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

Pull slave-dmaengine updates from Vinod Koul:
 "This time we have dmatest improvements from Andy along with dw_dmac
  fixes.  He has also done support for acpi for dmanegine.

  Also we have bunch of fixes going in DT support for dmanegine for
  various folks.  Then Haswell and other ioat changes from Dave and
  SUDMAC support from Shimoda."

* 'for-linus' of git://git.infradead.org/users/vkoul/slave-dma: (53 commits)
  dma: tegra: implement suspend/resume callbacks
  dma:of: Use a mutex to protect the of_dma_list
  dma: of: Fix of_node reference leak
  dmaengine: sirf: move driver init from module_init to subsys_initcall
  sudmac: add support for SUDMAC
  dma: sh: add Kconfig
  at_hdmac: move to generic DMA binding
  ioatdma: ioat3_alloc_sed can be static
  ioatdma: Adding write back descriptor error status support for ioatdma 3.3
  ioatdma: S1200 platforms ioatdma channel 2 and 3 falsely advertise RAID cap
  ioatdma: Adding support for 16 src PQ ops and super extended descriptors
  ioatdma: Removing hw bug workaround for CB3.x .2 and earlier
  dw_dmac: add ACPI support
  dmaengine: call acpi_dma_request_slave_channel as well
  dma: acpi-dma: introduce ACPI DMA helpers
  dma: of: Remove unnecessary list_empty check
  DMA: OF: Check properties value before running be32_to_cpup() on it
  DMA: of: Constant names
  ioatdma: skip silicon bug workaround for pq_align for cb3.3
  ioatdma: Removing PQ val disable for cb3.3
  ...

31 files changed, 2843 insertions, 499 deletions

diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index aeaea32bcfd..e9924898043 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -63,8 +63,6 @@ config INTEL_IOATDMA
 	depends on PCI && X86
 	select DMA_ENGINE
 	select DCA
-	select ASYNC_TX_DISABLE_PQ_VAL_DMA
-	select ASYNC_TX_DISABLE_XOR_VAL_DMA
 	help
 	  Enable support for the Intel(R) I/OAT DMA engine present
 	  in recent Intel Xeon chipsets.
@@ -174,15 +172,7 @@ config TEGRA20_APB_DMA
 	  This DMA controller transfers data from memory to peripheral fifo
 	  or vice versa. It does not support memory to memory data transfer.
 
-
-
-config SH_DMAE
-	tristate "Renesas SuperH DMAC support"
-	depends on (SUPERH && SH_DMA) || (ARM && ARCH_SHMOBILE)
-	depends on !SH_DMA_API
-	select DMA_ENGINE
-	help
-	  Enable support for the Renesas SuperH DMA controllers.
+source "drivers/dma/sh/Kconfig"
 
 config COH901318
 	bool "ST-Ericsson COH901318 DMA support"
@@ -328,6 +318,10 @@ config DMA_ENGINE
 config DMA_VIRTUAL_CHANNELS
 	tristate
 
+config DMA_ACPI
+	def_bool y
+	depends on ACPI
+
 config DMA_OF
 	def_bool y
 	depends on OF
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 488e3ff85b5..a2b0df591f9 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -3,6 +3,7 @@ ccflags-$(CONFIG_DMADEVICES_VDEBUG) += -DVERBOSE_DEBUG
 
 obj-$(CONFIG_DMA_ENGINE) += dmaengine.o
 obj-$(CONFIG_DMA_VIRTUAL_CHANNELS) += virt-dma.o
+obj-$(CONFIG_DMA_ACPI) += acpi-dma.o
 obj-$(CONFIG_DMA_OF) += of-dma.o
 
 obj-$(CONFIG_NET_DMA) += iovlock.o
@@ -18,7 +19,7 @@ obj-$(CONFIG_DW_DMAC) += dw_dmac.o
 obj-$(CONFIG_AT_HDMAC) += at_hdmac.o
 obj-$(CONFIG_MX3_IPU) += ipu/
 obj-$(CONFIG_TXX9_DMAC) += txx9dmac.o
-obj-$(CONFIG_SH_DMAE) += sh/
+obj-$(CONFIG_SH_DMAE_BASE) += sh/
 obj-$(CONFIG_COH901318) += coh901318.o coh901318_lli.o
 obj-$(CONFIG_AMCC_PPC440SPE_ADMA) += ppc4xx/
 obj-$(CONFIG_IMX_SDMA) += imx-sdma.o
diff --git a/drivers/dma/acpi-dma.c b/drivers/dma/acpi-dma.c
new file mode 100644
index 00000000000..ba6fc62e965
--- /dev/null
+++ b/drivers/dma/acpi-dma.c
@@ -0,0 +1,279 @@
+/*
+ * ACPI helpers for DMA request / controller
+ *
+ * Based on of-dma.c
+ *
+ * Copyright (C) 2013, Intel Corporation
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/acpi.h>
+#include <linux/acpi_dma.h>
+
+static LIST_HEAD(acpi_dma_list);
+static DEFINE_MUTEX(acpi_dma_lock);
+
+/**
+ * acpi_dma_controller_register - Register a DMA controller to ACPI DMA helpers
+ * @dev:		struct device of DMA controller
+ * @acpi_dma_xlate:	translation function which converts a dma specifier
+ *			into a dma_chan structure
+ * @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.
+ */
+int acpi_dma_controller_register(struct device *dev,
+		struct dma_chan *(*acpi_dma_xlate)
+		(struct acpi_dma_spec *, struct acpi_dma *),
+		void *data)
+{
+	struct acpi_device *adev;
+	struct acpi_dma	*adma;
+
+	if (!dev || !acpi_dma_xlate)
+		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))
+		return -EINVAL;
+
+	adma = kzalloc(sizeof(*adma), GFP_KERNEL);
+	if (!adma)
+		return -ENOMEM;
+
+	adma->dev = dev;
+	adma->acpi_dma_xlate = acpi_dma_xlate;
+	adma->data = data;
+
+	/* Now queue acpi_dma controller structure in list */
+	mutex_lock(&acpi_dma_lock);
+	list_add_tail(&adma->dma_controllers, &acpi_dma_list);
+	mutex_unlock(&acpi_dma_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(acpi_dma_controller_register);
+
+/**
+ * acpi_dma_controller_free - Remove a DMA controller from ACPI DMA helpers list
+ * @dev:	struct device of DMA controller
+ *
+ * Memory allocated by acpi_dma_controller_register() is freed here.
+ */
+int acpi_dma_controller_free(struct device *dev)
+{
+	struct acpi_dma *adma;
+
+	if (!dev)
+		return -EINVAL;
+
+	mutex_lock(&acpi_dma_lock);
+
+	list_for_each_entry(adma, &acpi_dma_list, dma_controllers)
+		if (adma->dev == dev) {
+			list_del(&adma->dma_controllers);
+			mutex_unlock(&acpi_dma_lock);
+			kfree(adma);
+			return 0;
+		}
+
+	mutex_unlock(&acpi_dma_lock);
+	return -ENODEV;
+}
+EXPORT_SYMBOL_GPL(acpi_dma_controller_free);
+
+static void devm_acpi_dma_release(struct device *dev, void *res)
+{
+	acpi_dma_controller_free(dev);
+}
+
+/**
+ * devm_acpi_dma_controller_register - resource managed acpi_dma_controller_register()
+ * @dev:		device that is registering this DMA controller
+ * @acpi_dma_xlate:	translation function
+ * @data		pointer to controller specific data
+ *
+ * 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.
+ */
+int devm_acpi_dma_controller_register(struct device *dev,
+		struct dma_chan *(*acpi_dma_xlate)
+		(struct acpi_dma_spec *, struct acpi_dma *),
+		void *data)
+{
+	void *res;
+	int ret;
+
+	res = devres_alloc(devm_acpi_dma_release, 0, GFP_KERNEL);
+	if (!res)
+		return -ENOMEM;
+
+	ret = acpi_dma_controller_register(dev, acpi_dma_xlate, data);
+	if (ret) {
+		devres_free(res);
+		return ret;
+	}
+	devres_add(dev, res);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(devm_acpi_dma_controller_register);
+
+/**
+ * devm_acpi_dma_controller_free - resource managed acpi_dma_controller_free()
+ *
+ * Unregister a DMA controller registered with
+ * devm_acpi_dma_controller_register(). Normally this function will not need to
+ * be called and the resource management code will ensure that the resource is
+ * freed.
+ */
+void devm_acpi_dma_controller_free(struct device *dev)
+{
+	WARN_ON(devres_destroy(dev, devm_acpi_dma_release, NULL, NULL));
+}
+EXPORT_SYMBOL_GPL(devm_acpi_dma_controller_free);
+
+struct acpi_dma_parser_data {
+	struct acpi_dma_spec dma_spec;
+	size_t index;
+	size_t n;
+};
+
+/**
+ * acpi_dma_parse_fixed_dma - Parse FixedDMA ACPI resources to a DMA specifier
+ * @res:	struct acpi_resource to get FixedDMA resources from
+ * @data:	pointer to a helper struct acpi_dma_parser_data
+ */
+static int acpi_dma_parse_fixed_dma(struct acpi_resource *res, void *data)
+{
+	struct acpi_dma_parser_data *pdata = data;
+
+	if (res->type == ACPI_RESOURCE_TYPE_FIXED_DMA) {
+		struct acpi_resource_fixed_dma *dma = &res->data.fixed_dma;
+
+		if (pdata->n++ == pdata->index) {
+			pdata->dma_spec.chan_id = dma->channels;
+			pdata->dma_spec.slave_id = dma->request_lines;
+		}
+	}
+
+	/* Tell the ACPI core to skip this resource */
+	return 1;
+}
+
+/**
+ * acpi_dma_request_slave_chan_by_index - Get the DMA slave channel
+ * @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.
+ */
+struct dma_chan *acpi_dma_request_slave_chan_by_index(struct device *dev,
+		size_t index)
+{
+	struct acpi_dma_parser_data pdata;
+	struct acpi_dma_spec *dma_spec = &pdata.dma_spec;
+	struct list_head resource_list;
+	struct acpi_device *adev;
+	struct acpi_dma *adma;
+	struct dma_chan *chan = NULL;
+
+	/* Check if the device was enumerated by ACPI */
+	if (!dev || !ACPI_HANDLE(dev))
+		return NULL;
+
+	if (acpi_bus_get_device(ACPI_HANDLE(dev), &adev))
+		return NULL;
+
+	memset(&pdata, 0, sizeof(pdata));
+	pdata.index = index;
+
+	/* Initial values for the request line and channel */
+	dma_spec->chan_id = -1;
+	dma_spec->slave_id = -1;
+
+	INIT_LIST_HEAD(&resource_list);
+	acpi_dev_get_resources(adev, &resource_list,
+			acpi_dma_parse_fixed_dma, &pdata);
+	acpi_dev_free_resource_list(&resource_list);
+
+	if (dma_spec->slave_id < 0 || dma_spec->chan_id < 0)
+		return NULL;
+
+	mutex_lock(&acpi_dma_lock);
+
+	list_for_each_entry(adma, &acpi_dma_list, dma_controllers) {
+		dma_spec->dev = adma->dev;
+		chan = adma->acpi_dma_xlate(dma_spec, adma);
+		if (chan)
+			break;
+	}
+
+	mutex_unlock(&acpi_dma_lock);
+	return chan;
+}
+EXPORT_SYMBOL_GPL(acpi_dma_request_slave_chan_by_index);
+
+/**
+ * acpi_dma_request_slave_chan_by_name - Get the DMA slave channel
+ * @dev:	struct device to get DMA request from
+ * @name:	represents corresponding FixedDMA descriptor for @dev
+ *
+ * In order to support both Device Tree and ACPI in a single driver we
+ * 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.
+ */
+struct dma_chan *acpi_dma_request_slave_chan_by_name(struct device *dev,
+		const char *name)
+{
+	size_t index;
+
+	if (!strcmp(name, "tx"))
+		index = 0;
+	else if (!strcmp(name, "rx"))
+		index = 1;
+	else
+		return NULL;
+
+	return acpi_dma_request_slave_chan_by_index(dev, index);
+}
+EXPORT_SYMBOL_GPL(acpi_dma_request_slave_chan_by_name);
+
+/**
+ * acpi_dma_simple_xlate - Simple ACPI DMA engine translation helper
+ * @dma_spec: pointer to ACPI DMA specifier
+ * @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.
+ */
+struct dma_chan *acpi_dma_simple_xlate(struct acpi_dma_spec *dma_spec,
+		struct acpi_dma *adma)
+{
+	struct acpi_dma_filter_info *info = adma->data;
+
+	if (!info || !info->filter_fn)
+		return NULL;
+
+	return dma_request_channel(info->dma_cap, info->filter_fn, dma_spec);
+}
+EXPORT_SYMBOL_GPL(acpi_dma_simple_xlate);
diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c
index 88cfc61329d..e923cda930f 100644
--- a/drivers/dma/at_hdmac.c
+++ b/drivers/dma/at_hdmac.c
@@ -24,6 +24,7 @@
 #include <linux/slab.h>
 #include <linux/of.h>
 #include <linux/of_device.h>
+#include <linux/of_dma.h>
 
 #include "at_hdmac_regs.h"
 #include "dmaengine.h"
@@ -677,7 +678,7 @@ atc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		ctrlb |=  ATC_DST_ADDR_MODE_FIXED
 			| ATC_SRC_ADDR_MODE_INCR
 			| ATC_FC_MEM2PER
-			| ATC_SIF(AT_DMA_MEM_IF) | ATC_DIF(AT_DMA_PER_IF);
+			| ATC_SIF(atchan->mem_if) | ATC_DIF(atchan->per_if);
 		reg = sconfig->dst_addr;
 		for_each_sg(sgl, sg, sg_len, i) {
 			struct at_desc	*desc;
@@ -716,7 +717,7 @@ atc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		ctrlb |=  ATC_DST_ADDR_MODE_INCR
 			| ATC_SRC_ADDR_MODE_FIXED
 			| ATC_FC_PER2MEM
-			| ATC_SIF(AT_DMA_PER_IF) | ATC_DIF(AT_DMA_MEM_IF);
+			| ATC_SIF(atchan->per_if) | ATC_DIF(atchan->mem_if);
 
 		reg = sconfig->src_addr;
 		for_each_sg(sgl, sg, sg_len, i) {
@@ -822,8 +823,8 @@ atc_dma_cyclic_fill_desc(struct dma_chan *chan, struct at_desc *desc,
 		desc->lli.ctrlb = ATC_DST_ADDR_MODE_FIXED
 				| ATC_SRC_ADDR_MODE_INCR
 				| ATC_FC_MEM2PER
-				| ATC_SIF(AT_DMA_MEM_IF)
-				| ATC_DIF(AT_DMA_PER_IF);
+				| ATC_SIF(atchan->mem_if)
+				| ATC_DIF(atchan->per_if);
 		break;
 
 	case DMA_DEV_TO_MEM:
@@ -833,8 +834,8 @@ atc_dma_cyclic_fill_desc(struct dma_chan *chan, struct at_desc *desc,
 		desc->lli.ctrlb = ATC_DST_ADDR_MODE_INCR
 				| ATC_SRC_ADDR_MODE_FIXED
 				| ATC_FC_PER2MEM
-				| ATC_SIF(AT_DMA_PER_IF)
-				| ATC_DIF(AT_DMA_MEM_IF);
+				| ATC_SIF(atchan->per_if)
+				| ATC_DIF(atchan->mem_if);
 		break;
 
 	default:
@@ -1188,6 +1189,67 @@ static void atc_free_chan_resources(struct dma_chan *chan)
 	dev_vdbg(chan2dev(chan), "free_chan_resources: done\n");
 }
 
+#ifdef CONFIG_OF
+static bool at_dma_filter(struct dma_chan *chan, void *slave)
+{
+	struct at_dma_slave *atslave = slave;
+
+	if (atslave->dma_dev == chan->device->dev) {
+		chan->private = atslave;
+		return true;
+	} else {
+		return false;
+	}
+}
+
+static struct dma_chan *at_dma_xlate(struct of_phandle_args *dma_spec,
+				     struct of_dma *of_dma)
+{
+	struct dma_chan *chan;
+	struct at_dma_chan *atchan;
+	struct at_dma_slave *atslave;
+	dma_cap_mask_t mask;
+	unsigned int per_id;
+	struct platform_device *dmac_pdev;
+
+	if (dma_spec->args_count != 2)
+		return NULL;
+
+	dmac_pdev = of_find_device_by_node(dma_spec->np);
+
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+
+	atslave = devm_kzalloc(&dmac_pdev->dev, sizeof(*atslave), GFP_KERNEL);
+	if (!atslave)
+		return NULL;
+	/*
+	 * We can fill both SRC_PER and DST_PER, one of these fields will be
+	 * ignored depending on DMA transfer direction.
+	 */
+	per_id = dma_spec->args[1];
+	atslave->cfg = ATC_FIFOCFG_HALFFIFO | ATC_DST_H2SEL_HW
+		      | ATC_SRC_H2SEL_HW | ATC_DST_PER(per_id)
+		      | ATC_SRC_PER(per_id);
+	atslave->dma_dev = &dmac_pdev->dev;
+
+	chan = dma_request_channel(mask, at_dma_filter, atslave);
+	if (!chan)
+		return NULL;
+
+	atchan = to_at_dma_chan(chan);
+	atchan->per_if = dma_spec->args[0] & 0xff;
+	atchan->mem_if = (dma_spec->args[0] >> 16) & 0xff;
+
+	return chan;
+}
+#else
+static struct dma_chan *at_dma_xlate(struct of_phandle_args *dma_spec,
+				     struct of_dma *of_dma)
+{
+	return NULL;
+}
+#endif
 
 /*--  Module Management  -----------------------------------------------*/
 
@@ -1342,6 +1404,8 @@ static int __init at_dma_probe(struct platform_device *pdev)
 	for (i = 0; i < plat_dat->nr_channels; i++) {
 		struct at_dma_chan	*atchan = &atdma->chan[i];
 
+		atchan->mem_if = AT_DMA_MEM_IF;
+		atchan->per_if = AT_DMA_PER_IF;
 		atchan->chan_common.device = &atdma->dma_common;
 		dma_cookie_init(&atchan->chan_common);
 		list_add_tail(&atchan->chan_common.device_node,
@@ -1388,8 +1452,25 @@ static int __init at_dma_probe(struct platform_device *pdev)
 
 	dma_async_device_register(&atdma->dma_common);
 
+	/*
+	 * Do not return an error if the dmac node is not present in order to
+	 * not break the existing way of requesting channel with
+	 * dma_request_channel().
+	 */
+	if (pdev->dev.of_node) {
+		err = of_dma_controller_register(pdev->dev.of_node,
+						 at_dma_xlate, atdma);
+		if (err) {
+			dev_err(&pdev->dev, "could not register of_dma_controller\n");
+			goto err_of_dma_controller_register;
+		}
+	}
+
 	return 0;
 
+err_of_dma_controller_register:
+	dma_async_device_unregister(&atdma->dma_common);
+	dma_pool_destroy(atdma->dma_desc_pool);
 err_pool_create:
 	platform_set_drvdata(pdev, NULL);
 	free_irq(platform_get_irq(pdev, 0), atdma);
@@ -1406,7 +1487,7 @@ err_kfree:
 	return err;
 }
 
-static int __exit at_dma_remove(struct platform_device *pdev)
+static int at_dma_remove(struct platform_device *pdev)
 {
 	struct at_dma		*atdma = platform_get_drvdata(pdev);
 	struct dma_chan		*chan, *_chan;
@@ -1564,7 +1645,7 @@ static const struct dev_pm_ops at_dma_dev_pm_ops = {
 };
 
 static struct platform_driver at_dma_driver = {
-	.remove		= __exit_p(at_dma_remove),
+	.remove		= at_dma_remove,
 	.shutdown	= at_dma_shutdown,
 	.id_table	= atdma_devtypes,
 	.driver = {
diff --git a/drivers/dma/at_hdmac_regs.h b/drivers/dma/at_hdmac_regs.h
index 0eb3c138866..c604d26fd4d 100644
--- a/drivers/dma/at_hdmac_regs.h
+++ b/drivers/dma/at_hdmac_regs.h
@@ -220,6 +220,8 @@ enum atc_status {
  * @device: parent device
  * @ch_regs: memory mapped register base
  * @mask: channel index in a mask
+ * @per_if: peripheral interface
+ * @mem_if: memory interface
  * @status: transmit status information from irq/prep* functions
  *                to tasklet (use atomic operations)
  * @tasklet: bottom half to finish transaction work
@@ -238,6 +240,8 @@ struct at_dma_chan {
 	struct at_dma		*device;
 	void __iomem		*ch_regs;
 	u8			mask;
+	u8			per_if;
+	u8			mem_if;
 	unsigned long		status;
 	struct tasklet_struct	tasklet;
 	u32			save_cfg;
diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c
index 797940e532f..3b23061cdb4 100644
--- a/drivers/dma/coh901318.c
+++ b/drivers/dma/coh901318.c
@@ -2748,7 +2748,7 @@ static int __init coh901318_probe(struct platform_device *pdev)
 	return err;
 }
 
-static int __exit coh901318_remove(struct platform_device *pdev)
+static int coh901318_remove(struct platform_device *pdev)
 {
 	struct coh901318_base *base = platform_get_drvdata(pdev);
 
@@ -2760,7 +2760,7 @@ static int __exit coh901318_remove(struct platform_device *pdev)
 
 
 static struct platform_driver coh901318_driver = {
-	.remove = __exit_p(coh901318_remove),
+	.remove = coh901318_remove,
 	.driver = {
 		.name	= "coh901318",
 	},
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index b2728d6ba2f..93f7992bee5 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -62,6 +62,8 @@
 #include <linux/rculist.h>
 #include <linux/idr.h>
 #include <linux/slab.h>
+#include <linux/acpi.h>
+#include <linux/acpi_dma.h>
 #include <linux/of_dma.h>
 
 static DEFINE_MUTEX(dma_list_mutex);
@@ -174,7 +176,8 @@ static struct class dma_devclass = {
 #define dma_device_satisfies_mask(device, mask) \
 	__dma_device_satisfies_mask((device), &(mask))
 static int
-__dma_device_satisfies_mask(struct dma_device *device, dma_cap_mask_t *want)
+__dma_device_satisfies_mask(struct dma_device *device,
+			    const dma_cap_mask_t *want)
 {
 	dma_cap_mask_t has;
 
@@ -463,7 +466,8 @@ static void dma_channel_rebalance(void)
 		}
 }
 
-static struct dma_chan *private_candidate(dma_cap_mask_t *mask, struct dma_device *dev,
+static struct dma_chan *private_candidate(const dma_cap_mask_t *mask,
+					  struct dma_device *dev,
 					  dma_filter_fn fn, void *fn_param)
 {
 	struct dma_chan *chan;
@@ -505,7 +509,8 @@ static struct dma_chan *private_candidate(dma_cap_mask_t *mask, struct dma_devic
  * @fn: optional callback to disposition available channels
  * @fn_param: opaque parameter to pass to dma_filter_fn
  */
-struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, void *fn_param)
+struct dma_chan *__dma_request_channel(const dma_cap_mask_t *mask,
+				       dma_filter_fn fn, void *fn_param)
 {
 	struct dma_device *device, *_d;
 	struct dma_chan *chan = NULL;
@@ -555,12 +560,16 @@ EXPORT_SYMBOL_GPL(__dma_request_channel);
  * @dev:	pointer to client device structure
  * @name:	slave channel name
  */
-struct dma_chan *dma_request_slave_channel(struct device *dev, char *name)
+struct dma_chan *dma_request_slave_channel(struct device *dev, const char *name)
 {
 	/* If device-tree is present get slave info from here */
 	if (dev->of_node)
 		return of_dma_request_slave_channel(dev->of_node, name);
 
