Merge tag 'dmaengine-4.4-rc1' of git://git.infradead.org/users/vkoul/slave-dma

Pull dmaengine updates from Vinod Koul:
 "This time we have a very typical update which is mostly fixes and
  updates to drivers and no new drivers.

   - the biggest change is coming from Peter for edma cleanup which even
     caused some last minute regression, things seem settled now
   - idma64 and dw updates
   - iotdma updates
   - module autoload fixes for various drivers
   - scatter gather support for hdmac"

* tag 'dmaengine-4.4-rc1' of git://git.infradead.org/users/vkoul/slave-dma: (77 commits)
  dmaengine: edma: Add dummy driver skeleton for edma3-tptc
  Revert "ARM: DTS: am33xx: Use the new DT bindings for the eDMA3"
  Revert "ARM: DTS: am437x: Use the new DT bindings for the eDMA3"
  dmaengine: dw: some Intel devices has no memcpy support
  dmaengine: dw: platform: provide platform data for Intel
  dmaengine: dw: don't override platform data with autocfg
  dmaengine: hdmac: Add scatter-gathered memset support
  dmaengine: hdmac: factorise memset descriptor allocation
  dmaengine: virt-dma: Fix kernel-doc annotations
  ARM: DTS: am437x: Use the new DT bindings for the eDMA3
  ARM: DTS: am33xx: Use the new DT bindings for the eDMA3
  dmaengine: edma: New device tree binding
  dmaengine: Kconfig: edma: Select TI_DMA_CROSSBAR in case of ARCH_OMAP
  dmaengine: ti-dma-crossbar: Add support for crossbar on AM33xx/AM43xx
  dmaengine: edma: Merge the of parsing functions
  dmaengine: edma: Do not allocate memory for edma_rsv_info in case of DT boot
  dmaengine: edma: Refactor the dma device and channel struct initialization
  dmaengine: edma: Get qDMA channel information from HW also
  dmaengine: edma: Merge map_dmach_to_queue into assign_channel_eventq
  dmaengine: edma: Correct PaRAM access function names (_parm_ to _param_)
  ...

Documentation/devicetree/bindings/dma/ti-dma-crossbar.txt

@@ -2,9 +2,10 @@ Texas Instruments DMA Crossbar (DMA request router)
 
 Required properties:
 - compatible:	"ti,dra7-dma-crossbar" for DRA7xx DMA crossbar
+		"ti,am335x-edma-crossbar" for AM335x and AM437x
 - reg:		Memory map for accessing module
-- #dma-cells:	Should be set to <1>.
-		Clients should use the crossbar request number (input)
+- #dma-cells:	Should be set to to match with the DMA controller's dma-cells
+		for ti,dra7-dma-crossbar and <3> for ti,am335x-edma-crossbar.
 - dma-requests:	Number of DMA requests the crossbar can receive
 - dma-masters:	phandle pointing to the DMA controller
 
@@ -14,6 +15,15 @@ The DMA controller node need to have the following poroperties:
 Optional properties:
 - ti,dma-safe-map: Safe routing value for unused request lines
 
+Notes:
+When requesting channel via ti,dra7-dma-crossbar, the DMA clinet must request
+the DMA event number as crossbar ID (input to the DMA crossbar).
+
+For ti,am335x-edma-crossbar: the meaning of parameters of dmas for clients:
+dmas = <&edma_xbar 12 0 1>; where <12> is the DMA request number, <0> is the TC
+the event should be assigned and <1> is the mux selection for in the crossbar.
+When mux 0 is used the DMA channel can be requested directly from edma node.
+
 Example:
 
 /* DMA controller */
@@ -47,6 +57,7 @@ uart1: serial@4806a000 {
 	ti,hwmods = "uart1";
 	clock-frequency = <48000000>;
 	status = "disabled";
+	/* Requesting crossbar input 49 and 50 */
 	dmas = <&sdma_xbar 49>, <&sdma_xbar 50>;
 	dma-names = "tx", "rx";
 };

Documentation/devicetree/bindings/dma/ti-edma.txt

-TI EDMA
+Texas Instruments eDMA
+
+The eDMA3 consists of two components: Channel controller (CC) and Transfer
+Controller(s) (TC). The CC is the main entry for DMA users since it is
+responsible for the DMA channel handling, while the TCs are responsible to
+execute the actual DMA tansfer.
+
+------------------------------------------------------------------------------
+eDMA3 Channel Controller
+
+Required properties:
+- compatible:	"ti,edma3-tpcc" for the channel controller(s)
+- #dma-cells:	Should be set to <2>. The first number is the DMA request
+		number and the second is the TC the channel is serviced on.
+- reg:		Memory map of eDMA CC
+- reg-names:	"edma3_cc"
+- interrupts:	Interrupt lines for CCINT, MPERR and CCERRINT.
+- interrupt-names: "edma3_ccint", "emda3_mperr" and "edma3_ccerrint"
+- ti,tptcs:	List of TPTCs associated with the eDMA in the following form:
+		<&tptc_phandle TC_priority_number>. The highest priority is 0.
+
+Optional properties:
+- ti,hwmods:	Name of the hwmods associated to the eDMA CC
+- ti,edma-memcpy-channels: List of channels allocated to be used for memcpy, iow
+		these channels will be SW triggered channels. The list must
+		contain 16 bits numbers, see example.
+- ti,edma-reserved-slot-ranges: PaRAM slot ranges which should not be used by
+		the driver, they are allocated to be used by for example the
+		DSP. See example.
+
+------------------------------------------------------------------------------
+eDMA3 Transfer Controller
+
+Required properties:
+- compatible:	"ti,edma3-tptc" for the transfer controller(s)
+- reg:		Memory map of eDMA TC
+- interrupts:	Interrupt number for TCerrint.
+
+Optional properties:
+- ti,hwmods:	Name of the hwmods associated to the given eDMA TC
+- interrupt-names: "edma3_tcerrint"
+
+------------------------------------------------------------------------------
+Example:
+
+edma: edma@49000000 {
+	compatible = "ti,edma3-tpcc";
+	ti,hwmods = "tpcc";
+	reg =	<0x49000000 0x10000>;
+	reg-names = "edma3_cc";
+	interrupts = <12 13 14>;
+	interrupt-names = "edma3_ccint", "emda3_mperr", "edma3_ccerrint";
+	dma-requests = <64>;
+	#dma-cells = <2>;
+
+	ti,tptcs = <&edma_tptc0 7>, <&edma_tptc1 7>, <&edma_tptc2 0>;
+
+	/* Channel 20 and 21 is allocated for memcpy */
+	ti,edma-memcpy-channels = /bits/ 16 <20 21>;
+	/* The following PaRAM slots are reserved: 35-45 and 100-110 */
+	ti,edma-reserved-slot-ranges = /bits/ 16 <35 10>,
+				       /bits/ 16 <100 10>;
+};
+
+edma_tptc0: tptc@49800000 {
+	compatible = "ti,edma3-tptc";
+	ti,hwmods = "tptc0";
+	reg =	<0x49800000 0x100000>;
+	interrupts = <112>;
+	interrupt-names = "edm3_tcerrint";
+};
+
+edma_tptc1: tptc@49900000 {
+	compatible = "ti,edma3-tptc";
+	ti,hwmods = "tptc1";
+	reg =	<0x49900000 0x100000>;
+	interrupts = <113>;
+	interrupt-names = "edm3_tcerrint";
+};
+
+edma_tptc2: tptc@49a00000 {
+	compatible = "ti,edma3-tptc";
+	ti,hwmods = "tptc2";
+	reg =	<0x49a00000 0x100000>;
+	interrupts = <114>;
+	interrupt-names = "edm3_tcerrint";
+};
+
+sham: sham@53100000 {
+	compatible = "ti,omap4-sham";
+	ti,hwmods = "sham";
+	reg = <0x53100000 0x200>;
+	interrupts = <109>;
+	/* DMA channel 36 executed on eDMA TC0 - low priority queue */
+	dmas = <&edma 36 0>;
+	dma-names = "rx";
+};
+
+mcasp0: mcasp@48038000 {
+	compatible = "ti,am33xx-mcasp-audio";
+	ti,hwmods = "mcasp0";
+	reg = <0x48038000 0x2000>,
+		<0x46000000 0x400000>;
+	reg-names = "mpu", "dat";
+	interrupts = <80>, <81>;
+	interrupt-names = "tx", "rx";
+	status = "disabled";
+	/* DMA channels 8 and 9 executed on eDMA TC2 - high priority queue */
+	dmas = <&edma 8 2>,
+	       <&edma 9 2>;
+	dma-names = "tx", "rx";
+};
+
+------------------------------------------------------------------------------
+DEPRECATED binding, new DTS files must use the ti,edma3-tpcc/ti,edma3-tptc
+binding.
 
 Required properties:
 - compatible : "ti,edma3"

arch/arm/Kconfig

@@ -737,7 +737,6 @@ config ARCH_DAVINCI
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_IRQ_CHIP
 	select HAVE_IDE
-	select TI_PRIV_EDMA
 	select USE_OF
 	select ZONE_DMA
 	help

arch/arm/common/Kconfig

@@ -17,6 +17,3 @@ config SHARP_PARAM
 
 config SHARP_SCOOP
 	bool
-
-config TI_PRIV_EDMA
-	bool

arch/arm/common/Makefile

@@ -15,6 +15,5 @@ obj-$(CONFIG_MCPM)		+= mcpm_head.o mcpm_entry.o mcpm_platsmp.o vlock.o
 CFLAGS_REMOVE_mcpm_entry.o	= -pg
 AFLAGS_mcpm_head.o		:= -march=armv7-a
 AFLAGS_vlock.o			:= -march=armv7-a
-obj-$(CONFIG_TI_PRIV_EDMA)	+= edma.o
 obj-$(CONFIG_BL_SWITCHER)	+= bL_switcher.o
 obj-$(CONFIG_BL_SWITCHER_DUMMY_IF) += bL_switcher_dummy_if.o

arch/arm/mach-davinci/devices-da8xx.c

@@ -147,150 +147,118 @@ static s8 da850_queue_priority_mapping[][2] = {
 	{-1, -1}
 };
 
-static struct edma_soc_info da830_edma_cc0_info = {
+static struct edma_soc_info da8xx_edma0_pdata = {
 	.queue_priority_mapping	= da8xx_queue_priority_mapping,
 	.default_queue		= EVENTQ_1,
 };
 
-static struct edma_soc_info *da830_edma_info[EDMA_MAX_CC] = {
-	&da830_edma_cc0_info,
+static struct edma_soc_info da850_edma1_pdata = {
+	.queue_priority_mapping	= da850_queue_priority_mapping,
+	.default_queue		= EVENTQ_0,
 };
 
-static struct edma_soc_info da850_edma_cc_info[] = {
+static struct resource da8xx_edma0_resources[] = {
 	{
-		.queue_priority_mapping	= da8xx_queue_priority_mapping,
-		.default_queue		= EVENTQ_1,
-	},
-	{
-		.queue_priority_mapping	= da850_queue_priority_mapping,
-		.default_queue		= EVENTQ_0,
-	},
-};
-
-static struct edma_soc_info *da850_edma_info[EDMA_MAX_CC] = {
-	&da850_edma_cc_info[0],
-	&da850_edma_cc_info[1],
-};
-
-static struct resource da830_edma_resources[] = {
-	{
-		.name	= "edma_cc0",
+		.name	= "edma3_cc",
 		.start	= DA8XX_TPCC_BASE,
 		.end	= DA8XX_TPCC_BASE + SZ_32K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc0",
+		.name	= "edma3_tc0",
 		.start	= DA8XX_TPTC0_BASE,
 		.end	= DA8XX_TPTC0_BASE + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc1",
+		.name	= "edma3_tc1",
 		.start	= DA8XX_TPTC1_BASE,
 		.end	= DA8XX_TPTC1_BASE + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma0",
+		.name	= "edma3_ccint",
 		.start	= IRQ_DA8XX_CCINT0,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
-		.name	= "edma0_err",
+		.name	= "edma3_ccerrint",
 		.start	= IRQ_DA8XX_CCERRINT,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
 
-static struct resource da850_edma_resources[] = {
-	{
-		.name	= "edma_cc0",
-		.start	= DA8XX_TPCC_BASE,
-		.end	= DA8XX_TPCC_BASE + SZ_32K - 1,
-		.flags	= IORESOURCE_MEM,
-	},
-	{
-		.name	= "edma_tc0",
-		.start	= DA8XX_TPTC0_BASE,
-		.end	= DA8XX_TPTC0_BASE + SZ_1K - 1,
-		.flags	= IORESOURCE_MEM,
-	},
-	{
-		.name	= "edma_tc1",
-		.start	= DA8XX_TPTC1_BASE,
-		.end	= DA8XX_TPTC1_BASE + SZ_1K - 1,
-		.flags	= IORESOURCE_MEM,
-	},
+static struct resource da850_edma1_resources[] = {
 	{
-		.name	= "edma_cc1",
+		.name	= "edma3_cc",
 		.start	= DA850_TPCC1_BASE,
 		.end	= DA850_TPCC1_BASE + SZ_32K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc2",
+		.name	= "edma3_tc0",
 		.start	= DA850_TPTC2_BASE,
 		.end	= DA850_TPTC2_BASE + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma0",
-		.start	= IRQ_DA8XX_CCINT0,
-		.flags	= IORESOURCE_IRQ,
-	},
-	{
-		.name	= "edma0_err",
-		.start	= IRQ_DA8XX_CCERRINT,
-		.flags	= IORESOURCE_IRQ,
-	},
-	{
-		.name	= "edma1",
+		.name	= "edma3_ccint",
 		.start	= IRQ_DA850_CCINT1,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
-		.name	= "edma1_err",
+		.name	= "edma3_ccerrint",
 		.start	= IRQ_DA850_CCERRINT1,
 		.flags	= IORESOURCE_IRQ,
 	},
 };
 
