Merge tag 'dmaengine-4.6-rc1' of git://git.infradead.org/users/vkoul/slave-dma

Pull dmaengine updates from Vinod Koul:
 "This is smallish update with minor changes to core and new driver and
  usual updates.  Nothing super exciting here..

   - We have made slave address as physical to enable driver to do the
     mapping.

   - We now expose the maxburst for slave dma as new capability so
     clients can know this and program accordingly

   - addition of device synchronize callbacks on omap and edma.

   - pl330 updates to support DMAFLUSHP for Rockchip platforms.

   - Updates and improved sg handling in Xilinx VDMA driver.

   - New hidma qualcomm dma driver, though some bits are still in
     progress"

* tag 'dmaengine-4.6-rc1' of git://git.infradead.org/users/vkoul/slave-dma: (40 commits)
  dmaengine: IOATDMA: revise channel reset workaround on CB3.3 platforms
  dmaengine: add Qualcomm Technologies HIDMA channel driver
  dmaengine: add Qualcomm Technologies HIDMA management driver
  dmaengine: hidma: Add Device Tree binding
  dmaengine: qcom_bam_dma: move to qcom directory
  dmaengine: tegra: Move of_device_id table near to its user
  dmaengine: xilinx_vdma: Remove unnecessary variable initializations
  dmaengine: sirf: use __maybe_unused to hide pm functions
  dmaengine: rcar-dmac: clear pertinence number of channels
  dmaengine: sh: shdmac: don't open code of_device_get_match_data()
  dmaengine: tegra: don't open code of_device_get_match_data()
  dmaengine: qcom_bam_dma: Make driver work for BE
  dmaengine: sun4i: support module autoloading
  dma/mic_x100_dma: IS_ERR() vs PTR_ERR() typo
  dmaengine: xilinx_vdma: Use readl_poll_timeout instead of do while loop's
  dmaengine: xilinx_vdma: Simplify spin lock handling
  dmaengine: xilinx_vdma: Fix issues with non-parking mode
  dmaengine: xilinx_vdma: Improve SG engine handling
  dmaengine: pl330: fix to support the burst mode
  dmaengine: make slave address physical
  ...