+	/* If device was enumerated by ACPI get slave info from here */
+	if (ACPI_HANDLE(dev))
+		return acpi_dma_request_slave_chan_by_name(dev, name);
+
 	return NULL;
 }
 EXPORT_SYMBOL_GPL(dma_request_slave_channel);
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index a2c8904b63e..d8ce4ecfef1 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -2,6 +2,7 @@
  * DMA Engine test module
  *
  * Copyright (C) 2007 Atmel Corporation
+ * Copyright (C) 2013 Intel Corporation
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -18,6 +19,10 @@
 #include <linux/random.h>
 #include <linux/slab.h>
 #include <linux/wait.h>
+#include <linux/ctype.h>
+#include <linux/debugfs.h>
+#include <linux/uaccess.h>
+#include <linux/seq_file.h>
 
 static unsigned int test_buf_size = 16384;
 module_param(test_buf_size, uint, S_IRUGO);
@@ -61,6 +66,9 @@ module_param(timeout, uint, S_IRUGO);
 MODULE_PARM_DESC(timeout, "Transfer Timeout in msec (default: 3000), "
 		 "Pass -1 for infinite timeout");
 
+/* Maximum amount of mismatched bytes in buffer to print */
+#define MAX_ERROR_COUNT		32
+
 /*
  * Initialization patterns. All bytes in the source buffer has bit 7
  * set, all bytes in the destination buffer has bit 7 cleared.
@@ -78,13 +86,65 @@ MODULE_PARM_DESC(timeout, "Transfer Timeout in msec (default: 3000), "
 #define PATTERN_OVERWRITE	0x20
 #define PATTERN_COUNT_MASK	0x1f
 
+enum dmatest_error_type {
+	DMATEST_ET_OK,
+	DMATEST_ET_MAP_SRC,
+	DMATEST_ET_MAP_DST,
+	DMATEST_ET_PREP,
+	DMATEST_ET_SUBMIT,
+	DMATEST_ET_TIMEOUT,
+	DMATEST_ET_DMA_ERROR,
+	DMATEST_ET_DMA_IN_PROGRESS,
+	DMATEST_ET_VERIFY,
+	DMATEST_ET_VERIFY_BUF,
+};
+
+struct dmatest_verify_buffer {
+	unsigned int	index;
+	u8		expected;
+	u8		actual;
+};
+
+struct dmatest_verify_result {
+	unsigned int			error_count;
+	struct dmatest_verify_buffer	data[MAX_ERROR_COUNT];
+	u8				pattern;
+	bool				is_srcbuf;
+};
+
+struct dmatest_thread_result {
+	struct list_head	node;
+	unsigned int		n;
+	unsigned int		src_off;
+	unsigned int		dst_off;
+	unsigned int		len;
+	enum dmatest_error_type	type;
+	union {
+		unsigned long			data;
+		dma_cookie_t			cookie;
+		enum dma_status			status;
+		int				error;
+		struct dmatest_verify_result	*vr;
+	};
+};
+
+struct dmatest_result {
+	struct list_head	node;
+	char			*name;
+	struct list_head	results;
+};
+
+struct dmatest_info;
+
 struct dmatest_thread {
 	struct list_head	node;
+	struct dmatest_info	*info;
 	struct task_struct	*task;
 	struct dma_chan		*chan;
 	u8			**srcs;
 	u8			**dsts;
 	enum dma_transaction_type type;
+	bool			done;
 };
 
 struct dmatest_chan {
@@ -93,25 +153,69 @@ struct dmatest_chan {
 	struct list_head	threads;
 };
 
-/*
- * These are protected by dma_list_mutex since they're only used by
- * the DMA filter function callback
+/**
+ * struct dmatest_params - test parameters.
+ * @buf_size:		size of the memcpy test buffer
+ * @channel:		bus ID of the channel to test
+ * @device:		bus ID of the DMA Engine to test
+ * @threads_per_chan:	number of threads to start per channel
+ * @max_channels:	maximum number of channels to use
+ * @iterations:		iterations before stopping test
+ * @xor_sources:	number of xor source buffers
+ * @pq_sources:		number of p+q source buffers
+ * @timeout:		transfer timeout in msec, -1 for infinite timeout
  */
-static LIST_HEAD(dmatest_channels);
-static unsigned int nr_channels;
+struct dmatest_params {
+	unsigned int	buf_size;
+	char		channel[20];
+	char		device[20];
+	unsigned int	threads_per_chan;
+	unsigned int	max_channels;
+	unsigned int	iterations;
+	unsigned int	xor_sources;
+	unsigned int	pq_sources;
+	int		timeout;
+};
 
-static bool dmatest_match_channel(struct dma_chan *chan)
+/**
+ * struct dmatest_info - test information.
+ * @params:		test parameters
+ * @lock:		access protection to the fields of this structure
+ */
+struct dmatest_info {
+	/* Test parameters */
+	struct dmatest_params	params;
+
+	/* Internal state */
+	struct list_head	channels;
+	unsigned int		nr_channels;
+	struct mutex		lock;
+
+	/* debugfs related stuff */
+	struct dentry		*root;
+	struct dmatest_params	dbgfs_params;
+
+	/* Test results */
+	struct list_head	results;
+	struct mutex		results_lock;
+};
+
+static struct dmatest_info test_info;
+
+static bool dmatest_match_channel(struct dmatest_params *params,
+		struct dma_chan *chan)
 {
-	if (test_channel[0] == '\0')
+	if (params->channel[0] == '\0')
 		return true;
-	return strcmp(dma_chan_name(chan), test_channel) == 0;
+	return strcmp(dma_chan_name(chan), params->channel) == 0;
 }
 
-static bool dmatest_match_device(struct dma_device *device)
+static bool dmatest_match_device(struct dmatest_params *params,
+		struct dma_device *device)
 {
-	if (test_device[0] == '\0')
+	if (params->device[0] == '\0')
 		return true;
-	return strcmp(dev_name(device->dev), test_device) == 0;
+	return strcmp(dev_name(device->dev), params->device) == 0;
 }
 
 static unsigned long dmatest_random(void)
@@ -122,7 +226,8 @@ static unsigned long dmatest_random(void)
 	return buf;
 }
 
-static void dmatest_init_srcs(u8 **bufs, unsigned int start, unsigned int len)
+static void dmatest_init_srcs(u8 **bufs, unsigned int start, unsigned int len,
+		unsigned int buf_size)
 {
 	unsigned int i;
 	u8 *buf;
@@ -133,13 +238,14 @@ static void dmatest_init_srcs(u8 **bufs, unsigned int start, unsigned int len)
 		for ( ; i < start + len; i++)
 			buf[i] = PATTERN_SRC | PATTERN_COPY
 				| (~i & PATTERN_COUNT_MASK);
-		for ( ; i < test_buf_size; i++)
+		for ( ; i < buf_size; i++)
 			buf[i] = PATTERN_SRC | (~i & PATTERN_COUNT_MASK);
 		buf++;
 	}
 }
 
-static void dmatest_init_dsts(u8 **bufs, unsigned int start, unsigned int len)
+static void dmatest_init_dsts(u8 **bufs, unsigned int start, unsigned int len,
+		unsigned int buf_size)
 {
 	unsigned int i;
 	u8 *buf;
@@ -150,40 +256,14 @@ static void dmatest_init_dsts(u8 **bufs, unsigned int start, unsigned int len)
 		for ( ; i < start + len; i++)
 			buf[i] = PATTERN_DST | PATTERN_OVERWRITE
 				| (~i & PATTERN_COUNT_MASK);
-		for ( ; i < test_buf_size; i++)
+		for ( ; i < buf_size; i++)
 			buf[i] = PATTERN_DST | (~i & PATTERN_COUNT_MASK);
 	}
 }
 
-static void dmatest_mismatch(u8 actual, u8 pattern, unsigned int index,
-		unsigned int counter, bool is_srcbuf)
-{
-	u8		diff = actual ^ pattern;
-	u8		expected = pattern | (~counter & PATTERN_COUNT_MASK);
-	const char	*thread_name = current->comm;
-
-	if (is_srcbuf)
-		pr_warning("%s: srcbuf[0x%x] overwritten!"
-				" Expected %02x, got %02x\n",
-				thread_name, index, expected, actual);
-	else if ((pattern & PATTERN_COPY)
-			&& (diff & (PATTERN_COPY | PATTERN_OVERWRITE)))
-		pr_warning("%s: dstbuf[0x%x] not copied!"
-				" Expected %02x, got %02x\n",
-				thread_name, index, expected, actual);
-	else if (diff & PATTERN_SRC)
-		pr_warning("%s: dstbuf[0x%x] was copied!"
-				" Expected %02x, got %02x\n",
-				thread_name, index, expected, actual);
-	else
-		pr_warning("%s: dstbuf[0x%x] mismatch!"
-				" Expected %02x, got %02x\n",
-				thread_name, index, expected, actual);
-}
-
-static unsigned int dmatest_verify(u8 **bufs, unsigned int start,
-		unsigned int end, unsigned int counter, u8 pattern,
-		bool is_srcbuf)
+static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
+		unsigned int start, unsigned int end, unsigned int counter,
+		u8 pattern, bool is_srcbuf)
 {
 	unsigned int i;
 	unsigned int error_count = 0;
@@ -191,6 +271,7 @@ static unsigned int dmatest_verify(u8 **bufs, unsigned int start,
 	u8 expected;
 	u8 *buf;
 	unsigned int counter_orig = counter;
+	struct dmatest_verify_buffer *vb;
 
 	for (; (buf = *bufs); bufs++) {
 		counter = counter_orig;
@@ -198,18 +279,21 @@ static unsigned int dmatest_verify(u8 **bufs, unsigned int start,
 			actual = buf[i];
 			expected = pattern | (~counter & PATTERN_COUNT_MASK);
 			if (actual != expected) {
-				if (error_count < 32)
-					dmatest_mismatch(actual, pattern, i,
-							 counter, is_srcbuf);
+				if (error_count < MAX_ERROR_COUNT && vr) {
+					vb = &vr->data[error_count];
+					vb->index = i;
+					vb->expected = expected;
+					vb->actual = actual;
+				}
 				error_count++;
 			}
 			counter++;
 		}
 	}
 
-	if (error_count > 32)
+	if (error_count > MAX_ERROR_COUNT)
 		pr_warning("%s: %u errors suppressed\n",
-			current->comm, error_count - 32);
+			current->comm, error_count - MAX_ERROR_COUNT);
 
 	return error_count;
 }
@@ -249,6 +333,170 @@ static unsigned int min_odd(unsigned int x, unsigned int y)
 	return val % 2 ? val : val - 1;
 }
 
+static char *verify_result_get_one(struct dmatest_verify_result *vr,
+		unsigned int i)
+{
+	struct dmatest_verify_buffer *vb = &vr->data[i];
+	u8 diff = vb->actual ^ vr->pattern;
+	static char buf[512];
+	char *msg;
+
+	if (vr->is_srcbuf)
+		msg = "srcbuf overwritten!";
+	else if ((vr->pattern & PATTERN_COPY)
+			&& (diff & (PATTERN_COPY | PATTERN_OVERWRITE)))
+		msg = "dstbuf not copied!";
+	else if (diff & PATTERN_SRC)
+		msg = "dstbuf was copied!";
+	else
+		msg = "dstbuf mismatch!";
+
+	snprintf(buf, sizeof(buf) - 1, "%s [0x%x] Expected %02x, got %02x", msg,
+		 vb->index, vb->expected, vb->actual);
+
+	return buf;
+}
+
+static char *thread_result_get(const char *name,
+		struct dmatest_thread_result *tr)
+{
+	static const char * const messages[] = {
+		[DMATEST_ET_OK]			= "No errors",
+		[DMATEST_ET_MAP_SRC]		= "src mapping error",
+		[DMATEST_ET_MAP_DST]		= "dst mapping error",
+		[DMATEST_ET_PREP]		= "prep error",
+		[DMATEST_ET_SUBMIT]		= "submit error",
+		[DMATEST_ET_TIMEOUT]		= "test timed out",
+		[DMATEST_ET_DMA_ERROR]		=
+			"got completion callback (DMA_ERROR)",
+		[DMATEST_ET_DMA_IN_PROGRESS]	=
+			"got completion callback (DMA_IN_PROGRESS)",
+		[DMATEST_ET_VERIFY]		= "errors",
+		[DMATEST_ET_VERIFY_BUF]		= "verify errors",
+	};
+	static char buf[512];
+
+	snprintf(buf, sizeof(buf) - 1,
+		 "%s: #%u: %s with src_off=0x%x ""dst_off=0x%x len=0x%x (%lu)",
+		 name, tr->n, messages[tr->type], tr->src_off, tr->dst_off,
+		 tr->len, tr->data);
+
+	return buf;
+}
+
+static int thread_result_add(struct dmatest_info *info,
+		struct dmatest_result *r, enum dmatest_error_type type,
+		unsigned int n, unsigned int src_off, unsigned int dst_off,
+		unsigned int len, unsigned long data)
+{
+	struct dmatest_thread_result *tr;
+
+	tr = kzalloc(sizeof(*tr), GFP_KERNEL);
+	if (!tr)
+		return -ENOMEM;
+
+	tr->type = type;
+	tr->n = n;
+	tr->src_off = src_off;
+	tr->dst_off = dst_off;
+	tr->len = len;
+	tr->data = data;
+
+	mutex_lock(&info->results_lock);
+	list_add_tail(&tr->node, &r->results);
+	mutex_unlock(&info->results_lock);
+
+	pr_warn("%s\n", thread_result_get(r->name, tr));
+	return 0;
+}
+
+static unsigned int verify_result_add(struct dmatest_info *info,
+		struct dmatest_result *r, unsigned int n,
+		unsigned int src_off, unsigned int dst_off, unsigned int len,
+		u8 **bufs, int whence, unsigned int counter, u8 pattern,
+		bool is_srcbuf)
+{
+	struct dmatest_verify_result *vr;
+	unsigned int error_count;
+	unsigned int buf_off = is_srcbuf ? src_off : dst_off;
+	unsigned int start, end;
+
+	if (whence < 0) {
+		start = 0;
+		end = buf_off;
+	} else if (whence > 0) {
+		start = buf_off + len;
+		end = info->params.buf_size;
+	} else {
+		start = buf_off;
+		end = buf_off + len;
+	}
+
+	vr = kmalloc(sizeof(*vr), GFP_KERNEL);
+	if (!vr) {
+		pr_warn("dmatest: No memory to store verify result\n");
+		return dmatest_verify(NULL, bufs, start, end, counter, pattern,
+				      is_srcbuf);
+	}
+
+	vr->pattern = pattern;
+	vr->is_srcbuf = is_srcbuf;
+
+	error_count = dmatest_verify(vr, bufs, start, end, counter, pattern,
+				     is_srcbuf);
+	if (error_count) {
+		vr->error_count = error_count;
+		thread_result_add(info, r, DMATEST_ET_VERIFY_BUF, n, src_off,
+				  dst_off, len, (unsigned long)vr);
+		return error_count;
+	}
+
+	kfree(vr);
+	return 0;
+}
+
+static void result_free(struct dmatest_info *info, const char *name)
+{
+	struct dmatest_result *r, *_r;
+
+	mutex_lock(&info->results_lock);
+	list_for_each_entry_safe(r, _r, &info->results, node) {
+		struct dmatest_thread_result *tr, *_tr;
+
+		if (name && strcmp(r->name, name))
+			continue;
+
+		list_for_each_entry_safe(tr, _tr, &r->results, node) {
+			if (tr->type == DMATEST_ET_VERIFY_BUF)
+				kfree(tr->vr);
+			list_del(&tr->node);
+			kfree(tr);
+		}
+
+		kfree(r->name);
+		list_del(&r->node);
+		kfree(r);
+	}
+
+	mutex_unlock(&info->results_lock);
+}
+
+static struct dmatest_result *result_init(struct dmatest_info *info,
+		const char *name)
+{
+	struct dmatest_result *r;
+
+	r = kzalloc(sizeof(*r), GFP_KERNEL);
+	if (r) {
+		r->name = kstrdup(name, GFP_KERNEL);
+		INIT_LIST_HEAD(&r->results);
+		mutex_lock(&info->results_lock);
+		list_add_tail(&r->node, &info->results);
+		mutex_unlock(&info->results_lock);
+	}
+	return r;
+}
+
 /*
  * This function repeatedly tests DMA transfers of various lengths and
  * offsets for a given operation type until it is told to exit by
@@ -268,6 +516,8 @@ static int dmatest_func(void *data)
 	DECLARE_WAIT_QUEUE_HEAD_ONSTACK(done_wait);
 	struct dmatest_thread	*thread = data;
 	struct dmatest_done	done = { .wait = &done_wait };
+	struct dmatest_info	*info;
+	struct dmatest_params	*params;
 	struct dma_chan		*chan;
 	struct dma_device	*dev;
 	const char		*thread_name;
@@ -278,11 +528,12 @@ static int dmatest_func(void *data)
 	dma_cookie_t		cookie;
 	enum dma_status		status;
 	enum dma_ctrl_flags 	flags;
-	u8			pq_coefs[pq_sources + 1];
+	u8			*pq_coefs = NULL;
 	int			ret;
 	int			src_cnt;
 	int			dst_cnt;
 	int			i;
+	struct dmatest_result	*result;
 
 	thread_name = current->comm;
 	set_freezable();
@@ -290,28 +541,39 @@ static int dmatest_func(void *data)
 	ret = -ENOMEM;
 
 	smp_rmb();
+	info = thread->info;
+	params = &info->params;
 	chan = thread->chan;
 	dev = chan->device;
 	if (thread->type == DMA_MEMCPY)
 		src_cnt = dst_cnt = 1;
 	else if (thread->type == DMA_XOR) {
 		/* force odd to ensure dst = src */
-		src_cnt = min_odd(xor_sources | 1, dev->max_xor);
+		src_cnt = min_odd(params->xor_sources | 1, dev->max_xor);
 		dst_cnt = 1;
 	} else if (thread->type == DMA_PQ) {
 		/* force odd to ensure dst = src */
-		src_cnt = min_odd(pq_sources | 1, dma_maxpq(dev, 0));
+		src_cnt = min_odd(params->pq_sources | 1, dma_maxpq(dev, 0));
 		dst_cnt = 2;
+
+		pq_coefs = kmalloc(params->pq_sources+1, GFP_KERNEL);
+		if (!pq_coefs)
+			goto err_thread_type;
+
 		for (i = 0; i < src_cnt; i++)
 			pq_coefs[i] = 1;
 	} else
+		goto err_thread_type;
+
+	result = result_init(info, thread_name);
+	if (!result)
 		goto err_srcs;
 
 	thread->srcs = kcalloc(src_cnt+1, sizeof(u8 *), GFP_KERNEL);
 	if (!thread->srcs)
 		goto err_srcs;
 	for (i = 0; i < src_cnt; i++) {
-		thread->srcs[i] = kmalloc(test_buf_size, GFP_KERNEL);
+		thread->srcs[i] = kmalloc(params->buf_size, GFP_KERNEL);
 		if (!thread->srcs[i])
 			goto err_srcbuf;
 	}
@@ -321,7 +583,7 @@ static int dmatest_func(void *data)
 	if (!thread->dsts)
 		goto err_dsts;
 	for (i = 0; i < dst_cnt; i++) {
-		thread->dsts[i] = kmalloc(test_buf_size, GFP_KERNEL);
+		thread->dsts[i] = kmalloc(params->buf_size, GFP_KERNEL);
 		if (!thread->dsts[i])
 			goto err_dstbuf;
 	}
@@ -337,7 +599,7 @@ static int dmatest_func(void *data)
 	      | DMA_COMPL_SKIP_DEST_UNMAP | DMA_COMPL_SRC_UNMAP_SINGLE;
 
 	while (!kthread_should_stop()
-	       && !(iterations && total_tests >= iterations)) {
+	       && !(params->iterations && total_tests >= params->iterations)) {
 		struct dma_async_tx_descriptor *tx = NULL;
 		dma_addr_t dma_srcs[src_cnt];
 		dma_addr_t dma_dsts[dst_cnt];
@@ -353,24 +615,24 @@ static int dmatest_func(void *data)
 		else if (thread->type == DMA_PQ)
 			align = dev->pq_align;
 
-		if (1 << align > test_buf_size) {
+		if (1 << align > params->buf_size) {
 			pr_err("%u-byte buffer too small for %d-byte alignment\n",
-			       test_buf_size, 1 << align);
+			       params->buf_size, 1 << align);
 			break;
 		}
 
-		len = dmatest_random() % test_buf_size + 1;
+		len = dmatest_random() % params->buf_size + 1;
 		len = (len >> align) << align;
 		if (!len)
 			len = 1 << align;
-		src_off = dmatest_random() % (test_buf_size - len + 1);
-		dst_off = dmatest_random() % (test_buf_size - len + 1);
+		src_off = dmatest_random() % (params->buf_size - len + 1);
+		dst_off = dmatest_random() % (params->buf_size - len + 1);
 
 		src_off = (src_off >> align) << align;
 		dst_off = (dst_off >> align) << align;
 
-		dmatest_init_srcs(thread->srcs, src_off, len);
-		dmatest_init_dsts(thread->dsts, dst_off, len);
+		dmatest_init_srcs(thread->srcs, src_off, len, params->buf_size);
+		dmatest_init_dsts(thread->dsts, dst_off, len, params->buf_size);
 
 		for (i = 0; i < src_cnt; i++) {
 			u8 *buf = thread->srcs[i] + src_off;
@@ -380,10 +642,10 @@ static int dmatest_func(void *data)
 			ret = dma_mapping_error(dev->dev, dma_srcs[i]);
 			if (ret) {
 				unmap_src(dev->dev, dma_srcs, len, i);
-				pr_warn("%s: #%u: mapping error %d with "
-					"src_off=0x%x len=0x%x\n",
-					thread_name, total_tests - 1, ret,
-					src_off, len);
+				thread_result_add(info, result,
+						  DMATEST_ET_MAP_SRC,
+						  total_tests, src_off, dst_off,
+						  len, ret);
 				failed_tests++;
 				continue;
 			}
@@ -391,16 +653,17 @@ static int dmatest_func(void *data)
 		/* map with DMA_BIDIRECTIONAL to force writeback/invalidate */
 		for (i = 0; i < dst_cnt; i++) {
 			dma_dsts[i] = dma_map_single(dev->dev, thread->dsts[i],
-						     test_buf_size,
+						     params->buf_size,
 						     DMA_BIDIRECTIONAL);
 			ret = dma_mapping_error(dev->dev, dma_dsts[i]);
 			if (ret) {
 				unmap_src(dev->dev, dma_srcs, len, src_cnt);
-				unmap_dst(dev->dev, dma_dsts, test_buf_size, i);
-				pr_warn("%s: #%u: mapping error %d with "
-					"dst_off=0x%x len=0x%x\n",
-					thread_name, total_tests - 1, ret,
-					dst_off, test_buf_size);
+				unmap_dst(dev->dev, dma_dsts, params->buf_size,
+					  i);
+				thread_result_add(info, result,
+						  DMATEST_ET_MAP_DST,
+						  total_tests, src_off, dst_off,
+						  len, ret);
 				failed_tests++;
 				continue;
 			}
@@ -428,11 +691,11 @@ static int dmatest_func(void *data)
 
 		if (!tx) {
 			unmap_src(dev->dev, dma_srcs, len, src_cnt);
-			unmap_dst(dev->dev, dma_dsts, test_buf_size, dst_cnt);
-			pr_warning("%s: #%u: prep error with src_off=0x%x "
-					"dst_off=0x%x len=0x%x\n",
-					thread_name, total_tests - 1,
-					src_off, dst_off, len);
+			unmap_dst(dev->dev, dma_dsts, params->buf_size,
+				  dst_cnt);
+			thread_result_add(info, result, DMATEST_ET_PREP,
+					  total_tests, src_off, dst_off,
+					  len, 0);
 			msleep(100);
 			failed_tests++;
 			continue;
@@ -444,18 +707,18 @@ static int dmatest_func(void *data)
 		cookie = tx->tx_submit(tx);
 
 		if (dma_submit_error(cookie)) {
-			pr_warning("%s: #%u: submit error %d with src_off=0x%x "
-					"dst_off=0x%x len=0x%x\n",
-					thread_name, total_tests - 1, cookie,
-					src_off, dst_off, len);
+			thread_result_add(info, result, DMATEST_ET_SUBMIT,
+					  total_tests, src_off, dst_off,
+					  len, cookie);
 			msleep(100);
 			failed_tests++;
 			continue;
 		}
 		dma_async_issue_pending(chan);
 