-static struct platform_device da830_edma_device = {
+static const struct platform_device_info da8xx_edma0_device __initconst = {
 	.name		= "edma",
-	.id		= -1,
-	.dev = {
-		.platform_data = da830_edma_info,
-	},
-	.num_resources	= ARRAY_SIZE(da830_edma_resources),
-	.resource	= da830_edma_resources,
+	.id		= 0,
+	.dma_mask	= DMA_BIT_MASK(32),
+	.res		= da8xx_edma0_resources,
+	.num_res	= ARRAY_SIZE(da8xx_edma0_resources),
+	.data		= &da8xx_edma0_pdata,
+	.size_data	= sizeof(da8xx_edma0_pdata),
 };
 
-static struct platform_device da850_edma_device = {
+static const struct platform_device_info da850_edma1_device __initconst = {
 	.name		= "edma",
-	.id		= -1,
-	.dev = {
-		.platform_data = da850_edma_info,
-	},
-	.num_resources	= ARRAY_SIZE(da850_edma_resources),
-	.resource	= da850_edma_resources,
+	.id		= 1,
+	.dma_mask	= DMA_BIT_MASK(32),
+	.res		= da850_edma1_resources,
+	.num_res	= ARRAY_SIZE(da850_edma1_resources),
+	.data		= &da850_edma1_pdata,
+	.size_data	= sizeof(da850_edma1_pdata),
 };
 
 int __init da830_register_edma(struct edma_rsv_info *rsv)
 {
-	da830_edma_cc0_info.rsv = rsv;
+	struct platform_device *edma_pdev;
+
+	da8xx_edma0_pdata.rsv = rsv;
 
-	return platform_device_register(&da830_edma_device);
+	edma_pdev = platform_device_register_full(&da8xx_edma0_device);
+	return IS_ERR(edma_pdev) ? PTR_ERR(edma_pdev) : 0;
 }
 
 int __init da850_register_edma(struct edma_rsv_info *rsv[2])
 {
+	struct platform_device *edma_pdev;
+
 	if (rsv) {
-		da850_edma_cc_info[0].rsv = rsv[0];
-		da850_edma_cc_info[1].rsv = rsv[1];
+		da8xx_edma0_pdata.rsv = rsv[0];
+		da850_edma1_pdata.rsv = rsv[1];
 	}
 
-	return platform_device_register(&da850_edma_device);
+	edma_pdev = platform_device_register_full(&da8xx_edma0_device);
+	if (IS_ERR(edma_pdev)) {
+		pr_warn("%s: Failed to register eDMA0\n", __func__);
+		return PTR_ERR(edma_pdev);
+	}
+	edma_pdev = platform_device_register_full(&da850_edma1_device);
+	return IS_ERR(edma_pdev) ? PTR_ERR(edma_pdev) : 0;
 }
 
 static struct resource da8xx_i2c_resources0[] = {

arch/arm/mach-davinci/dm355.c

@@ -569,61 +569,58 @@ static u8 dm355_default_priorities[DAVINCI_N_AINTC_IRQ] = {
 
 /*----------------------------------------------------------------------*/
 
-static s8
-queue_priority_mapping[][2] = {
+static s8 queue_priority_mapping[][2] = {
 	/* {event queue no, Priority} */
 	{0, 3},
 	{1, 7},
 	{-1, -1},
 };
 
-static struct edma_soc_info edma_cc0_info = {
+static struct edma_soc_info dm355_edma_pdata = {
 	.queue_priority_mapping	= queue_priority_mapping,
 	.default_queue		= EVENTQ_1,
 };
 
-static struct edma_soc_info *dm355_edma_info[EDMA_MAX_CC] = {
-       &edma_cc0_info,
-};
-
 static struct resource edma_resources[] = {
 	{
-		.name	= "edma_cc0",
+		.name	= "edma3_cc",
 		.start	= 0x01c00000,
 		.end	= 0x01c00000 + SZ_64K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc0",
+		.name	= "edma3_tc0",
 		.start	= 0x01c10000,
 		.end	= 0x01c10000 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc1",
+		.name	= "edma3_tc1",
 		.start	= 0x01c10400,
 		.end	= 0x01c10400 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma0",
+		.name	= "edma3_ccint",
 		.start	= IRQ_CCINT0,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
-		.name	= "edma0_err",
+		.name	= "edma3_ccerrint",
 		.start	= IRQ_CCERRINT,
 		.flags	= IORESOURCE_IRQ,
 	},
 	/* not using (or muxing) TC*_ERR */
 };
 
-static struct platform_device dm355_edma_device = {
-	.name			= "edma",
-	.id			= 0,
-	.dev.platform_data	= dm355_edma_info,
-	.num_resources		= ARRAY_SIZE(edma_resources),
-	.resource		= edma_resources,
+static const struct platform_device_info dm355_edma_device __initconst = {
+	.name		= "edma",
+	.id		= 0,
+	.dma_mask	= DMA_BIT_MASK(32),
+	.res		= edma_resources,
+	.num_res	= ARRAY_SIZE(edma_resources),
+	.data		= &dm355_edma_pdata,
+	.size_data	= sizeof(dm355_edma_pdata),
 };
 
 static struct resource dm355_asp1_resources[] = {
@@ -1062,13 +1059,18 @@ int __init dm355_init_video(struct vpfe_config *vpfe_cfg,
 
 static int __init dm355_init_devices(void)
 {
+	struct platform_device *edma_pdev;
 	int ret = 0;
 
 	if (!cpu_is_davinci_dm355())
 		return 0;
 
 	davinci_cfg_reg(DM355_INT_EDMA_CC);
-	platform_device_register(&dm355_edma_device);
+	edma_pdev = platform_device_register_full(&dm355_edma_device);
+	if (IS_ERR(edma_pdev)) {
+		pr_warn("%s: Failed to register eDMA\n", __func__);
+		return PTR_ERR(edma_pdev);
+	}
 
 	ret = davinci_init_wdt();
 	if (ret)

arch/arm/mach-davinci/dm365.c

@@ -853,8 +853,7 @@ static u8 dm365_default_priorities[DAVINCI_N_AINTC_IRQ] = {
 };
 
 /* Four Transfer Controllers on DM365 */
-static s8
-dm365_queue_priority_mapping[][2] = {
+static s8 dm365_queue_priority_mapping[][2] = {
 	/* {event queue no, Priority} */
 	{0, 7},
 	{1, 7},
@@ -863,53 +862,49 @@ dm365_queue_priority_mapping[][2] = {
 	{-1, -1},
 };
 
-static struct edma_soc_info edma_cc0_info = {
+static struct edma_soc_info dm365_edma_pdata = {
 	.queue_priority_mapping	= dm365_queue_priority_mapping,
 	.default_queue		= EVENTQ_3,
 };
 
-static struct edma_soc_info *dm365_edma_info[EDMA_MAX_CC] = {
-	&edma_cc0_info,
-};
-
 static struct resource edma_resources[] = {
 	{
-		.name	= "edma_cc0",
+		.name	= "edma3_cc",
 		.start	= 0x01c00000,
 		.end	= 0x01c00000 + SZ_64K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc0",
+		.name	= "edma3_tc0",
 		.start	= 0x01c10000,
 		.end	= 0x01c10000 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc1",
+		.name	= "edma3_tc1",
 		.start	= 0x01c10400,
 		.end	= 0x01c10400 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc2",
+		.name	= "edma3_tc2",
 		.start	= 0x01c10800,
 		.end	= 0x01c10800 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc3",
+		.name	= "edma3_tc3",
 		.start	= 0x01c10c00,
 		.end	= 0x01c10c00 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma0",
+		.name	= "edma3_ccint",
 		.start	= IRQ_CCINT0,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
-		.name	= "edma0_err",
+		.name	= "edma3_ccerrint",
 		.start	= IRQ_CCERRINT,
 		.flags	= IORESOURCE_IRQ,
 	},
@@ -919,7 +914,7 @@ static struct resource edma_resources[] = {
 static struct platform_device dm365_edma_device = {
 	.name			= "edma",
 	.id			= 0,
-	.dev.platform_data	= dm365_edma_info,
+	.dev.platform_data	= &dm365_edma_pdata,
 	.num_resources		= ARRAY_SIZE(edma_resources),
 	.resource		= edma_resources,
 };

arch/arm/mach-davinci/dm644x.c

@@ -498,61 +498,58 @@ static u8 dm644x_default_priorities[DAVINCI_N_AINTC_IRQ] = {
 
 /*----------------------------------------------------------------------*/
 
-static s8
-queue_priority_mapping[][2] = {
+static s8 queue_priority_mapping[][2] = {
 	/* {event queue no, Priority} */
 	{0, 3},
 	{1, 7},
 	{-1, -1},
 };
 
-static struct edma_soc_info edma_cc0_info = {
+static struct edma_soc_info dm644x_edma_pdata = {
 	.queue_priority_mapping	= queue_priority_mapping,
 	.default_queue		= EVENTQ_1,
 };
 
-static struct edma_soc_info *dm644x_edma_info[EDMA_MAX_CC] = {
-	&edma_cc0_info,
-};
-
 static struct resource edma_resources[] = {
 	{
-		.name	= "edma_cc0",
+		.name	= "edma3_cc",
 		.start	= 0x01c00000,
 		.end	= 0x01c00000 + SZ_64K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc0",
+		.name	= "edma3_tc0",
 		.start	= 0x01c10000,
 		.end	= 0x01c10000 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc1",
+		.name	= "edma3_tc1",
 		.start	= 0x01c10400,
 		.end	= 0x01c10400 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma0",
+		.name	= "edma3_ccint",
 		.start	= IRQ_CCINT0,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
-		.name	= "edma0_err",
+		.name	= "edma3_ccerrint",
 		.start	= IRQ_CCERRINT,
 		.flags	= IORESOURCE_IRQ,
 	},
 	/* not using TC*_ERR */
 };
 
-static struct platform_device dm644x_edma_device = {
-	.name			= "edma",
-	.id			= 0,
-	.dev.platform_data	= dm644x_edma_info,
-	.num_resources		= ARRAY_SIZE(edma_resources),
-	.resource		= edma_resources,
+static const struct platform_device_info dm644x_edma_device __initconst = {
+	.name		= "edma",
+	.id		= 0,
+	.dma_mask	= DMA_BIT_MASK(32),
+	.res		= edma_resources,
+	.num_res	= ARRAY_SIZE(edma_resources),
+	.data		= &dm644x_edma_pdata,
+	.size_data	= sizeof(dm644x_edma_pdata),
 };
 
 /* DM6446 EVM uses ASP0; line-out is a pair of RCA jacks */
@@ -950,12 +947,17 @@ int __init dm644x_init_video(struct vpfe_config *vpfe_cfg,
 
 static int __init dm644x_init_devices(void)
 {
+	struct platform_device *edma_pdev;
 	int ret = 0;
 
 	if (!cpu_is_davinci_dm644x())
 		return 0;
 
-	platform_device_register(&dm644x_edma_device);
+	edma_pdev = platform_device_register_full(&dm644x_edma_device);
+	if (IS_ERR(edma_pdev)) {
+		pr_warn("%s: Failed to register eDMA\n", __func__);
+		return PTR_ERR(edma_pdev);
+	}
 
 	platform_device_register(&dm644x_mdio_device);
 	platform_device_register(&dm644x_emac_device);

arch/arm/mach-davinci/dm646x.c

@@ -531,8 +531,7 @@ static u8 dm646x_default_priorities[DAVINCI_N_AINTC_IRQ] = {
 /*----------------------------------------------------------------------*/
 
 /* Four Transfer Controllers on DM646x */
-static s8
-dm646x_queue_priority_mapping[][2] = {
+static s8 dm646x_queue_priority_mapping[][2] = {
 	/* {event queue no, Priority} */
 	{0, 4},
 	{1, 0},
@@ -541,65 +540,63 @@ dm646x_queue_priority_mapping[][2] = {
 	{-1, -1},
 };
 
-static struct edma_soc_info edma_cc0_info = {
+static struct edma_soc_info dm646x_edma_pdata = {
 	.queue_priority_mapping	= dm646x_queue_priority_mapping,
 	.default_queue		= EVENTQ_1,
 };
 
-static struct edma_soc_info *dm646x_edma_info[EDMA_MAX_CC] = {
-	&edma_cc0_info,
-};
-
 static struct resource edma_resources[] = {
 	{
-		.name	= "edma_cc0",
+		.name	= "edma3_cc",
 		.start	= 0x01c00000,
 		.end	= 0x01c00000 + SZ_64K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc0",
+		.name	= "edma3_tc0",
 		.start	= 0x01c10000,
 		.end	= 0x01c10000 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc1",
+		.name	= "edma3_tc1",
 		.start	= 0x01c10400,
 		.end	= 0x01c10400 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc2",
+		.name	= "edma3_tc2",
 		.start	= 0x01c10800,
 		.end	= 0x01c10800 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma_tc3",
+		.name	= "edma3_tc3",
 		.start	= 0x01c10c00,
 		.end	= 0x01c10c00 + SZ_1K - 1,
 		.flags	= IORESOURCE_MEM,
 	},
 	{
-		.name	= "edma0",
+		.name	= "edma3_ccint",
 		.start	= IRQ_CCINT0,
 		.flags	= IORESOURCE_IRQ,
 	},
 	{
-		.name	= "edma0_err",
+		.name	= "edma3_ccerrint",
 		.start	= IRQ_CCERRINT,
 		.flags	= IORESOURCE_IRQ,
 	},
 	/* not using TC*_ERR */
 };
 