37 files changed, 2146 insertions, 464 deletions

diff --git a/Documentation/ABI/testing/sysfs-platform-hidma-mgmt b/Documentation/ABI/testing/sysfs-platform-hidma-mgmt
new file mode 100644
index 000000000000..c2fb5d033f0e
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-platform-hidma-mgmt
@@ -0,0 +1,97 @@
+What:		/sys/devices/platform/hidma-mgmt*/chanops/chan*/priority
+		/sys/devices/platform/QCOM8060:*/chanops/chan*/priority
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Contains either 0 or 1 and indicates if the DMA channel is a
+		low priority (0) or high priority (1) channel.
+
+What:		/sys/devices/platform/hidma-mgmt*/chanops/chan*/weight
+		/sys/devices/platform/QCOM8060:*/chanops/chan*/weight
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Contains 0..15 and indicates the weight of the channel among
+		equal priority channels during round robin scheduling.
+
+What:		/sys/devices/platform/hidma-mgmt*/chreset_timeout_cycles
+		/sys/devices/platform/QCOM8060:*/chreset_timeout_cycles
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Contains the platform specific cycle value to wait after a
+		reset command is issued. If the value is chosen too short,
+		then the HW will issue a reset failure interrupt. The value
+		is platform specific and should not be changed without
+		consultance.
+
+What:		/sys/devices/platform/hidma-mgmt*/dma_channels
+		/sys/devices/platform/QCOM8060:*/dma_channels
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Contains the number of dma channels supported by one instance
+		of HIDMA hardware. The value may change from chip to chip.
+
+What:		/sys/devices/platform/hidma-mgmt*/hw_version_major
+		/sys/devices/platform/QCOM8060:*/hw_version_major
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Version number major for the hardware.
+
+What:		/sys/devices/platform/hidma-mgmt*/hw_version_minor
+		/sys/devices/platform/QCOM8060:*/hw_version_minor
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Version number minor for the hardware.
+
+What:		/sys/devices/platform/hidma-mgmt*/max_rd_xactions
+		/sys/devices/platform/QCOM8060:*/max_rd_xactions
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Contains a value between 0 and 31. Maximum number of
+		read transactions that can be issued back to back.
+		Choosing a higher number gives better performance but
+		can also cause performance reduction to other peripherals
+		sharing the same bus.
+
+What:		/sys/devices/platform/hidma-mgmt*/max_read_request
+		/sys/devices/platform/QCOM8060:*/max_read_request
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Size of each read request. The value needs to be a power
+		of two and can be between 128 and 1024.
+
+What:		/sys/devices/platform/hidma-mgmt*/max_wr_xactions
+		/sys/devices/platform/QCOM8060:*/max_wr_xactions
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Contains a value between 0 and 31. Maximum number of
+		write transactions that can be issued back to back.
+		Choosing a higher number gives better performance but
+		can also cause performance reduction to other peripherals
+		sharing the same bus.
+
+
+What:		/sys/devices/platform/hidma-mgmt*/max_write_request
+		/sys/devices/platform/QCOM8060:*/max_write_request
+Date:		Nov 2015
+KernelVersion:	4.4
+Contact:	"Sinan Kaya <okaya@cudeaurora.org>"
+Description:
+		Size of each write request. The value needs to be a power
+		of two and can be between 128 and 1024.
diff --git a/Documentation/devicetree/bindings/dma/arm-pl330.txt b/Documentation/devicetree/bindings/dma/arm-pl330.txt
index 267565894db9..db7e2260f9c5 100644
--- a/Documentation/devicetree/bindings/dma/arm-pl330.txt
+++ b/Documentation/devicetree/bindings/dma/arm-pl330.txt
@@ -15,6 +15,7 @@ Optional properties:
     cells in the dmas property of client device.
   - dma-channels: contains the total number of DMA channels supported by the DMAC
   - dma-requests: contains the total number of DMA requests supported by the DMAC
+  - arm,pl330-broken-no-flushp: quirk for avoiding to execute DMAFLUSHP
 
 Example:
 