-		wait_event_freezable_timeout(done_wait, done.done,
-					     msecs_to_jiffies(timeout));
+		wait_event_freezable_timeout(done_wait,
+					     done.done || kthread_should_stop(),
+					     msecs_to_jiffies(params->timeout));
 
 		status = dma_async_is_tx_complete(chan, cookie, NULL, NULL);
 
@@ -468,56 +731,57 @@ static int dmatest_func(void *data)
 			 * free it this time?" dancing.  For now, just
 			 * leave it dangling.
 			 */
-			pr_warning("%s: #%u: test timed out\n",
-				   thread_name, total_tests - 1);
+			thread_result_add(info, result, DMATEST_ET_TIMEOUT,
+					  total_tests, src_off, dst_off,
+					  len, 0);
 			failed_tests++;
 			continue;
 		} else if (status != DMA_SUCCESS) {
-			pr_warning("%s: #%u: got completion callback,"
-				   " but status is \'%s\'\n",
-				   thread_name, total_tests - 1,
-				   status == DMA_ERROR ? "error" : "in progress");
+			enum dmatest_error_type type = (status == DMA_ERROR) ?
+				DMATEST_ET_DMA_ERROR : DMATEST_ET_DMA_IN_PROGRESS;
+			thread_result_add(info, result, type,
+					  total_tests, src_off, dst_off,
+					  len, status);
 			failed_tests++;
 			continue;
 		}
 
 		/* Unmap by myself (see DMA_COMPL_SKIP_DEST_UNMAP above) */
-		unmap_dst(dev->dev, dma_dsts, test_buf_size, dst_cnt);
+		unmap_dst(dev->dev, dma_dsts, params->buf_size, dst_cnt);
 
 		error_count = 0;
 
 		pr_debug("%s: verifying source buffer...\n", thread_name);
-		error_count += dmatest_verify(thread->srcs, 0, src_off,
+		error_count += verify_result_add(info, result, total_tests,
+				src_off, dst_off, len, thread->srcs, -1,
 				0, PATTERN_SRC, true);
-		error_count += dmatest_verify(thread->srcs, src_off,
-				src_off + len, src_off,
-				PATTERN_SRC | PATTERN_COPY, true);
-		error_count += dmatest_verify(thread->srcs, src_off + len,
-				test_buf_size, src_off + len,
-				PATTERN_SRC, true);
-
-		pr_debug("%s: verifying dest buffer...\n",
-				thread->task->comm);
-		error_count += dmatest_verify(thread->dsts, 0, dst_off,
+		error_count += verify_result_add(info, result, total_tests,
+				src_off, dst_off, len, thread->srcs, 0,
+				src_off, PATTERN_SRC | PATTERN_COPY, true);
+		error_count += verify_result_add(info, result, total_tests,
+				src_off, dst_off, len, thread->srcs, 1,
+				src_off + len, PATTERN_SRC, true);
+
+		pr_debug("%s: verifying dest buffer...\n", thread_name);
+		error_count += verify_result_add(info, result, total_tests,
+				src_off, dst_off, len, thread->dsts, -1,
 				0, PATTERN_DST, false);
-		error_count += dmatest_verify(thread->dsts, dst_off,
-				dst_off + len, src_off,
-				PATTERN_SRC | PATTERN_COPY, false);
-		error_count += dmatest_verify(thread->dsts, dst_off + len,
-				test_buf_size, dst_off + len,
-				PATTERN_DST, false);
+		error_count += verify_result_add(info, result, total_tests,
+				src_off, dst_off, len, thread->dsts, 0,
+				src_off, PATTERN_SRC | PATTERN_COPY, false);
+		error_count += verify_result_add(info, result, total_tests,
+				src_off, dst_off, len, thread->dsts, 1,
+				dst_off + len, PATTERN_DST, false);
 
 		if (error_count) {
-			pr_warning("%s: #%u: %u errors with "
-				"src_off=0x%x dst_off=0x%x len=0x%x\n",
-				thread_name, total_tests - 1, error_count,
-				src_off, dst_off, len);
+			thread_result_add(info, result, DMATEST_ET_VERIFY,
+					  total_tests, src_off, dst_off,
+					  len, error_count);
 			failed_tests++;
 		} else {
-			pr_debug("%s: #%u: No errors with "
-				"src_off=0x%x dst_off=0x%x len=0x%x\n",
-				thread_name, total_tests - 1,
-				src_off, dst_off, len);
+			thread_result_add(info, result, DMATEST_ET_OK,
+					  total_tests, src_off, dst_off,
+					  len, 0);
 		}
 	}
 
@@ -532,6 +796,8 @@ err_dsts:
 err_srcbuf:
 	kfree(thread->srcs);
 err_srcs:
+	kfree(pq_coefs);
+err_thread_type:
 	pr_notice("%s: terminating after %u tests, %u failures (status %d)\n",
 			thread_name, total_tests, failed_tests, ret);
 
@@ -539,7 +805,9 @@ err_srcs:
 	if (ret)
 		dmaengine_terminate_all(chan);
 
-	if (iterations > 0)
+	thread->done = true;
+
+	if (params->iterations > 0)
 		while (!kthread_should_stop()) {
 			DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wait_dmatest_exit);
 			interruptible_sleep_on(&wait_dmatest_exit);
@@ -568,8 +836,10 @@ static void dmatest_cleanup_channel(struct dmatest_chan *dtc)
 	kfree(dtc);
 }
 
-static int dmatest_add_threads(struct dmatest_chan *dtc, enum dma_transaction_type type)
+static int dmatest_add_threads(struct dmatest_info *info,
+		struct dmatest_chan *dtc, enum dma_transaction_type type)
 {
+	struct dmatest_params *params = &info->params;
 	struct dmatest_thread *thread;
 	struct dma_chan *chan = dtc->chan;
 	char *op;
@@ -584,7 +854,7 @@ static int dmatest_add_threads(struct dmatest_chan *dtc, enum dma_transaction_ty
 	else
 		return -EINVAL;
 
-	for (i = 0; i < threads_per_chan; i++) {
+	for (i = 0; i < params->threads_per_chan; i++) {
 		thread = kzalloc(sizeof(struct dmatest_thread), GFP_KERNEL);
 		if (!thread) {
 			pr_warning("dmatest: No memory for %s-%s%u\n",
@@ -592,6 +862,7 @@ static int dmatest_add_threads(struct dmatest_chan *dtc, enum dma_transaction_ty
 
 			break;
 		}
+		thread->info = info;
 		thread->chan = dtc->chan;
 		thread->type = type;
 		smp_wmb();
@@ -612,7 +883,8 @@ static int dmatest_add_threads(struct dmatest_chan *dtc, enum dma_transaction_ty
 	return i;
 }
 
-static int dmatest_add_channel(struct dma_chan *chan)
+static int dmatest_add_channel(struct dmatest_info *info,
+		struct dma_chan *chan)
 {
 	struct dmatest_chan	*dtc;
 	struct dma_device	*dma_dev = chan->device;
@@ -629,75 +901,418 @@ static int dmatest_add_channel(struct dma_chan *chan)
 	INIT_LIST_HEAD(&dtc->threads);
 
 	if (dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask)) {
-		cnt = dmatest_add_threads(dtc, DMA_MEMCPY);
+		cnt = dmatest_add_threads(info, dtc, DMA_MEMCPY);
 		thread_count += cnt > 0 ? cnt : 0;
 	}
 	if (dma_has_cap(DMA_XOR, dma_dev->cap_mask)) {
-		cnt = dmatest_add_threads(dtc, DMA_XOR);
+		cnt = dmatest_add_threads(info, dtc, DMA_XOR);
 		thread_count += cnt > 0 ? cnt : 0;
 	}
 	if (dma_has_cap(DMA_PQ, dma_dev->cap_mask)) {
-		cnt = dmatest_add_threads(dtc, DMA_PQ);
+		cnt = dmatest_add_threads(info, dtc, DMA_PQ);
 		thread_count += cnt > 0 ? cnt : 0;
 	}
 
 	pr_info("dmatest: Started %u threads using %s\n",
 		thread_count, dma_chan_name(chan));
 
-	list_add_tail(&dtc->node, &dmatest_channels);
-	nr_channels++;
+	list_add_tail(&dtc->node, &info->channels);
+	info->nr_channels++;
 
 	return 0;
 }
 
 static bool filter(struct dma_chan *chan, void *param)
 {
-	if (!dmatest_match_channel(chan) || !dmatest_match_device(chan->device))
+	struct dmatest_params *params = param;
+
+	if (!dmatest_match_channel(params, chan) ||
+	    !dmatest_match_device(params, chan->device))
 		return false;
 	else
 		return true;
 }
 
-static int __init dmatest_init(void)
+static int __run_threaded_test(struct dmatest_info *info)
 {
 	dma_cap_mask_t mask;
 	struct dma_chan *chan;
+	struct dmatest_params *params = &info->params;
 	int err = 0;
 
 	dma_cap_zero(mask);
 	dma_cap_set(DMA_MEMCPY, mask);
 	for (;;) {
-		chan = dma_request_channel(mask, filter, NULL);
+		chan = dma_request_channel(mask, filter, params);
 		if (chan) {
-			err = dmatest_add_channel(chan);
+			err = dmatest_add_channel(info, chan);
 			if (err) {
 				dma_release_channel(chan);
 				break; /* add_channel failed, punt */
 			}
 		} else
 			break; /* no more channels available */
-		if (max_channels && nr_channels >= max_channels)
+		if (params->max_channels &&
+		    info->nr_channels >= params->max_channels)
 			break; /* we have all we need */
 	}
-
 	return err;
 }
-/* when compiled-in wait for drivers to load first */
-late_initcall(dmatest_init);
 
-static void __exit dmatest_exit(void)
+#ifndef MODULE
+static int run_threaded_test(struct dmatest_info *info)
+{
+	int ret;
+
+	mutex_lock(&info->lock);
+	ret = __run_threaded_test(info);
+	mutex_unlock(&info->lock);
+	return ret;
+}
+#endif
+
+static void __stop_threaded_test(struct dmatest_info *info)
 {
 	struct dmatest_chan *dtc, *_dtc;
 	struct dma_chan *chan;
 
-	list_for_each_entry_safe(dtc, _dtc, &dmatest_channels, node) {
+	list_for_each_entry_safe(dtc, _dtc, &info->channels, node) {
 		list_del(&dtc->node);
 		chan = dtc->chan;
 		dmatest_cleanup_channel(dtc);
-		pr_debug("dmatest: dropped channel %s\n",
-			 dma_chan_name(chan));
+		pr_debug("dmatest: dropped channel %s\n", dma_chan_name(chan));
 		dma_release_channel(chan);
 	}
+
+	info->nr_channels = 0;
+}
+
+static void stop_threaded_test(struct dmatest_info *info)
+{
+	mutex_lock(&info->lock);
+	__stop_threaded_test(info);
+	mutex_unlock(&info->lock);
+}
+
+static int __restart_threaded_test(struct dmatest_info *info, bool run)
+{
+	struct dmatest_params *params = &info->params;
+	int ret;
+
+	/* Stop any running test first */
+	__stop_threaded_test(info);
+
+	if (run == false)
+		return 0;
+
+	/* Clear results from previous run */
+	result_free(info, NULL);
+
+	/* Copy test parameters */
+	memcpy(params, &info->dbgfs_params, sizeof(*params));
+
+	/* Run test with new parameters */
+	ret = __run_threaded_test(info);
+	if (ret) {
+		__stop_threaded_test(info);
+		pr_err("dmatest: Can't run test\n");
+	}
+
+	return ret;
+}
+
+static ssize_t dtf_write_string(void *to, size_t available, loff_t *ppos,
+		const void __user *from, size_t count)
+{
+	char tmp[20];
+	ssize_t len;
+
+	len = simple_write_to_buffer(tmp, sizeof(tmp) - 1, ppos, from, count);
+	if (len >= 0) {
+		tmp[len] = '\0';
+		strlcpy(to, strim(tmp), available);
+	}
+
+	return len;
+}
+
+static ssize_t dtf_read_channel(struct file *file, char __user *buf,
+		size_t count, loff_t *ppos)
+{
+	struct dmatest_info *info = file->private_data;
+	return simple_read_from_buffer(buf, count, ppos,
+			info->dbgfs_params.channel,
+			strlen(info->dbgfs_params.channel));
+}
+
+static ssize_t dtf_write_channel(struct file *file, const char __user *buf,
+		size_t size, loff_t *ppos)
+{
+	struct dmatest_info *info = file->private_data;
+	return dtf_write_string(info->dbgfs_params.channel,
+				sizeof(info->dbgfs_params.channel),
+				ppos, buf, size);
+}
+
+static const struct file_operations dtf_channel_fops = {
+	.read	= dtf_read_channel,
+	.write	= dtf_write_channel,
+	.open	= simple_open,
+	.llseek	= default_llseek,
+};
+
+static ssize_t dtf_read_device(struct file *file, char __user *buf,
+		size_t count, loff_t *ppos)
+{
+	struct dmatest_info *info = file->private_data;
+	return simple_read_from_buffer(buf, count, ppos,
+			info->dbgfs_params.device,
+			strlen(info->dbgfs_params.device));
+}
+
+static ssize_t dtf_write_device(struct file *file, const char __user *buf,
+		size_t size, loff_t *ppos)
+{
+	struct dmatest_info *info = file->private_data;
+	return dtf_write_string(info->dbgfs_params.device,
+				sizeof(info->dbgfs_params.device),
+				ppos, buf, size);
+}
+
+static const struct file_operations dtf_device_fops = {
+	.read	= dtf_read_device,
+	.write	= dtf_write_device,
+	.open	= simple_open,
+	.llseek	= default_llseek,
+};
+
+static ssize_t dtf_read_run(struct file *file, char __user *user_buf,
+		size_t count, loff_t *ppos)
+{
+	struct dmatest_info *info = file->private_data;
+	char buf[3];
+	struct dmatest_chan *dtc;
+	bool alive = false;
+
+	mutex_lock(&info->lock);
+	list_for_each_entry(dtc, &info->channels, node) {
+		struct dmatest_thread *thread;
+
+		list_for_each_entry(thread, &dtc->threads, node) {
+			if (!thread->done) {
+				alive = true;
+				break;
+			}
+		}
+	}
+
+	if (alive) {
+		buf[0] = 'Y';
+	} else {
+		__stop_threaded_test(info);
+		buf[0] = 'N';
+	}
+
+	mutex_unlock(&info->lock);
+	buf[1] = '\n';
+	buf[2] = 0x00;
+	return simple_read_from_buffer(user_buf, count, ppos, buf, 2);
+}
+
+static ssize_t dtf_write_run(struct file *file, const char __user *user_buf,
+		size_t count, loff_t *ppos)
+{
+	struct dmatest_info *info = file->private_data;
+	char buf[16];
+	bool bv;
+	int ret = 0;
+
+	if (copy_from_user(buf, user_buf, min(count, (sizeof(buf) - 1))))
+		return -EFAULT;
+
+	if (strtobool(buf, &bv) == 0) {
+		mutex_lock(&info->lock);
+		ret = __restart_threaded_test(info, bv);
+		mutex_unlock(&info->lock);
+	}
+
+	return ret ? ret : count;
+}
+
+static const struct file_operations dtf_run_fops = {
+	.read	= dtf_read_run,
+	.write	= dtf_write_run,
+	.open	= simple_open,
+	.llseek	= default_llseek,
+};
+
+static int dtf_results_show(struct seq_file *sf, void *data)
+{
+	struct dmatest_info *info = sf->private;
+	struct dmatest_result *result;
+	struct dmatest_thread_result *tr;
+	unsigned int i;
+
+	mutex_lock(&info->results_lock);
+	list_for_each_entry(result, &info->results, node) {
+		list_for_each_entry(tr, &result->results, node) {
+			seq_printf(sf, "%s\n",
+				thread_result_get(result->name, tr));
+			if (tr->type == DMATEST_ET_VERIFY_BUF) {
+				for (i = 0; i < tr->vr->error_count; i++) {
+					seq_printf(sf, "\t%s\n",
+						verify_result_get_one(tr->vr, i));
+				}
+			}
+		}
+	}
+
+	mutex_unlock(&info->results_lock);
+	return 0;
+}
+
+static int dtf_results_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dtf_results_show, inode->i_private);
+}
+
+static const struct file_operations dtf_results_fops = {
+	.open		= dtf_results_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int dmatest_register_dbgfs(struct dmatest_info *info)
+{
+	struct dentry *d;
+	struct dmatest_params *params = &info->dbgfs_params;
+	int ret = -ENOMEM;
+
+	d = debugfs_create_dir("dmatest", NULL);
+	if (IS_ERR(d))
+		return PTR_ERR(d);
+	if (!d)
+		goto err_root;
+
+	info->root = d;
+
+	/* Copy initial values */
+	memcpy(params, &info->params, sizeof(*params));
+
+	/* Test parameters */
+
+	d = debugfs_create_u32("test_buf_size", S_IWUSR | S_IRUGO, info->root,
+			       (u32 *)&params->buf_size);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_file("channel", S_IRUGO | S_IWUSR, info->root,
+				info, &dtf_channel_fops);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_file("device", S_IRUGO | S_IWUSR, info->root,
+				info, &dtf_device_fops);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_u32("threads_per_chan", S_IWUSR | S_IRUGO, info->root,
+			       (u32 *)&params->threads_per_chan);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_u32("max_channels", S_IWUSR | S_IRUGO, info->root,
+			       (u32 *)&params->max_channels);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_u32("iterations", S_IWUSR | S_IRUGO, info->root,
+			       (u32 *)&params->iterations);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_u32("xor_sources", S_IWUSR | S_IRUGO, info->root,
+			       (u32 *)&params->xor_sources);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_u32("pq_sources", S_IWUSR | S_IRUGO, info->root,
+			       (u32 *)&params->pq_sources);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	d = debugfs_create_u32("timeout", S_IWUSR | S_IRUGO, info->root,
+			       (u32 *)&params->timeout);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	/* Run or stop threaded test */
+	d = debugfs_create_file("run", S_IWUSR | S_IRUGO, info->root,
+				info, &dtf_run_fops);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	/* Results of test in progress */
+	d = debugfs_create_file("results", S_IRUGO, info->root, info,
+				&dtf_results_fops);
+	if (IS_ERR_OR_NULL(d))
+		goto err_node;
+
+	return 0;
+
+err_node:
+	debugfs_remove_recursive(info->root);
+err_root:
+	pr_err("dmatest: Failed to initialize debugfs\n");
+	return ret;
+}
+
+static int __init dmatest_init(void)
+{
+	struct dmatest_info *info = &test_info;
+	struct dmatest_params *params = &info->params;
+	int ret;
+
+	memset(info, 0, sizeof(*info));
+
+	mutex_init(&info->lock);
+	INIT_LIST_HEAD(&info->channels);
+
+	mutex_init(&info->results_lock);
+	INIT_LIST_HEAD(&info->results);
+
+	/* Set default parameters */
+	params->buf_size = test_buf_size;
+	strlcpy(params->channel, test_channel, sizeof(params->channel));
+	strlcpy(params->device, test_device, sizeof(params->device));
+	params->threads_per_chan = threads_per_chan;
+	params->max_channels = max_channels;
+	params->iterations = iterations;
+	params->xor_sources = xor_sources;
+	params->pq_sources = pq_sources;
+	params->timeout = timeout;
+
+	ret = dmatest_register_dbgfs(info);
+	if (ret)
+		return ret;
+
+#ifdef MODULE
+	return 0;
+#else
+	return run_threaded_test(info);
+#endif
+}
+/* when compiled-in wait for drivers to load first */
+late_initcall(dmatest_init);
+
+static void __exit dmatest_exit(void)
+{
+	struct dmatest_info *info = &test_info;
+
+	debugfs_remove_recursive(info->root);
+	stop_threaded_test(info);
+	result_free(info, NULL);
 }
 module_exit(dmatest_exit);
 
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
index 43a5329d448..2e5deaa82b6 100644
--- a/drivers/dma/dw_dmac.c
+++ b/drivers/dma/dw_dmac.c
@@ -25,6 +25,8 @@
 #include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
+#include <linux/acpi.h>
+#include <linux/acpi_dma.h>
 
 #include "dw_dmac_regs.h"
 #include "dmaengine.h"
@@ -49,29 +51,22 @@ static inline unsigned int dwc_get_sms(struct dw_dma_slave *slave)
 	return slave ? slave->src_master : 1;
 }
 
-#define SRC_MASTER	0
-#define DST_MASTER	1
-
-static inline unsigned int dwc_get_master(struct dma_chan *chan, int master)
+static inline void dwc_set_masters(struct dw_dma_chan *dwc)
 {
-	struct dw_dma *dw = to_dw_dma(chan->device);
-	struct dw_dma_slave *dws = chan->private;
-	unsigned int m;
-
-	if (master == SRC_MASTER)
-		m = dwc_get_sms(dws);
-	else
-		m = dwc_get_dms(dws);
+	struct dw_dma *dw = to_dw_dma(dwc->chan.device);
+	struct dw_dma_slave *dws = dwc->chan.private;
+	unsigned char mmax = dw->nr_masters - 1;
 
-	return min_t(unsigned int, dw->nr_masters - 1, m);
+	if (dwc->request_line == ~0) {
+		dwc->src_master = min_t(unsigned char, mmax, dwc_get_sms(dws));
+		dwc->dst_master = min_t(unsigned char, mmax, dwc_get_dms(dws));
+	}
 }
 
 #define DWC_DEFAULT_CTLLO(_chan) ({				\
 		struct dw_dma_chan *_dwc = to_dw_dma_chan(_chan);	\
 		struct dma_slave_config	*_sconfig = &_dwc->dma_sconfig;	\
 		bool _is_slave = is_slave_direction(_dwc->direction);	\
-		int _dms = dwc_get_master(_chan, DST_MASTER);		\
-		int _sms = dwc_get_master(_chan, SRC_MASTER);		\
 		u8 _smsize = _is_slave ? _sconfig->src_maxburst :	\
 			DW_DMA_MSIZE_16;			\
 		u8 _dmsize = _is_slave ? _sconfig->dst_maxburst :	\
@@ -81,8 +76,8 @@ static inline unsigned int dwc_get_master(struct dma_chan *chan, int master)
 		 | DWC_CTLL_SRC_MSIZE(_smsize)			\
 		 | DWC_CTLL_LLP_D_EN				\
 		 | DWC_CTLL_LLP_S_EN				\
-		 | DWC_CTLL_DMS(_dms)				\
-		 | DWC_CTLL_SMS(_sms));				\
+		 | DWC_CTLL_DMS(_dwc->dst_master)		\
+		 | DWC_CTLL_SMS(_dwc->src_master));		\
 	})
 
 /*
@@ -92,13 +87,6 @@ static inline unsigned int dwc_get_master(struct dma_chan *chan, int master)
  */
 #define NR_DESCS_PER_CHANNEL	64
 
-static inline unsigned int dwc_get_data_width(struct dma_chan *chan, int master)
-{
-	struct dw_dma *dw = to_dw_dma(chan->device);
-
-	return dw->data_width[dwc_get_master(chan, master)];
-}
-
 /*----------------------------------------------------------------------*/
 
 static struct device *chan2dev(struct dma_chan *chan)
@@ -172,13 +160,7 @@ static void dwc_initialize(struct dw_dma_chan *dwc)
 	if (dwc->initialized == true)
 		return;
 
