Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (33 commits)
  x86: poll waiting for I/OAT DMA channel status
  maintainers: add dma engine tree details
  dmaengine: add TODO items for future work on dma drivers
  dmaengine: Add API documentation for slave dma usage
  dmaengine/dw_dmac: Update maintainer-ship
  dmaengine: move link order
  dmaengine/dw_dmac: implement pause and resume in dwc_control
  dmaengine/dw_dmac: Replace spin_lock* with irqsave variants and enable submission from callback
  dmaengine/dw_dmac: Divide one sg to many desc, if sg len is greater than DWC_MAX_COUNT
  dmaengine/dw_dmac: set residue as total len in dwc_tx_status if status is !DMA_SUCCESS
  dmaengine/dw_dmac: don't call callback routine in case dmaengine_terminate_all() is called
  dmaengine: at_hdmac: pause: no need to wait for FIFO empty
  pch_dma: modify pci device table definition
  pch_dma: Support new device ML7223 IOH
  pch_dma: Support I2S for ML7213 IOH
  pch_dma: Fix DMA setting issue
  pch_dma: modify for checkpatch
  pch_dma: fix dma direction issue for ML7213 IOH video-in
  dmaengine: at_hdmac: use descriptor chaining help function
  dmaengine: at_hdmac: implement pause and resume in atc_control
  ...

Fix up trivial conflict in drivers/dma/dw_dmac.c

Documentation/dmaengine.txt

-See Documentation/crypto/async-tx-api.txt
+			DMA Engine API Guide
+			====================
+
+		 Vinod Koul <vinod dot koul at intel.com>
+
+NOTE: For DMA Engine usage in async_tx please see:
+	Documentation/crypto/async-tx-api.txt
+
+
+Below is a guide to device driver writers on how to use the Slave-DMA API of the
+DMA Engine. This is applicable only for slave DMA usage only.
+
+The slave DMA usage consists of following steps
+1. Allocate a DMA slave channel
+2. Set slave and controller specific parameters
+3. Get a descriptor for transaction
+4. Submit the transaction and wait for callback notification
+
+1. Allocate a DMA slave channel
+Channel allocation is slightly different in the slave DMA context, client
+drivers typically need a channel from a particular DMA controller only and even
+in some cases a specific channel is desired. To request a channel
+dma_request_channel() API is used.
+
+Interface:
+struct dma_chan *dma_request_channel(dma_cap_mask_t mask,
+		dma_filter_fn filter_fn,
+		void *filter_param);
+where dma_filter_fn is defined as:
+typedef bool (*dma_filter_fn)(struct dma_chan *chan, void *filter_param);
+
+When the optional 'filter_fn' parameter is set to NULL dma_request_channel
+simply returns the first channel that satisfies the capability mask.  Otherwise,
+when the mask parameter is insufficient for specifying the necessary channel,
+the filter_fn routine can be used to disposition the available channels in the
+system. The filter_fn routine is called once for each free channel in the
+system.  Upon seeing a suitable channel filter_fn returns DMA_ACK which flags
+that channel to be the return value from dma_request_channel.  A channel
+allocated via this interface is exclusive to the caller, until
+dma_release_channel() is called.
+
+2. Set slave and controller specific parameters
+Next step is always to pass some specific information to the DMA driver. Most of
+the generic information which a slave DMA can use is in struct dma_slave_config.
+It allows the clients to specify DMA direction, DMA addresses, bus widths, DMA
+burst lengths etc. If some DMA controllers have more parameters to be sent then
+they should try to embed struct dma_slave_config in their controller specific
+structure. That gives flexibility to client to pass more parameters, if
+required.
+
+Interface:
+int dmaengine_slave_config(struct dma_chan *chan,
+					  struct dma_slave_config *config)
+
+3. Get a descriptor for transaction
+For slave usage the various modes of slave transfers supported by the
+DMA-engine are:
+slave_sg	- DMA a list of scatter gather buffers from/to a peripheral
+dma_cyclic	- Perform a cyclic DMA operation from/to a peripheral till the
+		  operation is explicitly stopped.
+The non NULL return of this transfer API represents a "descriptor" for the given
+transaction.
+
+Interface:
+struct dma_async_tx_descriptor *(*chan->device->device_prep_dma_sg)(
+		struct dma_chan *chan,
+		struct scatterlist *dst_sg, unsigned int dst_nents,
+		struct scatterlist *src_sg, unsigned int src_nents,
+		unsigned long flags);
+struct dma_async_tx_descriptor *(*chan->device->device_prep_dma_cyclic)(
+		struct dma_chan *chan, dma_addr_t buf_addr, size_t buf_len,
+		size_t period_len, enum dma_data_direction direction);
+
+4. Submit the transaction and wait for callback notification
+To schedule the transaction to be scheduled by dma device, the "descriptor"
+returned in above (3) needs to be submitted.
+To tell the dma driver that a transaction is ready to be serviced, the
+descriptor->submit() callback needs to be invoked. This chains the descriptor to
+the pending queue.
+The transactions in the pending queue can be activated by calling the
+issue_pending API. If channel is idle then the first transaction in queue is
+started and subsequent ones queued up.
+On completion of the DMA operation the next in queue is submitted and a tasklet
+triggered. The tasklet would then call the client driver completion callback
+routine for notification, if set.
+Interface:
+void dma_async_issue_pending(struct dma_chan *chan);
+
+==============================================================================
+
+Additional usage notes for dma driver writers
+1/ Although DMA engine specifies that completion callback routines cannot submit
+any new operations, but typically for slave DMA subsequent transaction may not
+be available for submit prior to callback routine being called. This requirement
+is not a requirement for DMA-slave devices. But they should take care to drop
+the spin-lock they might be holding before calling the callback routine