-static struct platform_device dm646x_edma_device = {
-	.name			= "edma",
-	.id			= 0,
-	.dev.platform_data	= dm646x_edma_info,
-	.num_resources		= ARRAY_SIZE(edma_resources),
-	.resource		= edma_resources,
+static const struct platform_device_info dm646x_edma_device __initconst = {
+	.name		= "edma",
+	.id		= 0,
+	.dma_mask	= DMA_BIT_MASK(32),
+	.res		= edma_resources,
+	.num_res	= ARRAY_SIZE(edma_resources),
+	.data		= &dm646x_edma_pdata,
+	.size_data	= sizeof(dm646x_edma_pdata),
 };
 
 static struct resource dm646x_mcasp0_resources[] = {
@@ -936,9 +933,12 @@ void dm646x_setup_vpif(struct vpif_display_config *display_config,
 
 int __init dm646x_init_edma(struct edma_rsv_info *rsv)
 {
-	edma_cc0_info.rsv = rsv;
+	struct platform_device *edma_pdev;
+
+	dm646x_edma_pdata.rsv = rsv;
 
-	return platform_device_register(&dm646x_edma_device);
+	edma_pdev = platform_device_register_full(&dm646x_edma_device);
+	return IS_ERR(edma_pdev) ? PTR_ERR(edma_pdev) : 0;
 }
 
 void __init dm646x_init(void)

arch/arm/mach-omap2/Kconfig

@@ -96,7 +96,6 @@ config ARCH_OMAP2PLUS
 	select OMAP_GPMC
 	select PINCTRL
 	select SOC_BUS
-	select TI_PRIV_EDMA
 	select OMAP_IRQCHIP
 	help
 	  Systems based on OMAP2, OMAP3, OMAP4 or OMAP5

arch/avr32/mach-at32ap/at32ap700x.c

@@ -603,18 +603,11 @@ static void __init genclk_init_parent(struct clk *clk)
 	clk->parent = parent;
 }
 
-static struct dw_dma_platform_data dw_dmac0_data = {
-	.nr_channels	= 3,
-	.block_size	= 4095U,
-	.nr_masters	= 2,
-	.data_width	= { 2, 2 },
-};
-
 static struct resource dw_dmac0_resource[] = {
 	PBMEM(0xff200000),
 	IRQ(2),
 };
-DEFINE_DEV_DATA(dw_dmac, 0);
+DEFINE_DEV(dw_dmac, 0);
 DEV_CLK(hclk, dw_dmac0, hsb, 10);
 
 /* --------------------------------------------------------------------

drivers/dma/Kconfig

@@ -229,7 +229,7 @@ config IMX_SDMA
 	  Support the i.MX SDMA engine. This engine is integrated into
 	  Freescale i.MX25/31/35/51/53/6 chips.
 
-config IDMA64
+config INTEL_IDMA64
 	tristate "Intel integrated DMA 64-bit support"
 	select DMA_ENGINE
 	select DMA_VIRTUAL_CHANNELS
@@ -486,7 +486,7 @@ config TI_EDMA
 	depends on ARCH_DAVINCI || ARCH_OMAP || ARCH_KEYSTONE
 	select DMA_ENGINE
 	select DMA_VIRTUAL_CHANNELS
-	select TI_PRIV_EDMA
+	select TI_DMA_CROSSBAR if ARCH_OMAP
 	default n
 	help
 	  Enable support for the TI EDMA controller. This DMA

drivers/dma/Makefile

@@ -34,7 +34,7 @@ obj-$(CONFIG_HSU_DMA) += hsu/
 obj-$(CONFIG_IMG_MDC_DMA) += img-mdc-dma.o
 obj-$(CONFIG_IMX_DMA) += imx-dma.o
 obj-$(CONFIG_IMX_SDMA) += imx-sdma.o
-obj-$(CONFIG_IDMA64) += idma64.o
+obj-$(CONFIG_INTEL_IDMA64) += idma64.o
 obj-$(CONFIG_INTEL_IOATDMA) += ioat/
 obj-$(CONFIG_INTEL_IOP_ADMA) += iop-adma.o
 obj-$(CONFIG_INTEL_MIC_X100_DMA) += mic_x100_dma.o

drivers/dma/acpi-dma.c

@@ -161,10 +161,8 @@ int acpi_dma_controller_register(struct device *dev,
 		return -EINVAL;
 
 	/* Check if the device was enumerated by ACPI */
-	if (!ACPI_HANDLE(dev))
-		return -EINVAL;
-
-	if (acpi_bus_get_device(ACPI_HANDLE(dev), &adev))
+	adev = ACPI_COMPANION(dev);
+	if (!adev)
 		return -EINVAL;
 
 	adma = kzalloc(sizeof(*adma), GFP_KERNEL);
@@ -359,10 +357,11 @@ struct dma_chan *acpi_dma_request_slave_chan_by_index(struct device *dev,
 	int found;
 
 	/* Check if the device was enumerated by ACPI */
-	if (!dev || !ACPI_HANDLE(dev))
+	if (!dev)
 		return ERR_PTR(-ENODEV);
 
-	if (acpi_bus_get_device(ACPI_HANDLE(dev), &adev))
+	adev = ACPI_COMPANION(dev);
+	if (!adev)
 		return ERR_PTR(-ENODEV);
 
 	memset(&pdata, 0, sizeof(pdata));

drivers/dma/at_hdmac.c

@@ -458,10 +458,10 @@ atc_chain_complete(struct at_dma_chan *atchan, struct at_desc *desc)
 		dma_cookie_complete(txd);
 
 	/* If the transfer was a memset, free our temporary buffer */
-	if (desc->memset) {
+	if (desc->memset_buffer) {
 		dma_pool_free(atdma->memset_pool, desc->memset_vaddr,
 			      desc->memset_paddr);
-		desc->memset = false;
+		desc->memset_buffer = false;
 	}
 
 	/* move children to free_list */
@@ -881,6 +881,46 @@ atc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 	return NULL;
 }
 
+static struct at_desc *atc_create_memset_desc(struct dma_chan *chan,
+					      dma_addr_t psrc,
+					      dma_addr_t pdst,
+					      size_t len)
+{
+	struct at_dma_chan *atchan = to_at_dma_chan(chan);
+	struct at_desc *desc;
+	size_t xfer_count;
+
+	u32 ctrla = ATC_SRC_WIDTH(2) | ATC_DST_WIDTH(2);
+	u32 ctrlb = ATC_DEFAULT_CTRLB | ATC_IEN |
+		ATC_SRC_ADDR_MODE_FIXED |
+		ATC_DST_ADDR_MODE_INCR |
+		ATC_FC_MEM2MEM;
+
+	xfer_count = len >> 2;
+	if (xfer_count > ATC_BTSIZE_MAX) {
+		dev_err(chan2dev(chan), "%s: buffer is too big\n",
+			__func__);
+		return NULL;
+	}
+
+	desc = atc_desc_get(atchan);
+	if (!desc) {
+		dev_err(chan2dev(chan), "%s: can't get a descriptor\n",
+			__func__);
+		return NULL;
+	}
+
+	desc->lli.saddr = psrc;
+	desc->lli.daddr = pdst;
+	desc->lli.ctrla = ctrla | xfer_count;
+	desc->lli.ctrlb = ctrlb;
+
+	desc->txd.cookie = 0;
+	desc->len = len;
+
+	return desc;
+}
+
 /**
  * atc_prep_dma_memset - prepare a memcpy operation
  * @chan: the channel to prepare operation on
@@ -893,12 +933,10 @@ static struct dma_async_tx_descriptor *
 atc_prep_dma_memset(struct dma_chan *chan, dma_addr_t dest, int value,
 		    size_t len, unsigned long flags)
 {
-	struct at_dma_chan	*atchan = to_at_dma_chan(chan);
 	struct at_dma		*atdma = to_at_dma(chan->device);
-	struct at_desc		*desc = NULL;
-	size_t			xfer_count;
-	u32			ctrla;
-	u32			ctrlb;
+	struct at_desc		*desc;
+	void __iomem		*vaddr;
+	dma_addr_t		paddr;
 
 	dev_vdbg(chan2dev(chan), "%s: d0x%x v0x%x l0x%zx f0x%lx\n", __func__,
 		dest, value, len, flags);
@@ -914,61 +952,117 @@ atc_prep_dma_memset(struct dma_chan *chan, dma_addr_t dest, int value,
 		return NULL;
 	}
 
-	xfer_count = len >> 2;
-	if (xfer_count > ATC_BTSIZE_MAX) {
-		dev_err(chan2dev(chan), "%s: buffer is too big\n",
+	vaddr = dma_pool_alloc(atdma->memset_pool, GFP_ATOMIC, &paddr);
+	if (!vaddr) {
+		dev_err(chan2dev(chan), "%s: couldn't allocate buffer\n",
 			__func__);
 		return NULL;
 	}
+	*(u32*)vaddr = value;
 
-	ctrlb =   ATC_DEFAULT_CTRLB | ATC_IEN
-		| ATC_SRC_ADDR_MODE_FIXED
-		| ATC_DST_ADDR_MODE_INCR
-		| ATC_FC_MEM2MEM;
+	desc = atc_create_memset_desc(chan, paddr, dest, len);
+	if (!desc) {
+		dev_err(chan2dev(chan), "%s: couldn't get a descriptor\n",
+			__func__);
+		goto err_free_buffer;
+	}
 
-	ctrla = ATC_SRC_WIDTH(2) |
-		ATC_DST_WIDTH(2);
+	desc->memset_paddr = paddr;
+	desc->memset_vaddr = vaddr;
+	desc->memset_buffer = true;
 
-	desc = atc_desc_get(atchan);
-	if (!desc) {
-		dev_err(chan2dev(chan), "%s: can't get a descriptor\n",
+	desc->txd.cookie = -EBUSY;
+	desc->total_len = len;
+
+	/* set end-of-link on the descriptor */
+	set_desc_eol(desc);
+
+	desc->txd.flags = flags;
+
+	return &desc->txd;
+
+err_free_buffer:
+	dma_pool_free(atdma->memset_pool, vaddr, paddr);
+	return NULL;
+}
+
+static struct dma_async_tx_descriptor *
+atc_prep_dma_memset_sg(struct dma_chan *chan,
+		       struct scatterlist *sgl,
+		       unsigned int sg_len, int value,
+		       unsigned long flags)
+{
+	struct at_dma_chan	*atchan = to_at_dma_chan(chan);
+	struct at_dma		*atdma = to_at_dma(chan->device);
+	struct at_desc		*desc = NULL, *first = NULL, *prev = NULL;
+	struct scatterlist	*sg;
+	void __iomem		*vaddr;
+	dma_addr_t		paddr;
+	size_t			total_len = 0;
+	int			i;
+
+	dev_vdbg(chan2dev(chan), "%s: v0x%x l0x%zx f0x%lx\n", __func__,
+		 value, sg_len, flags);
+
+	if (unlikely(!sgl || !sg_len)) {
+		dev_dbg(chan2dev(chan), "%s: scatterlist is empty!\n",
 			__func__);
 		return NULL;
 	}
 
-	desc->memset_vaddr = dma_pool_alloc(atdma->memset_pool, GFP_ATOMIC,
-					    &desc->memset_paddr);
-	if (!desc->memset_vaddr) {
+	vaddr = dma_pool_alloc(atdma->memset_pool, GFP_ATOMIC, &paddr);
+	if (!vaddr) {
 		dev_err(chan2dev(chan), "%s: couldn't allocate buffer\n",
 			__func__);
-		goto err_put_desc;
+		return NULL;
 	}
+	*(u32*)vaddr = value;
 
-	*desc->memset_vaddr = value;
-	desc->memset = true;
+	for_each_sg(sgl, sg, sg_len, i) {
+		dma_addr_t dest = sg_dma_address(sg);
+		size_t len = sg_dma_len(sg);
 
-	desc->lli.saddr = desc->memset_paddr;
-	desc->lli.daddr = dest;
-	desc->lli.ctrla = ctrla | xfer_count;
-	desc->lli.ctrlb = ctrlb;
+		dev_vdbg(chan2dev(chan), "%s: d0x%08x, l0x%zx\n",
+			 __func__, dest, len);
 
-	desc->txd.cookie = -EBUSY;
-	desc->len = len;
-	desc->total_len = len;
+		if (!is_dma_fill_aligned(chan->device, dest, 0, len)) {
+			dev_err(chan2dev(chan), "%s: buffer is not aligned\n",
+				__func__);
+			goto err_put_desc;
+		}
+
+		desc = atc_create_memset_desc(chan, paddr, dest, len);
+		if (!desc)
+			goto err_put_desc;
+
+		atc_desc_chain(&first, &prev, desc);
+
+		total_len += len;
+	}
+
+	/*
+	 * Only set the buffer pointers on the last descriptor to
+	 * avoid free'ing while we have our transfer still going
+	 */
+	desc->memset_paddr = paddr;
+	desc->memset_vaddr = vaddr;
+	desc->memset_buffer = true;
+
+	first->txd.cookie = -EBUSY;
+	first->total_len = total_len;
 
 	/* set end-of-link on the descriptor */
 	set_desc_eol(desc);
 