diff --git a/Documentation/devicetree/bindings/dma/qcom_hidma_mgmt.txt b/Documentation/devicetree/bindings/dma/qcom_hidma_mgmt.txt
new file mode 100644
index 000000000000..fd5618bd8fbc
--- /dev/null
+++ b/Documentation/devicetree/bindings/dma/qcom_hidma_mgmt.txt
@@ -0,0 +1,89 @@
+Qualcomm Technologies HIDMA Management interface
+
+Qualcomm Technologies HIDMA is a high speed DMA device. It only supports
+memcpy and memset capabilities. It has been designed for virtualized
+environments.
+
+Each HIDMA HW instance consists of multiple DMA channels. These channels
+share the same bandwidth. The bandwidth utilization can be parititioned
+among channels based on the priority and weight assignments.
+
+There are only two priority levels and 15 weigh assignments possible.
+
+Other parameters here determine how much of the system bus this HIDMA
+instance can use like maximum read/write request and and number of bytes to
+read/write in a single burst.
+
+Main node required properties:
+- compatible: "qcom,hidma-mgmt-1.0";
+- reg: Address range for DMA device
+- dma-channels: Number of channels supported by this DMA controller.
+- max-write-burst-bytes: Maximum write burst in bytes that HIDMA can
+  occupy the bus for in a single transaction. A memcpy requested is
+  fragmented to multiples of this amount. This parameter is used while
+  writing into destination memory. Setting this value incorrectly can
+  starve other peripherals in the system.
+- max-read-burst-bytes: Maximum read burst in bytes that HIDMA can
+  occupy the bus for in a single transaction. A memcpy request is
+  fragmented to multiples of this amount. This parameter is used while
+  reading the source memory. Setting this value incorrectly can starve
+  other peripherals in the system.
+- max-write-transactions: This value is how many times a write burst is
+  applied back to back while writing to the destination before yielding
+  the bus.
+- max-read-transactions: This value is how many times a read burst is
+  applied back to back while reading the source before yielding the bus.
+- channel-reset-timeout-cycles: Channel reset timeout in cycles for this SOC.
+  Once a reset is applied to the HW, HW starts a timer for reset operation
+  to confirm. If reset is not completed within this time, HW reports reset
+  failure.
+
+Sub-nodes:
+
+HIDMA has one or more DMA channels that are used to move data from one
+memory location to another.
+
+When the OS is not in control of the management interface (i.e. it's a guest),
+the channel nodes appear on their own, not under a management node.
+
+Required properties:
+- compatible: must contain "qcom,hidma-1.0"
+- reg: Addresses for the transfer and event channel
+- interrupts: Should contain the event interrupt
+- desc-count: Number of asynchronous requests this channel can handle
+- iommus: required a iommu node
+
+Example:
+
+Hypervisor OS configuration:
+
+	hidma-mgmt@f9984000 = {
+		compatible = "qcom,hidma-mgmt-1.0";
+		reg = <0xf9984000 0x15000>;
+		dma-channels = <6>;
+		max-write-burst-bytes = <1024>;
+		max-read-burst-bytes = <1024>;
+		max-write-transactions = <31>;
+		max-read-transactions = <31>;
+		channel-reset-timeout-cycles = <0x500>;
+
+		hidma_24: dma-controller@0x5c050000 {
+			compatible = "qcom,hidma-1.0";
+			reg = <0 0x5c050000 0x0 0x1000>,
+			      <0 0x5c0b0000 0x0 0x1000>;
+			interrupts = <0 389 0>;
+			desc-count = <10>;
+			iommus = <&system_mmu>;
+		};
+	};
+
+Guest OS configuration:
+
+	hidma_24: dma-controller@0x5c050000 {
+		compatible = "qcom,hidma-1.0";
+		reg = <0 0x5c050000 0x0 0x1000>,
+		      <0 0x5c0b0000 0x0 0x1000>;
+		interrupts = <0 389 0>;
+		desc-count = <10>;
+		iommus = <&system_mmu>;
+	};
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 79b1390f2016..d96d87c56f2e 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -341,12 +341,13 @@ config MV_XOR
 
 config MXS_DMA
 	bool "MXS DMA support"
-	depends on SOC_IMX23 || SOC_IMX28 || SOC_IMX6Q
+	depends on SOC_IMX23 || SOC_IMX28 || SOC_IMX6Q || SOC_IMX6UL
 	select STMP_DEVICE
 	select DMA_ENGINE
 	help
 	  Support the MXS DMA engine. This engine including APBH-DMA
-	  and APBX-DMA is integrated into Freescale i.MX23/28/MX6Q/MX6DL chips.
+	  and APBX-DMA is integrated into Freescale
+	  i.MX23/28/MX6Q/MX6DL/MX6UL chips.
 
 config MX3_IPU
 	bool "MX3x Image Processing Unit support"
@@ -408,15 +409,6 @@ config PXA_DMA
 	  16 to 32 channels for peripheral to memory or memory to memory
 	  transfers.
 
-config QCOM_BAM_DMA
-	tristate "QCOM BAM DMA support"
-	depends on ARCH_QCOM || (COMPILE_TEST && OF && ARM)
-	select DMA_ENGINE
-	select DMA_VIRTUAL_CHANNELS
-	---help---
-	  Enable support for the QCOM BAM DMA controller.  This controller
-	  provides DMA capabilities for a variety of on-chip devices.
-
 config SIRF_DMA
 	tristate "CSR SiRFprimaII/SiRFmarco DMA support"
 	depends on ARCH_SIRF
@@ -539,6 +531,8 @@ config ZX_DMA
 # driver files
 source "drivers/dma/bestcomm/Kconfig"
 