MAINTAINERS

@@ -2178,6 +2178,8 @@ M:	Dan Williams <dan.j.williams@intel.com>
 S:	Supported
 F:	drivers/dma/
 F:	include/linux/dma*
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx.git
+T:	git git://git.infradead.org/users/vkoul/slave-dma.git (slave-dma)
 
 DME1737 HARDWARE MONITOR DRIVER
 M:	Juerg Haefliger <juergh@gmail.com>
@@ -5451,6 +5453,13 @@ L:	linux-serial@vger.kernel.org
 S:	Maintained
 F:	drivers/tty/serial
 
+SYNOPSYS DESIGNWARE DMAC DRIVER
+M:	Viresh Kumar <viresh.kumar@st.com>
+S:	Maintained
+F:	include/linux/dw_dmac.h
+F:	drivers/dma/dw_dmac_regs.h
+F:	drivers/dma/dw_dmac.c
+
 TIMEKEEPING, NTP
 M:	John Stultz <johnstul@us.ibm.com>
 M:	Thomas Gleixner <tglx@linutronix.de>

drivers/Makefile

@@ -17,6 +17,9 @@ obj-$(CONFIG_SFI)		+= sfi/
 # was used and do nothing if so
 obj-$(CONFIG_PNP)		+= pnp/
 obj-$(CONFIG_ARM_AMBA)		+= amba/
+# Many drivers will want to use DMA so this has to be made available
+# really early.
+obj-$(CONFIG_DMA_ENGINE)	+= dma/
 
 obj-$(CONFIG_VIRTIO)		+= virtio/
 obj-$(CONFIG_XEN)		+= xen/
@@ -92,7 +95,6 @@ obj-$(CONFIG_EISA)		+= eisa/
 obj-y				+= lguest/
 obj-$(CONFIG_CPU_FREQ)		+= cpufreq/
 obj-$(CONFIG_CPU_IDLE)		+= cpuidle/
-obj-$(CONFIG_DMA_ENGINE)	+= dma/
 obj-$(CONFIG_MMC)		+= mmc/
 obj-$(CONFIG_MEMSTICK)		+= memstick/
 obj-y				+= leds/

drivers/dma/Kconfig

@@ -200,16 +200,18 @@ config PL330_DMA
 	  platform_data for a dma-pl330 device.
 