-	if (dws && dws->cfg_hi == ~0 && dws->cfg_lo == ~0) {
-		/* autoconfigure based on request line from DT */
-		if (dwc->direction == DMA_MEM_TO_DEV)
-			cfghi = DWC_CFGH_DST_PER(dwc->request_line);
-		else if (dwc->direction == DMA_DEV_TO_MEM)
-			cfghi = DWC_CFGH_SRC_PER(dwc->request_line);
-	} else if (dws) {
+	if (dws) {
 		/*
 		 * We need controller-specific data to set up slave
 		 * transfers.
@@ -189,9 +171,9 @@ static void dwc_initialize(struct dw_dma_chan *dwc)
 		cfglo |= dws->cfg_lo & ~DWC_CFGL_CH_PRIOR_MASK;
 	} else {
 		if (dwc->direction == DMA_MEM_TO_DEV)
-			cfghi = DWC_CFGH_DST_PER(dwc->dma_sconfig.slave_id);
+			cfghi = DWC_CFGH_DST_PER(dwc->request_line);
 		else if (dwc->direction == DMA_DEV_TO_MEM)
-			cfghi = DWC_CFGH_SRC_PER(dwc->dma_sconfig.slave_id);
+			cfghi = DWC_CFGH_SRC_PER(dwc->request_line);
 	}
 
 	channel_writel(dwc, CFG_LO, cfglo);
@@ -473,16 +455,16 @@ static void dwc_scan_descriptors(struct dw_dma *dw, struct dw_dma_chan *dwc)
 			(unsigned long long)llp);
 
 	list_for_each_entry_safe(desc, _desc, &dwc->active_list, desc_node) {
-		/* initial residue value */
+		/* Initial residue value */
 		dwc->residue = desc->total_len;
 
-		/* check first descriptors addr */
+		/* Check first descriptors addr */
 		if (desc->txd.phys == llp) {
 			spin_unlock_irqrestore(&dwc->lock, flags);
 			return;
 		}
 
-		/* check first descriptors llp */
+		/* Check first descriptors llp */
 		if (desc->lli.llp == llp) {
 			/* This one is currently in progress */
 			dwc->residue -= dwc_get_sent(dwc);
@@ -588,7 +570,7 @@ inline dma_addr_t dw_dma_get_dst_addr(struct dma_chan *chan)
 }
 EXPORT_SYMBOL(dw_dma_get_dst_addr);
 
-/* called with dwc->lock held and all DMAC interrupts disabled */
+/* Called with dwc->lock held and all DMAC interrupts disabled */
 static void dwc_handle_cyclic(struct dw_dma *dw, struct dw_dma_chan *dwc,
 		u32 status_err, u32 status_xfer)
 {
@@ -626,7 +608,7 @@ static void dwc_handle_cyclic(struct dw_dma *dw, struct dw_dma_chan *dwc,
 
 		dwc_chan_disable(dw, dwc);
 
-		/* make sure DMA does not restart by loading a new list */
+		/* Make sure DMA does not restart by loading a new list */
 		channel_writel(dwc, LLP, 0);
 		channel_writel(dwc, CTL_LO, 0);
 		channel_writel(dwc, CTL_HI, 0);
@@ -745,6 +727,7 @@ dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 		size_t len, unsigned long flags)
 {
 	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
 	struct dw_desc		*desc;
 	struct dw_desc		*first;
 	struct dw_desc		*prev;
@@ -767,8 +750,8 @@ dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 
 	dwc->direction = DMA_MEM_TO_MEM;
 
-	data_width = min_t(unsigned int, dwc_get_data_width(chan, SRC_MASTER),
-			   dwc_get_data_width(chan, DST_MASTER));
+	data_width = min_t(unsigned int, dw->data_width[dwc->src_master],
+			   dw->data_width[dwc->dst_master]);
 
 	src_width = dst_width = min_t(unsigned int, data_width,
 				      dwc_fast_fls(src | dest | len));
@@ -826,6 +809,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		unsigned long flags, void *context)
 {
 	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
 	struct dma_slave_config	*sconfig = &dwc->dma_sconfig;
 	struct dw_desc		*prev;
 	struct dw_desc		*first;
@@ -859,7 +843,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		ctllo |= sconfig->device_fc ? DWC_CTLL_FC(DW_DMA_FC_P_M2P) :
 			DWC_CTLL_FC(DW_DMA_FC_D_M2P);
 
-		data_width = dwc_get_data_width(chan, SRC_MASTER);
+		data_width = dw->data_width[dwc->src_master];
 
 		for_each_sg(sgl, sg, sg_len, i) {
 			struct dw_desc	*desc;
@@ -919,7 +903,7 @@ slave_sg_todev_fill_desc:
 		ctllo |= sconfig->device_fc ? DWC_CTLL_FC(DW_DMA_FC_P_P2M) :
 			DWC_CTLL_FC(DW_DMA_FC_D_P2M);
 
-		data_width = dwc_get_data_width(chan, DST_MASTER);
+		data_width = dw->data_width[dwc->dst_master];
 
 		for_each_sg(sgl, sg, sg_len, i) {
 			struct dw_desc	*desc;
@@ -1001,13 +985,6 @@ static inline void convert_burst(u32 *maxburst)
 		*maxburst = 0;
 }
 
-static inline void convert_slave_id(struct dw_dma_chan *dwc)
-{
-	struct dw_dma *dw = to_dw_dma(dwc->chan.device);
-
-	dwc->dma_sconfig.slave_id -= dw->request_line_base;
-}
-
 static int
 set_runtime_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
 {
@@ -1020,9 +997,12 @@ set_runtime_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
 	memcpy(&dwc->dma_sconfig, sconfig, sizeof(*sconfig));
 	dwc->direction = sconfig->direction;
 
+	/* Take the request line from slave_id member */
+	if (dwc->request_line == ~0)
+		dwc->request_line = sconfig->slave_id;
+
 	convert_burst(&dwc->dma_sconfig.src_maxburst);
 	convert_burst(&dwc->dma_sconfig.dst_maxburst);
-	convert_slave_id(dwc);
 
 	return 0;
 }
@@ -1030,10 +1010,11 @@ set_runtime_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
 static inline void dwc_chan_pause(struct dw_dma_chan *dwc)
 {
 	u32 cfglo = channel_readl(dwc, CFG_LO);
+	unsigned int count = 20;	/* timeout iterations */
 
 	channel_writel(dwc, CFG_LO, cfglo | DWC_CFGL_CH_SUSP);
-	while (!(channel_readl(dwc, CFG_LO) & DWC_CFGL_FIFO_EMPTY))
-		cpu_relax();
+	while (!(channel_readl(dwc, CFG_LO) & DWC_CFGL_FIFO_EMPTY) && count--)
+		udelay(2);
 
 	dwc->paused = true;
 }
@@ -1169,6 +1150,8 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan)
 	 * doesn't mean what you think it means), and status writeback.
 	 */
 
+	dwc_set_masters(dwc);
+
 	spin_lock_irqsave(&dwc->lock, flags);
 	i = dwc->descs_allocated;
 	while (dwc->descs_allocated < NR_DESCS_PER_CHANNEL) {
@@ -1226,6 +1209,7 @@ static void dwc_free_chan_resources(struct dma_chan *chan)
 	list_splice_init(&dwc->free_list, &list);
 	dwc->descs_allocated = 0;
 	dwc->initialized = false;
+	dwc->request_line = ~0;
 
 	/* Disable interrupts */
 	channel_clear_bit(dw, MASK.XFER, dwc->mask);
@@ -1241,42 +1225,36 @@ static void dwc_free_chan_resources(struct dma_chan *chan)
 	dev_vdbg(chan2dev(chan), "%s: done\n", __func__);
 }
 
-struct dw_dma_filter_args {
+/*----------------------------------------------------------------------*/
+
+struct dw_dma_of_filter_args {
 	struct dw_dma *dw;
 	unsigned int req;
 	unsigned int src;
 	unsigned int dst;
 };
 
-static bool dw_dma_generic_filter(struct dma_chan *chan, void *param)
+static bool dw_dma_of_filter(struct dma_chan *chan, void *param)
 {
 	struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
-	struct dw_dma *dw = to_dw_dma(chan->device);
-	struct dw_dma_filter_args *fargs = param;
-	struct dw_dma_slave *dws = &dwc->slave;
+	struct dw_dma_of_filter_args *fargs = param;
 
-	/* ensure the device matches our channel */
+	/* Ensure the device matches our channel */
         if (chan->device != &fargs->dw->dma)
                 return false;
 
-	dws->dma_dev	= dw->dma.dev;
-	dws->cfg_hi	= ~0;
-	dws->cfg_lo	= ~0;
-	dws->src_master	= fargs->src;
-	dws->dst_master	= fargs->dst;
-
 	dwc->request_line = fargs->req;
-
-	chan->private = dws;
+	dwc->src_master	= fargs->src;
+	dwc->dst_master	= fargs->dst;
 
 	return true;
 }
 
-static struct dma_chan *dw_dma_xlate(struct of_phandle_args *dma_spec,
-					 struct of_dma *ofdma)
+static struct dma_chan *dw_dma_of_xlate(struct of_phandle_args *dma_spec,
+					struct of_dma *ofdma)
 {
 	struct dw_dma *dw = ofdma->of_dma_data;
-	struct dw_dma_filter_args fargs = {
+	struct dw_dma_of_filter_args fargs = {
 		.dw = dw,
 	};
 	dma_cap_mask_t cap;
@@ -1297,8 +1275,48 @@ static struct dma_chan *dw_dma_xlate(struct of_phandle_args *dma_spec,
 	dma_cap_set(DMA_SLAVE, cap);
 
 	/* TODO: there should be a simpler way to do this */
-	return dma_request_channel(cap, dw_dma_generic_filter, &fargs);
+	return dma_request_channel(cap, dw_dma_of_filter, &fargs);
+}
+
+#ifdef CONFIG_ACPI
+static bool dw_dma_acpi_filter(struct dma_chan *chan, void *param)
+{
+	struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
+	struct acpi_dma_spec *dma_spec = param;
+
+	if (chan->device->dev != dma_spec->dev ||
+	    chan->chan_id != dma_spec->chan_id)
+		return false;
+
+	dwc->request_line = dma_spec->slave_id;
+	dwc->src_master = dwc_get_sms(NULL);
+	dwc->dst_master = dwc_get_dms(NULL);
+
+	return true;
+}
+
+static void dw_dma_acpi_controller_register(struct dw_dma *dw)
+{
+	struct device *dev = dw->dma.dev;
+	struct acpi_dma_filter_info *info;
+	int ret;
+
+	info = devm_kzalloc(dev, sizeof(*info), GFP_KERNEL);
+	if (!info)
+		return;
+
+	dma_cap_zero(info->dma_cap);
+	dma_cap_set(DMA_SLAVE, info->dma_cap);
+	info->filter_fn = dw_dma_acpi_filter;
+
+	ret = devm_acpi_dma_controller_register(dev, acpi_dma_simple_xlate,
+						info);
+	if (ret)
+		dev_err(dev, "could not register acpi_dma_controller\n");
 }
+#else /* !CONFIG_ACPI */
+static inline void dw_dma_acpi_controller_register(struct dw_dma *dw) {}
+#endif /* !CONFIG_ACPI */
 
 /* --------------------- Cyclic DMA API extensions -------------------- */
 
@@ -1322,7 +1340,7 @@ int dw_dma_cyclic_start(struct dma_chan *chan)
 
 	spin_lock_irqsave(&dwc->lock, flags);
 
-	/* assert channel is idle */
+	/* Assert channel is idle */
 	if (dma_readl(dw, CH_EN) & dwc->mask) {
 		dev_err(chan2dev(&dwc->chan),
 			"BUG: Attempted to start non-idle channel\n");
@@ -1334,7 +1352,7 @@ int dw_dma_cyclic_start(struct dma_chan *chan)
 	dma_writel(dw, CLEAR.ERROR, dwc->mask);
 	dma_writel(dw, CLEAR.XFER, dwc->mask);
 
-	/* setup DMAC channel registers */
+	/* Setup DMAC channel registers */
 	channel_writel(dwc, LLP, dwc->cdesc->desc[0]->txd.phys);
 	channel_writel(dwc, CTL_LO, DWC_CTLL_LLP_D_EN | DWC_CTLL_LLP_S_EN);
 	channel_writel(dwc, CTL_HI, 0);
@@ -1501,7 +1519,7 @@ struct dw_cyclic_desc *dw_dma_cyclic_prep(struct dma_chan *chan,
 		last = desc;
 	}
 
-	/* lets make a cyclic list */
+	/* 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 "
@@ -1636,7 +1654,6 @@ dw_dma_parse_dt(struct platform_device *pdev)
 
 static int dw_probe(struct platform_device *pdev)
 {
-	const struct platform_device_id *match;
 	struct dw_dma_platform_data *pdata;
 	struct resource		*io;
 	struct dw_dma		*dw;
@@ -1706,7 +1723,7 @@ static int dw_probe(struct platform_device *pdev)
 
 	dw->regs = regs;
 
-	/* get hardware configuration parameters */
+	/* Get hardware configuration parameters */
 	if (autocfg) {
 		max_blk_size = dma_readl(dw, MAX_BLK_SIZE);
 
@@ -1720,18 +1737,13 @@ static int dw_probe(struct platform_device *pdev)
 		memcpy(dw->data_width, pdata->data_width, 4);
 	}
 
-	/* Get the base request line if set */
-	match = platform_get_device_id(pdev);
-	if (match)
-		dw->request_line_base = (unsigned int)match->driver_data;
-
 	/* Calculate all channel mask before DMA setup */
 	dw->all_chan_mask = (1 << nr_channels) - 1;
 
-	/* force dma off, just in case */
+	/* Force dma off, just in case */
 	dw_dma_off(dw);
 
-	/* disable BLOCK interrupts as well */
+	/* Disable BLOCK interrupts as well */
 	channel_clear_bit(dw, MASK.BLOCK, dw->all_chan_mask);
 
 	err = devm_request_irq(&pdev->dev, irq, dw_dma_interrupt, 0,
@@ -1741,7 +1753,7 @@ static int dw_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, dw);
 
-	/* create a pool of consistent memory blocks for hardware descriptors */
+	/* Create a pool of consistent memory blocks for hardware descriptors */
 	dw->desc_pool = dmam_pool_create("dw_dmac_desc_pool", &pdev->dev,
 					 sizeof(struct dw_desc), 4, 0);
 	if (!dw->desc_pool) {
@@ -1781,8 +1793,9 @@ static int dw_probe(struct platform_device *pdev)
 		channel_clear_bit(dw, CH_EN, dwc->mask);
 
 		dwc->direction = DMA_TRANS_NONE;
+		dwc->request_line = ~0;
 
-		/* hardware configuration */
+		/* Hardware configuration */
 		if (autocfg) {
 			unsigned int dwc_params;
 
@@ -1842,12 +1855,15 @@ static int dw_probe(struct platform_device *pdev)
 
 	if (pdev->dev.of_node) {
 		err = of_dma_controller_register(pdev->dev.of_node,
-						 dw_dma_xlate, dw);
-		if (err && err != -ENODEV)
+						 dw_dma_of_xlate, dw);
+		if (err)
 			dev_err(&pdev->dev,
 				"could not register of_dma_controller\n");
 	}
 
+	if (ACPI_HANDLE(&pdev->dev))
+		dw_dma_acpi_controller_register(dw);
+
 	return 0;
 }
 
@@ -1912,18 +1928,19 @@ static const struct dev_pm_ops dw_dev_pm_ops = {
 };
 
 #ifdef CONFIG_OF
-static const struct of_device_id dw_dma_id_table[] = {
+static const struct of_device_id dw_dma_of_id_table[] = {
 	{ .compatible = "snps,dma-spear1340" },
 	{}
 };
-MODULE_DEVICE_TABLE(of, dw_dma_id_table);
+MODULE_DEVICE_TABLE(of, dw_dma_of_id_table);
 #endif
 
-static const struct platform_device_id dw_dma_ids[] = {
-	/* Name,	Request Line Base */
-	{ "INTL9C60",	(kernel_ulong_t)16 },
+#ifdef CONFIG_ACPI
+static const struct acpi_device_id dw_dma_acpi_id_table[] = {
+	{ "INTL9C60", 0 },
 	{ }
 };
+#endif
 
 static struct platform_driver dw_driver = {
 	.probe		= dw_probe,
@@ -1932,9 +1949,9 @@ static struct platform_driver dw_driver = {
 	.driver = {
 		.name	= "dw_dmac",
 		.pm	= &dw_dev_pm_ops,
-		.of_match_table = of_match_ptr(dw_dma_id_table),
+		.of_match_table = of_match_ptr(dw_dma_of_id_table),
+		.acpi_match_table = ACPI_PTR(dw_dma_acpi_id_table),
 	},
-	.id_table	= dw_dma_ids,
 };
 
 static int __init dw_init(void)
diff --git a/drivers/dma/dw_dmac_regs.h b/drivers/dma/dw_dmac_regs.h
index 4d02c3669b7..9d417200bd5 100644
--- a/drivers/dma/dw_dmac_regs.h
+++ b/drivers/dma/dw_dmac_regs.h
@@ -212,8 +212,11 @@ struct dw_dma_chan {
 	/* hardware configuration */
 	unsigned int		block_size;
 	bool			nollp;
+
+	/* custom slave configuration */
 	unsigned int		request_line;
-	struct dw_dma_slave	slave;
+	unsigned char		src_master;
+	unsigned char		dst_master;
 
 	/* configuration passed via DMA_SLAVE_CONFIG */
 	struct dma_slave_config dma_sconfig;
@@ -247,7 +250,6 @@ struct dw_dma {
 	/* hardware configuration */
 	unsigned char		nr_masters;
 	unsigned char		data_width[4];
-	unsigned int		request_line_base;
 
 	struct dw_dma_chan	chan[0];
 };
diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
index 70b8975d107..f28583370d0 100644
--- a/drivers/dma/imx-dma.c
+++ b/drivers/dma/imx-dma.c
@@ -859,8 +859,7 @@ static struct dma_async_tx_descriptor *imxdma_prep_dma_cyclic(
 
 	desc = list_first_entry(&imxdmac->ld_free, struct imxdma_desc, node);
 
-	if (imxdmac->sg_list)
-		kfree(imxdmac->sg_list);
+	kfree(imxdmac->sg_list);
 
 	imxdmac->sg_list = kcalloc(periods + 1,
 			sizeof(struct scatterlist), GFP_KERNEL);
@@ -1145,7 +1144,7 @@ err:
 	return ret;
 }
 
-static int __exit imxdma_remove(struct platform_device *pdev)
+static int imxdma_remove(struct platform_device *pdev)
 {
 	struct imxdma_engine *imxdma = platform_get_drvdata(pdev);
 
@@ -1162,7 +1161,7 @@ static struct platform_driver imxdma_driver = {
 		.name	= "imx-dma",
 	},
 	.id_table	= imx_dma_devtype,
-	.remove		= __exit_p(imxdma_remove),
+	.remove		= imxdma_remove,
 };
 
 static int __init imxdma_module_init(void)
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index f082aa3a918..092867bf795 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -1462,7 +1462,7 @@ err_irq:
 	return ret;
 }
 
-static int __exit sdma_remove(struct platform_device *pdev)
+static int sdma_remove(struct platform_device *pdev)
 {
 	return -EBUSY;
 }
@@ -1473,7 +1473,7 @@ static struct platform_driver sdma_driver = {
 		.of_match_table = sdma_dt_ids,
 	},
 	.id_table	= sdma_devtypes,
-	.remove		= __exit_p(sdma_remove),
+	.remove		= sdma_remove,
 };
 
 static int __init sdma_module_init(void)
diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index 1879a5942bf..17a2393b3e2 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -892,7 +892,7 @@ MODULE_PARM_DESC(ioat_interrupt_style,
  * ioat_dma_setup_interrupts - setup interrupt handler
  * @device: ioat device
  */
-static int ioat_dma_setup_interrupts(struct ioatdma_device *device)
+int ioat_dma_setup_interrupts(struct ioatdma_device *device)
 {
 	struct ioat_chan_common *chan;
 	struct pci_dev *pdev = device->pdev;
@@ -941,6 +941,7 @@ msix:
 		}
 	}
 	intrctrl |= IOAT_INTRCTRL_MSIX_VECTOR_CONTROL;
+	device->irq_mode = IOAT_MSIX;
 	goto done;
 
 msix_single_vector:
@@ -956,6 +957,7 @@ msix_single_vector:
 		pci_disable_msix(pdev);
 		goto msi;
 	}
+	device->irq_mode = IOAT_MSIX_SINGLE;
 	goto done;
 
 msi:
@@ -969,6 +971,7 @@ msi:
 		pci_disable_msi(pdev);
 		goto intx;
 	}
+	device->irq_mode = IOAT_MSIX;
 	goto done;
 
 intx:
@@ -977,6 +980,7 @@ intx:
 	if (err)
 		goto err_no_irq;
 
+	device->irq_mode = IOAT_INTX;
 done:
 	if (device->intr_quirk)
 		device->intr_quirk(device);
@@ -987,9 +991,11 @@ done:
 err_no_irq:
 	/* Disable all interrupt generation */
 	writeb(0, device->reg_base + IOAT_INTRCTRL_OFFSET);
+	device->irq_mode = IOAT_NOIRQ;
 	dev_err(dev, "no usable interrupts\n");
 	return err;
 }
+EXPORT_SYMBOL(ioat_dma_setup_interrupts);
 
 static void ioat_disable_interrupts(struct ioatdma_device *device)
 {
diff --git a/drivers/dma/ioat/dma.h b/drivers/dma/ioat/dma.h
index 53a4cbb78f4..54fb7b9ff9a 100644
--- a/drivers/dma/ioat/dma.h
+++ b/drivers/dma/ioat/dma.h
@@ -39,6 +39,7 @@
 #define to_ioat_desc(lh) container_of(lh, struct ioat_desc_sw, node)
 #define tx_to_ioat_desc(tx) container_of(tx, struct ioat_desc_sw, txd)
 #define to_dev(ioat_chan) (&(ioat_chan)->device->pdev->dev)
+#define to_pdev(ioat_chan) ((ioat_chan)->device->pdev)
 
 #define chan_num(ch) ((int)((ch)->reg_base - (ch)->device->reg_base) / 0x80)
 
@@ -48,6 +49,14 @@
  */
 #define NULL_DESC_BUFFER_SIZE 1
 
+enum ioat_irq_mode {
+	IOAT_NOIRQ = 0,
+	IOAT_MSIX,
+	IOAT_MSIX_SINGLE,
+	IOAT_MSI,
+	IOAT_INTX
+};
+
 /**
  * struct ioatdma_device - internal representation of a IOAT device
  * @pdev: PCI-Express device
@@ -72,11 +81,16 @@ struct ioatdma_device {
 	void __iomem *reg_base;
 	struct pci_pool *dma_pool;
 	struct pci_pool *completion_pool;
+#define MAX_SED_POOLS	5
+	struct dma_pool *sed_hw_pool[MAX_SED_POOLS];
+	struct kmem_cache *sed_pool;
 	struct dma_device common;
 	u8 version;
 	struct msix_entry msix_entries[4];
 	struct ioat_chan_common *idx[4];
 	struct dca_provider *dca;
+	enum ioat_irq_mode irq_mode;
+	u32 cap;
 	void (*intr_quirk)(struct ioatdma_device *device);
 	int (*enumerate_channels)(struct ioatdma_device *device);
 	int (*reset_hw)(struct ioat_chan_common *chan);
@@ -131,6 +145,20 @@ struct ioat_dma_chan {
 	u16 active;
 };
 
+/**
+ * struct ioat_sed_ent - wrapper around super extended hardware descriptor
+ * @hw: hardware SED
+ * @sed_dma: dma address for the SED
+ * @list: list member
+ * @parent: point to the dma descriptor that's the parent
+ */
+struct ioat_sed_ent {
+	struct ioat_sed_raw_descriptor *hw;
+	dma_addr_t dma;
+	struct ioat_ring_ent *parent;
+	unsigned int hw_pool;
+};
+
 static inline struct ioat_chan_common *to_chan_common(struct dma_chan *c)
 {
 	return container_of(c, struct ioat_chan_common, common);
@@ -179,7 +207,7 @@ __dump_desc_dbg(struct ioat_chan_common *chan, struct ioat_dma_descriptor *hw,
 	struct device *dev = to_dev(chan);
 
 	dev_dbg(dev, "desc[%d]: (%#llx->%#llx) cookie: %d flags: %#x"
-		" ctl: %#x (op: %d int_en: %d compl: %d)\n", id,
+		" ctl: %#10.8x (op: %#x int_en: %d compl: %d)\n", id,
 		(unsigned long long) tx->phys,
 		(unsigned long long) hw->next, tx->cookie, tx->flags,
 		hw->ctl, hw->ctl_f.op, hw->ctl_f.int_en, hw->ctl_f.compl_write);
@@ -201,7 +229,7 @@ ioat_chan_by_index(struct ioatdma_device *device, int index)
 	return device->idx[index];
 }
 