+source "drivers/dma/qcom/Kconfig"
+
 source "drivers/dma/dw/Kconfig"
 
 source "drivers/dma/hsu/Kconfig"
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 2dd0a067a0ca..6084127c1486 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -52,7 +52,6 @@ obj-$(CONFIG_PCH_DMA) += pch_dma.o
 obj-$(CONFIG_PL330_DMA) += pl330.o
 obj-$(CONFIG_PPC_BESTCOMM) += bestcomm/
 obj-$(CONFIG_PXA_DMA) += pxa_dma.o
-obj-$(CONFIG_QCOM_BAM_DMA) += qcom_bam_dma.o
 obj-$(CONFIG_RENESAS_DMA) += sh/
 obj-$(CONFIG_SIRF_DMA) += sirf-dma.o
 obj-$(CONFIG_STE_DMA40) += ste_dma40.o ste_dma40_ll.o
@@ -67,4 +66,5 @@ obj-$(CONFIG_TI_EDMA) += edma.o
 obj-$(CONFIG_XGENE_DMA) += xgene-dma.o
 obj-$(CONFIG_ZX_DMA) += zx296702_dma.o
 
+obj-y += qcom/
 obj-y += xilinx/
diff --git a/drivers/dma/acpi-dma.c b/drivers/dma/acpi-dma.c
index eed6bda01790..4a748c3435d7 100644
--- a/drivers/dma/acpi-dma.c
+++ b/drivers/dma/acpi-dma.c
@@ -438,7 +438,7 @@ struct dma_chan *acpi_dma_request_slave_chan_by_name(struct device *dev,
 			return ERR_PTR(-ENODEV);
 	}
 
-	dev_dbg(dev, "found DMA channel \"%s\" at index %d\n", name, index);
+	dev_dbg(dev, "Looking for DMA channel \"%s\" at index %d...\n", name, index);
 	return acpi_dma_request_slave_chan_by_index(dev, index);
 }
 EXPORT_SYMBOL_GPL(acpi_dma_request_slave_chan_by_name);
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index c50a247be2e0..0cb259c59916 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -496,6 +496,7 @@ int dma_get_slave_caps(struct dma_chan *chan, struct dma_slave_caps *caps)
 	caps->src_addr_widths = device->src_addr_widths;
 	caps->dst_addr_widths = device->dst_addr_widths;
 	caps->directions = device->directions;
+	caps->max_burst = device->max_burst;
 	caps->residue_granularity = device->residue_granularity;
 	caps->descriptor_reuse = device->descriptor_reuse;
 
diff --git a/drivers/dma/dw/regs.h b/drivers/dma/dw/regs.h
index 241ff2b1402b..0a50c18d85b8 100644
--- a/drivers/dma/dw/regs.h
+++ b/drivers/dma/dw/regs.h
@@ -150,7 +150,7 @@ enum dw_dma_msize {
 #define DWC_CTLL_DST_INC	(0<<7)		/* DAR update/not */
 #define DWC_CTLL_DST_DEC	(1<<7)
 #define DWC_CTLL_DST_FIX	(2<<7)
-#define DWC_CTLL_SRC_INC	(0<<7)		/* SAR update/not */
+#define DWC_CTLL_SRC_INC	(0<<9)		/* SAR update/not */
 #define DWC_CTLL_SRC_DEC	(1<<9)
 #define DWC_CTLL_SRC_FIX	(2<<9)
 #define DWC_CTLL_DST_MSIZE(n)	((n)<<11)	/* burst, #elements */
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index e3d7fcb69b4c..ee3463e774f8 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -869,6 +869,13 @@ static int edma_terminate_all(struct dma_chan *chan)
 	return 0;
 }
 
+static void edma_synchronize(struct dma_chan *chan)
+{
+	struct edma_chan *echan = to_edma_chan(chan);
+
+	vchan_synchronize(&echan->vchan);
+}
+
 static int edma_slave_config(struct dma_chan *chan,
 	struct dma_slave_config *cfg)
 {
@@ -1365,36 +1372,36 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
 static void edma_completion_handler(struct edma_chan *echan)
 {
 	struct device *dev = echan->vchan.chan.device->dev;
-	struct edma_desc *edesc = echan->edesc;
-
-	if (!edesc)
-		return;
+	struct edma_desc *edesc;
 