 config PCH_DMA
-	tristate "Intel EG20T PCH / OKI SEMICONDUCTOR ML7213 IOH DMA support"
+	tristate "Intel EG20T PCH / OKI Semi IOH(ML7213/ML7223) DMA support"
 	depends on PCI && X86
 	select DMA_ENGINE
 	help
 	  Enable support for Intel EG20T PCH DMA engine.
 
-	  This driver also can be used for OKI SEMICONDUCTOR ML7213 IOH(Input/
-	  Output Hub) which is for IVI(In-Vehicle Infotainment) use.
-	  ML7213 is companion chip for Intel Atom E6xx series.
-	  ML7213 is completely compatible for Intel EG20T PCH.
+	  This driver also can be used for OKI SEMICONDUCTOR IOH(Input/
+	  Output Hub), ML7213 and ML7223.
+	  ML7213 IOH is for IVI(In-Vehicle Infotainment) use and ML7223 IOH is
+	  for MP(Media Phone) use.
+	  ML7213/ML7223 is companion chip for Intel Atom E6xx series.
+	  ML7213/ML7223 is completely compatible for Intel EG20T PCH.
 
 config IMX_SDMA
 	tristate "i.MX SDMA support"

drivers/dma/TODO

+TODO for slave dma
+
+1. Move remaining drivers to use new slave interface
+2. Remove old slave pointer machansim
+3. Make issue_pending to start the transaction in below drivers
+	- mpc512x_dma
+	- imx-dma
+	- imx-sdma
+	- mxs-dma.c
+	- dw_dmac
+	- intel_mid_dma
+	- ste_dma40
+4. Check other subsystems for dma drivers and merge/move to dmaengine
+5. Remove dma_slave_config's dma direction.

drivers/dma/at_hdmac_regs.h

@@ -103,6 +103,10 @@
 /* Bitfields in CTRLB */
 #define	ATC_SIF(i)		(0x3 & (i))	/* Src tx done via AHB-Lite Interface i */
 #define	ATC_DIF(i)		((0x3 & (i)) <<  4)	/* Dst tx done via AHB-Lite Interface i */
+				  /* Specify AHB interfaces */
+#define AT_DMA_MEM_IF		0 /* interface 0 as memory interface */
+#define AT_DMA_PER_IF		1 /* interface 1 as peripheral interface */
+
 #define	ATC_SRC_PIP		(0x1 <<  8)	/* Source Picture-in-Picture enabled */
 #define	ATC_DST_PIP		(0x1 << 12)	/* Destination Picture-in-Picture enabled */
 #define	ATC_SRC_DSCR_DIS	(0x1 << 16)	/* Src Descriptor fetch disable */
@@ -180,13 +184,24 @@ txd_to_at_desc(struct dma_async_tx_descriptor *txd)
 
 /*--  Channels  --------------------------------------------------------*/
 
+/**
+ * atc_status - information bits stored in channel status flag
+ *
+ * Manipulated with atomic operations.
+ */
+enum atc_status {
+	ATC_IS_ERROR = 0,
+	ATC_IS_PAUSED = 1,
+	ATC_IS_CYCLIC = 24,
+};
+
 /**
  * struct at_dma_chan - internal representation of an Atmel HDMAC channel
  * @chan_common: common dmaengine channel object members
  * @device: parent device
  * @ch_regs: memory mapped register base
  * @mask: channel index in a mask
- * @error_status: transmit error status information from irq handler
+ * @status: transmit status information from irq/prep* functions
  *                to tasklet (use atomic operations)
  * @tasklet: bottom half to finish transaction work
  * @lock: serializes enqueue/dequeue operations to descriptors lists
@@ -201,7 +216,7 @@ struct at_dma_chan {
 	struct at_dma		*device;
 	void __iomem		*ch_regs;
 	u8			mask;
-	unsigned long		error_status;
+	unsigned long		status;
 	struct tasklet_struct	tasklet;
 
 	spinlock_t		lock;
@@ -309,8 +324,8 @@ static void atc_setup_irq(struct at_dma_chan *atchan, int on)
 	struct at_dma	*atdma = to_at_dma(atchan->chan_common.device);
 	u32		ebci;
 
