as even numbers due to dma script limitation.
+			 */
+			ssi_private->dma_params_tx.maxburst &= ~0x1;
+			ssi_private->dma_params_rx.maxburst &= ~0x1;
+		}
 
 		shared = of_device_is_compatible(of_get_parent(np),
 			    "fsl,spba-bus");