 	spin_lock(&echan->vchan.lock);
-	if (edesc->cyclic) {
-		vchan_cyclic_callback(&edesc->vdesc);
-		spin_unlock(&echan->vchan.lock);
-		return;
-	} else if (edesc->processed == edesc->pset_nr) {
-		edesc->residue = 0;
-		edma_stop(echan);
-		vchan_cookie_complete(&edesc->vdesc);
-		echan->edesc = NULL;
-
-		dev_dbg(dev, "Transfer completed on channel %d\n",
-			echan->ch_num);
-	} else {
-		dev_dbg(dev, "Sub transfer completed on channel %d\n",
-			echan->ch_num);
-
-		edma_pause(echan);
-
-		/* Update statistics for tx_status */
-		edesc->residue -= edesc->sg_len;
-		edesc->residue_stat = edesc->residue;
-		edesc->processed_stat = edesc->processed;
+	edesc = echan->edesc;
+	if (edesc) {
+		if (edesc->cyclic) {
+			vchan_cyclic_callback(&edesc->vdesc);
+			spin_unlock(&echan->vchan.lock);
+			return;
+		} else if (edesc->processed == edesc->pset_nr) {
+			edesc->residue = 0;
+			edma_stop(echan);
+			vchan_cookie_complete(&edesc->vdesc);
+			echan->edesc = NULL;
+
+			dev_dbg(dev, "Transfer completed on channel %d\n",
+				echan->ch_num);
+		} else {
+			dev_dbg(dev, "Sub transfer completed on channel %d\n",
+				echan->ch_num);
+
+			edma_pause(echan);
+
+			/* Update statistics for tx_status */
+			edesc->residue -= edesc->sg_len;
+			edesc->residue_stat = edesc->residue;
+			edesc->processed_stat = edesc->processed;
+		}
+		edma_execute(echan);
 	}
-	edma_execute(echan);
 
 	spin_unlock(&echan->vchan.lock);
 }
@@ -1837,6 +1844,7 @@ static void edma_dma_init(struct edma_cc *ecc, bool legacy_mode)
 	s_ddev->device_pause = edma_dma_pause;
 	s_ddev->device_resume = edma_dma_resume;
 	s_ddev->device_terminate_all = edma_terminate_all;
+	s_ddev->device_synchronize = edma_synchronize;
 
 	s_ddev->src_addr_widths = EDMA_DMA_BUSWIDTHS;
 	s_ddev->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
@@ -1862,6 +1870,7 @@ static void edma_dma_init(struct edma_cc *ecc, bool legacy_mode)
 		m_ddev->device_pause = edma_dma_pause;
 		m_ddev->device_resume = edma_dma_resume;
 		m_ddev->device_terminate_all = edma_terminate_all;
+		m_ddev->device_synchronize = edma_synchronize;
 
 		m_ddev->src_addr_widths = EDMA_DMA_BUSWIDTHS;
 		m_ddev->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
diff --git a/drivers/dma/ep93xx_dma.c b/drivers/dma/ep93xx_dma.c
index 57ff46284f15..21f08cc3352b 100644
--- a/drivers/dma/ep93xx_dma.c
+++ b/drivers/dma/ep93xx_dma.c
@@ -421,23 +421,25 @@ static int m2p_hw_interrupt(struct ep93xx_dma_chan *edmac)
 			desc->size);
 	}
 
-	switch (irq_status & (M2P_INTERRUPT_STALL | M2P_INTERRUPT_NFB)) {
-	case M2P_INTERRUPT_STALL:
-		/* Disable interrupts */
-		control = readl(edmac->regs + M2P_CONTROL);
-		control &= ~(M2P_CONTROL_STALLINT | M2P_CONTROL_NFBINT);
-		m2p_set_control(edmac, control);
-
-		return INTERRUPT_DONE;
-
-	case M2P_INTERRUPT_NFB:
-		if (ep93xx_dma_advance_active(edmac))
-			m2p_fill_desc(edmac);
+	/*
+	 * Even latest E2 silicon revision sometimes assert STALL interrupt
+	 * instead of NFB. Therefore we treat them equally, basing on the
+	 * amount of data we still have to transfer.
+	 */
+	if (!(irq_status & (M2P_INTERRUPT_STALL | M2P_INTERRUPT_NFB)))
+		return INTERRUPT_UNKNOWN;
 