-static inline u64 ioat_chansts(struct ioat_chan_common *chan)
+static inline u64 ioat_chansts_32(struct ioat_chan_common *chan)
 {
 	u8 ver = chan->device->version;
 	u64 status;
@@ -218,6 +246,26 @@ static inline u64 ioat_chansts(struct ioat_chan_common *chan)
 	return status;
 }
 
+#if BITS_PER_LONG == 64
+
+static inline u64 ioat_chansts(struct ioat_chan_common *chan)
+{
+	u8 ver = chan->device->version;
+	u64 status;
+
+	 /* With IOAT v3.3 the status register is 64bit.  */
+	if (ver >= IOAT_VER_3_3)
+		status = readq(chan->reg_base + IOAT_CHANSTS_OFFSET(ver));
+	else
+		status = ioat_chansts_32(chan);
+
+	return status;
+}
+
+#else
+#define ioat_chansts ioat_chansts_32
+#endif
+
 static inline void ioat_start(struct ioat_chan_common *chan)
 {
 	u8 ver = chan->device->version;
@@ -321,6 +369,7 @@ bool ioat_cleanup_preamble(struct ioat_chan_common *chan,
 			   dma_addr_t *phys_complete);
 void ioat_kobject_add(struct ioatdma_device *device, struct kobj_type *type);
 void ioat_kobject_del(struct ioatdma_device *device);
+int ioat_dma_setup_interrupts(struct ioatdma_device *device);
 extern const struct sysfs_ops ioat_sysfs_ops;
 extern struct ioat_sysfs_entry ioat_version_attr;
 extern struct ioat_sysfs_entry ioat_cap_attr;
diff --git a/drivers/dma/ioat/dma_v2.h b/drivers/dma/ioat/dma_v2.h
index e100f644e34..29bf9448035 100644
--- a/drivers/dma/ioat/dma_v2.h
+++ b/drivers/dma/ioat/dma_v2.h
@@ -137,6 +137,7 @@ struct ioat_ring_ent {
 	#ifdef DEBUG
 	int id;
 	#endif
+	struct ioat_sed_ent *sed;
 };
 
 static inline struct ioat_ring_ent *
@@ -157,6 +158,7 @@ static inline void ioat2_set_chainaddr(struct ioat2_dma_chan *ioat, u64 addr)
 
 int ioat2_dma_probe(struct ioatdma_device *dev, int dca);
 int ioat3_dma_probe(struct ioatdma_device *dev, int dca);
+void ioat3_dma_remove(struct ioatdma_device *dev);
 struct dca_provider *ioat2_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 int ioat2_check_space_lock(struct ioat2_dma_chan *ioat, int num_descs);
diff --git a/drivers/dma/ioat/dma_v3.c b/drivers/dma/ioat/dma_v3.c
index e8336cce360..ca6ea9b3551 100644
--- a/drivers/dma/ioat/dma_v3.c
+++ b/drivers/dma/ioat/dma_v3.c
@@ -55,7 +55,7 @@
 /*
  * Support routines for v3+ hardware
  */
-
+#include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/gfp.h>
 #include <linux/dmaengine.h>
@@ -70,6 +70,10 @@
 /* ioat hardware assumes at least two sources for raid operations */
 #define src_cnt_to_sw(x) ((x) + 2)
 #define src_cnt_to_hw(x) ((x) - 2)
+#define ndest_to_sw(x) ((x) + 1)
+#define ndest_to_hw(x) ((x) - 1)
+#define src16_cnt_to_sw(x) ((x) + 9)
+#define src16_cnt_to_hw(x) ((x) - 9)
 
 /* provide a lookup table for setting the source address in the base or
  * extended descriptor of an xor or pq descriptor
@@ -77,7 +81,20 @@
 static const u8 xor_idx_to_desc = 0xe0;
 static const u8 xor_idx_to_field[] = { 1, 4, 5, 6, 7, 0, 1, 2 };
 static const u8 pq_idx_to_desc = 0xf8;
+static const u8 pq16_idx_to_desc[] = { 0, 0, 1, 1, 1, 1, 1, 1, 1,
+				       2, 2, 2, 2, 2, 2, 2 };
 static const u8 pq_idx_to_field[] = { 1, 4, 5, 0, 1, 2, 4, 5 };
+static const u8 pq16_idx_to_field[] = { 1, 4, 1, 2, 3, 4, 5, 6, 7,
+					0, 1, 2, 3, 4, 5, 6 };
+
+/*
+ * technically sources 1 and 2 do not require SED, but the op will have
+ * at least 9 descriptors so that's irrelevant.
+ */
+static const u8 pq16_idx_to_sed[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0,
+				      1, 1, 1, 1, 1, 1, 1 };
+
+static void ioat3_eh(struct ioat2_dma_chan *ioat);
 
 static dma_addr_t xor_get_src(struct ioat_raw_descriptor *descs[2], int idx)
 {
@@ -101,6 +118,13 @@ static dma_addr_t pq_get_src(struct ioat_raw_descriptor *descs[2], int idx)
 	return raw->field[pq_idx_to_field[idx]];
 }
 
+static dma_addr_t pq16_get_src(struct ioat_raw_descriptor *desc[3], int idx)
+{
+	struct ioat_raw_descriptor *raw = desc[pq16_idx_to_desc[idx]];
+
+	return raw->field[pq16_idx_to_field[idx]];
+}
+
 static void pq_set_src(struct ioat_raw_descriptor *descs[2],
 		       dma_addr_t addr, u32 offset, u8 coef, int idx)
 {
@@ -111,6 +135,167 @@ static void pq_set_src(struct ioat_raw_descriptor *descs[2],
 	pq->coef[idx] = coef;
 }
 
+static int sed_get_pq16_pool_idx(int src_cnt)
+{
+
+	return pq16_idx_to_sed[src_cnt];
+}
+
+static bool is_jf_ioat(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF0:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF1:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF2:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF3:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF4:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF5:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF6:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF7:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF8:
+	case PCI_DEVICE_ID_INTEL_IOAT_JSF9:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool is_snb_ioat(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB0:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB1:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB2:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB3:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB4:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB5:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB6:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB7:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB8:
+	case PCI_DEVICE_ID_INTEL_IOAT_SNB9:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool is_ivb_ioat(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB0:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB1:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB2:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB3:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB4:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB5:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB6:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB7:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB8:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB9:
+		return true;
+	default:
+		return false;
+	}
+
+}
+
+static bool is_hsw_ioat(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW0:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW1:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW2:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW3:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW4:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW5:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW6:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW7:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW8:
+	case PCI_DEVICE_ID_INTEL_IOAT_HSW9:
+		return true;
+	default:
+		return false;
+	}
+
+}
+
+static bool is_xeon_cb32(struct pci_dev *pdev)
+{
+	return is_jf_ioat(pdev) || is_snb_ioat(pdev) || is_ivb_ioat(pdev) ||
+		is_hsw_ioat(pdev);
+}
+
+static bool is_bwd_ioat(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_INTEL_IOAT_BWD0:
+	case PCI_DEVICE_ID_INTEL_IOAT_BWD1:
+	case PCI_DEVICE_ID_INTEL_IOAT_BWD2:
+	case PCI_DEVICE_ID_INTEL_IOAT_BWD3:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool is_bwd_noraid(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_INTEL_IOAT_BWD2:
+	case PCI_DEVICE_ID_INTEL_IOAT_BWD3:
+		return true;
+	default:
+		return false;
+	}
+
+}
+
+static void pq16_set_src(struct ioat_raw_descriptor *desc[3],
+			dma_addr_t addr, u32 offset, u8 coef, int idx)
+{
+	struct ioat_pq_descriptor *pq = (struct ioat_pq_descriptor *)desc[0];
+	struct ioat_pq16a_descriptor *pq16 =
+		(struct ioat_pq16a_descriptor *)desc[1];
+	struct ioat_raw_descriptor *raw = desc[pq16_idx_to_desc[idx]];
+
+	raw->field[pq16_idx_to_field[idx]] = addr + offset;
+
+	if (idx < 8)
+		pq->coef[idx] = coef;
+	else
+		pq16->coef[idx - 8] = coef;
+}
+
+static struct ioat_sed_ent *
+ioat3_alloc_sed(struct ioatdma_device *device, unsigned int hw_pool)
+{
+	struct ioat_sed_ent *sed;
+	gfp_t flags = __GFP_ZERO | GFP_ATOMIC;
+
+	sed = kmem_cache_alloc(device->sed_pool, flags);
+	if (!sed)
+		return NULL;
+
+	sed->hw_pool = hw_pool;
+	sed->hw = dma_pool_alloc(device->sed_hw_pool[hw_pool],
+				 flags, &sed->dma);
+	if (!sed->hw) {
+		kmem_cache_free(device->sed_pool, sed);
+		return NULL;
+	}
+
+	return sed;
+}
+
+static void ioat3_free_sed(struct ioatdma_device *device, struct ioat_sed_ent *sed)
+{
+	if (!sed)
+		return;
+
+	dma_pool_free(device->sed_hw_pool[sed->hw_pool], sed->hw, sed->dma);
+	kmem_cache_free(device->sed_pool, sed);
+}
+
 static void ioat3_dma_unmap(struct ioat2_dma_chan *ioat,
 			    struct ioat_ring_ent *desc, int idx)
 {
@@ -223,6 +408,54 @@ static void ioat3_dma_unmap(struct ioat2_dma_chan *ioat,
 		}
 		break;
 	}
+	case IOAT_OP_PQ_16S:
+	case IOAT_OP_PQ_VAL_16S: {
+		struct ioat_pq_descriptor *pq = desc->pq;
+		int src_cnt = src16_cnt_to_sw(pq->ctl_f.src_cnt);
+		struct ioat_raw_descriptor *descs[4];
+		int i;
+
+		/* in the 'continue' case don't unmap the dests as sources */
+		if (dmaf_p_disabled_continue(flags))
+			src_cnt--;
+		else if (dmaf_continue(flags))
+			src_cnt -= 3;
+
+		if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
+			descs[0] = (struct ioat_raw_descriptor *)pq;
+			descs[1] = (struct ioat_raw_descriptor *)(desc->sed->hw);
+			descs[2] = (struct ioat_raw_descriptor *)(&desc->sed->hw->b[0]);
+			for (i = 0; i < src_cnt; i++) {
+				dma_addr_t src = pq16_get_src(descs, i);
+
+				ioat_unmap(pdev, src - offset, len,
+					   PCI_DMA_TODEVICE, flags, 0);
+			}
+
+			/* the dests are sources in pq validate operations */
+			if (pq->ctl_f.op == IOAT_OP_XOR_VAL) {
+				if (!(flags & DMA_PREP_PQ_DISABLE_P))
+					ioat_unmap(pdev, pq->p_addr - offset,
+						   len, PCI_DMA_TODEVICE,
+						   flags, 0);
+				if (!(flags & DMA_PREP_PQ_DISABLE_Q))
+					ioat_unmap(pdev, pq->q_addr - offset,
+						   len, PCI_DMA_TODEVICE,
+						   flags, 0);
+				break;
+			}
+		}
+
+		if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
+			if (!(flags & DMA_PREP_PQ_DISABLE_P))
+				ioat_unmap(pdev, pq->p_addr - offset, len,
+					   PCI_DMA_BIDIRECTIONAL, flags, 1);
+			if (!(flags & DMA_PREP_PQ_DISABLE_Q))
+				ioat_unmap(pdev, pq->q_addr - offset, len,
+					   PCI_DMA_BIDIRECTIONAL, flags, 1);
+		}
+		break;
+	}
 	default:
 		dev_err(&pdev->dev, "%s: unknown op type: %#x\n",
 			__func__, desc->hw->ctl_f.op);
@@ -250,6 +483,63 @@ static bool desc_has_ext(struct ioat_ring_ent *desc)
 	return false;
 }
 
+static u64 ioat3_get_current_completion(struct ioat_chan_common *chan)
+{
+	u64 phys_complete;
+	u64 completion;
+
+	completion = *chan->completion;
+	phys_complete = ioat_chansts_to_addr(completion);
+
+	dev_dbg(to_dev(chan), "%s: phys_complete: %#llx\n", __func__,
+		(unsigned long long) phys_complete);
+
+	return phys_complete;
+}
+
+static bool ioat3_cleanup_preamble(struct ioat_chan_common *chan,
+				   u64 *phys_complete)
+{
+	*phys_complete = ioat3_get_current_completion(chan);
+	if (*phys_complete == chan->last_completion)
+		return false;
+
+	clear_bit(IOAT_COMPLETION_ACK, &chan->state);
+	mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+
+	return true;
+}
+
+static void
+desc_get_errstat(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc)
+{
+	struct ioat_dma_descriptor *hw = desc->hw;
+
+	switch (hw->ctl_f.op) {
+	case IOAT_OP_PQ_VAL:
+	case IOAT_OP_PQ_VAL_16S:
+	{
+		struct ioat_pq_descriptor *pq = desc->pq;
+
+		/* check if there's error written */
+		if (!pq->dwbes_f.wbes)
+			return;
+
+		/* need to set a chanerr var for checking to clear later */
+
+		if (pq->dwbes_f.p_val_err)
+			*desc->result |= SUM_CHECK_P_RESULT;
+
+		if (pq->dwbes_f.q_val_err)
+			*desc->result |= SUM_CHECK_Q_RESULT;
+
+		return;
+	}
+	default:
+		return;
+	}
+}
+
 /**
  * __cleanup - reclaim used descriptors
  * @ioat: channel (ring) to clean
@@ -260,6 +550,7 @@ static bool desc_has_ext(struct ioat_ring_ent *desc)
 static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
 {
 	struct ioat_chan_common *chan = &ioat->base;
+	struct ioatdma_device *device = chan->device;
 	struct ioat_ring_ent *desc;
 	bool seen_current = false;
 	int idx = ioat->tail, i;
@@ -268,6 +559,16 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
 	dev_dbg(to_dev(chan), "%s: head: %#x tail: %#x issued: %#x\n",
 		__func__, ioat->head, ioat->tail, ioat->issued);
 
+	/*
+	 * At restart of the channel, the completion address and the
+	 * channel status will be 0 due to starting a new chain. Since
+	 * it's new chain and the first descriptor "fails", there is
+	 * nothing to clean up. We do not want to reap the entire submitted
+	 * chain due to this 0 address value and then BUG.
+	 */
+	if (!phys_complete)
+		return;
+
 	active = ioat2_ring_active(ioat);
 	for (i = 0; i < active && !seen_current; i++) {
 		struct dma_async_tx_descriptor *tx;
@@ -276,6 +577,11 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
 		prefetch(ioat2_get_ring_ent(ioat, idx + i + 1));
 		desc = ioat2_get_ring_ent(ioat, idx + i);
 		dump_desc_dbg(ioat, desc);
+
+		/* set err stat if we are using dwbes */
+		if (device->cap & IOAT_CAP_DWBES)
+			desc_get_errstat(ioat, desc);
+
 		tx = &desc->txd;
 		if (tx->cookie) {
 			dma_cookie_complete(tx);
@@ -294,6 +600,12 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
 			BUG_ON(i + 1 >= active);
 			i++;
 		}
+
+		/* cleanup super extended descriptors */
+		if (desc->sed) {
+			ioat3_free_sed(device, desc->sed);
+			desc->sed = NULL;
+		}
 	}
 	smp_mb(); /* finish all descriptor reads before incrementing tail */
 	ioat->tail = idx + i;
@@ -314,11 +626,22 @@ static void __cleanup(struct ioat2_dma_chan *ioat, dma_addr_t phys_complete)
 static void ioat3_cleanup(struct ioat2_dma_chan *ioat)
 {
 	struct ioat_chan_common *chan = &ioat->base;
-	dma_addr_t phys_complete;
+	u64 phys_complete;
 
 	spin_lock_bh(&chan->cleanup_lock);
-	if (ioat_cleanup_preamble(chan, &phys_complete))
+
+	if (ioat3_cleanup_preamble(chan, &phys_complete))
 		__cleanup(ioat, phys_complete);
+
+	if (is_ioat_halted(*chan->completion)) {
+		u32 chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+
+		if (chanerr & IOAT_CHANERR_HANDLE_MASK) {
+			mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
+			ioat3_eh(ioat);
+		}
+	}
+
 	spin_unlock_bh(&chan->cleanup_lock);
 }
 
@@ -333,15 +656,78 @@ static void ioat3_cleanup_event(unsigned long data)
 static void ioat3_restart_channel(struct ioat2_dma_chan *ioat)
 {
 	struct ioat_chan_common *chan = &ioat->base;
-	dma_addr_t phys_complete;
+	u64 phys_complete;
 
 	ioat2_quiesce(chan, 0);
-	if (ioat_cleanup_preamble(chan, &phys_complete))
+	if (ioat3_cleanup_preamble(chan, &phys_complete))
 		__cleanup(ioat, phys_complete);
 
 	__ioat2_restart_chan(ioat);
 }
 
+static void ioat3_eh(struct ioat2_dma_chan *ioat)
+{
+	struct ioat_chan_common *chan = &ioat->base;
+	struct pci_dev *pdev = to_pdev(chan);
+	struct ioat_dma_descriptor *hw;
+	u64 phys_complete;
+	struct ioat_ring_ent *desc;
+	u32 err_handled = 0;
+	u32 chanerr_int;
+	u32 chanerr;
+
+	/* cleanup so tail points to descriptor that caused the error */
+	if (ioat3_cleanup_preamble(chan, &phys_complete))
+		__cleanup(ioat, phys_complete);
+
+	chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+	pci_read_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, &chanerr_int);
+
+	dev_dbg(to_dev(chan), "%s: error = %x:%x\n",
+		__func__, chanerr, chanerr_int);
+
+	desc = ioat2_get_ring_ent(ioat, ioat->tail);
+	hw = desc->hw;
+	dump_desc_dbg(ioat, desc);
+
+	switch (hw->ctl_f.op) {
+	case IOAT_OP_XOR_VAL:
+		if (chanerr & IOAT_CHANERR_XOR_P_OR_CRC_ERR) {
+			*desc->result |= SUM_CHECK_P_RESULT;
+			err_handled |= IOAT_CHANERR_XOR_P_OR_CRC_ERR;
+		}
+		break;
+	case IOAT_OP_PQ_VAL:
+	case IOAT_OP_PQ_VAL_16S:
+		if (chanerr & IOAT_CHANERR_XOR_P_OR_CRC_ERR) {
+			*desc->result |= SUM_CHECK_P_RESULT;
+			err_handled |= IOAT_CHANERR_XOR_P_OR_CRC_ERR;
+		}
+		if (chanerr & IOAT_CHANERR_XOR_Q_ERR) {
+			*desc->result |= SUM_CHECK_Q_RESULT;
+			err_handled |= IOAT_CHANERR_XOR_Q_ERR;
+		}
+		break;
+	}
+
+	/* fault on unhandled error or spurious halt */
+	if (chanerr ^ err_handled || chanerr == 0) {
+		dev_err(to_dev(chan), "%s: fatal error (%x:%x)\n",
+			__func__, chanerr, err_handled);
+		BUG();
+	}
+
+	writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
+	pci_write_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, chanerr_int);
+
+	/* mark faulting descriptor as complete */
+	*chan->completion = desc->txd.phys;
+
+	spin_lock_bh(&ioat->prep_lock);
+	ioat3_restart_channel(ioat);
+	spin_unlock_bh(&ioat->prep_lock);
+}
+
 static void check_active(struct ioat2_dma_chan *ioat)
 {
 	struct ioat_chan_common *chan = &ioat->base;
@@ -605,7 +991,8 @@ dump_pq_desc_dbg(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc, struct
 	int i;
 
 	dev_dbg(dev, "desc[%d]: (%#llx->%#llx) flags: %#x"
-		" sz: %#x ctl: %#x (op: %d int: %d compl: %d pq: '%s%s' src_cnt: %d)\n",
+		" sz: %#10.8x ctl: %#x (op: %#x int: %d compl: %d pq: '%s%s'"
+		" src_cnt: %d)\n",
 		desc_id(desc), (unsigned long long) desc->txd.phys,
 		(unsigned long long) (pq_ex ? pq_ex->next : pq->next),
 		desc->txd.flags, pq->size, pq->ctl, pq->ctl_f.op, pq->ctl_f.int_en,
@@ -617,6 +1004,42 @@ dump_pq_desc_dbg(struct ioat2_dma_chan *ioat, struct ioat_ring_ent *desc, struct
 			(unsigned long long) pq_get_src(descs, i), pq->coef[i]);
 	dev_dbg(dev, "\tP: %#llx\n", pq->p_addr);
 	dev_dbg(dev, "\tQ: %#llx\n", pq->q_addr);
+	dev_dbg(dev, "\tNEXT: %#llx\n", pq->next);
+}
+
+static void dump_pq16_desc_dbg(struct ioat2_dma_chan *ioat,
+			       struct ioat_ring_ent *desc)
+{
+	struct device *dev = to_dev(&ioat->base);
+	struct ioat_pq_descriptor *pq = desc->pq;
+	struct ioat_raw_descriptor *descs[] = { (void *)pq,
+						(void *)pq,
+						(void *)pq };
+	int src_cnt = src16_cnt_to_sw(pq->ctl_f.src_cnt);
+	int i;
+
+	if (desc->sed) {
+		descs[1] = (void *)desc->sed->hw;
+		descs[2] = (void *)desc->sed->hw + 64;
+	}
+
+	dev_dbg(dev, "desc[%d]: (%#llx->%#llx) flags: %#x"
+		" sz: %#x ctl: %#x (op: %#x int: %d compl: %d pq: '%s%s'"
+		" src_cnt: %d)\n",
+		desc_id(desc), (unsigned long long) desc->txd.phys,
+		(unsigned long long) pq->next,
+		desc->txd.flags, pq->size, pq->ctl,
+		pq->ctl_f.op, pq->ctl_f.int_en,
+		pq->ctl_f.compl_write,
+		pq->ctl_f.p_disable ? "" : "p", pq->ctl_f.q_disable ? "" : "q",
+		pq->ctl_f.src_cnt);
+	for (i = 0; i < src_cnt; i++) {
+		dev_dbg(dev, "\tsrc[%d]: %#llx coef: %#x\n", i,
+			(unsigned long long) pq16_get_src(descs, i),
+			pq->coef[i]);
+	}
+	dev_dbg(dev, "\tP: %#llx\n", pq->p_addr);
+	dev_dbg(dev, "\tQ: %#llx\n", pq->q_addr);
 }
 
 static struct dma_async_tx_descriptor *
@@ -627,6 +1050,7 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
 {
 	struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
 	struct ioat_chan_common *chan = &ioat->base;
+	struct ioatdma_device *device = chan->device;
 	struct ioat_ring_ent *compl_desc;
 	struct ioat_ring_ent *desc;
 	struct ioat_ring_ent *ext;
@@ -637,6 +1061,7 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
 	u32 offset = 0;
 	u8 op = result ? IOAT_OP_PQ_VAL : IOAT_OP_PQ;
 	int i, s, idx, with_ext, num_descs;
+	int cb32 = (device->version < IOAT_VER_3_3) ? 1 : 0;
 
 	dev_dbg(to_dev(chan), "%s\n", __func__);
 	/* the engine requires at least two sources (we provide
@@ -662,7 +1087,7 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
 	 * order.
 	 */
 	if (likely(num_descs) &&
-	    ioat2_check_space_lock(ioat, num_descs+1) == 0)
+	    ioat2_check_space_lock(ioat, num_descs + cb32) == 0)
 		idx = ioat->head;
 	else
 		return NULL;
@@ -700,6 +1125,9 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
 		pq->q_addr = dst[1] + offset;
 		pq->ctl = 0;
 		pq->ctl_f.op = op;
+		/* we turn on descriptor write back error status */
+		if (device->cap & IOAT_CAP_DWBES)
+			pq->ctl_f.wb_en = result ? 1 : 0;
 		pq->ctl_f.src_cnt = src_cnt_to_hw(s);
 		pq->ctl_f.p_disable = !!(flags & DMA_PREP_PQ_DISABLE_P);
 		pq->ctl_f.q_disable = !!(flags & DMA_PREP_PQ_DISABLE_Q);
@@ -716,26 +1144,140 @@ __ioat3_prep_pq_lock(struct dma_chan *c, enum sum_check_flags *result,
 	pq->ctl_f.fence = !!(flags & DMA_PREP_FENCE);
 	dump_pq_desc_dbg(ioat, desc, ext);
 