-	desc->txd.flags = flags;
+	first->txd.flags = flags;
 
-	return &desc->txd;
+	return &first->txd;
 
 err_put_desc:
-	atc_desc_put(atchan, desc);
+	atc_desc_put(atchan, first);
 	return NULL;
 }
 
-
 /**
  * atc_prep_slave_sg - prepare descriptors for a DMA_SLAVE transaction
  * @chan: DMA channel
@@ -1851,6 +1945,7 @@ static int __init at_dma_probe(struct platform_device *pdev)
 	dma_cap_set(DMA_INTERLEAVE, at91sam9g45_config.cap_mask);
 	dma_cap_set(DMA_MEMCPY, at91sam9g45_config.cap_mask);
 	dma_cap_set(DMA_MEMSET, at91sam9g45_config.cap_mask);
+	dma_cap_set(DMA_MEMSET_SG, at91sam9g45_config.cap_mask);
 	dma_cap_set(DMA_PRIVATE, at91sam9g45_config.cap_mask);
 	dma_cap_set(DMA_SLAVE, at91sam9g45_config.cap_mask);
 	dma_cap_set(DMA_SG, at91sam9g45_config.cap_mask);
@@ -1972,6 +2067,7 @@ static int __init at_dma_probe(struct platform_device *pdev)
 
 	if (dma_has_cap(DMA_MEMSET, atdma->dma_common.cap_mask)) {
 		atdma->dma_common.device_prep_dma_memset = atc_prep_dma_memset;
+		atdma->dma_common.device_prep_dma_memset_sg = atc_prep_dma_memset_sg;
 		atdma->dma_common.fill_align = DMAENGINE_ALIGN_4_BYTES;
 	}
 

drivers/dma/at_hdmac_regs.h

@@ -202,7 +202,7 @@ struct at_desc {
 	size_t				src_hole;
 
 	/* Memset temporary buffer */
-	bool				memset;
+	bool				memset_buffer;
 	dma_addr_t			memset_paddr;
 	int				*memset_vaddr;
 };

drivers/dma/at_xdmac.c

@@ -938,13 +938,19 @@ at_xdmac_prep_interleaved(struct dma_chan *chan,
 {
 	struct at_xdmac_chan	*atchan = to_at_xdmac_chan(chan);
 	struct at_xdmac_desc	*prev = NULL, *first = NULL;
-	struct data_chunk	*chunk, *prev_chunk = NULL;
 	dma_addr_t		dst_addr, src_addr;
-	size_t			dst_skip, src_skip, len = 0;
-	size_t			prev_dst_icg = 0, prev_src_icg = 0;
+	size_t			src_skip = 0, dst_skip = 0, len = 0;
+	struct data_chunk	*chunk;
 	int			i;
 
-	if (!xt || (xt->numf != 1) || (xt->dir != DMA_MEM_TO_MEM))
+	if (!xt || !xt->numf || (xt->dir != DMA_MEM_TO_MEM))
+		return NULL;
+
+	/*
+	 * TODO: Handle the case where we have to repeat a chain of
+	 * descriptors...
+	 */
+	if ((xt->numf > 1) && (xt->frame_size > 1))
 		return NULL;
 
 	dev_dbg(chan2dev(chan), "%s: src=0x%08x, dest=0x%08x, numf=%d, frame_size=%d, flags=0x%lx\n",
@@ -954,66 +960,60 @@ at_xdmac_prep_interleaved(struct dma_chan *chan,
 	src_addr = xt->src_start;
 	dst_addr = xt->dst_start;
 
-	for (i = 0; i < xt->frame_size; i++) {
-		struct at_xdmac_desc *desc;
-		size_t src_icg, dst_icg;
+	if (xt->numf > 1) {
+		first = at_xdmac_interleaved_queue_desc(chan, atchan,
+							NULL,
+							src_addr, dst_addr,
+							xt, xt->sgl);
+		for (i = 0; i < xt->numf; i++)
+			at_xdmac_increment_block_count(chan, first);
 
-		chunk = xt->sgl + i;
+		dev_dbg(chan2dev(chan), "%s: add desc 0x%p to descs_list 0x%p\n",
+			__func__, first, first);
+		list_add_tail(&first->desc_node, &first->descs_list);
+	} else {
+		for (i = 0; i < xt->frame_size; i++) {
+			size_t src_icg = 0, dst_icg = 0;
+			struct at_xdmac_desc *desc;
 
-		dst_icg = dmaengine_get_dst_icg(xt, chunk);
-		src_icg = dmaengine_get_src_icg(xt, chunk);
+			chunk = xt->sgl + i;
 
-		src_skip = chunk->size + src_icg;
-		dst_skip = chunk->size + dst_icg;
+			dst_icg = dmaengine_get_dst_icg(xt, chunk);
+			src_icg = dmaengine_get_src_icg(xt, chunk);
 
-		dev_dbg(chan2dev(chan),
-			"%s: chunk size=%d, src icg=%d, dst icg=%d\n",
-			__func__, chunk->size, src_icg, dst_icg);
+			src_skip = chunk->size + src_icg;
+			dst_skip = chunk->size + dst_icg;
 
-		/*
-		 * Handle the case where we just have the same
-		 * transfer to setup, we can just increase the
-		 * block number and reuse the same descriptor.
-		 */
-		if (prev_chunk && prev &&
-		    (prev_chunk->size == chunk->size) &&
-		    (prev_src_icg == src_icg) &&
-		    (prev_dst_icg == dst_icg)) {
 			dev_dbg(chan2dev(chan),
-				"%s: same configuration that the previous chunk, merging the descriptors...\n",
-				__func__);
-			at_xdmac_increment_block_count(chan, prev);
-			continue;
-		}
-
-		desc = at_xdmac_interleaved_queue_desc(chan, atchan,
-						       prev,
-						       src_addr, dst_addr,
-						       xt, chunk);
-		if (!desc) {
-			list_splice_init(&first->descs_list,
-					 &atchan->free_descs_list);
-			return NULL;
-		}
+				"%s: chunk size=%d, src icg=%d, dst icg=%d\n",
+				__func__, chunk->size, src_icg, dst_icg);
+
+			desc = at_xdmac_interleaved_queue_desc(chan, atchan,
+							       prev,
+							       src_addr, dst_addr,
+							       xt, chunk);
+			if (!desc) {
+				list_splice_init(&first->descs_list,
+						 &atchan->free_descs_list);
+				return NULL;
+			}
 
-		if (!first)
-			first = desc;
+			if (!first)
+				first = desc;
 
-		dev_dbg(chan2dev(chan), "%s: add desc 0x%p to descs_list 0x%p\n",
-			__func__, desc, first);
-		list_add_tail(&desc->desc_node, &first->descs_list);
+			dev_dbg(chan2dev(chan), "%s: add desc 0x%p to descs_list 0x%p\n",
+				__func__, desc, first);
+			list_add_tail(&desc->desc_node, &first->descs_list);
 
-		if (xt->src_sgl)
-			src_addr += src_skip;
+			if (xt->src_sgl)
+				src_addr += src_skip;
 
-		if (xt->dst_sgl)
-			dst_addr += dst_skip;
+			if (xt->dst_sgl)
+				dst_addr += dst_skip;
 
-		len += chunk->size;
-		prev_chunk = chunk;
-		prev_dst_icg = dst_icg;
-		prev_src_icg = src_icg;
-		prev = desc;
+			len += chunk->size;
+			prev = desc;
+		}
 	}
 
 	first->tx_dma_desc.cookie = -EBUSY;

drivers/dma/dmaengine.c

@@ -1074,11 +1074,9 @@ static void dmaengine_destroy_unmap_pool(void)
 	for (i = 0; i < ARRAY_SIZE(unmap_pool); i++) {
 		struct dmaengine_unmap_pool *p = &unmap_pool[i];
 
-		if (p->pool)
-			mempool_destroy(p->pool);
+		mempool_destroy(p->pool);
 		p->pool = NULL;
-		if (p->cache)
-			kmem_cache_destroy(p->cache);
+		kmem_cache_destroy(p->cache);
 		p->cache = NULL;
 	}
 }

drivers/dma/dw/core.c

@@ -163,7 +163,7 @@ static void dwc_initialize(struct dw_dma_chan *dwc)
 
 /*----------------------------------------------------------------------*/
 
-static inline unsigned int dwc_fast_fls(unsigned long long v)
+static inline unsigned int dwc_fast_ffs(unsigned long long v)
 {
 	/*
 	 * We can be a lot more clever here, but this should take care
@@ -712,7 +712,7 @@ dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 			   dw->data_width[dwc->dst_master]);
 
 	src_width = dst_width = min_t(unsigned int, data_width,
-				      dwc_fast_fls(src | dest | len));
+				      dwc_fast_ffs(src | dest | len));
 
 	ctllo = DWC_DEFAULT_CTLLO(chan)
 			| DWC_CTLL_DST_WIDTH(dst_width)
@@ -791,7 +791,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 
 	switch (direction) {
 	case DMA_MEM_TO_DEV:
-		reg_width = __fls(sconfig->dst_addr_width);
+		reg_width = __ffs(sconfig->dst_addr_width);
 		reg = sconfig->dst_addr;
 		ctllo = (DWC_DEFAULT_CTLLO(chan)
 				| DWC_CTLL_DST_WIDTH(reg_width)
@@ -811,7 +811,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 			len = sg_dma_len(sg);
 
 			mem_width = min_t(unsigned int,
-					  data_width, dwc_fast_fls(mem | len));
+					  data_width, dwc_fast_ffs(mem | len));
 
 slave_sg_todev_fill_desc:
 			desc = dwc_desc_get(dwc);
@@ -848,7 +848,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		}
 		break;
 	case DMA_DEV_TO_MEM:
-		reg_width = __fls(sconfig->src_addr_width);
+		reg_width = __ffs(sconfig->src_addr_width);
 		reg = sconfig->src_addr;
 		ctllo = (DWC_DEFAULT_CTLLO(chan)
 				| DWC_CTLL_SRC_WIDTH(reg_width)
@@ -868,7 +868,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 			len = sg_dma_len(sg);
 
 			mem_width = min_t(unsigned int,
-					  data_width, dwc_fast_fls(mem | len));
+					  data_width, dwc_fast_ffs(mem | len));
 
 slave_sg_fromdev_fill_desc:
 			desc = dwc_desc_get(dwc);
@@ -1499,9 +1499,8 @@ EXPORT_SYMBOL(dw_dma_cyclic_free);
 int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 {
 	struct dw_dma		*dw;
-	bool			autocfg;
+	bool			autocfg = false;
 	unsigned int		dw_params;
-	unsigned int		nr_channels;
 	unsigned int		max_blk_size = 0;
 	int			err;
 	int			i;
@@ -1515,33 +1514,42 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 
 	pm_runtime_get_sync(chip->dev);
 
-	dw_params = dma_read_byaddr(chip->regs, DW_PARAMS);
-	autocfg = dw_params >> DW_PARAMS_EN & 0x1;
+	if (!pdata) {
+		dw_params = dma_read_byaddr(chip->regs, DW_PARAMS);
+		dev_dbg(chip->dev, "DW_PARAMS: 0x%08x\n", dw_params);
 
-	dev_dbg(chip->dev, "DW_PARAMS: 0x%08x\n", dw_params);
+		autocfg = dw_params >> DW_PARAMS_EN & 1;
+		if (!autocfg) {
+			err = -EINVAL;
+			goto err_pdata;
+		}
 
-	if (!pdata && autocfg) {
 		pdata = devm_kzalloc(chip->dev, sizeof(*pdata), GFP_KERNEL);
 		if (!pdata) {
 			err = -ENOMEM;
 			goto err_pdata;
 		}
 
+		/* Get hardware configuration parameters */
+		pdata->nr_channels = (dw_params >> DW_PARAMS_NR_CHAN & 7) + 1;
+		pdata->nr_masters = (dw_params >> DW_PARAMS_NR_MASTER & 3) + 1;
+		for (i = 0; i < pdata->nr_masters; i++) {
+			pdata->data_width[i] =
+				(dw_params >> DW_PARAMS_DATA_WIDTH(i) & 3) + 2;
+		}
+		max_blk_size = dma_readl(dw, MAX_BLK_SIZE);
+
 		/* Fill platform data with the default values */
 		pdata->is_private = true;
+		pdata->is_memcpy = true;
 		pdata->chan_allocation_order = CHAN_ALLOCATION_ASCENDING;
 		pdata->chan_priority = CHAN_PRIORITY_ASCENDING;
-	} else if (!pdata || pdata->nr_channels > DW_DMA_MAX_NR_CHANNELS) {
+	} else if (pdata->nr_channels > DW_DMA_MAX_NR_CHANNELS) {
 		err = -EINVAL;
 		goto err_pdata;
 	}
 
-	if (autocfg)
-		nr_channels = (dw_params >> DW_PARAMS_NR_CHAN & 0x7) + 1;
-	else
-		nr_channels = pdata->nr_channels;
-
-	dw->chan = devm_kcalloc(chip->dev, nr_channels, sizeof(*dw->chan),
+	dw->chan = devm_kcalloc(chip->dev, pdata->nr_channels, sizeof(*dw->chan),
 				GFP_KERNEL);
 	if (!dw->chan) {
 		err = -ENOMEM;
@@ -1549,22 +1557,12 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 	}
 