+	if (ep93xx_dma_advance_active(edmac)) {
+		m2p_fill_desc(edmac);
 		return INTERRUPT_NEXT_BUFFER;
 	}
 
-	return INTERRUPT_UNKNOWN;
+	/* Disable interrupts */
+	control = readl(edmac->regs + M2P_CONTROL);
+	control &= ~(M2P_CONTROL_STALLINT | M2P_CONTROL_NFBINT);
+	m2p_set_control(edmac, control);
+
+	return INTERRUPT_DONE;
 }
 
 /*
diff --git a/drivers/dma/idma64.c b/drivers/dma/idma64.c
index 3cb7b2c78197..1953e57505f4 100644
--- a/drivers/dma/idma64.c
+++ b/drivers/dma/idma64.c
@@ -289,6 +289,9 @@ static void idma64_desc_fill(struct idma64_chan *idma64c,
 
 	/* Trigger an interrupt after the last block is transfered */
 	lli->ctllo |= IDMA64C_CTLL_INT_EN;
+
+	/* Disable LLP transfer in the last block */
+	lli->ctllo &= ~(IDMA64C_CTLL_LLP_S_EN | IDMA64C_CTLL_LLP_D_EN);
 }
 
 static struct dma_async_tx_descriptor *idma64_prep_slave_sg(
diff --git a/drivers/dma/idma64.h b/drivers/dma/idma64.h
index 8423f13ed0da..dc6874424188 100644
--- a/drivers/dma/idma64.h
+++ b/drivers/dma/idma64.h
@@ -71,7 +71,7 @@
 #define IDMA64C_CFGH_SRC_PER(x)		((x) << 0)	/* src peripheral */
 #define IDMA64C_CFGH_DST_PER(x)		((x) << 4)	/* dst peripheral */
 #define IDMA64C_CFGH_RD_ISSUE_THD(x)	((x) << 8)
-#define IDMA64C_CFGH_RW_ISSUE_THD(x)	((x) << 18)
+#define IDMA64C_CFGH_WR_ISSUE_THD(x)	((x) << 18)
 
 /* Interrupt registers */
 
diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index 21539d5c54c3..bd09961443b1 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -31,6 +31,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/workqueue.h>
 #include <linux/prefetch.h>
+#include <linux/sizes.h>
 #include "dma.h"
 #include "registers.h"
 #include "hw.h"
@@ -290,24 +291,30 @@ static dma_cookie_t ioat_tx_submit_unlock(struct dma_async_tx_descriptor *tx)
 }
 
 static struct ioat_ring_ent *
-ioat_alloc_ring_ent(struct dma_chan *chan, gfp_t flags)
+ioat_alloc_ring_ent(struct dma_chan *chan, int idx, gfp_t flags)
 {
 	struct ioat_dma_descriptor *hw;
 	struct ioat_ring_ent *desc;
 	struct ioatdma_device *ioat_dma;
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(chan);
+	int chunk;
 	dma_addr_t phys;
+	u8 *pos;
+	off_t offs;
 
 	ioat_dma = to_ioatdma_device(chan->device);
-	hw = pci_pool_alloc(ioat_dma->dma_pool, flags, &phys);
-	if (!hw)
-		return NULL;
+
+	chunk = idx / IOAT_DESCS_PER_2M;
+	idx &= (IOAT_DESCS_PER_2M - 1);
+	offs = idx * IOAT_DESC_SZ;
+	pos = (u8 *)ioat_chan->descs[chunk].virt + offs;
+	phys = ioat_chan->descs[chunk].hw + offs;
+	hw = (struct ioat_dma_descriptor *)pos;
 	memset(hw, 0, sizeof(*hw));
 