-	/* enable interrupts on buffer chain completion & error */
-	ebci =    AT_DMA_CBTC(atchan->chan_common.chan_id)
+	/* enable interrupts on buffer transfer completion & error */
+	ebci =    AT_DMA_BTC(atchan->chan_common.chan_id)
 		| AT_DMA_ERR(atchan->chan_common.chan_id);
 	if (on)
 		dma_writel(atdma, EBCIER, ebci);
@@ -347,7 +362,12 @@ static inline int atc_chan_is_enabled(struct at_dma_chan *atchan)
  */
 static void set_desc_eol(struct at_desc *desc)
 {
-	desc->lli.ctrlb |= ATC_SRC_DSCR_DIS | ATC_DST_DSCR_DIS;
+	u32 ctrlb = desc->lli.ctrlb;
+
+	ctrlb &= ~ATC_IEN;
+	ctrlb |= ATC_SRC_DSCR_DIS | ATC_DST_DSCR_DIS;
+
+	desc->lli.ctrlb = ctrlb;
 	desc->lli.dscr = 0;
 }
 

drivers/dma/coh901318.c

@@ -1610,7 +1610,7 @@ int __init coh901318_init(void)
 {
 	return platform_driver_probe(&coh901318_driver, coh901318_probe);
 }
-arch_initcall(coh901318_init);
+subsys_initcall(coh901318_init);
 
 void __exit coh901318_exit(void)
 {

drivers/dma/dw_dmac_regs.h

@@ -2,6 +2,7 @@
  * Driver for the Synopsys DesignWare AHB DMA Controller
  *
  * Copyright (C) 2005-2007 Atmel Corporation
+ * Copyright (C) 2010-2011 ST Microelectronics
  *
  * 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
@@ -138,6 +139,7 @@ struct dw_dma_chan {
 	void __iomem		*ch_regs;
 	u8			mask;
 	u8			priority;
+	bool			paused;
 
 	spinlock_t		lock;
 

drivers/dma/intel_mid_dma.c

@@ -1292,8 +1292,7 @@ static int __devinit intel_mid_dma_probe(struct pci_dev *pdev,
 	if (err)
 		goto err_dma;
 
-	pm_runtime_set_active(&pdev->dev);
-	pm_runtime_enable(&pdev->dev);
+	pm_runtime_put_noidle(&pdev->dev);
 	pm_runtime_allow(&pdev->dev);
 	return 0;
 
@@ -1322,6 +1321,9 @@ static int __devinit intel_mid_dma_probe(struct pci_dev *pdev,
 static void __devexit intel_mid_dma_remove(struct pci_dev *pdev)
 {
 	struct middma_device *device = pci_get_drvdata(pdev);
+
+	pm_runtime_get_noresume(&pdev->dev);
+	pm_runtime_forbid(&pdev->dev);
 	middma_shutdown(pdev);
 	pci_dev_put(pdev);
 	kfree(device);
@@ -1385,13 +1387,20 @@ int dma_resume(struct pci_dev *pci)
 static int dma_runtime_suspend(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	return dma_suspend(pci_dev, PMSG_SUSPEND);
+	struct middma_device *device = pci_get_drvdata(pci_dev);
+
+	device->state = SUSPENDED;
+	return 0;
 }
 
 static int dma_runtime_resume(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	return dma_resume(pci_dev);
+	struct middma_device *device = pci_get_drvdata(pci_dev);
+
+	device->state = RUNNING;
+	iowrite32(REG_BIT0, device->dma_base + DMA_CFG);
+	return 0;
 }
 
 static int dma_runtime_idle(struct device *dev)

drivers/dma/ioat/dma_v2.c

@@ -508,6 +508,7 @@ int ioat2_alloc_chan_resources(struct dma_chan *c)
 	struct ioat_ring_ent **ring;
 	u64 status;
 	int order;
+	int i = 0;
 