-	/* completion descriptor carries interrupt bit */
-	compl_desc = ioat2_get_ring_ent(ioat, idx + i);
-	compl_desc->txd.flags = flags & DMA_PREP_INTERRUPT;
-	hw = compl_desc->hw;
-	hw->ctl = 0;
-	hw->ctl_f.null = 1;
-	hw->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
-	hw->ctl_f.compl_write = 1;
-	hw->size = NULL_DESC_BUFFER_SIZE;
-	dump_desc_dbg(ioat, compl_desc);
+	if (!cb32) {
+		pq->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
+		pq->ctl_f.compl_write = 1;
+		compl_desc = desc;
+	} else {
+		/* completion descriptor carries interrupt bit */
+		compl_desc = ioat2_get_ring_ent(ioat, idx + i);
+		compl_desc->txd.flags = flags & DMA_PREP_INTERRUPT;
+		hw = compl_desc->hw;
+		hw->ctl = 0;
+		hw->ctl_f.null = 1;
+		hw->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
+		hw->ctl_f.compl_write = 1;
+		hw->size = NULL_DESC_BUFFER_SIZE;
+		dump_desc_dbg(ioat, compl_desc);
+	}
+
 
 	/* we leave the channel locked to ensure in order submission */
 	return &compl_desc->txd;
 }
 
 static struct dma_async_tx_descriptor *
+__ioat3_prep_pq16_lock(struct dma_chan *c, enum sum_check_flags *result,
+		       const dma_addr_t *dst, const dma_addr_t *src,
+		       unsigned int src_cnt, const unsigned char *scf,
+		       size_t len, unsigned long flags)
+{
+	struct ioat2_dma_chan *ioat = to_ioat2_chan(c);
+	struct ioat_chan_common *chan = &ioat->base;
+	struct ioatdma_device *device = chan->device;
+	struct ioat_ring_ent *desc;
+	size_t total_len = len;
+	struct ioat_pq_descriptor *pq;
+	u32 offset = 0;
+	u8 op;
+	int i, s, idx, num_descs;
+
+	/* this function only handles src_cnt 9 - 16 */
+	BUG_ON(src_cnt < 9);
+
+	/* this function is only called with 9-16 sources */
+	op = result ? IOAT_OP_PQ_VAL_16S : IOAT_OP_PQ_16S;
+
+	dev_dbg(to_dev(chan), "%s\n", __func__);
+
+	num_descs = ioat2_xferlen_to_descs(ioat, len);
+
+	/*
+	 * 16 source pq is only available on cb3.3 and has no completion
+	 * write hw bug.
+	 */
+	if (num_descs && ioat2_check_space_lock(ioat, num_descs) == 0)
+		idx = ioat->head;
+	else
+		return NULL;
+
+	i = 0;
+
+	do {
+		struct ioat_raw_descriptor *descs[4];
+		size_t xfer_size = min_t(size_t, len, 1 << ioat->xfercap_log);
+
+		desc = ioat2_get_ring_ent(ioat, idx + i);
+		pq = desc->pq;
+
+		descs[0] = (struct ioat_raw_descriptor *) pq;
+
+		desc->sed = ioat3_alloc_sed(device,
+					    sed_get_pq16_pool_idx(src_cnt));
+		if (!desc->sed) {
+			dev_err(to_dev(chan),
+				"%s: no free sed entries\n", __func__);
+			return NULL;
+		}
+
+		pq->sed_addr = desc->sed->dma;
+		desc->sed->parent = desc;
+
+		descs[1] = (struct ioat_raw_descriptor *)desc->sed->hw;
+		descs[2] = (void *)descs[1] + 64;
+
+		for (s = 0; s < src_cnt; s++)
+			pq16_set_src(descs, src[s], offset, scf[s], s);
+
+		/* see the comment for dma_maxpq in include/linux/dmaengine.h */
+		if (dmaf_p_disabled_continue(flags))
+			pq16_set_src(descs, dst[1], offset, 1, s++);
+		else if (dmaf_continue(flags)) {
+			pq16_set_src(descs, dst[0], offset, 0, s++);
+			pq16_set_src(descs, dst[1], offset, 1, s++);
+			pq16_set_src(descs, dst[1], offset, 0, s++);
+		}
+
+		pq->size = xfer_size;
+		pq->p_addr = dst[0] + offset;
+		pq->q_addr = dst[1] + offset;
+		pq->ctl = 0;
+		pq->ctl_f.op = op;
+		pq->ctl_f.src_cnt = src16_cnt_to_hw(s);
+		/* we turn on descriptor write back error status */
+		if (device->cap & IOAT_CAP_DWBES)
+			pq->ctl_f.wb_en = result ? 1 : 0;
+		pq->ctl_f.p_disable = !!(flags & DMA_PREP_PQ_DISABLE_P);
+		pq->ctl_f.q_disable = !!(flags & DMA_PREP_PQ_DISABLE_Q);
+
+		len -= xfer_size;
+		offset += xfer_size;
+	} while (++i < num_descs);
+
+	/* last pq descriptor carries the unmap parameters and fence bit */
+	desc->txd.flags = flags;
+	desc->len = total_len;
+	if (result)
+		desc->result = result;
+	pq->ctl_f.fence = !!(flags & DMA_PREP_FENCE);
+
+	/* with cb3.3 we should be able to do completion w/o a null desc */
+	pq->ctl_f.int_en = !!(flags & DMA_PREP_INTERRUPT);
+	pq->ctl_f.compl_write = 1;
+
+	dump_pq16_desc_dbg(ioat, desc);
+
+	/* we leave the channel locked to ensure in order submission */
+	return &desc->txd;
+}
+
+static struct dma_async_tx_descriptor *
 ioat3_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 dma_device *dma = chan->device;
+
 	/* specify valid address for disabled result */
 	if (flags & DMA_PREP_PQ_DISABLE_P)
 		dst[0] = dst[1];
@@ -755,11 +1297,20 @@ ioat3_prep_pq(struct dma_chan *chan, dma_addr_t *dst, dma_addr_t *src,
 		single_source_coef[0] = scf[0];
 		single_source_coef[1] = 0;
 
-		return __ioat3_prep_pq_lock(chan, NULL, dst, single_source, 2,
-					    single_source_coef, len, flags);
-	} else
-		return __ioat3_prep_pq_lock(chan, NULL, dst, src, src_cnt, scf,
-					    len, flags);
+		return (src_cnt > 8) && (dma->max_pq > 8) ?
+			__ioat3_prep_pq16_lock(chan, NULL, dst, single_source,
+					       2, single_source_coef, len,
+					       flags) :
+			__ioat3_prep_pq_lock(chan, NULL, dst, single_source, 2,
+					     single_source_coef, len, flags);
+
+	} else {
+		return (src_cnt > 8) && (dma->max_pq > 8) ?
+			__ioat3_prep_pq16_lock(chan, NULL, dst, src, src_cnt,
+					       scf, len, flags) :
+			__ioat3_prep_pq_lock(chan, NULL, dst, src, src_cnt,
+					     scf, len, flags);
+	}
 }
 
 struct dma_async_tx_descriptor *
@@ -767,6 +1318,8 @@ ioat3_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 dma_device *dma = chan->device;
+
 	/* specify valid address for disabled result */
 	if (flags & DMA_PREP_PQ_DISABLE_P)
 		pq[0] = pq[1];
@@ -778,14 +1331,18 @@ ioat3_prep_pq_val(struct dma_chan *chan, dma_addr_t *pq, dma_addr_t *src,
 	 */
 	*pqres = 0;
 
-	return __ioat3_prep_pq_lock(chan, pqres, pq, src, src_cnt, scf, len,
-				    flags);
+	return (src_cnt > 8) && (dma->max_pq > 8) ?
+		__ioat3_prep_pq16_lock(chan, pqres, pq, src, src_cnt, scf, len,
+				       flags) :
+		__ioat3_prep_pq_lock(chan, pqres, pq, src, src_cnt, scf, len,
+				     flags);
 }
 
 static struct dma_async_tx_descriptor *
 ioat3_prep_pqxor(struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
 		 unsigned int src_cnt, size_t len, unsigned long flags)
 {
+	struct dma_device *dma = chan->device;
 	unsigned char scf[src_cnt];
 	dma_addr_t pq[2];
 
@@ -794,8 +1351,11 @@ ioat3_prep_pqxor(struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
 	flags |= DMA_PREP_PQ_DISABLE_Q;
 	pq[1] = dst; /* specify valid address for disabled result */
 
-	return __ioat3_prep_pq_lock(chan, NULL, pq, src, src_cnt, scf, len,
-				    flags);
+	return (src_cnt > 8) && (dma->max_pq > 8) ?
+		__ioat3_prep_pq16_lock(chan, NULL, pq, src, src_cnt, scf, len,
+				       flags) :
+		__ioat3_prep_pq_lock(chan, NULL, pq, src, src_cnt, scf, len,
+				     flags);
 }
 
 struct dma_async_tx_descriptor *
@@ -803,6 +1363,7 @@ ioat3_prep_pqxor_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 dma_device *dma = chan->device;
 	unsigned char scf[src_cnt];
 	dma_addr_t pq[2];
 
@@ -816,8 +1377,12 @@ ioat3_prep_pqxor_val(struct dma_chan *chan, dma_addr_t *src,
 	flags |= DMA_PREP_PQ_DISABLE_Q;
 	pq[1] = pq[0]; /* specify valid address for disabled result */
 
-	return __ioat3_prep_pq_lock(chan, result, pq, &src[1], src_cnt - 1, scf,
-				    len, flags);
+
+	return (src_cnt > 8) && (dma->max_pq > 8) ?
+		__ioat3_prep_pq16_lock(chan, result, pq, &src[1], src_cnt - 1,
+				       scf, len, flags) :
+		__ioat3_prep_pq_lock(chan, result, pq, &src[1], src_cnt - 1,
+				     scf, len, flags);
 }
 
 static struct dma_async_tx_descriptor *
@@ -1167,6 +1732,56 @@ static int ioat3_dma_self_test(struct ioatdma_device *device)
 	return 0;
 }
 
+static int ioat3_irq_reinit(struct ioatdma_device *device)
+{
+	int msixcnt = device->common.chancnt;
+	struct pci_dev *pdev = device->pdev;
+	int i;
+	struct msix_entry *msix;
+	struct ioat_chan_common *chan;
+	int err = 0;
+
+	switch (device->irq_mode) {
+	case IOAT_MSIX:
+
+		for (i = 0; i < msixcnt; i++) {
+			msix = &device->msix_entries[i];
+			chan = ioat_chan_by_index(device, i);
+			devm_free_irq(&pdev->dev, msix->vector, chan);
+		}
+
+		pci_disable_msix(pdev);
+		break;
+
+	case IOAT_MSIX_SINGLE:
+		msix = &device->msix_entries[0];
+		chan = ioat_chan_by_index(device, 0);
+		devm_free_irq(&pdev->dev, msix->vector, chan);
+		pci_disable_msix(pdev);
+		break;
+
+	case IOAT_MSI:
+		chan = ioat_chan_by_index(device, 0);
+		devm_free_irq(&pdev->dev, pdev->irq, chan);
+		pci_disable_msi(pdev);
+		break;
+
+	case IOAT_INTX:
+		chan = ioat_chan_by_index(device, 0);
+		devm_free_irq(&pdev->dev, pdev->irq, chan);
+		break;
+
+	default:
+		return 0;
+	}
+
+	device->irq_mode = IOAT_NOIRQ;
+
+	err = ioat_dma_setup_interrupts(device);
+
+	return err;
+}
+
 static int ioat3_reset_hw(struct ioat_chan_common *chan)
 {
 	/* throw away whatever the channel was doing and get it
@@ -1183,80 +1798,65 @@ static int ioat3_reset_hw(struct ioat_chan_common *chan)
 	chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
 	writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
 
-	/* clear any pending errors */
-	err = pci_read_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, &chanerr);
+	if (device->version < IOAT_VER_3_3) {
+		/* clear any pending errors */
+		err = pci_read_config_dword(pdev,
+				IOAT_PCI_CHANERR_INT_OFFSET, &chanerr);
+		if (err) {
+			dev_err(&pdev->dev,
+				"channel error register unreachable\n");
+			return err;
+		}
+		pci_write_config_dword(pdev,
+				IOAT_PCI_CHANERR_INT_OFFSET, chanerr);
+
+		/* Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
+		 * (workaround for spurious config parity error after restart)
+		 */
+		pci_read_config_word(pdev, IOAT_PCI_DEVICE_ID_OFFSET, &dev_id);
+		if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0) {
+			pci_write_config_dword(pdev,
+					       IOAT_PCI_DMAUNCERRSTS_OFFSET,
+					       0x10);
+		}
+	}
+
+	err = ioat2_reset_sync(chan, msecs_to_jiffies(200));
 	if (err) {
-		dev_err(&pdev->dev, "channel error register unreachable\n");
+		dev_err(&pdev->dev, "Failed to reset!\n");
 		return err;
 	}
-	pci_write_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, chanerr);
 
-	/* Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
-	 * (workaround for spurious config parity error after restart)
-	 */
-	pci_read_config_word(pdev, IOAT_PCI_DEVICE_ID_OFFSET, &dev_id);
-	if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0)
-		pci_write_config_dword(pdev, IOAT_PCI_DMAUNCERRSTS_OFFSET, 0x10);
+	if (device->irq_mode != IOAT_NOIRQ && is_bwd_ioat(pdev))
+		err = ioat3_irq_reinit(device);
 
-	return ioat2_reset_sync(chan, msecs_to_jiffies(200));
+	return err;
 }
 
-static bool is_jf_ioat(struct pci_dev *pdev)
+static void ioat3_intr_quirk(struct ioatdma_device *device)
 {
-	switch (pdev->device) {
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF0:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF1:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF2:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF3:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF4:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF5:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF6:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF7:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF8:
-	case PCI_DEVICE_ID_INTEL_IOAT_JSF9:
-		return true;
-	default:
-		return false;
-	}
-}
+	struct dma_device *dma;
+	struct dma_chan *c;
+	struct ioat_chan_common *chan;
+	u32 errmask;
 
-static bool is_snb_ioat(struct pci_dev *pdev)
-{
-	switch (pdev->device) {
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB0:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB1:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB2:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB3:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB4:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB5:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB6:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB7:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB8:
-	case PCI_DEVICE_ID_INTEL_IOAT_SNB9:
-		return true;
-	default:
-		return false;
-	}
-}
+	dma = &device->common;
 
-static bool is_ivb_ioat(struct pci_dev *pdev)
-{
-	switch (pdev->device) {
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB0:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB1:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB2:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB3:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB4:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB5:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB6:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB7:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB8:
-	case PCI_DEVICE_ID_INTEL_IOAT_IVB9:
-		return true;
-	default:
-		return false;
+	/*
+	 * if we have descriptor write back error status, we mask the
+	 * error interrupts
+	 */
+	if (device->cap & IOAT_CAP_DWBES) {
+		list_for_each_entry(c, &dma->channels, device_node) {
+			chan = to_chan_common(c);
+			errmask = readl(chan->reg_base +
+					IOAT_CHANERR_MASK_OFFSET);
+			errmask |= IOAT_CHANERR_XOR_P_OR_CRC_ERR |
+				   IOAT_CHANERR_XOR_Q_ERR;
+			writel(errmask, chan->reg_base +
+					IOAT_CHANERR_MASK_OFFSET);
+		}
 	}
-
 }
 
 int ioat3_dma_probe(struct ioatdma_device *device, int dca)
@@ -1268,30 +1868,33 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
 	struct ioat_chan_common *chan;
 	bool is_raid_device = false;
 	int err;
-	u32 cap;
 
 	device->enumerate_channels = ioat2_enumerate_channels;
 	device->reset_hw = ioat3_reset_hw;
 	device->self_test = ioat3_dma_self_test;
+	device->intr_quirk = ioat3_intr_quirk;
 	dma = &device->common;
 	dma->device_prep_dma_memcpy = ioat2_dma_prep_memcpy_lock;
 	dma->device_issue_pending = ioat2_issue_pending;
 	dma->device_alloc_chan_resources = ioat2_alloc_chan_resources;
 	dma->device_free_chan_resources = ioat2_free_chan_resources;
 
-	if (is_jf_ioat(pdev) || is_snb_ioat(pdev) || is_ivb_ioat(pdev))
+	if (is_xeon_cb32(pdev))
 		dma->copy_align = 6;
 
 	dma_cap_set(DMA_INTERRUPT, dma->cap_mask);
 	dma->device_prep_dma_interrupt = ioat3_prep_interrupt_lock;
 
-	cap = readl(device->reg_base + IOAT_DMA_CAP_OFFSET);
+	device->cap = readl(device->reg_base + IOAT_DMA_CAP_OFFSET);
+
+	if (is_bwd_noraid(pdev))
+		device->cap &= ~(IOAT_CAP_XOR | IOAT_CAP_PQ | IOAT_CAP_RAID16SS);
 
 	/* dca is incompatible with raid operations */
-	if (dca_en && (cap & (IOAT_CAP_XOR|IOAT_CAP_PQ)))
-		cap &= ~(IOAT_CAP_XOR|IOAT_CAP_PQ);
+	if (dca_en && (device->cap & (IOAT_CAP_XOR|IOAT_CAP_PQ)))
+		device->cap &= ~(IOAT_CAP_XOR|IOAT_CAP_PQ);
 
-	if (cap & IOAT_CAP_XOR) {
+	if (device->cap & IOAT_CAP_XOR) {
 		is_raid_device = true;
 		dma->max_xor = 8;
 		dma->xor_align = 6;
@@ -1302,53 +1905,86 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
 		dma_cap_set(DMA_XOR_VAL, dma->cap_mask);
 		dma->device_prep_dma_xor_val = ioat3_prep_xor_val;
 	}
-	if (cap & IOAT_CAP_PQ) {
+
+	if (device->cap & IOAT_CAP_PQ) {
 		is_raid_device = true;
-		dma_set_maxpq(dma, 8, 0);
-		dma->pq_align = 6;
 
-		dma_cap_set(DMA_PQ, dma->cap_mask);
 		dma->device_prep_dma_pq = ioat3_prep_pq;
-
-		dma_cap_set(DMA_PQ_VAL, dma->cap_mask);
 		dma->device_prep_dma_pq_val = ioat3_prep_pq_val;
+		dma_cap_set(DMA_PQ, dma->cap_mask);
+		dma_cap_set(DMA_PQ_VAL, dma->cap_mask);
 
-		if (!(cap & IOAT_CAP_XOR)) {
-			dma->max_xor = 8;
-			dma->xor_align = 6;
+		if (device->cap & IOAT_CAP_RAID16SS) {
+			dma_set_maxpq(dma, 16, 0);
+			dma->pq_align = 0;
+		} else {
+			dma_set_maxpq(dma, 8, 0);
+			if (is_xeon_cb32(pdev))
+				dma->pq_align = 6;
+			else
+				dma->pq_align = 0;
+		}
 
-			dma_cap_set(DMA_XOR, dma->cap_mask);
+		if (!(device->cap & IOAT_CAP_XOR)) {
 			dma->device_prep_dma_xor = ioat3_prep_pqxor;
-
-			dma_cap_set(DMA_XOR_VAL, dma->cap_mask);
 			dma->device_prep_dma_xor_val = ioat3_prep_pqxor_val;
+			dma_cap_set(DMA_XOR, dma->cap_mask);
+			dma_cap_set(DMA_XOR_VAL, dma->cap_mask);
+
+			if (device->cap & IOAT_CAP_RAID16SS) {
+				dma->max_xor = 16;
+				dma->xor_align = 0;
+			} else {
+				dma->max_xor = 8;
+				if (is_xeon_cb32(pdev))
+					dma->xor_align = 6;
+				else
+					dma->xor_align = 0;
+			}
 		}
 	}
-	if (is_raid_device && (cap & IOAT_CAP_FILL_BLOCK)) {
+
+	if (is_raid_device && (device->cap & IOAT_CAP_FILL_BLOCK)) {
 		dma_cap_set(DMA_MEMSET, dma->cap_mask);
 		dma->device_prep_dma_memset = ioat3_prep_memset_lock;
 	}
 
 
-	if (is_raid_device) {
-		dma->device_tx_status = ioat3_tx_status;
-		device->cleanup_fn = ioat3_cleanup_event;
-		device->timer_fn = ioat3_timer_event;
-	} else {
-		dma->device_tx_status = ioat_dma_tx_status;
-		device->cleanup_fn = ioat2_cleanup_event;
-		device->timer_fn = ioat2_timer_event;
+	dma->device_tx_status = ioat3_tx_status;
+	device->cleanup_fn = ioat3_cleanup_event;
+	device->timer_fn = ioat3_timer_event;
+
+	if (is_xeon_cb32(pdev)) {
+		dma_cap_clear(DMA_XOR_VAL, dma->cap_mask);
+		dma->device_prep_dma_xor_val = NULL;
+
+		dma_cap_clear(DMA_PQ_VAL, dma->cap_mask);
+		dma->device_prep_dma_pq_val = NULL;
 	}
 
-	#ifdef CONFIG_ASYNC_TX_DISABLE_PQ_VAL_DMA
-	dma_cap_clear(DMA_PQ_VAL, dma->cap_mask);
-	dma->device_prep_dma_pq_val = NULL;
-	#endif
+	/* starting with CB3.3 super extended descriptors are supported */
+	if (device->cap & IOAT_CAP_RAID16SS) {
+		char pool_name[14];
+		int i;
+
+		/* allocate sw descriptor pool for SED */
+		device->sed_pool = kmem_cache_create("ioat_sed",
+				sizeof(struct ioat_sed_ent), 0, 0, NULL);
+		if (!device->sed_pool)
+			return -ENOMEM;
+
+		for (i = 0; i < MAX_SED_POOLS; i++) {
+			snprintf(pool_name, 14, "ioat_hw%d_sed", i);
 
-	#ifdef CONFIG_ASYNC_TX_DISABLE_XOR_VAL_DMA
-	dma_cap_clear(DMA_XOR_VAL, dma->cap_mask);
-	dma->device_prep_dma_xor_val = NULL;
-	#endif
+			/* allocate SED DMA pool */
+			device->sed_hw_pool[i] = dma_pool_create(pool_name,
+					&pdev->dev,
+					SED_SIZE * (i + 1), 64, 0);
+			if (!device->sed_hw_pool[i])
+				goto sed_pool_cleanup;
+
+		}
+	}
 
 	err = ioat_probe(device);
 	if (err)
@@ -1371,4 +2007,28 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
 		device->dca = ioat3_dca_init(pdev, device->reg_base);
 
 	return 0;
+
+sed_pool_cleanup:
+	if (device->sed_pool) {
+		int i;
+		kmem_cache_destroy(device->sed_pool);
+
+		for (i = 0; i < MAX_SED_POOLS; i++)
+			if (device->sed_hw_pool[i])
+				dma_pool_destroy(device->sed_hw_pool[i]);
+	}
+
+	return -ENOMEM;
+}
+
+void ioat3_dma_remove(struct ioatdma_device *device)
+{
+	if (device->sed_pool) {
+		int i;
+		kmem_cache_destroy(device->sed_pool);
+
+		for (i = 0; i < MAX_SED_POOLS; i++)
+			if (device->sed_hw_pool[i])
+				dma_pool_destroy(device->sed_hw_pool[i]);
+	}
 }
diff --git a/drivers/dma/ioat/hw.h b/drivers/dma/ioat/hw.h
index 7cb74c62c71..5ee57d402a6 100644
--- a/drivers/dma/ioat/hw.h
+++ b/drivers/dma/ioat/hw.h
@@ -30,11 +30,6 @@
 #define IOAT_PCI_DID_SCNB       0x65FF
 #define IOAT_PCI_DID_SNB        0x402F
 