 	desc = kmem_cache_zalloc(ioat_cache, flags);
-	if (!desc) {
-		pci_pool_free(ioat_dma->dma_pool, hw, phys);
+	if (!desc)
 		return NULL;
-	}
 
 	dma_async_tx_descriptor_init(&desc->txd, chan);
 	desc->txd.tx_submit = ioat_tx_submit_unlock;
@@ -318,32 +325,63 @@ ioat_alloc_ring_ent(struct dma_chan *chan, gfp_t flags)
 
 void ioat_free_ring_ent(struct ioat_ring_ent *desc, struct dma_chan *chan)
 {
-	struct ioatdma_device *ioat_dma;
-
-	ioat_dma = to_ioatdma_device(chan->device);
-	pci_pool_free(ioat_dma->dma_pool, desc->hw, desc->txd.phys);
 	kmem_cache_free(ioat_cache, desc);
 }
 
 struct ioat_ring_ent **
 ioat_alloc_ring(struct dma_chan *c, int order, gfp_t flags)
 {
+	struct ioatdma_chan *ioat_chan = to_ioat_chan(c);
 	struct ioat_ring_ent **ring;
-	int descs = 1 << order;
-	int i;
-
-	if (order > ioat_get_max_alloc_order())
-		return NULL;
+	int total_descs = 1 << order;
+	int i, chunks;
 
 	/* allocate the array to hold the software ring */
-	ring = kcalloc(descs, sizeof(*ring), flags);
+	ring = kcalloc(total_descs, sizeof(*ring), flags);
 	if (!ring)
 		return NULL;
-	for (i = 0; i < descs; i++) {
-		ring[i] = ioat_alloc_ring_ent(c, flags);
+
+	ioat_chan->desc_chunks = chunks = (total_descs * IOAT_DESC_SZ) / SZ_2M;
+
+	for (i = 0; i < chunks; i++) {
+		struct ioat_descs *descs = &ioat_chan->descs[i];
+
+		descs->virt = dma_alloc_coherent(to_dev(ioat_chan),
+						 SZ_2M, &descs->hw, flags);
+		if (!descs->virt && (i > 0)) {
+			int idx;
+
+			for (idx = 0; idx < i; idx++) {
+				dma_free_coherent(to_dev(ioat_chan), SZ_2M,
+						  descs->virt, descs->hw);
+				descs->virt = NULL;
+				descs->hw = 0;
+			}
+
+			ioat_chan->desc_chunks = 0;
+			kfree(ring);
+			return NULL;
+		}
+	}
+
+	for (i = 0; i < total_descs; i++) {
+		ring[i] = ioat_alloc_ring_ent(c, i, flags);
 		if (!ring[i]) {
+			int idx;
+
 			while (i--)
 				ioat_free_ring_ent(ring[i], c);
+
+			for (idx = 0; idx < ioat_chan->desc_chunks; idx++) {
+				dma_free_coherent(to_dev(ioat_chan),
+						  SZ_2M,
+						  ioat_chan->descs[idx].virt,
+						  ioat_chan->descs[idx].hw);
+				ioat_chan->descs[idx].virt = NULL;
+				ioat_chan->descs[idx].hw = 0;
+			}
+
+			ioat_chan->desc_chunks = 0;
 			kfree(ring);
 			return NULL;
 		}
@@ -351,7 +389,7 @@ ioat_alloc_ring(struct dma_chan *c, int order, gfp_t flags)
 	}
 
 	/* link descs */
-	for (i = 0; i < descs-1; i++) {
+	for (i = 0; i < total_descs-1; i++) {
 		struct ioat_ring_ent *next = ring[i+1];
 		struct ioat_dma_descriptor *hw = ring[i]->hw;
 
@@ -362,114 +400,6 @@ ioat_alloc_ring(struct dma_chan *c, int order, gfp_t flags)
 	return ring;
 }
 