 	/* have we already been set up? */
 	if (ioat->ring)
@@ -548,8 +549,11 @@ int ioat2_alloc_chan_resources(struct dma_chan *c)
 	ioat2_start_null_desc(ioat);
 
 	/* check that we got off the ground */
-	udelay(5);
-	status = ioat_chansts(chan);
+	do {
+		udelay(1);
+		status = ioat_chansts(chan);
+	} while (i++ < 20 && !is_ioat_active(status) && !is_ioat_idle(status));
+
 	if (is_ioat_active(status) || is_ioat_idle(status)) {
 		set_bit(IOAT_RUN, &chan->state);
 		return 1 << ioat->alloc_order;

drivers/dma/iop-adma.c

@@ -619,7 +619,7 @@ iop_adma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dma_dest,
 
 	if (unlikely(!len))
 		return NULL;
-	BUG_ON(unlikely(len > IOP_ADMA_MAX_BYTE_COUNT));
+	BUG_ON(len > IOP_ADMA_MAX_BYTE_COUNT);
 
 	dev_dbg(iop_chan->device->common.dev, "%s len: %u\n",
 		__func__, len);
@@ -652,7 +652,7 @@ iop_adma_prep_dma_memset(struct dma_chan *chan, dma_addr_t dma_dest,
 
 	if (unlikely(!len))
 		return NULL;
-	BUG_ON(unlikely(len > IOP_ADMA_MAX_BYTE_COUNT));
+	BUG_ON(len > IOP_ADMA_MAX_BYTE_COUNT);
 
 	dev_dbg(iop_chan->device->common.dev, "%s len: %u\n",
 		__func__, len);
@@ -686,7 +686,7 @@ iop_adma_prep_dma_xor(struct dma_chan *chan, dma_addr_t dma_dest,
 
 	if (unlikely(!len))
 		return NULL;
-	BUG_ON(unlikely(len > IOP_ADMA_XOR_MAX_BYTE_COUNT));
+	BUG_ON(len > IOP_ADMA_XOR_MAX_BYTE_COUNT);
 
 	dev_dbg(iop_chan->device->common.dev,
 		"%s src_cnt: %d len: %u flags: %lx\n",

drivers/dma/mv_xor.c

@@ -671,7 +671,7 @@ mv_xor_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 	if (unlikely(len < MV_XOR_MIN_BYTE_COUNT))
 		return NULL;
 
-	BUG_ON(unlikely(len > MV_XOR_MAX_BYTE_COUNT));
+	BUG_ON(len > MV_XOR_MAX_BYTE_COUNT);
 
 	spin_lock_bh(&mv_chan->lock);
 	slot_cnt = mv_chan_memcpy_slot_count(len);
@@ -710,7 +710,7 @@ mv_xor_prep_dma_memset(struct dma_chan *chan, dma_addr_t dest, int value,
 	if (unlikely(len < MV_XOR_MIN_BYTE_COUNT))
 		return NULL;
 
-	BUG_ON(unlikely(len > MV_XOR_MAX_BYTE_COUNT));
+	BUG_ON(len > MV_XOR_MAX_BYTE_COUNT);
 
 	spin_lock_bh(&mv_chan->lock);
 	slot_cnt = mv_chan_memset_slot_count(len);
@@ -744,7 +744,7 @@ mv_xor_prep_dma_xor(struct dma_chan *chan, dma_addr_t dest, dma_addr_t *src,
 	if (unlikely(len < MV_XOR_MIN_BYTE_COUNT))
 		return NULL;
 