 	/* Get hardware configuration parameters */
-	if (autocfg) {
-		max_blk_size = dma_readl(dw, MAX_BLK_SIZE);
-
-		dw->nr_masters = (dw_params >> DW_PARAMS_NR_MASTER & 3) + 1;
-		for (i = 0; i < dw->nr_masters; i++) {
-			dw->data_width[i] =
-				(dw_params >> DW_PARAMS_DATA_WIDTH(i) & 3) + 2;
-		}
-	} else {
-		dw->nr_masters = pdata->nr_masters;
-		for (i = 0; i < dw->nr_masters; i++)
-			dw->data_width[i] = pdata->data_width[i];
-	}
+	dw->nr_masters = pdata->nr_masters;
+	for (i = 0; i < dw->nr_masters; i++)
+		dw->data_width[i] = pdata->data_width[i];
 
 	/* Calculate all channel mask before DMA setup */
-	dw->all_chan_mask = (1 << nr_channels) - 1;
+	dw->all_chan_mask = (1 << pdata->nr_channels) - 1;
 
 	/* Force dma off, just in case */
 	dw_dma_off(dw);
@@ -1589,7 +1587,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 		goto err_pdata;
 
 	INIT_LIST_HEAD(&dw->dma.channels);
-	for (i = 0; i < nr_channels; i++) {
+	for (i = 0; i < pdata->nr_channels; i++) {
 		struct dw_dma_chan	*dwc = &dw->chan[i];
 
 		dwc->chan.device = &dw->dma;
@@ -1602,7 +1600,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 
 		/* 7 is highest priority & 0 is lowest. */
 		if (pdata->chan_priority == CHAN_PRIORITY_ASCENDING)
-			dwc->priority = nr_channels - i - 1;
+			dwc->priority = pdata->nr_channels - i - 1;
 		else
 			dwc->priority = i;
 
@@ -1656,10 +1654,13 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 	dma_writel(dw, CLEAR.DST_TRAN, dw->all_chan_mask);
 	dma_writel(dw, CLEAR.ERROR, dw->all_chan_mask);
 
-	dma_cap_set(DMA_MEMCPY, dw->dma.cap_mask);
+	/* Set capabilities */
 	dma_cap_set(DMA_SLAVE, dw->dma.cap_mask);
 	if (pdata->is_private)
 		dma_cap_set(DMA_PRIVATE, dw->dma.cap_mask);
+	if (pdata->is_memcpy)
+		dma_cap_set(DMA_MEMCPY, dw->dma.cap_mask);
+
 	dw->dma.dev = chip->dev;
 	dw->dma.device_alloc_chan_resources = dwc_alloc_chan_resources;
 	dw->dma.device_free_chan_resources = dwc_free_chan_resources;
@@ -1687,7 +1688,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
 		goto err_dma_register;
 
 	dev_info(chip->dev, "DesignWare DMA Controller, %d channels\n",
-		 nr_channels);
+		 pdata->nr_channels);
 
 	pm_runtime_put_sync_suspend(chip->dev);
 

drivers/dma/dw/pci.c

@@ -15,12 +15,6 @@
 
 #include "internal.h"
 
-static struct dw_dma_platform_data dw_pci_pdata = {
-	.is_private = 1,
-	.chan_allocation_order = CHAN_ALLOCATION_ASCENDING,
-	.chan_priority = CHAN_PRIORITY_ASCENDING,
-};
-
 static int dw_pci_probe(struct pci_dev *pdev, const struct pci_device_id *pid)
 {
 	struct dw_dma_chip *chip;
@@ -101,19 +95,19 @@ static const struct dev_pm_ops dw_pci_dev_pm_ops = {
 
 static const struct pci_device_id dw_pci_id_table[] = {
 	/* Medfield */
-	{ PCI_VDEVICE(INTEL, 0x0827), (kernel_ulong_t)&dw_pci_pdata },
-	{ PCI_VDEVICE(INTEL, 0x0830), (kernel_ulong_t)&dw_pci_pdata },
+	{ PCI_VDEVICE(INTEL, 0x0827) },
+	{ PCI_VDEVICE(INTEL, 0x0830) },
 
 	/* BayTrail */
-	{ PCI_VDEVICE(INTEL, 0x0f06), (kernel_ulong_t)&dw_pci_pdata },
-	{ PCI_VDEVICE(INTEL, 0x0f40), (kernel_ulong_t)&dw_pci_pdata },
+	{ PCI_VDEVICE(INTEL, 0x0f06) },
+	{ PCI_VDEVICE(INTEL, 0x0f40) },
 
 	/* Braswell */
-	{ PCI_VDEVICE(INTEL, 0x2286), (kernel_ulong_t)&dw_pci_pdata },
-	{ PCI_VDEVICE(INTEL, 0x22c0), (kernel_ulong_t)&dw_pci_pdata },
+	{ PCI_VDEVICE(INTEL, 0x2286) },
+	{ PCI_VDEVICE(INTEL, 0x22c0) },
 
 	/* Haswell */
-	{ PCI_VDEVICE(INTEL, 0x9c60), (kernel_ulong_t)&dw_pci_pdata },
+	{ PCI_VDEVICE(INTEL, 0x9c60) },
 	{ }
 };
 MODULE_DEVICE_TABLE(pci, dw_pci_id_table);

drivers/dma/dw/platform.c

@@ -155,6 +155,7 @@ static int dw_probe(struct platform_device *pdev)
 	struct dw_dma_chip *chip;
 	struct device *dev = &pdev->dev;
 	struct resource *mem;
+	const struct acpi_device_id *id;
 	struct dw_dma_platform_data *pdata;
 	int err;
 
@@ -178,6 +179,11 @@ static int dw_probe(struct platform_device *pdev)
 	pdata = dev_get_platdata(dev);
 	if (!pdata)
 		pdata = dw_dma_parse_dt(pdev);
+	if (!pdata && has_acpi_companion(dev)) {
+		id = acpi_match_device(dev->driver->acpi_match_table, dev);
+		if (id)
+			pdata = (struct dw_dma_platform_data *)id->driver_data;
+	}
 
 	chip->dev = dev;
 
@@ -246,8 +252,17 @@ MODULE_DEVICE_TABLE(of, dw_dma_of_id_table);
 #endif
 
 #ifdef CONFIG_ACPI
+static struct dw_dma_platform_data dw_dma_acpi_pdata = {
+	.nr_channels = 8,
+	.is_private = true,
+	.chan_allocation_order = CHAN_ALLOCATION_ASCENDING,
+	.chan_priority = CHAN_PRIORITY_ASCENDING,
+	.block_size = 4095,
+	.nr_masters = 2,
+};
+
 static const struct acpi_device_id dw_dma_acpi_id_table[] = {
-	{ "INTL9C60", 0 },
+	{ "INTL9C60", (kernel_ulong_t)&dw_dma_acpi_pdata },
 	{ }
 };
 MODULE_DEVICE_TABLE(acpi, dw_dma_acpi_id_table);

drivers/dma/fsldma.c

@@ -1512,6 +1512,7 @@ static const struct of_device_id fsldma_of_ids[] = {
 	{ .compatible = "fsl,elo-dma", },
 	{}
 };
+MODULE_DEVICE_TABLE(of, fsldma_of_ids);
 
 static struct platform_driver fsldma_of_driver = {
 	.driver = {

drivers/dma/idma64.c

@@ -65,9 +65,6 @@ static void idma64_chan_init(struct idma64 *idma64, struct idma64_chan *idma64c)
 	u32 cfghi = IDMA64C_CFGH_SRC_PER(1) | IDMA64C_CFGH_DST_PER(0);
 	u32 cfglo = 0;
 
-	/* Enforce FIFO drain when channel is suspended */
-	cfglo |= IDMA64C_CFGL_CH_DRAIN;
-
 	/* Set default burst alignment */
 	cfglo |= IDMA64C_CFGL_DST_BURST_ALIGN | IDMA64C_CFGL_SRC_BURST_ALIGN;
 
@@ -257,15 +254,15 @@ static u64 idma64_hw_desc_fill(struct idma64_hw_desc *hw,
 		dar = config->dst_addr;
 		ctllo |= IDMA64C_CTLL_DST_FIX | IDMA64C_CTLL_SRC_INC |
 			 IDMA64C_CTLL_FC_M2P;
-		src_width = min_t(u32, 2, __fls(sar | hw->len));
-		dst_width = __fls(config->dst_addr_width);
+		src_width = __ffs(sar | hw->len | 4);
+		dst_width = __ffs(config->dst_addr_width);
 	} else {	/* DMA_DEV_TO_MEM */
 		sar = config->src_addr;
 		dar = hw->phys;
 		ctllo |= IDMA64C_CTLL_DST_INC | IDMA64C_CTLL_SRC_FIX |
 			 IDMA64C_CTLL_FC_P2M;
-		src_width = __fls(config->src_addr_width);
-		dst_width = min_t(u32, 2, __fls(dar | hw->len));
+		src_width = __ffs(config->src_addr_width);
+		dst_width = __ffs(dar | hw->len | 4);
 	}
 
 	lli->sar = sar;
@@ -428,12 +425,17 @@ static int idma64_slave_config(struct dma_chan *chan,
 	return 0;
 }
 
-static void idma64_chan_deactivate(struct idma64_chan *idma64c)
+static void idma64_chan_deactivate(struct idma64_chan *idma64c, bool drain)
 {
 	unsigned short count = 100;
 	u32 cfglo;
 
 	cfglo = channel_readl(idma64c, CFG_LO);
+	if (drain)
+		cfglo |= IDMA64C_CFGL_CH_DRAIN;
+	else
+		cfglo &= ~IDMA64C_CFGL_CH_DRAIN;
+
 	channel_writel(idma64c, CFG_LO, cfglo | IDMA64C_CFGL_CH_SUSP);
 	do {
 		udelay(1);
@@ -456,7 +458,7 @@ static int idma64_pause(struct dma_chan *chan)
 
 	spin_lock_irqsave(&idma64c->vchan.lock, flags);
 	if (idma64c->desc && idma64c->desc->status == DMA_IN_PROGRESS) {
-		idma64_chan_deactivate(idma64c);
+		idma64_chan_deactivate(idma64c, false);
 		idma64c->desc->status = DMA_PAUSED;
 	}
 	spin_unlock_irqrestore(&idma64c->vchan.lock, flags);
@@ -486,7 +488,7 @@ static int idma64_terminate_all(struct dma_chan *chan)
 	LIST_HEAD(head);
 
 	spin_lock_irqsave(&idma64c->vchan.lock, flags);
-	idma64_chan_deactivate(idma64c);
+	idma64_chan_deactivate(idma64c, true);
 	idma64_stop_transfer(idma64c);
 	if (idma64c->desc) {
 		idma64_vdesc_free(&idma64c->desc->vdesc);

drivers/dma/idma64.h

@@ -16,6 +16,8 @@
 #include <linux/spinlock.h>
 #include <linux/types.h>
 
+#include <asm-generic/io-64-nonatomic-lo-hi.h>
+
 #include "virt-dma.h"
 
 /* Channel registers */
@@ -166,19 +168,13 @@ static inline void idma64c_writel(struct idma64_chan *idma64c, int offset,
 
 static inline u64 idma64c_readq(struct idma64_chan *idma64c, int offset)
 {
-	u64 l, h;
-
-	l = idma64c_readl(idma64c, offset);
-	h = idma64c_readl(idma64c, offset + 4);
-
-	return l | (h << 32);
+	return lo_hi_readq(idma64c->regs + offset);
 }
 
 static inline void idma64c_writeq(struct idma64_chan *idma64c, int offset,
 				  u64 value)
 {
-	idma64c_writel(idma64c, offset, value);
-	idma64c_writel(idma64c, offset + 4, value >> 32);
+	lo_hi_writeq(value, idma64c->regs + offset);
 }
 
 #define channel_readq(idma64c, reg)		\
@@ -217,7 +213,7 @@ static inline void idma64_writel(struct idma64 *idma64, int offset, u32 value)
 	idma64_writel(idma64, IDMA64_##reg, (value))
 
 /**
- * struct idma64_chip - representation of DesignWare DMA controller hardware
+ * struct idma64_chip - representation of iDMA 64-bit controller hardware
  * @dev:		struct device of the DMA controller
  * @irq:		irq line
  * @regs:		memory mapped I/O space

drivers/dma/imx-sdma.c

@@ -1478,7 +1478,7 @@ static int __init sdma_event_remap(struct sdma_engine *sdma)
 	event_remap = of_find_property(np, propname, NULL);
 	num_map = event_remap ? (event_remap->length / sizeof(u32)) : 0;
 	if (!num_map) {
-		dev_warn(sdma->dev, "no event needs to be remapped\n");
+		dev_dbg(sdma->dev, "no event needs to be remapped\n");
 		goto out;
 	} else if (num_map % EVENT_REMAP_CELLS) {
 		dev_err(sdma->dev, "the property %s must modulo %d\n",
@@ -1826,8 +1826,6 @@ static int sdma_probe(struct platform_device *pdev)
 		of_node_put(spba_bus);
 	}
 