-static bool reshape_ring(struct ioatdma_chan *ioat_chan, int order)
-{
-	/* reshape differs from normal ring allocation in that we want
-	 * to allocate a new software ring while only
-	 * extending/truncating the hardware ring
-	 */
-	struct dma_chan *c = &ioat_chan->dma_chan;
-	const u32 curr_size = ioat_ring_size(ioat_chan);
-	const u16 active = ioat_ring_active(ioat_chan);
-	const u32 new_size = 1 << order;
-	struct ioat_ring_ent **ring;
-	u32 i;
-
-	if (order > ioat_get_max_alloc_order())
-		return false;
-
-	/* double check that we have at least 1 free descriptor */
-	if (active == curr_size)
-		return false;
-
-	/* when shrinking, verify that we can hold the current active
-	 * set in the new ring
-	 */
-	if (active >= new_size)
-		return false;
-
-	/* allocate the array to hold the software ring */
-	ring = kcalloc(new_size, sizeof(*ring), GFP_NOWAIT);
-	if (!ring)
-		return false;
-
-	/* allocate/trim descriptors as needed */
-	if (new_size > curr_size) {
-		/* copy current descriptors to the new ring */
-		for (i = 0; i < curr_size; i++) {
-			u16 curr_idx = (ioat_chan->tail+i) & (curr_size-1);
-			u16 new_idx = (ioat_chan->tail+i) & (new_size-1);
-
-			ring[new_idx] = ioat_chan->ring[curr_idx];
-			set_desc_id(ring[new_idx], new_idx);
-		}
-
-		/* add new descriptors to the ring */
-		for (i = curr_size; i < new_size; i++) {
-			u16 new_idx = (ioat_chan->tail+i) & (new_size-1);
-
-			ring[new_idx] = ioat_alloc_ring_ent(c, GFP_NOWAIT);
-			if (!ring[new_idx]) {
-				while (i--) {
-					u16 new_idx = (ioat_chan->tail+i) &
-						       (new_size-1);
-
-					ioat_free_ring_ent(ring[new_idx], c);
-				}
-				kfree(ring);
-				return false;
-			}
-			set_desc_id(ring[new_idx], new_idx);
-		}
-
-		/* hw link new descriptors */
-		for (i = curr_size-1; i < new_size; i++) {
-			u16 new_idx = (ioat_chan->tail+i) & (new_size-1);
-			struct ioat_ring_ent *next =
-				ring[(new_idx+1) & (new_size-1)];
-			struct ioat_dma_descriptor *hw = ring[new_idx]->hw;
-
-			hw->next = next->txd.phys;
-		}
-	} else {
-		struct ioat_dma_descriptor *hw;
-		struct ioat_ring_ent *next;
-
-		/* copy current descriptors to the new ring, dropping the
-		 * removed descriptors
-		 */
-		for (i = 0; i < new_size; i++) {
-			u16 curr_idx = (ioat_chan->tail+i) & (curr_size-1);
-			u16 new_idx = (ioat_chan->tail+i) & (new_size-1);
-
-			ring[new_idx] = ioat_chan->ring[curr_idx];
-			set_desc_id(ring[new_idx], new_idx);
-		}
-
-		/* free deleted descriptors */
-		for (i = new_size; i < curr_size; i++) {
-			struct ioat_ring_ent *ent;
-
-			ent = ioat_get_ring_ent(ioat_chan, ioat_chan->tail+i);
-			ioat_free_ring_ent(ent, c);
-		}
-
-		/* fix up hardware ring */
-		hw = ring[(ioat_chan->tail+new_size-1) & (new_size-1)]->hw;
-		next = ring[(ioat_chan->tail+new_size) & (new_size-1)];
-		hw->next = next->txd.phys;
-	}
-
-	dev_dbg(to_dev(ioat_chan), "%s: allocated %d descriptors\n",
-		__func__, new_size);
-
-	kfree(ioat_chan->ring);
-	ioat_chan->ring = ring;
-	ioat_chan->alloc_order = order;
-
-	return true;
-}
-
 /**
  * ioat_check_space_lock - verify space and grab ring producer lock
  * @ioat: ioat,3 channel (ring) to operate on
@@ -478,9 +408,6 @@ static bool reshape_ring(struct ioatdma_chan *ioat_chan, int order)