-#define IOAT_VER_1_2            0x12    /* Version 1.2 */
-#define IOAT_VER_2_0            0x20    /* Version 2.0 */
-#define IOAT_VER_3_0            0x30    /* Version 3.0 */
-#define IOAT_VER_3_2            0x32    /* Version 3.2 */
-
 #define PCI_DEVICE_ID_INTEL_IOAT_IVB0	0x0e20
 #define PCI_DEVICE_ID_INTEL_IOAT_IVB1	0x0e21
 #define PCI_DEVICE_ID_INTEL_IOAT_IVB2	0x0e22
@@ -46,6 +41,29 @@
 #define PCI_DEVICE_ID_INTEL_IOAT_IVB8	0x0e2e
 #define PCI_DEVICE_ID_INTEL_IOAT_IVB9	0x0e2f
 
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW0	0x2f20
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW1	0x2f21
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW2	0x2f22
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW3	0x2f23
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW4	0x2f24
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW5	0x2f25
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW6	0x2f26
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW7	0x2f27
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW8	0x2f2e
+#define PCI_DEVICE_ID_INTEL_IOAT_HSW9	0x2f2f
+
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD0	0x0C50
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD1	0x0C51
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD2	0x0C52
+#define PCI_DEVICE_ID_INTEL_IOAT_BWD3	0x0C53
+
+#define IOAT_VER_1_2            0x12    /* Version 1.2 */
+#define IOAT_VER_2_0            0x20    /* Version 2.0 */
+#define IOAT_VER_3_0            0x30    /* Version 3.0 */
+#define IOAT_VER_3_2            0x32    /* Version 3.2 */
+#define IOAT_VER_3_3            0x33    /* Version 3.3 */
+
+
 int system_has_dca_enabled(struct pci_dev *pdev);
 
 struct ioat_dma_descriptor {
@@ -147,7 +165,17 @@ struct ioat_xor_ext_descriptor {
 };
 
 struct ioat_pq_descriptor {
-	uint32_t	size;
+	union {
+		uint32_t	size;
+		uint32_t	dwbes;
+		struct {
+			unsigned int rsvd:25;
+			unsigned int p_val_err:1;
+			unsigned int q_val_err:1;
+			unsigned int rsvd1:4;
+			unsigned int wbes:1;
+		} dwbes_f;
+	};
 	union {
 		uint32_t ctl;
 		struct {
@@ -162,9 +190,14 @@ struct ioat_pq_descriptor {
 			unsigned int hint:1;
 			unsigned int p_disable:1;
 			unsigned int q_disable:1;
-			unsigned int rsvd:11;
+			unsigned int rsvd2:2;
+			unsigned int wb_en:1;
+			unsigned int prl_en:1;
+			unsigned int rsvd3:7;
 			#define IOAT_OP_PQ 0x89
 			#define IOAT_OP_PQ_VAL 0x8a
+			#define IOAT_OP_PQ_16S 0xa0
+			#define IOAT_OP_PQ_VAL_16S 0xa1
 			unsigned int op:8;
 		} ctl_f;
 	};
@@ -172,7 +205,10 @@ struct ioat_pq_descriptor {
 	uint64_t	p_addr;
 	uint64_t	next;
 	uint64_t	src_addr2;
-	uint64_t	src_addr3;
+	union {
+		uint64_t	src_addr3;
+		uint64_t	sed_addr;
+	};
 	uint8_t		coef[8];
 	uint64_t	q_addr;
 };
@@ -221,4 +257,40 @@ struct ioat_pq_update_descriptor {
 struct ioat_raw_descriptor {
 	uint64_t	field[8];
 };
+
+struct ioat_pq16a_descriptor {
+	uint8_t coef[8];
+	uint64_t src_addr3;
+	uint64_t src_addr4;
+	uint64_t src_addr5;
+	uint64_t src_addr6;
+	uint64_t src_addr7;
+	uint64_t src_addr8;
+	uint64_t src_addr9;
+};
+
+struct ioat_pq16b_descriptor {
+	uint64_t src_addr10;
+	uint64_t src_addr11;
+	uint64_t src_addr12;
+	uint64_t src_addr13;
+	uint64_t src_addr14;
+	uint64_t src_addr15;
+	uint64_t src_addr16;
+	uint64_t rsvd;
+};
+
+union ioat_sed_pq_descriptor {
+	struct ioat_pq16a_descriptor a;
+	struct ioat_pq16b_descriptor b;
+};
+
+#define SED_SIZE	64
+
+struct ioat_sed_raw_descriptor {
+	uint64_t	a[8];
+	uint64_t	b[8];
+	uint64_t	c[8];
+};
+
 #endif
diff --git a/drivers/dma/ioat/pci.c b/drivers/dma/ioat/pci.c
index 71c7ecd80fa..2c8d560e633 100644
--- a/drivers/dma/ioat/pci.c
+++ b/drivers/dma/ioat/pci.c
@@ -94,6 +94,23 @@ static struct pci_device_id ioat_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_IVB8) },
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_IVB9) },
 
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW0) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW1) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW2) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW3) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW4) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW5) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW6) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW7) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW8) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_HSW9) },
+
+	/* I/OAT v3.3 platforms */
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD0) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD1) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD2) },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_BWD3) },
+
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, ioat_pci_tbl);
@@ -190,6 +207,9 @@ static void ioat_remove(struct pci_dev *pdev)
 	if (!device)
 		return;
 
+	if (device->version >= IOAT_VER_3_0)
+		ioat3_dma_remove(device);
+
 	dev_err(&pdev->dev, "Removing dma and dca services\n");
 	if (device->dca) {
 		unregister_dca_provider(device->dca, &pdev->dev);
diff --git a/drivers/dma/ioat/registers.h b/drivers/dma/ioat/registers.h
index 1391798542b..2f1cfa0f1f4 100644
--- a/drivers/dma/ioat/registers.h
+++ b/drivers/dma/ioat/registers.h
@@ -79,6 +79,8 @@
 #define IOAT_CAP_APIC				0x00000080
 #define IOAT_CAP_XOR				0x00000100
 #define IOAT_CAP_PQ				0x00000200
+#define IOAT_CAP_DWBES				0x00002000
+#define IOAT_CAP_RAID16SS			0x00020000
 
 #define IOAT_CHANNEL_MMIO_SIZE			0x80	/* Each Channel MMIO space is this size */
 
@@ -93,6 +95,8 @@
 #define IOAT_CHANCTRL_ERR_COMPLETION_EN		0x0004
 #define IOAT_CHANCTRL_INT_REARM			0x0001
 #define IOAT_CHANCTRL_RUN			(IOAT_CHANCTRL_INT_REARM |\
+						 IOAT_CHANCTRL_ERR_INT_EN |\
+						 IOAT_CHANCTRL_ERR_COMPLETION_EN |\
 						 IOAT_CHANCTRL_ANY_ERR_ABORT_EN)
 
 #define IOAT_DMA_COMP_OFFSET			0x02	/* 16-bit DMA channel compatibility */
diff --git a/drivers/dma/ipu/ipu_idmac.c b/drivers/dma/ipu/ipu_idmac.c
index 8c61d17a86b..d39c2cd0795 100644
--- a/drivers/dma/ipu/ipu_idmac.c
+++ b/drivers/dma/ipu/ipu_idmac.c
@@ -1642,7 +1642,7 @@ static int __init ipu_idmac_init(struct ipu *ipu)
 	return dma_async_device_register(&idmac->dma);
 }
 
-static void __exit ipu_idmac_exit(struct ipu *ipu)
+static void ipu_idmac_exit(struct ipu *ipu)
 {
 	int i;
 	struct idmac *idmac = &ipu->idmac;
@@ -1756,7 +1756,7 @@ err_noirq:
 	return ret;
 }
 
-static int __exit ipu_remove(struct platform_device *pdev)
+static int ipu_remove(struct platform_device *pdev)
 {
 	struct ipu *ipu = platform_get_drvdata(pdev);
 
@@ -1781,7 +1781,7 @@ static struct platform_driver ipu_platform_driver = {
 		.name	= "ipu-core",
 		.owner	= THIS_MODULE,
 	},
-	.remove		= __exit_p(ipu_remove),
+	.remove		= ipu_remove,
 };
 
 static int __init ipu_init(void)
diff --git a/drivers/dma/of-dma.c b/drivers/dma/of-dma.c
index 69d04d28b1e..7aa0864cd48 100644
--- a/drivers/dma/of-dma.c
+++ b/drivers/dma/of-dma.c
@@ -13,43 +13,31 @@
 #include <linux/device.h>
 #include <linux/err.h>
 #include <linux/module.h>
-#include <linux/rculist.h>
+#include <linux/mutex.h>
 #include <linux/slab.h>
 #include <linux/of.h>
 #include <linux/of_dma.h>
 
 static LIST_HEAD(of_dma_list);
-static DEFINE_SPINLOCK(of_dma_lock);
+static DEFINE_MUTEX(of_dma_lock);
 
 /**
- * of_dma_get_controller - Get a DMA controller in DT DMA helpers list
+ * of_dma_find_controller - Get a DMA controller in DT DMA helpers list
  * @dma_spec:	pointer to DMA specifier as found in the device tree
  *
  * Finds a DMA controller with matching device node and number for dma cells
- * in a list of registered DMA controllers. If a match is found the use_count
- * variable is increased and a valid pointer to the DMA data stored is retuned.
- * A NULL pointer is returned if no match is found.
+ * in a list of registered DMA controllers. If a match is found a valid pointer
+ * to the DMA data stored is retuned. A NULL pointer is returned if no match is
+ * found.
  */
-static struct of_dma *of_dma_get_controller(struct of_phandle_args *dma_spec)
+static struct of_dma *of_dma_find_controller(struct of_phandle_args *dma_spec)
 {
 	struct of_dma *ofdma;
 
-	spin_lock(&of_dma_lock);
-
-	if (list_empty(&of_dma_list)) {
-		spin_unlock(&of_dma_lock);
-		return NULL;
-	}
-
 	list_for_each_entry(ofdma, &of_dma_list, of_dma_controllers)
 		if ((ofdma->of_node == dma_spec->np) &&
-		    (ofdma->of_dma_nbcells == dma_spec->args_count)) {
-			ofdma->use_count++;
-			spin_unlock(&of_dma_lock);
+		    (ofdma->of_dma_nbcells == dma_spec->args_count))
 			return ofdma;
-		}
-
-	spin_unlock(&of_dma_lock);
 
 	pr_debug("%s: can't find DMA controller %s\n", __func__,
 		 dma_spec->np->full_name);
@@ -58,22 +46,6 @@ static struct of_dma *of_dma_get_controller(struct of_phandle_args *dma_spec)
 }
 
 /**
- * of_dma_put_controller - Decrement use count for a registered DMA controller
- * @of_dma:	pointer to DMA controller data
- *
- * Decrements the use_count variable in the DMA data structure. This function
- * should be called only when a valid pointer is returned from
- * of_dma_get_controller() and no further accesses to data referenced by that
- * pointer are needed.
- */
-static void of_dma_put_controller(struct of_dma *ofdma)
-{
-	spin_lock(&of_dma_lock);
-	ofdma->use_count--;
-	spin_unlock(&of_dma_lock);
-}
-
-/**
  * of_dma_controller_register - Register a DMA controller to DT DMA helpers
  * @np:			device node of DMA controller
  * @of_dma_xlate:	translation function which converts a phandle
@@ -93,6 +65,7 @@ int of_dma_controller_register(struct device_node *np,
 {
 	struct of_dma	*ofdma;
 	int		nbcells;
+	const __be32	*prop;
 
 	if (!np || !of_dma_xlate) {
 		pr_err("%s: not enough information provided\n", __func__);
@@ -103,8 +76,11 @@ int of_dma_controller_register(struct device_node *np,
 	if (!ofdma)
 		return -ENOMEM;
 
-	nbcells = be32_to_cpup(of_get_property(np, "#dma-cells", NULL));
-	if (!nbcells) {
+	prop = of_get_property(np, "#dma-cells", NULL);
+	if (prop)
+		nbcells = be32_to_cpup(prop);
+
+	if (!prop || !nbcells) {
 		pr_err("%s: #dma-cells property is missing or invalid\n",
 		       __func__);
 		kfree(ofdma);
@@ -115,12 +91,11 @@ int of_dma_controller_register(struct device_node *np,
 	ofdma->of_dma_nbcells = nbcells;
 	ofdma->of_dma_xlate = of_dma_xlate;
 	ofdma->of_dma_data = data;
-	ofdma->use_count = 0;
 
 	/* Now queue of_dma controller structure in list */
-	spin_lock(&of_dma_lock);
+	mutex_lock(&of_dma_lock);
 	list_add_tail(&ofdma->of_dma_controllers, &of_dma_list);
-	spin_unlock(&of_dma_lock);
+	mutex_unlock(&of_dma_lock);
 
 	return 0;
 }
@@ -132,32 +107,20 @@ EXPORT_SYMBOL_GPL(of_dma_controller_register);
  *
  * Memory allocated by of_dma_controller_register() is freed here.
  */
-int of_dma_controller_free(struct device_node *np)
+void of_dma_controller_free(struct device_node *np)
 {
 	struct of_dma *ofdma;
 
-	spin_lock(&of_dma_lock);
-
-	if (list_empty(&of_dma_list)) {
-		spin_unlock(&of_dma_lock);
-		return -ENODEV;
-	}
+	mutex_lock(&of_dma_lock);
 
 	list_for_each_entry(ofdma, &of_dma_list, of_dma_controllers)
 		if (ofdma->of_node == np) {
-			if (ofdma->use_count) {
-				spin_unlock(&of_dma_lock);
-				return -EBUSY;
-			}
-
 			list_del(&ofdma->of_dma_controllers);
-			spin_unlock(&of_dma_lock);
 			kfree(ofdma);
-			return 0;
+			break;
 		}
 
-	spin_unlock(&of_dma_lock);
-	return -ENODEV;
+	mutex_unlock(&of_dma_lock);
 }
 EXPORT_SYMBOL_GPL(of_dma_controller_free);
 
@@ -172,8 +135,8 @@ EXPORT_SYMBOL_GPL(of_dma_controller_free);
  * specifiers, matches the name provided. Returns 0 if the name matches and
  * a valid pointer to the DMA specifier is found. Otherwise returns -ENODEV.
  */
-static int of_dma_match_channel(struct device_node *np, char *name, int index,
-				struct of_phandle_args *dma_spec)
+static int of_dma_match_channel(struct device_node *np, const char *name,
+				int index, struct of_phandle_args *dma_spec)
 {
 	const char *s;
 
@@ -198,7 +161,7 @@ static int of_dma_match_channel(struct device_node *np, char *name, int index,
  * Returns pointer to appropriate dma channel on success or NULL on error.
  */
 struct dma_chan *of_dma_request_slave_channel(struct device_node *np,
-					      char *name)
+					      const char *name)
 {
 	struct of_phandle_args	dma_spec;
 	struct of_dma		*ofdma;
@@ -220,14 +183,15 @@ struct dma_chan *of_dma_request_slave_channel(struct device_node *np,
 		if (of_dma_match_channel(np, name, i, &dma_spec))
 			continue;
 
-		ofdma = of_dma_get_controller(&dma_spec);
-
-		if (!ofdma)
-			continue;
+		mutex_lock(&of_dma_lock);
+		ofdma = of_dma_find_controller(&dma_spec);
 
-		chan = ofdma->of_dma_xlate(&dma_spec, ofdma);
+		if (ofdma)
+			chan = ofdma->of_dma_xlate(&dma_spec, ofdma);
+		else
+			chan = NULL;
 
-		of_dma_put_controller(ofdma);
+		mutex_unlock(&of_dma_lock);
 
 		of_node_put(dma_spec.np);
 
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index 08b43bf3715..ec3fc4fd916 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -16,6 +16,8 @@
 #include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
+#include <linux/of_dma.h>
+#include <linux/of_device.h>
 
 #include "virt-dma.h"
 
@@ -67,6 +69,10 @@ static const unsigned es_bytes[] = {
 	[OMAP_DMA_DATA_TYPE_S32] = 4,
 };
 
+static struct of_dma_filter_info omap_dma_info = {
+	.filter_fn = omap_dma_filter_fn,
+};
+
 static inline struct omap_dmadev *to_omap_dma_dev(struct dma_device *d)
 {
 	return container_of(d, struct omap_dmadev, ddev);
@@ -629,8 +635,22 @@ static int omap_dma_probe(struct platform_device *pdev)
 		pr_warn("OMAP-DMA: failed to register slave DMA engine device: %d\n",
 			rc);
 		omap_dma_free(od);
-	} else {
-		platform_set_drvdata(pdev, od);
+		return rc;
+	}
+
+	platform_set_drvdata(pdev, od);
+
+	if (pdev->dev.of_node) {
+		omap_dma_info.dma_cap = od->ddev.cap_mask;
+
+		/* Device-tree DMA controller registration */
+		rc = of_dma_controller_register(pdev->dev.of_node,
+				of_dma_simple_xlate, &omap_dma_info);
+		if (rc) {
+			pr_warn("OMAP-DMA: failed to register DMA controller\n");
+			dma_async_device_unregister(&od->ddev);
+			omap_dma_free(od);
+		}
 	}
 
 	dev_info(&pdev->dev, "OMAP DMA engine driver\n");
@@ -642,18 +662,32 @@ static int omap_dma_remove(struct platform_device *pdev)
 {
 	struct omap_dmadev *od = platform_get_drvdata(pdev);
 
+	if (pdev->dev.of_node)
+		of_dma_controller_free(pdev->dev.of_node);
+
 	dma_async_device_unregister(&od->ddev);
 	omap_dma_free(od);
 
 	return 0;
 }
 
+static const struct of_device_id omap_dma_match[] = {
+	{ .compatible = "ti,omap2420-sdma", },
+	{ .compatible = "ti,omap2430-sdma", },
+	{ .compatible = "ti,omap3430-sdma", },
+	{ .compatible = "ti,omap3630-sdma", },
+	{ .compatible = "ti,omap4430-sdma", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, omap_dma_match);
+
 static struct platform_driver omap_dma_driver = {
 	.probe	= omap_dma_probe,
 	.remove	= omap_dma_remove,
 	.driver = {
 		.name = "omap-dma-engine",
 		.owner = THIS_MODULE,
+		.of_match_table = of_match_ptr(omap_dma_match),
 	},
 };
 
diff --git a/drivers/dma/pch_dma.c b/drivers/dma/pch_dma.c
index d01faeb0f27..ce3dc3e9688 100644
--- a/drivers/dma/pch_dma.c
+++ b/drivers/dma/pch_dma.c
@@ -476,7 +476,7 @@ static struct pch_dma_desc *pdc_desc_get(struct pch_dma_chan *pd_chan)
 	dev_dbg(chan2dev(&pd_chan->chan), "scanned %d descriptors\n", i);
 
 	if (!ret) {
-		ret = pdc_alloc_desc(&pd_chan->chan, GFP_NOIO);
+		ret = pdc_alloc_desc(&pd_chan->chan, GFP_ATOMIC);
 		if (ret) {
 			spin_lock(&pd_chan->lock);
 			pd_chan->descs_allocated++;
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 5dbc5946c4c..a17553f7c02 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -26,6 +26,7 @@
 #include <linux/scatterlist.h>
 #include <linux/of.h>
 #include <linux/of_dma.h>
+#include <linux/err.h>
 
 #include "dmaengine.h"
 #define PL330_MAX_CHAN		8
@@ -2288,13 +2289,12 @@ static inline void fill_queue(struct dma_pl330_chan *pch)
 
 		/* If already submitted */
 		if (desc->status == BUSY)
-			break;
+			continue;
 
 		ret = pl330_submit_req(pch->pl330_chid,
 						&desc->req);
 		if (!ret) {
 			desc->status = BUSY;
-			break;
 		} else if (ret == -EAGAIN) {
 			/* QFull or DMAC Dying */
 			break;
@@ -2904,9 +2904,9 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 	pi->mcbufsz = pdat ? pdat->mcbuf_sz : 0;
 
 	res = &adev->res;
-	pi->base = devm_request_and_ioremap(&adev->dev, res);
-	if (!pi->base)
-		return -ENXIO;
+	pi->base = devm_ioremap_resource(&adev->dev, res);
+	if (IS_ERR(pi->base))
+		return PTR_ERR(pi->base);
 
 	amba_set_drvdata(adev, pdmac);
 