-	dev_info(sdma->dev, "initialized\n");
-
 	return 0;
 
 err_register:
@@ -1852,7 +1850,6 @@ static int sdma_remove(struct platform_device *pdev)
 	}
 
 	platform_set_drvdata(pdev, NULL);
-	dev_info(&pdev->dev, "Removed...\n");
 	return 0;
 }
 

drivers/dma/ioat/dma.c

@@ -197,7 +197,8 @@ static void __ioat_start_null_desc(struct ioatdma_chan *ioat_chan)
 void ioat_start_null_desc(struct ioatdma_chan *ioat_chan)
 {
 	spin_lock_bh(&ioat_chan->prep_lock);
-	__ioat_start_null_desc(ioat_chan);
+	if (!test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		__ioat_start_null_desc(ioat_chan);
 	spin_unlock_bh(&ioat_chan->prep_lock);
 }
 

drivers/dma/ioat/dma.h

@@ -82,8 +82,9 @@ struct ioatdma_device {
 	struct dma_pool *sed_hw_pool[MAX_SED_POOLS];
 	struct dma_device dma_dev;
 	u8 version;
-	struct msix_entry msix_entries[4];
-	struct ioatdma_chan *idx[4];
+#define IOAT_MAX_CHANS 4
+	struct msix_entry msix_entries[IOAT_MAX_CHANS];
+	struct ioatdma_chan *idx[IOAT_MAX_CHANS];
 	struct dca_provider *dca;
 	enum ioat_irq_mode irq_mode;
 	u32 cap;
@@ -95,6 +96,7 @@ struct ioatdma_chan {
 	dma_addr_t last_completion;
 	spinlock_t cleanup_lock;
 	unsigned long state;
+	#define IOAT_CHAN_DOWN 0
 	#define IOAT_COMPLETION_ACK 1
 	#define IOAT_RESET_PENDING 2
 	#define IOAT_KOBJ_INIT_FAIL 3

drivers/dma/ioat/init.c

@@ -27,6 +27,7 @@
 #include <linux/workqueue.h>
 #include <linux/prefetch.h>
 #include <linux/dca.h>
+#include <linux/aer.h>
 #include "dma.h"
 #include "registers.h"
 #include "hw.h"
@@ -1186,13 +1187,116 @@ static int ioat3_dma_probe(struct ioatdma_device *ioat_dma, int dca)
 	return 0;
 }
 
+static void ioat_shutdown(struct pci_dev *pdev)
+{
+	struct ioatdma_device *ioat_dma = pci_get_drvdata(pdev);
+	struct ioatdma_chan *ioat_chan;
+	int i;
+
+	if (!ioat_dma)
+		return;
+
+	for (i = 0; i < IOAT_MAX_CHANS; i++) {
+		ioat_chan = ioat_dma->idx[i];
+		if (!ioat_chan)
+			continue;
+
+		spin_lock_bh(&ioat_chan->prep_lock);
+		set_bit(IOAT_CHAN_DOWN, &ioat_chan->state);
+		del_timer_sync(&ioat_chan->timer);
+		spin_unlock_bh(&ioat_chan->prep_lock);
+		/* this should quiesce then reset */
+		ioat_reset_hw(ioat_chan);
+	}
+
+	ioat_disable_interrupts(ioat_dma);
+}
+
+void ioat_resume(struct ioatdma_device *ioat_dma)
+{
+	struct ioatdma_chan *ioat_chan;
+	u32 chanerr;
+	int i;
+
+	for (i = 0; i < IOAT_MAX_CHANS; i++) {
+		ioat_chan = ioat_dma->idx[i];
+		if (!ioat_chan)
+			continue;
+
+		spin_lock_bh(&ioat_chan->prep_lock);
+		clear_bit(IOAT_CHAN_DOWN, &ioat_chan->state);
+		spin_unlock_bh(&ioat_chan->prep_lock);
+
+		chanerr = readl(ioat_chan->reg_base + IOAT_CHANERR_OFFSET);
+		writel(chanerr, ioat_chan->reg_base + IOAT_CHANERR_OFFSET);
+
+		/* no need to reset as shutdown already did that */
+	}
+}
+
 #define DRV_NAME "ioatdma"
 
+static pci_ers_result_t ioat_pcie_error_detected(struct pci_dev *pdev,
+						 enum pci_channel_state error)
+{
+	dev_dbg(&pdev->dev, "%s: PCIe AER error %d\n", DRV_NAME, error);
+
+	/* quiesce and block I/O */
+	ioat_shutdown(pdev);
+
+	return PCI_ERS_RESULT_NEED_RESET;
+}
+
+static pci_ers_result_t ioat_pcie_error_slot_reset(struct pci_dev *pdev)
+{
+	pci_ers_result_t result = PCI_ERS_RESULT_RECOVERED;
+	int err;
+
+	dev_dbg(&pdev->dev, "%s post reset handling\n", DRV_NAME);
+
+	if (pci_enable_device_mem(pdev) < 0) {
+		dev_err(&pdev->dev,
+			"Failed to enable PCIe device after reset.\n");
+		result = PCI_ERS_RESULT_DISCONNECT;
+	} else {
+		pci_set_master(pdev);
+		pci_restore_state(pdev);
+		pci_save_state(pdev);
+		pci_wake_from_d3(pdev, false);
+	}
+
+	err = pci_cleanup_aer_uncorrect_error_status(pdev);
+	if (err) {
+		dev_err(&pdev->dev,
+			"AER uncorrect error status clear failed: %#x\n", err);
+	}
+
+	return result;
+}
+
+static void ioat_pcie_error_resume(struct pci_dev *pdev)
+{
+	struct ioatdma_device *ioat_dma = pci_get_drvdata(pdev);
+
+	dev_dbg(&pdev->dev, "%s: AER handling resuming\n", DRV_NAME);
+
+	/* initialize and bring everything back */
+	ioat_resume(ioat_dma);
+}
+
+static const struct pci_error_handlers ioat_err_handler = {
+	.error_detected = ioat_pcie_error_detected,
+	.slot_reset = ioat_pcie_error_slot_reset,
+	.resume = ioat_pcie_error_resume,
+};
+
 static struct pci_driver ioat_pci_driver = {
 	.name		= DRV_NAME,
 	.id_table	= ioat_pci_tbl,
 	.probe		= ioat_pci_probe,
 	.remove		= ioat_remove,
+	.shutdown	= ioat_shutdown,
+	.err_handler	= &ioat_err_handler,
 };
 
 static struct ioatdma_device *
@@ -1245,13 +1349,17 @@ static int ioat_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	pci_set_drvdata(pdev, device);
 
 	device->version = readb(device->reg_base + IOAT_VER_OFFSET);
-	if (device->version >= IOAT_VER_3_0)
+	if (device->version >= IOAT_VER_3_0) {
 		err = ioat3_dma_probe(device, ioat_dca_enabled);
-	else
+
+		if (device->version >= IOAT_VER_3_3)
+			pci_enable_pcie_error_reporting(pdev);
+	} else
 		return -ENODEV;
 
 	if (err) {
 		dev_err(dev, "Intel(R) I/OAT DMA Engine init failed\n");
+		pci_disable_pcie_error_reporting(pdev);
 		return -ENODEV;
 	}
 
@@ -1271,6 +1379,8 @@ static void ioat_remove(struct pci_dev *pdev)
 		free_dca_provider(device->dca);
 		device->dca = NULL;
 	}
+
+	pci_disable_pcie_error_reporting(pdev);
 	ioat_dma_remove(device);
 }
 

drivers/dma/ioat/prep.c

@@ -121,6 +121,9 @@ ioat_dma_prep_memcpy_lock(struct dma_chan *c, dma_addr_t dma_dest,
 	size_t total_len = len;
 	int num_descs, idx, i;
 
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
+
 	num_descs = ioat_xferlen_to_descs(ioat_chan, len);
 	if (likely(num_descs) &&
 	    ioat_check_space_lock(ioat_chan, num_descs) == 0)
@@ -254,6 +257,11 @@ struct dma_async_tx_descriptor *
 ioat_prep_xor(struct dma_chan *chan, dma_addr_t dest, dma_addr_t *src,
 	       unsigned int src_cnt, size_t len, unsigned long flags)
 {
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(chan);
+
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
+
 	return __ioat_prep_xor_lock(chan, NULL, dest, src, src_cnt, len, flags);
 }
 
@@ -262,6 +270,11 @@ ioat_prep_xor_val(struct dma_chan *chan, dma_addr_t *src,
 		    unsigned int src_cnt, size_t len,
 		    enum sum_check_flags *result, unsigned long flags)
 {
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(chan);
+
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
+
 	/* the cleanup routine only sets bits on validate failure, it
 	 * does not clear bits on validate success... so clear it here
 	 */
@@ -574,6 +587,11 @@ ioat_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
 	      unsigned int src_cnt, const unsigned char *scf, size_t len,
 	      unsigned long flags)
 {
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(chan);
+
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
+
 	/* specify valid address for disabled result */
 	if (flags & DMA_PREP_PQ_DISABLE_P)
 		dst[0] = dst[1];
@@ -614,6 +632,11 @@ ioat_prep_pq_val(struct dma_chan *chan, dma_addr_t *pq, dma_addr_t *src,
 		  unsigned int src_cnt, const unsigned char *scf, size_t len,
 		  enum sum_check_flags *pqres, unsigned long flags)
 {
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(chan);
+
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
+
 	/* specify valid address for disabled result */
 	if (flags & DMA_PREP_PQ_DISABLE_P)
 		pq[0] = pq[1];
@@ -638,6 +661,10 @@ ioat_prep_pqxor(struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
 {
 	unsigned char scf[MAX_SCF];
 	dma_addr_t pq[2];
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(chan);
+
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
 
 	if (src_cnt > MAX_SCF)
 		return NULL;
@@ -661,6 +688,10 @@ ioat_prep_pqxor_val(struct dma_chan *chan, dma_addr_t *src,
 {
 	unsigned char scf[MAX_SCF];
 	dma_addr_t pq[2];
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(chan);
+
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
 
 	if (src_cnt > MAX_SCF)
 		return NULL;
@@ -689,6 +720,9 @@ ioat_prep_interrupt_lock(struct dma_chan *c, unsigned long flags)
 	struct ioat_ring_ent *desc;
 	struct ioat_dma_descriptor *hw;
 
+	if (test_bit(IOAT_CHAN_DOWN, &ioat_chan->state))
+		return NULL;
+
 	if (ioat_check_space_lock(ioat_chan, 1) == 0)
 		desc = ioat_get_ring_ent(ioat_chan, ioat_chan->head);
 	else

drivers/dma/moxart-dma.c

@@ -652,6 +652,7 @@ static const struct of_device_id moxart_dma_match[] = {
 	{ .compatible = "moxa,moxart-dma" },
 	{ }
 };
+MODULE_DEVICE_TABLE(of, moxart_dma_match);
 
 static struct platform_driver moxart_driver = {
 	.probe	= moxart_probe,

drivers/dma/mpc512x_dma.c

@@ -1073,6 +1073,7 @@ static const struct of_device_id mpc_dma_match[] = {
 	{ .compatible = "fsl,mpc8308-dma", },
 	{},
 };
+MODULE_DEVICE_TABLE(of, mpc_dma_match);
 
 static struct platform_driver mpc_dma_driver = {
 	.probe		= mpc_dma_probe,

drivers/dma/omap-dma.c

@@ -935,8 +935,12 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
 		else
 			d->ccr |= CCR_SYNC_ELEMENT;
 
-		if (dir == DMA_DEV_TO_MEM)
+		if (dir == DMA_DEV_TO_MEM) {
 			d->ccr |= CCR_TRIGGER_SRC;
+			d->csdp |= CSDP_DST_PACKED;
+		} else {
+			d->csdp |= CSDP_SRC_PACKED;
+		}
 
 		d->cicr |= CICR_MISALIGNED_ERR_IE | CICR_TRANS_ERR_IE;
 

drivers/dma/sirf-dma.c

@@ -1149,6 +1149,7 @@ static const struct of_device_id sirfsoc_dma_match[] = {
 	{ .compatible = "sirf,atlas7-dmac-v2", .data = &sirfsoc_dmadata_a7v2,},
 	{},
 };
+MODULE_DEVICE_TABLE(of, sirfsoc_dma_match);
 
 static struct platform_driver sirfsoc_dma_driver = {
 	.probe		= sirfsoc_dma_probe,

drivers/dma/ste_dma40.c

@@ -2907,7 +2907,7 @@ static int __init d40_dmaengine_init(struct d40_base *base,
 
 	if (err) {
 		d40_err(base->dev,
-			"Failed to regsiter memcpy only channels\n");
+			"Failed to register memcpy only channels\n");
 		goto failure2;
 	}
 

drivers/dma/sun6i-dma.c

@@ -908,6 +908,7 @@ static const struct of_device_id sun6i_dma_match[] = {
 	{ .compatible = "allwinner,sun8i-h3-dma", .data = &sun8i_h3_dma_cfg },
 	{ /* sentinel */ }
 };
+MODULE_DEVICE_TABLE(of, sun6i_dma_match);
 
 static int sun6i_dma_probe(struct platform_device *pdev)
 {

drivers/dma/ti-dma-crossbar.c

@@ -17,13 +17,184 @@
 #include <linux/of_device.h>
 #include <linux/of_dma.h>
 
-#define TI_XBAR_OUTPUTS	127
-#define TI_XBAR_INPUTS	256
+#define TI_XBAR_DRA7		0
+#define TI_XBAR_AM335X		1
+
+static const struct of_device_id ti_dma_xbar_match[] = {
+	{
+		.compatible = "ti,dra7-dma-crossbar",
+		.data = (void *)TI_XBAR_DRA7,
+	},
+	{
+		.compatible = "ti,am335x-edma-crossbar",
+		.data = (void *)TI_XBAR_AM335X,
+	},
+	{},
+};
+
+/* Crossbar on AM335x/AM437x family */
+#define TI_AM335X_XBAR_LINES	64
+
+struct ti_am335x_xbar_data {
+	void __iomem *iomem;
+
+	struct dma_router dmarouter;
+
+	u32 xbar_events; /* maximum number of events to select in xbar */
+	u32 dma_requests; /* number of DMA requests on eDMA */
+};
+
+struct ti_am335x_xbar_map {
+	u16 dma_line;
+	u16 mux_val;
+};
+
+static inline void ti_am335x_xbar_write(void __iomem *iomem, int event, u16 val)
+{
+	writeb_relaxed(val & 0x1f, iomem + event);
+}
+
+static void ti_am335x_xbar_free(struct device *dev, void *route_data)
+{
+	struct ti_am335x_xbar_data *xbar = dev_get_drvdata(dev);
+	struct ti_am335x_xbar_map *map = route_data;
+
+	dev_dbg(dev, "Unmapping XBAR event %u on channel %u\n",
+		map->mux_val, map->dma_line);
+
+	ti_am335x_xbar_write(xbar->iomem, map->dma_line, 0);
+	kfree(map);
+}
+
+static void *ti_am335x_xbar_route_allocate(struct of_phandle_args *dma_spec,
+					   struct of_dma *ofdma)
+{
+	struct platform_device *pdev = of_find_device_by_node(ofdma->of_node);
+	struct ti_am335x_xbar_data *xbar = platform_get_drvdata(pdev);
+	struct ti_am335x_xbar_map *map;
+
+	if (dma_spec->args_count != 3)
+		return ERR_PTR(-EINVAL);
+
+	if (dma_spec->args[2] >= xbar->xbar_events) {
+		dev_err(&pdev->dev, "Invalid XBAR event number: %d\n",
+			dma_spec->args[2]);
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (dma_spec->args[0] >= xbar->dma_requests) {
+		dev_err(&pdev->dev, "Invalid DMA request line number: %d\n",
+			dma_spec->args[0]);
+		return ERR_PTR(-EINVAL);
+	}
+
+	/* The of_node_put() will be done in the core for the node */
+	dma_spec->np = of_parse_phandle(ofdma->of_node, "dma-masters", 0);
+	if (!dma_spec->np) {
+		dev_err(&pdev->dev, "Can't get DMA master\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	map = kzalloc(sizeof(*map), GFP_KERNEL);
+	if (!map) {
+		of_node_put(dma_spec->np);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	map->dma_line = (u16)dma_spec->args[0];
+	map->mux_val = (u16)dma_spec->args[2];
+
+	dma_spec->args[2] = 0;
+	dma_spec->args_count = 2;
+
+	dev_dbg(&pdev->dev, "Mapping XBAR event%u to DMA%u\n",
+		map->mux_val, map->dma_line);
+
+	ti_am335x_xbar_write(xbar->iomem, map->dma_line, map->mux_val);
+
+	return map;
+}
+
+static const struct of_device_id ti_am335x_master_match[] = {
+	{ .compatible = "ti,edma3-tpcc", },
+	{},
+};
+
+static int ti_am335x_xbar_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	const struct of_device_id *match;
+	struct device_node *dma_node;
+	struct ti_am335x_xbar_data *xbar;
+	struct resource *res;
+	void __iomem *iomem;
+	int i, ret;
+
+	if (!node)
+		return -ENODEV;
+
+	xbar = devm_kzalloc(&pdev->dev, sizeof(*xbar), GFP_KERNEL);
+	if (!xbar)
+		return -ENOMEM;
+
+	dma_node = of_parse_phandle(node, "dma-masters", 0);
+	if (!dma_node) {
+		dev_err(&pdev->dev, "Can't get DMA master node\n");
+		return -ENODEV;
+	}
+
+	match = of_match_node(ti_am335x_master_match, dma_node);
+	if (!match) {
+		dev_err(&pdev->dev, "DMA master is not supported\n");
+		return -EINVAL;
+	}
+
+	if (of_property_read_u32(dma_node, "dma-requests",
+				 &xbar->dma_requests)) {
+		dev_info(&pdev->dev,
+			 "Missing XBAR output information, using %u.\n",
+			 TI_AM335X_XBAR_LINES);
+		xbar->dma_requests = TI_AM335X_XBAR_LINES;
+	}
+	of_node_put(dma_node);
+
+	if (of_property_read_u32(node, "dma-requests", &xbar->xbar_events)) {
+		dev_info(&pdev->dev,
+			 "Missing XBAR input information, using %u.\n",
+			 TI_AM335X_XBAR_LINES);
+		xbar->xbar_events = TI_AM335X_XBAR_LINES;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	iomem = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(iomem))
+		return PTR_ERR(iomem);
+
+	xbar->iomem = iomem;
+
+	xbar->dmarouter.dev = &pdev->dev;
+	xbar->dmarouter.route_free = ti_am335x_xbar_free;
+
+	platform_set_drvdata(pdev, xbar);
+
+	/* Reset the crossbar */
+	for (i = 0; i < xbar->dma_requests; i++)
+		ti_am335x_xbar_write(xbar->iomem, i, 0);
+
+	ret = of_dma_router_register(node, ti_am335x_xbar_route_allocate,
+				     &xbar->dmarouter);
+
+	return ret;
+}
+
+/* Crossbar on DRA7xx family */
+#define TI_DRA7_XBAR_OUTPUTS	127
+#define TI_DRA7_XBAR_INPUTS	256
 
 #define TI_XBAR_EDMA_OFFSET	0
 #define TI_XBAR_SDMA_OFFSET	1
 
-struct ti_dma_xbar_data {
+struct ti_dra7_xbar_data {
 	void __iomem *iomem;
 
 	struct dma_router dmarouter;
@@ -35,35 +206,35 @@ struct ti_dma_xbar_data {
 	u32 dma_offset;
 };
 
-struct ti_dma_xbar_map {
+struct ti_dra7_xbar_map {
 	u16 xbar_in;
 	int xbar_out;
 };
 
-static inline void ti_dma_xbar_write(void __iomem *iomem, int xbar, u16 val)
+static inline void ti_dra7_xbar_write(void __iomem *iomem, int xbar, u16 val)
 {
 	writew_relaxed(val, iomem + (xbar * 2));
 }
 
-static void ti_dma_xbar_free(struct device *dev, void *route_data)
+static void ti_dra7_xbar_free(struct device *dev, void *route_data)
 {
-	struct ti_dma_xbar_data *xbar = dev_get_drvdata(dev);
-	struct ti_dma_xbar_map *map = route_data;
+	struct ti_dra7_xbar_data *xbar = dev_get_drvdata(dev);
+	struct ti_dra7_xbar_map *map = route_data;
 
 	dev_dbg(dev, "Unmapping XBAR%u (was routed to %d)\n",
 		map->xbar_in, map->xbar_out);
 
-	ti_dma_xbar_write(xbar->iomem, map->xbar_out, xbar->safe_val);
+	ti_dra7_xbar_write(xbar->iomem, map->xbar_out, xbar->safe_val);
 	idr_remove(&xbar->map_idr, map->xbar_out);
 	kfree(map);
 }
 
-static void *ti_dma_xbar_route_allocate(struct of_phandle_args *dma_spec,
-					struct of_dma *ofdma)
+static void *ti_dra7_xbar_route_allocate(struct of_phandle_args *dma_spec,
+					 struct of_dma *ofdma)
 {
 	struct platform_device *pdev = of_find_device_by_node(ofdma->of_node);
-	struct ti_dma_xbar_data *xbar = platform_get_drvdata(pdev);
-	struct ti_dma_xbar_map *map;
+	struct ti_dra7_xbar_data *xbar = platform_get_drvdata(pdev);
+	struct ti_dra7_xbar_map *map;
 
 	if (dma_spec->args[0] >= xbar->xbar_requests) {
 		dev_err(&pdev->dev, "Invalid XBAR request number: %d\n",
@@ -93,12 +264,12 @@ static void *ti_dma_xbar_route_allocate(struct of_phandle_args *dma_spec,
 	dev_dbg(&pdev->dev, "Mapping XBAR%u to DMA%d\n",
 		map->xbar_in, map->xbar_out);
 
-	ti_dma_xbar_write(xbar->iomem, map->xbar_out, map->xbar_in);
+	ti_dra7_xbar_write(xbar->iomem, map->xbar_out, map->xbar_in);
 
 	return map;
 }
 
-static const struct of_device_id ti_dma_master_match[] = {
+static const struct of_device_id ti_dra7_master_match[] = {
 	{
 		.compatible = "ti,omap4430-sdma",
 		.data = (void *)TI_XBAR_SDMA_OFFSET,
@@ -110,12 +281,12 @@ static const struct of_device_id ti_dma_master_match[] = {
 	{},
 };
 
-static int ti_dma_xbar_probe(struct platform_device *pdev)
+static int ti_dra7_xbar_probe(struct platform_device *pdev)
 {
 	struct device_node *node = pdev->dev.of_node;
 	const struct of_device_id *match;
 	struct device_node *dma_node;
-	struct ti_dma_xbar_data *xbar;
+	struct ti_dra7_xbar_data *xbar;
 	struct resource *res;
 	u32 safe_val;
 	void __iomem *iomem;
@@ -136,7 +307,7 @@ static int ti_dma_xbar_probe(struct platform_device *pdev)
 		return -ENODEV;
 	}
 
-	match = of_match_node(ti_dma_master_match, dma_node);
+	match = of_match_node(ti_dra7_master_match, dma_node);
 	if (!match) {
 		dev_err(&pdev->dev, "DMA master is not supported\n");
 		return -EINVAL;
@@ -146,16 +317,16 @@ static int ti_dma_xbar_probe(struct platform_device *pdev)
 				 &xbar->dma_requests)) {
 		dev_info(&pdev->dev,
 			 "Missing XBAR output information, using %u.\n",
-			 TI_XBAR_OUTPUTS);
-		xbar->dma_requests = TI_XBAR_OUTPUTS;
+			 TI_DRA7_XBAR_OUTPUTS);
+		xbar->dma_requests = TI_DRA7_XBAR_OUTPUTS;
 	}
 	of_node_put(dma_node);
 
 	if (of_property_read_u32(node, "dma-requests", &xbar->xbar_requests)) {
 		dev_info(&pdev->dev,
 			 "Missing XBAR input information, using %u.\n",
-			 TI_XBAR_INPUTS);
-		xbar->xbar_requests = TI_XBAR_INPUTS;
+			 TI_DRA7_XBAR_INPUTS);
+		xbar->xbar_requests = TI_DRA7_XBAR_INPUTS;
 	}
 
 	if (!of_property_read_u32(node, "ti,dma-safe-map", &safe_val))
@@ -169,30 +340,50 @@ static int ti_dma_xbar_probe(struct platform_device *pdev)
 	xbar->iomem = iomem;
 
 	xbar->dmarouter.dev = &pdev->dev;
-	xbar->dmarouter.route_free = ti_dma_xbar_free;
+	xbar->dmarouter.route_free = ti_dra7_xbar_free;
 	xbar->dma_offset = (u32)match->data;
 
 	platform_set_drvdata(pdev, xbar);
 
 	/* Reset the crossbar */
 	for (i = 0; i < xbar->dma_requests; i++)
-		ti_dma_xbar_write(xbar->iomem, i, xbar->safe_val);
+		ti_dra7_xbar_write(xbar->iomem, i, xbar->safe_val);
 
-	ret = of_dma_router_register(node, ti_dma_xbar_route_allocate,
+	ret = of_dma_router_register(node, ti_dra7_xbar_route_allocate,
 				     &xbar->dmarouter);
 	if (ret) {
 		/* Restore the defaults for the crossbar */
 		for (i = 0; i < xbar->dma_requests; i++)
-			ti_dma_xbar_write(xbar->iomem, i, i);
+			ti_dra7_xbar_write(xbar->iomem, i, i);
 	}
 
 	return ret;
 }
 
-static const struct of_device_id ti_dma_xbar_match[] = {
-	{ .compatible = "ti,dra7-dma-crossbar" },
-	{},
-};
+static int ti_dma_xbar_probe(struct platform_device *pdev)
+{
+	const struct of_device_id *match;
+	int ret;
+
+	match = of_match_node(ti_dma_xbar_match, pdev->dev.of_node);
+	if (unlikely(!match))
+		return -EINVAL;
+
+	switch ((u32)match->data) {
+	case TI_XBAR_DRA7:
+		ret = ti_dra7_xbar_probe(pdev);
+		break;
+	case TI_XBAR_AM335X:
+		ret = ti_am335x_xbar_probe(pdev);
+		break;
+	default:
+		dev_err(&pdev->dev, "Unsupported crossbar\n");
+		ret = -ENODEV;
+		break;
+	}
+
+	return ret;
+}
 
 static struct platform_driver ti_dma_xbar_driver = {
 	.driver = {

drivers/dma/virt-dma.h

@@ -47,9 +47,9 @@ struct virt_dma_desc *vchan_find_desc(struct virt_dma_chan *, dma_cookie_t);
 
 /**
  * vchan_tx_prep - prepare a descriptor
- * vc: virtual channel allocating this descriptor
- * vd: virtual descriptor to prepare
- * tx_flags: flags argument passed in to prepare function
+ * @vc: virtual channel allocating this descriptor
+ * @vd: virtual descriptor to prepare
+ * @tx_flags: flags argument passed in to prepare function
  */
 static inline struct dma_async_tx_descriptor *vchan_tx_prep(struct virt_dma_chan *vc,
 	struct virt_dma_desc *vd, unsigned long tx_flags)
@@ -65,7 +65,7 @@ static inline struct dma_async_tx_descriptor *vchan_tx_prep(struct virt_dma_chan
 