diff --git a/drivers/dma/sh/Kconfig b/drivers/dma/sh/Kconfig
new file mode 100644
index 00000000000..5c1dee20c13
--- /dev/null
+++ b/drivers/dma/sh/Kconfig
@@ -0,0 +1,24 @@
+#
+# DMA engine configuration for sh
+#
+
+config SH_DMAE_BASE
+	bool "Renesas SuperH DMA Engine support"
+	depends on (SUPERH && SH_DMA) || (ARM && ARCH_SHMOBILE)
+	depends on !SH_DMA_API
+	default y
+	select DMA_ENGINE
+	help
+	  Enable support for the Renesas SuperH DMA controllers.
+
+config SH_DMAE
+	tristate "Renesas SuperH DMAC support"
+	depends on SH_DMAE_BASE
+	help
+	  Enable support for the Renesas SuperH DMA controllers.
+
+config SUDMAC
+	tristate "Renesas SUDMAC support"
+	depends on SH_DMAE_BASE
+	help
+	  Enable support for the Renesas SUDMAC controllers.
diff --git a/drivers/dma/sh/Makefile b/drivers/dma/sh/Makefile
index 54ae9572b0a..c07ca4612e4 100644
--- a/drivers/dma/sh/Makefile
+++ b/drivers/dma/sh/Makefile
@@ -1,2 +1,3 @@
-obj-$(CONFIG_SH_DMAE) += shdma-base.o
+obj-$(CONFIG_SH_DMAE_BASE) += shdma-base.o
 obj-$(CONFIG_SH_DMAE) += shdma.o
+obj-$(CONFIG_SUDMAC) += sudmac.o
diff --git a/drivers/dma/sh/sudmac.c b/drivers/dma/sh/sudmac.c
new file mode 100644
index 00000000000..e7c94bbddb5
--- /dev/null
+++ b/drivers/dma/sh/sudmac.c
@@ -0,0 +1,428 @@
+/*
+ * Renesas SUDMAC support
+ *
+ * Copyright (C) 2013 Renesas Solutions Corp.
+ *
+ * based on drivers/dma/sh/shdma.c:
+ * Copyright (C) 2011-2012 Guennadi Liakhovetski <g.liakhovetski@gmx.de>
+ * Copyright (C) 2009 Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
+ * Copyright (C) 2009 Renesas Solutions, Inc. All rights reserved.
+ * Copyright (C) 2007 Freescale Semiconductor, Inc. All rights reserved.
+ *
+ * This is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/dmaengine.h>
+#include <linux/platform_device.h>
+#include <linux/sudmac.h>
+
+struct sudmac_chan {
+	struct shdma_chan shdma_chan;
+	void __iomem *base;
+	char dev_id[16];	/* unique name per DMAC of channel */
+
+	u32 offset;		/* for CFG, BA, BBC, CA, CBC, DEN */
+	u32 cfg;
+	u32 dint_end_bit;
+};
+
+struct sudmac_device {
+	struct shdma_dev shdma_dev;
+	struct sudmac_pdata *pdata;
+	void __iomem *chan_reg;
+};
+
+struct sudmac_regs {
+	u32 base_addr;
+	u32 base_byte_count;
+};
+
+struct sudmac_desc {
+	struct sudmac_regs hw;
+	struct shdma_desc shdma_desc;
+};
+
+#define to_chan(schan) container_of(schan, struct sudmac_chan, shdma_chan)
+#define to_desc(sdesc) container_of(sdesc, struct sudmac_desc, shdma_desc)
+#define to_sdev(sc) container_of(sc->shdma_chan.dma_chan.device, \
+				 struct sudmac_device, shdma_dev.dma_dev)
+
+/* SUDMAC register */
+#define SUDMAC_CH0CFG		0x00
+#define SUDMAC_CH0BA		0x10
+#define SUDMAC_CH0BBC		0x18
+#define SUDMAC_CH0CA		0x20
+#define SUDMAC_CH0CBC		0x28
+#define SUDMAC_CH0DEN		0x30
+#define SUDMAC_DSTSCLR		0x38
+#define SUDMAC_DBUFCTRL		0x3C
+#define SUDMAC_DINTCTRL		0x40
+#define SUDMAC_DINTSTS		0x44
+#define SUDMAC_DINTSTSCLR	0x48
+#define SUDMAC_CH0SHCTRL	0x50
+
+/* Definitions for the sudmac_channel.config */
+#define SUDMAC_SENDBUFM	0x1000 /* b12: Transmit Buffer Mode */
+#define SUDMAC_RCVENDM	0x0100 /* b8: Receive Data Transfer End Mode */
+#define SUDMAC_LBA_WAIT	0x0030 /* b5-4: Local Bus Access Wait */
+
+/* Definitions for the sudmac_channel.dint_end_bit */
+#define SUDMAC_CH1ENDE	0x0002 /* b1: Ch1 DMA Transfer End Int Enable */
+#define SUDMAC_CH0ENDE	0x0001 /* b0: Ch0 DMA Transfer End Int Enable */
+
+#define SUDMAC_DRV_NAME "sudmac"
+
+static void sudmac_writel(struct sudmac_chan *sc, u32 data, u32 reg)
+{
+	iowrite32(data, sc->base + reg);
+}
+
+static u32 sudmac_readl(struct sudmac_chan *sc, u32 reg)
+{
+	return ioread32(sc->base + reg);
+}
+
+static bool sudmac_is_busy(struct sudmac_chan *sc)
+{
+	u32 den = sudmac_readl(sc, SUDMAC_CH0DEN + sc->offset);
+
+	if (den)
+		return true; /* working */
+
+	return false; /* waiting */
+}
+
+static void sudmac_set_reg(struct sudmac_chan *sc, struct sudmac_regs *hw,
+			   struct shdma_desc *sdesc)
+{
+	sudmac_writel(sc, sc->cfg, SUDMAC_CH0CFG + sc->offset);
+	sudmac_writel(sc, hw->base_addr, SUDMAC_CH0BA + sc->offset);
+	sudmac_writel(sc, hw->base_byte_count, SUDMAC_CH0BBC + sc->offset);
+}
+
+static void sudmac_start(struct sudmac_chan *sc)
+{
+	u32 dintctrl = sudmac_readl(sc, SUDMAC_DINTCTRL);
+
+	sudmac_writel(sc, dintctrl | sc->dint_end_bit, SUDMAC_DINTCTRL);
+	sudmac_writel(sc, 1, SUDMAC_CH0DEN + sc->offset);
+}
+
+static void sudmac_start_xfer(struct shdma_chan *schan,
+			      struct shdma_desc *sdesc)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+	struct sudmac_desc *sd = to_desc(sdesc);
+
+	sudmac_set_reg(sc, &sd->hw, sdesc);
+	sudmac_start(sc);
+}
+
+static bool sudmac_channel_busy(struct shdma_chan *schan)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+
+	return sudmac_is_busy(sc);
+}
+
+static void sudmac_setup_xfer(struct shdma_chan *schan, int slave_id)
+{
+}
+
+static const struct sudmac_slave_config *sudmac_find_slave(
+	struct sudmac_chan *sc, int slave_id)
+{
+	struct sudmac_device *sdev = to_sdev(sc);
+	struct sudmac_pdata *pdata = sdev->pdata;
+	const struct sudmac_slave_config *cfg;
+	int i;
+
+	for (i = 0, cfg = pdata->slave; i < pdata->slave_num; i++, cfg++)
+		if (cfg->slave_id == slave_id)
+			return cfg;
+
+	return NULL;
+}
+
+static int sudmac_set_slave(struct shdma_chan *schan, int slave_id, bool try)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+	const struct sudmac_slave_config *cfg = sudmac_find_slave(sc, slave_id);
+
+	if (!cfg)
+		return -ENODEV;
+
+	return 0;
+}
+
+static inline void sudmac_dma_halt(struct sudmac_chan *sc)
+{
+	u32 dintctrl = sudmac_readl(sc, SUDMAC_DINTCTRL);
+
+	sudmac_writel(sc, 0, SUDMAC_CH0DEN + sc->offset);
+	sudmac_writel(sc, dintctrl & ~sc->dint_end_bit, SUDMAC_DINTCTRL);
+	sudmac_writel(sc, sc->dint_end_bit, SUDMAC_DINTSTSCLR);
+}
+
+static int sudmac_desc_setup(struct shdma_chan *schan,
+			     struct shdma_desc *sdesc,
+			     dma_addr_t src, dma_addr_t dst, size_t *len)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+	struct sudmac_desc *sd = to_desc(sdesc);
+
+	dev_dbg(sc->shdma_chan.dev, "%s: src=%x, dst=%x, len=%d\n",
+		__func__, src, dst, *len);
+
+	if (*len > schan->max_xfer_len)
+		*len = schan->max_xfer_len;
+
+	if (dst)
+		sd->hw.base_addr = dst;
+	else if (src)
+		sd->hw.base_addr = src;
+	sd->hw.base_byte_count = *len;
+
+	return 0;
+}
+
+static void sudmac_halt(struct shdma_chan *schan)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+
+	sudmac_dma_halt(sc);
+}
+
+static bool sudmac_chan_irq(struct shdma_chan *schan, int irq)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+	u32 dintsts = sudmac_readl(sc, SUDMAC_DINTSTS);
+
+	if (!(dintsts & sc->dint_end_bit))
+		return false;
+
+	/* DMA stop */
+	sudmac_dma_halt(sc);
+
+	return true;
+}
+
+static size_t sudmac_get_partial(struct shdma_chan *schan,
+				 struct shdma_desc *sdesc)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+	struct sudmac_desc *sd = to_desc(sdesc);
+	u32 current_byte_count = sudmac_readl(sc, SUDMAC_CH0CBC + sc->offset);
+
+	return sd->hw.base_byte_count - current_byte_count;
+}
+
+static bool sudmac_desc_completed(struct shdma_chan *schan,
+				  struct shdma_desc *sdesc)
+{
+	struct sudmac_chan *sc = to_chan(schan);
+	struct sudmac_desc *sd = to_desc(sdesc);
+	u32 current_addr = sudmac_readl(sc, SUDMAC_CH0CA + sc->offset);
+
+	return sd->hw.base_addr + sd->hw.base_byte_count == current_addr;
+}
+
+static int sudmac_chan_probe(struct sudmac_device *su_dev, int id, int irq,
+			     unsigned long flags)
+{
+	struct shdma_dev *sdev = &su_dev->shdma_dev;
+	struct platform_device *pdev = to_platform_device(sdev->dma_dev.dev);
+	struct sudmac_chan *sc;
+	struct shdma_chan *schan;
+	int err;
+
+	sc = devm_kzalloc(&pdev->dev, sizeof(struct sudmac_chan), GFP_KERNEL);
+	if (!sc) {
+		dev_err(sdev->dma_dev.dev,
+			"No free memory for allocating dma channels!\n");
+		return -ENOMEM;
+	}
+
+	schan = &sc->shdma_chan;
+	schan->max_xfer_len = 64 * 1024 * 1024 - 1;
+
+	shdma_chan_probe(sdev, schan, id);
+
+	sc->base = su_dev->chan_reg;
+
+	/* get platform_data */
+	sc->offset = su_dev->pdata->channel->offset;
+	if (su_dev->pdata->channel->config & SUDMAC_TX_BUFFER_MODE)
+		sc->cfg |= SUDMAC_SENDBUFM;
+	if (su_dev->pdata->channel->config & SUDMAC_RX_END_MODE)
+		sc->cfg |= SUDMAC_RCVENDM;
+	sc->cfg |= (su_dev->pdata->channel->wait << 4) & SUDMAC_LBA_WAIT;
+
+	if (su_dev->pdata->channel->dint_end_bit & SUDMAC_DMA_BIT_CH0)
+		sc->dint_end_bit |= SUDMAC_CH0ENDE;
+	if (su_dev->pdata->channel->dint_end_bit & SUDMAC_DMA_BIT_CH1)
+		sc->dint_end_bit |= SUDMAC_CH1ENDE;
+
+	/* set up channel irq */
+	if (pdev->id >= 0)
+		snprintf(sc->dev_id, sizeof(sc->dev_id), "sudmac%d.%d",
+			 pdev->id, id);
+	else
+		snprintf(sc->dev_id, sizeof(sc->dev_id), "sudmac%d", id);
+
+	err = shdma_request_irq(schan, irq, flags, sc->dev_id);
+	if (err) {
+		dev_err(sdev->dma_dev.dev,
+			"DMA channel %d request_irq failed %d\n", id, err);
+		goto err_no_irq;
+	}
+
+	return 0;
+
+err_no_irq:
+	/* remove from dmaengine device node */
+	shdma_chan_remove(schan);
+	return err;
+}
+
+static void sudmac_chan_remove(struct sudmac_device *su_dev)
+{
+	struct dma_device *dma_dev = &su_dev->shdma_dev.dma_dev;
+	struct shdma_chan *schan;
+	int i;
+
+	shdma_for_each_chan(schan, &su_dev->shdma_dev, i) {
+		struct sudmac_chan *sc = to_chan(schan);
+
+		BUG_ON(!schan);
+
+		shdma_free_irq(&sc->shdma_chan);
+		shdma_chan_remove(schan);
+	}
+	dma_dev->chancnt = 0;
+}
+
+static dma_addr_t sudmac_slave_addr(struct shdma_chan *schan)
+{
+	/* SUDMAC doesn't need the address */
+	return 0;
+}
+
+static struct shdma_desc *sudmac_embedded_desc(void *buf, int i)
+{
+	return &((struct sudmac_desc *)buf)[i].shdma_desc;
+}
+
+static const struct shdma_ops sudmac_shdma_ops = {
+	.desc_completed = sudmac_desc_completed,
+	.halt_channel = sudmac_halt,
+	.channel_busy = sudmac_channel_busy,
+	.slave_addr = sudmac_slave_addr,
+	.desc_setup = sudmac_desc_setup,
+	.set_slave = sudmac_set_slave,
+	.setup_xfer = sudmac_setup_xfer,
+	.start_xfer = sudmac_start_xfer,
+	.embedded_desc = sudmac_embedded_desc,
+	.chan_irq = sudmac_chan_irq,
+	.get_partial = sudmac_get_partial,
+};
+
+static int sudmac_probe(struct platform_device *pdev)
+{
+	struct sudmac_pdata *pdata = pdev->dev.platform_data;
+	int err, i;
+	struct sudmac_device *su_dev;
+	struct dma_device *dma_dev;
+	struct resource *chan, *irq_res;
+
+	/* get platform data */
+	if (!pdata)
+		return -ENODEV;
+
+	chan = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	irq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!chan || !irq_res)
+		return -ENODEV;
+
+	err = -ENOMEM;
+	su_dev = devm_kzalloc(&pdev->dev, sizeof(struct sudmac_device),
+			      GFP_KERNEL);
+	if (!su_dev) {
+		dev_err(&pdev->dev, "Not enough memory\n");
+		return err;
+	}
+
+	dma_dev = &su_dev->shdma_dev.dma_dev;
+
+	su_dev->chan_reg = devm_request_and_ioremap(&pdev->dev, chan);
+	if (!su_dev->chan_reg)
+		return err;
+
+	dma_cap_set(DMA_SLAVE, dma_dev->cap_mask);
+
+	su_dev->shdma_dev.ops = &sudmac_shdma_ops;
+	su_dev->shdma_dev.desc_size = sizeof(struct sudmac_desc);
+	err = shdma_init(&pdev->dev, &su_dev->shdma_dev, pdata->channel_num);
+	if (err < 0)
+		return err;
+
+	/* platform data */
+	su_dev->pdata = pdev->dev.platform_data;
+
+	platform_set_drvdata(pdev, su_dev);
+
+	/* Create DMA Channel */
+	for (i = 0; i < pdata->channel_num; i++) {
+		err = sudmac_chan_probe(su_dev, i, irq_res->start, IRQF_SHARED);
+		if (err)
+			goto chan_probe_err;
+	}
+
+	err = dma_async_device_register(&su_dev->shdma_dev.dma_dev);
+	if (err < 0)
+		goto chan_probe_err;
+
+	return err;
+
+chan_probe_err:
+	sudmac_chan_remove(su_dev);
+
+	platform_set_drvdata(pdev, NULL);
+	shdma_cleanup(&su_dev->shdma_dev);
+
+	return err;
+}
+
+static int sudmac_remove(struct platform_device *pdev)
+{
+	struct sudmac_device *su_dev = platform_get_drvdata(pdev);
+	struct dma_device *dma_dev = &su_dev->shdma_dev.dma_dev;
+
+	dma_async_device_unregister(dma_dev);
+	sudmac_chan_remove(su_dev);
+	shdma_cleanup(&su_dev->shdma_dev);
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+static struct platform_driver sudmac_driver = {
+	.driver		= {
+		.owner	= THIS_MODULE,
+		.name	= SUDMAC_DRV_NAME,
+	},
+	.probe		= sudmac_probe,
+	.remove		= sudmac_remove,
+};
+module_platform_driver(sudmac_driver);
+
+MODULE_AUTHOR("Yoshihiro Shimoda");
+MODULE_DESCRIPTION("Renesas SUDMAC driver");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:" SUDMAC_DRV_NAME);
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index 1d627e2391f..1765a0a2736 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -16,6 +16,7 @@
 #include <linux/of_address.h>
 #include <linux/of_device.h>
 #include <linux/of_platform.h>
+#include <linux/clk.h>
 #include <linux/sirfsoc_dma.h>
 
 #include "dmaengine.h"
@@ -78,6 +79,7 @@ struct sirfsoc_dma {
 	struct sirfsoc_dma_chan		channels[SIRFSOC_DMA_CHANNELS];
 	void __iomem			*base;
 	int				irq;
+	struct clk			*clk;
 	bool				is_marco;
 };
 
@@ -639,6 +641,12 @@ static int sirfsoc_dma_probe(struct platform_device *op)
 		return -EINVAL;
 	}
 
+	sdma->clk = devm_clk_get(dev, NULL);
+	if (IS_ERR(sdma->clk)) {
+		dev_err(dev, "failed to get a clock.\n");
+		return PTR_ERR(sdma->clk);
+	}
+
 	ret = of_address_to_resource(dn, 0, &res);
 	if (ret) {
 		dev_err(dev, "Error parsing memory region!\n");
@@ -698,6 +706,8 @@ static int sirfsoc_dma_probe(struct platform_device *op)
 
 	tasklet_init(&sdma->tasklet, sirfsoc_dma_tasklet, (unsigned long)sdma);
 
+	clk_prepare_enable(sdma->clk);
+
 	/* Register DMA engine */
 	dev_set_drvdata(dev, sdma);
 	ret = dma_async_device_register(dma);
@@ -720,6 +730,7 @@ static int sirfsoc_dma_remove(struct platform_device *op)
 	struct device *dev = &op->dev;
 	struct sirfsoc_dma *sdma = dev_get_drvdata(dev);
 
+	clk_disable_unprepare(sdma->clk);
 	dma_async_device_unregister(&sdma->dma);
 	free_irq(sdma->irq, sdma);
 	irq_dispose_mapping(sdma->irq);
@@ -742,7 +753,18 @@ static struct platform_driver sirfsoc_dma_driver = {
 	},
 };
 
-module_platform_driver(sirfsoc_dma_driver);
+static __init int sirfsoc_dma_init(void)
+{
+	return platform_driver_register(&sirfsoc_dma_driver);
+}
+
+static void __exit sirfsoc_dma_exit(void)
+{
+	platform_driver_unregister(&sirfsoc_dma_driver);
+}
+
+subsys_initcall(sirfsoc_dma_init);
+module_exit(sirfsoc_dma_exit);
 
 MODULE_AUTHOR("Rongjun Ying <rongjun.ying@csr.com>, "
 	"Barry Song <baohua.song@csr.com>");
diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index fcee27eae1f..ce193409ebd 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c
@@ -30,6 +30,7 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/platform_device.h>
+#include <linux/pm.h>
 #include <linux/pm_runtime.h>
 #include <linux/slab.h>
 #include <linux/clk/tegra.h>
@@ -199,6 +200,7 @@ struct tegra_dma_channel {
 
 	/* Channel-slave specific configuration */
 	struct dma_slave_config dma_sconfig;
+	struct tegra_dma_channel_regs	channel_reg;
 };
 
 /* tegra_dma: Tegra DMA specific information */
@@ -1213,7 +1215,6 @@ static const struct tegra_dma_chip_data tegra20_dma_chip_data = {
 	.support_channel_pause	= false,
 };
 
-#if defined(CONFIG_OF)
 /* Tegra30 specific DMA controller information */
 static const struct tegra_dma_chip_data tegra30_dma_chip_data = {
 	.nr_channels		= 32,
@@ -1243,7 +1244,6 @@ static const struct of_device_id tegra_dma_of_match[] = {
 	},
 };
 MODULE_DEVICE_TABLE(of, tegra_dma_of_match);
-#endif
 
 static int tegra_dma_probe(struct platform_device *pdev)
 {
@@ -1252,20 +1252,14 @@ static int tegra_dma_probe(struct platform_device *pdev)
 	int ret;
 	int i;
 	const struct tegra_dma_chip_data *cdata = NULL;
+	const struct of_device_id *match;
 
-	if (pdev->dev.of_node) {
-		const struct of_device_id *match;
-		match = of_match_device(of_match_ptr(tegra_dma_of_match),
-					&pdev->dev);
-		if (!match) {
-			dev_err(&pdev->dev, "Error: No device match found\n");
-			return -ENODEV;
-		}
-		cdata = match->data;
-	} else {
-		/* If no device tree then fallback to tegra20 */
-		cdata = &tegra20_dma_chip_data;
+	match = of_match_device(tegra_dma_of_match, &pdev->dev);
+	if (!match) {
+		dev_err(&pdev->dev, "Error: No device match found\n");
+		return -ENODEV;
 	}
+	cdata = match->data;
 
 	tdma = devm_kzalloc(&pdev->dev, sizeof(*tdma) + cdata->nr_channels *
 			sizeof(struct tegra_dma_channel), GFP_KERNEL);
@@ -1448,11 +1442,74 @@ static int tegra_dma_runtime_resume(struct device *dev)
 	return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int tegra_dma_pm_suspend(struct device *dev)
+{
+	struct tegra_dma *tdma = dev_get_drvdata(dev);
+	int i;
+	int ret;
+
+	/* Enable clock before accessing register */
+	ret = tegra_dma_runtime_resume(dev);
+	if (ret < 0)
+		return ret;
+
+	tdma->reg_gen = tdma_read(tdma, TEGRA_APBDMA_GENERAL);
+	for (i = 0; i < tdma->chip_data->nr_channels; i++) {
+		struct tegra_dma_channel *tdc = &tdma->channels[i];
+		struct tegra_dma_channel_regs *ch_reg = &tdc->channel_reg;
+
+		ch_reg->csr = tdc_read(tdc, TEGRA_APBDMA_CHAN_CSR);
+		ch_reg->ahb_ptr = tdc_read(tdc, TEGRA_APBDMA_CHAN_AHBPTR);
+		ch_reg->apb_ptr = tdc_read(tdc, TEGRA_APBDMA_CHAN_APBPTR);
+		ch_reg->ahb_seq = tdc_read(tdc, TEGRA_APBDMA_CHAN_AHBSEQ);
+		ch_reg->apb_seq = tdc_read(tdc, TEGRA_APBDMA_CHAN_APBSEQ);
+	}
+
+	/* Disable clock */
+	tegra_dma_runtime_suspend(dev);
+	return 0;
+}
+
+static int tegra_dma_pm_resume(struct device *dev)
+{
+	struct tegra_dma *tdma = dev_get_drvdata(dev);
+	int i;
+	int ret;
+
+	/* Enable clock before accessing register */
+	ret = tegra_dma_runtime_resume(dev);
+	if (ret < 0)
+		return ret;
+
+	tdma_write(tdma, TEGRA_APBDMA_GENERAL, tdma->reg_gen);
+	tdma_write(tdma, TEGRA_APBDMA_CONTROL, 0);
+	tdma_write(tdma, TEGRA_APBDMA_IRQ_MASK_SET, 0xFFFFFFFFul);
+
+	for (i = 0; i < tdma->chip_data->nr_channels; i++) {
+		struct tegra_dma_channel *tdc = &tdma->channels[i];
+		struct tegra_dma_channel_regs *ch_reg = &tdc->channel_reg;
+
+		tdc_write(tdc, TEGRA_APBDMA_CHAN_APBSEQ, ch_reg->apb_seq);
+		tdc_write(tdc, TEGRA_APBDMA_CHAN_APBPTR, ch_reg->apb_ptr);
+		tdc_write(tdc, TEGRA_APBDMA_CHAN_AHBSEQ, ch_reg->ahb_seq);
+		tdc_write(tdc, TEGRA_APBDMA_CHAN_AHBPTR, ch_reg->ahb_ptr);
+		tdc_write(tdc, TEGRA_APBDMA_CHAN_CSR,
+			(ch_reg->csr & ~TEGRA_APBDMA_CSR_ENB));
+	}
+
+	/* Disable clock */
+	tegra_dma_runtime_suspend(dev);
+	return 0;
+}
+#endif
+
 static const struct dev_pm_ops tegra_dma_dev_pm_ops = {
 #ifdef CONFIG_PM_RUNTIME
 	.runtime_suspend = tegra_dma_runtime_suspend,
 	.runtime_resume = tegra_dma_runtime_resume,
 #endif
+	SET_SYSTEM_SLEEP_PM_OPS(tegra_dma_pm_suspend, tegra_dma_pm_resume)
 };
 
 static struct platform_driver tegra_dmac_driver = {
@@ -1460,7 +1517,7 @@ static struct platform_driver tegra_dmac_driver = {
 		.name	= "tegra-apbdma",
 		.owner = THIS_MODULE,
 		.pm	= &tegra_dma_dev_pm_ops,
-		.of_match_table = of_match_ptr(tegra_dma_of_match),
+		.of_match_table = tegra_dma_of_match,
 	},
 	.probe		= tegra_dma_probe,
 	.remove		= tegra_dma_remove,
diff --git a/drivers/dma/timb_dma.c b/drivers/dma/timb_dma.c
index 952f823901a..26107ba6edb 100644
--- a/drivers/dma/timb_dma.c
+++ b/drivers/dma/timb_dma.c
@@ -823,7 +823,7 @@ static struct platform_driver td_driver = {
 		.owner  = THIS_MODULE,
 	},
 	.probe	= td_probe,
-	.remove	= __exit_p(td_remove),
+	.remove	= td_remove,
 };
 
 module_platform_driver(td_driver);
diff --git a/drivers/dma/txx9dmac.c b/drivers/dma/txx9dmac.c
index 913f55c76c9..a59fb4841d4 100644
--- a/drivers/dma/txx9dmac.c
+++ b/drivers/dma/txx9dmac.c
@@ -1190,7 +1190,7 @@ static int __init txx9dmac_chan_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int __exit txx9dmac_chan_remove(struct platform_device *pdev)
+static int txx9dmac_chan_remove(struct platform_device *pdev)
 {
 	struct txx9dmac_chan *dc = platform_get_drvdata(pdev);
 
@@ -1252,7 +1252,7 @@ static int __init txx9dmac_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int __exit txx9dmac_remove(struct platform_device *pdev)
+static int txx9dmac_remove(struct platform_device *pdev)
 {
 	struct txx9dmac_dev *ddev = platform_get_drvdata(pdev);
 
@@ -1299,14 +1299,14 @@ static const struct dev_pm_ops txx9dmac_dev_pm_ops = {
 };
 
 static struct platform_driver txx9dmac_chan_driver = {
-	.remove		= __exit_p(txx9dmac_chan_remove),
+	.remove		= txx9dmac_chan_remove,
 	.driver = {
 		.name	= "txx9dmac-chan",
 	},
 };
 
 static struct platform_driver txx9dmac_driver = {
-	.remove		= __exit_p(txx9dmac_remove),
+	.remove		= txx9dmac_remove,
 	.shutdown	= txx9dmac_shutdown,
 	.driver = {
 		.name	= "txx9dmac",