 /**
  * vchan_issue_pending - move submitted descriptors to issued list
- * vc: virtual channel to update
+ * @vc: virtual channel to update
  *
  * vc.lock must be held by caller
  */
@@ -77,7 +77,7 @@ static inline bool vchan_issue_pending(struct virt_dma_chan *vc)
 
 /**
  * vchan_cookie_complete - report completion of a descriptor
- * vd: virtual descriptor to update
+ * @vd: virtual descriptor to update
  *
  * vc.lock must be held by caller
  */
@@ -97,7 +97,7 @@ static inline void vchan_cookie_complete(struct virt_dma_desc *vd)
 
 /**
  * vchan_cyclic_callback - report the completion of a period
- * vd: virtual descriptor
+ * @vd: virtual descriptor
  */
 static inline void vchan_cyclic_callback(struct virt_dma_desc *vd)
 {
@@ -109,7 +109,7 @@ static inline void vchan_cyclic_callback(struct virt_dma_desc *vd)
 
 /**
  * vchan_next_desc - peek at the next descriptor to be processed
- * vc: virtual channel to obtain descriptor from
+ * @vc: virtual channel to obtain descriptor from
  *
  * vc.lock must be held by caller
  */
@@ -123,8 +123,8 @@ static inline struct virt_dma_desc *vchan_next_desc(struct virt_dma_chan *vc)
 
 /**
  * vchan_get_all_descriptors - obtain all submitted and issued descriptors
- * vc: virtual channel to get descriptors from
- * head: list of descriptors found
+ * @vc: virtual channel to get descriptors from
+ * @head: list of descriptors found
  *
  * vc.lock must be held by caller
  *

drivers/dma/xgene-dma.c

@@ -547,14 +547,12 @@ static struct xgene_dma_desc_sw *xgene_dma_alloc_descriptor(
 	struct xgene_dma_desc_sw *desc;
 	dma_addr_t phys;
 
-	desc = dma_pool_alloc(chan->desc_pool, GFP_NOWAIT, &phys);
+	desc = dma_pool_zalloc(chan->desc_pool, GFP_NOWAIT, &phys);
 	if (!desc) {
 		chan_err(chan, "Failed to allocate LDs\n");
 		return NULL;
 	}
 
-	memset(desc, 0, sizeof(*desc));
-
 	INIT_LIST_HEAD(&desc->tx_list);
 	desc->tx.phys = phys;
 	desc->tx.tx_submit = xgene_dma_tx_submit;
@@ -894,60 +892,6 @@ static void xgene_dma_free_chan_resources(struct dma_chan *dchan)
 	chan->desc_pool = NULL;
 }
 
-static struct dma_async_tx_descriptor *xgene_dma_prep_memcpy(
-	struct dma_chan *dchan, dma_addr_t dst, dma_addr_t src,
-	size_t len, unsigned long flags)
-{
-	struct xgene_dma_desc_sw *first = NULL, *new;
-	struct xgene_dma_chan *chan;
-	size_t copy;
-
-	if (unlikely(!dchan || !len))
-		return NULL;
-
-	chan = to_dma_chan(dchan);
-
-	do {
-		/* Allocate the link descriptor from DMA pool */
-		new = xgene_dma_alloc_descriptor(chan);
-		if (!new)
-			goto fail;
-
-		/* Create the largest transaction possible */
-		copy = min_t(size_t, len, XGENE_DMA_MAX_64B_DESC_BYTE_CNT);
-
-		/* Prepare DMA descriptor */
-		xgene_dma_prep_cpy_desc(chan, new, dst, src, copy);
-
-		if (!first)
-			first = new;
-
-		new->tx.cookie = 0;
-		async_tx_ack(&new->tx);
-
-		/* Update metadata */
-		len -= copy;
-		dst += copy;
-		src += copy;
-
-		/* Insert the link descriptor to the LD ring */
-		list_add_tail(&new->node, &first->tx_list);
-	} while (len);
-
-	new->tx.flags = flags; /* client is in control of this ack */
-	new->tx.cookie = -EBUSY;
-	list_splice(&first->tx_list, &new->tx_list);
-
-	return &new->tx;
-
-fail:
-	if (!first)
-		return NULL;
-
-	xgene_dma_free_desc_list(chan, &first->tx_list);
-	return NULL;
-}
-
 static struct dma_async_tx_descriptor *xgene_dma_prep_sg(
 	struct dma_chan *dchan, struct scatterlist *dst_sg,
 	u32 dst_nents, struct scatterlist *src_sg,
@@ -1707,7 +1651,6 @@ static void xgene_dma_set_caps(struct xgene_dma_chan *chan,
 	dma_cap_zero(dma_dev->cap_mask);
 
 	/* Set DMA device capability */
-	dma_cap_set(DMA_MEMCPY, dma_dev->cap_mask);
 	dma_cap_set(DMA_SG, dma_dev->cap_mask);
 
 	/* Basically here, the X-Gene SoC DMA engine channel 0 supports XOR
@@ -1734,7 +1677,6 @@ static void xgene_dma_set_caps(struct xgene_dma_chan *chan,
 	dma_dev->device_free_chan_resources = xgene_dma_free_chan_resources;
 	dma_dev->device_issue_pending = xgene_dma_issue_pending;
 	dma_dev->device_tx_status = xgene_dma_tx_status;
-	dma_dev->device_prep_dma_memcpy = xgene_dma_prep_memcpy;
 	dma_dev->device_prep_dma_sg = xgene_dma_prep_sg;
 
 	if (dma_has_cap(DMA_XOR, dma_dev->cap_mask)) {
@@ -1787,8 +1729,7 @@ static int xgene_dma_async_register(struct xgene_dma *pdma, int id)
 
 	/* DMA capability info */
 	dev_info(pdma->dev,
-		 "%s: CAPABILITY ( %s%s%s%s)\n", dma_chan_name(&chan->dma_chan),
-		 dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask) ? "MEMCPY " : "",
+		 "%s: CAPABILITY ( %s%s%s)\n", dma_chan_name(&chan->dma_chan),
 		 dma_has_cap(DMA_SG, dma_dev->cap_mask) ? "SGCPY " : "",
 		 dma_has_cap(DMA_XOR, dma_dev->cap_mask) ? "XOR " : "",
 		 dma_has_cap(DMA_PQ, dma_dev->cap_mask) ? "PQ " : "");

drivers/dma/xilinx/xilinx_vdma.c

@@ -1349,6 +1349,7 @@ static const struct of_device_id xilinx_vdma_of_ids[] = {
 	{ .compatible = "xlnx,axi-vdma-1.00.a",},
 	{}
 };
+MODULE_DEVICE_TABLE(of, xilinx_vdma_of_ids);
 
 static struct platform_driver xilinx_vdma_driver = {
 	.driver = {

drivers/dma/zx296702_dma.c

@@ -441,7 +441,7 @@ static struct zx_dma_desc_sw *zx_alloc_desc_resource(int num,
 		kfree(ds);
 		return NULL;
 	}
-	memset(ds->desc_hw, sizeof(struct zx_desc_hw) * num, 0);
+	memset(ds->desc_hw, 0, sizeof(struct zx_desc_hw) * num);
 	ds->desc_num = num;
 	return ds;
 }

include/linux/of_dma.h

@@ -34,7 +34,7 @@ struct of_dma_filter_info {
 	dma_filter_fn	filter_fn;
 };
 
-#ifdef CONFIG_OF
+#ifdef CONFIG_DMA_OF
 extern int of_dma_controller_register(struct device_node *np,
 		struct dma_chan *(*of_dma_xlate)
 		(struct of_phandle_args *, struct of_dma *),

include/linux/platform_data/dma-dw.h

@@ -37,6 +37,7 @@ struct dw_dma_slave {
  * @nr_channels: Number of channels supported by hardware (max 8)
  * @is_private: The device channels should be marked as private and not for
  *	by the general purpose DMA channel allocator.
+ * @is_memcpy: The device channels do support memory-to-memory transfers.
  * @chan_allocation_order: Allocate channels starting from 0 or 7
  * @chan_priority: Set channel priority increasing from 0 to 7 or 7 to 0.
  * @block_size: Maximum block size supported by the controller
@@ -47,6 +48,7 @@ struct dw_dma_slave {
 struct dw_dma_platform_data {
 	unsigned int	nr_channels;
 	bool		is_private;
+	bool		is_memcpy;
 #define CHAN_ALLOCATION_ASCENDING	0	/* zero to seven */
 #define CHAN_ALLOCATION_DESCENDING	1	/* seven to zero */
 	unsigned char	chan_allocation_order;

include/linux/platform_data/edma.h

@@ -41,51 +41,6 @@
 #ifndef EDMA_H_
 #define EDMA_H_
 
-/* PaRAM slots are laid out like this */
-struct edmacc_param {
-	u32 opt;
-	u32 src;
-	u32 a_b_cnt;
-	u32 dst;
-	u32 src_dst_bidx;
-	u32 link_bcntrld;
-	u32 src_dst_cidx;
-	u32 ccnt;
-} __packed;
-
-/* fields in edmacc_param.opt */
-#define SAM		BIT(0)
-#define DAM		BIT(1)
-#define SYNCDIM		BIT(2)
-#define STATIC		BIT(3)
-#define EDMA_FWID	(0x07 << 8)
-#define TCCMODE		BIT(11)
-#define EDMA_TCC(t)	((t) << 12)
-#define TCINTEN		BIT(20)
-#define ITCINTEN	BIT(21)
-#define TCCHEN		BIT(22)
-#define ITCCHEN		BIT(23)
-
-/*ch_status paramater of callback function possible values*/
-#define EDMA_DMA_COMPLETE 1
-#define EDMA_DMA_CC_ERROR 2
-#define EDMA_DMA_TC1_ERROR 3
-#define EDMA_DMA_TC2_ERROR 4
-
-enum address_mode {
-	INCR = 0,
-	FIFO = 1
-};
-
-enum fifo_width {
-	W8BIT = 0,
-	W16BIT = 1,
-	W32BIT = 2,
-	W64BIT = 3,
-	W128BIT = 4,
-	W256BIT = 5
-};
-
 enum dma_event_q {
 	EVENTQ_0 = 0,
 	EVENTQ_1 = 1,
@@ -94,64 +49,10 @@ enum dma_event_q {
 	EVENTQ_DEFAULT = -1
 };
 
-enum sync_dimension {
-	ASYNC = 0,
-	ABSYNC = 1
-};
-
 #define EDMA_CTLR_CHAN(ctlr, chan)	(((ctlr) << 16) | (chan))
 #define EDMA_CTLR(i)			((i) >> 16)
 #define EDMA_CHAN_SLOT(i)		((i) & 0xffff)
 
-#define EDMA_CHANNEL_ANY		-1	/* for edma_alloc_channel() */
-#define EDMA_SLOT_ANY			-1	/* for edma_alloc_slot() */
-#define EDMA_CONT_PARAMS_ANY		 1001
-#define EDMA_CONT_PARAMS_FIXED_EXACT	 1002
-#define EDMA_CONT_PARAMS_FIXED_NOT_EXACT 1003
-
-#define EDMA_MAX_CC               2
-
-/* alloc/free DMA channels and their dedicated parameter RAM slots */
-int edma_alloc_channel(int channel,
-	void (*callback)(unsigned channel, u16 ch_status, void *data),
-	void *data, enum dma_event_q);
-void edma_free_channel(unsigned channel);
-
-/* alloc/free parameter RAM slots */
-int edma_alloc_slot(unsigned ctlr, int slot);
-void edma_free_slot(unsigned slot);
-
-/* alloc/free a set of contiguous parameter RAM slots */
-int edma_alloc_cont_slots(unsigned ctlr, unsigned int id, int slot, int count);
-int edma_free_cont_slots(unsigned slot, int count);
-
-/* calls that operate on part of a parameter RAM slot */
-void edma_set_src(unsigned slot, dma_addr_t src_port,
-				enum address_mode mode, enum fifo_width);
-void edma_set_dest(unsigned slot, dma_addr_t dest_port,
-				 enum address_mode mode, enum fifo_width);
-dma_addr_t edma_get_position(unsigned slot, bool dst);
-void edma_set_src_index(unsigned slot, s16 src_bidx, s16 src_cidx);
-void edma_set_dest_index(unsigned slot, s16 dest_bidx, s16 dest_cidx);
-void edma_set_transfer_params(unsigned slot, u16 acnt, u16 bcnt, u16 ccnt,
-		u16 bcnt_rld, enum sync_dimension sync_mode);
-void edma_link(unsigned from, unsigned to);
-void edma_unlink(unsigned from);
-
-/* calls that operate on an entire parameter RAM slot */
-void edma_write_slot(unsigned slot, const struct edmacc_param *params);
-void edma_read_slot(unsigned slot, struct edmacc_param *params);
-
-/* channel control operations */
-int edma_start(unsigned channel);
-void edma_stop(unsigned channel);
-void edma_clean_channel(unsigned channel);
-void edma_clear_event(unsigned channel);
-void edma_pause(unsigned channel);
-void edma_resume(unsigned channel);
-
-void edma_assign_channel_eventq(unsigned channel, enum dma_event_q eventq_no);
-
 struct edma_rsv_info {
 
 	const s16	(*rsv_chans)[2];
@@ -170,10 +71,11 @@ struct edma_soc_info {
 	/* Resource reservation for other cores */
 	struct edma_rsv_info	*rsv;
 
+	/* List of channels allocated for memcpy, terminated with -1 */
+	s16			*memcpy_channels;
+
 	s8	(*queue_priority_mapping)[2];
 	const s16	(*xbar_chans)[2];
 };
 
-int edma_trigger_channel(unsigned);
-
 #endif