-	BUG_ON(unlikely(len > MV_XOR_MAX_BYTE_COUNT));
+	BUG_ON(len > MV_XOR_MAX_BYTE_COUNT);
 
 	dev_dbg(mv_chan->device->common.dev,
 		"%s src_cnt: %d len: dest %x %u flags: %ld\n",

drivers/dma/pch_dma.c

@@ -77,10 +77,10 @@ struct pch_dma_regs {
 	u32	dma_ctl0;
 	u32	dma_ctl1;
 	u32	dma_ctl2;
-	u32	reserved1;
+	u32	dma_ctl3;
 	u32	dma_sts0;
 	u32	dma_sts1;
-	u32	reserved2;
+	u32	dma_sts2;
 	u32	reserved3;
 	struct pch_dma_desc_regs desc[MAX_CHAN_NR];
 };
@@ -130,6 +130,7 @@ struct pch_dma {
 #define PCH_DMA_CTL0	0x00
 #define PCH_DMA_CTL1	0x04
 #define PCH_DMA_CTL2	0x08
+#define PCH_DMA_CTL3	0x0C
 #define PCH_DMA_STS0	0x10
 #define PCH_DMA_STS1	0x14
 
@@ -138,7 +139,8 @@ struct pch_dma {
 #define dma_writel(pd, name, val) \
 	writel((val), (pd)->membase + PCH_DMA_##name)
 
-static inline struct pch_dma_desc *to_pd_desc(struct dma_async_tx_descriptor *txd)
+static inline
+struct pch_dma_desc *to_pd_desc(struct dma_async_tx_descriptor *txd)
 {
 	return container_of(txd, struct pch_dma_desc, txd);
 }
@@ -163,13 +165,15 @@ static inline struct device *chan2parent(struct dma_chan *chan)
 	return chan->dev->device.parent;
 }
 
-static inline struct pch_dma_desc *pdc_first_active(struct pch_dma_chan *pd_chan)
+static inline
+struct pch_dma_desc *pdc_first_active(struct pch_dma_chan *pd_chan)
 {
 	return list_first_entry(&pd_chan->active_list,
 				struct pch_dma_desc, desc_node);
 }
 
-static inline struct pch_dma_desc *pdc_first_queued(struct pch_dma_chan *pd_chan)
+static inline
+struct pch_dma_desc *pdc_first_queued(struct pch_dma_chan *pd_chan)
 {
 	return list_first_entry(&pd_chan->queue,
 				struct pch_dma_desc, desc_node);
@@ -199,16 +203,30 @@ static void pdc_set_dir(struct dma_chan *chan)
 	struct pch_dma *pd = to_pd(chan->device);
 	u32 val;
 
-	val = dma_readl(pd, CTL0);
+	if (chan->chan_id < 8) {
+		val = dma_readl(pd, CTL0);
 
-	if (pd_chan->dir == DMA_TO_DEVICE)
-		val |= 0x1 << (DMA_CTL0_BITS_PER_CH * chan->chan_id +
-			       DMA_CTL0_DIR_SHIFT_BITS);
-	else
-		val &= ~(0x1 << (DMA_CTL0_BITS_PER_CH * chan->chan_id +
-				 DMA_CTL0_DIR_SHIFT_BITS));
+		if (pd_chan->dir == DMA_TO_DEVICE)
+			val |= 0x1 << (DMA_CTL0_BITS_PER_CH * chan->chan_id +
+				       DMA_CTL0_DIR_SHIFT_BITS);
+		else
+			val &= ~(0x1 << (DMA_CTL0_BITS_PER_CH * chan->chan_id +
+					 DMA_CTL0_DIR_SHIFT_BITS));
+
+		dma_writel(pd, CTL0, val);
+	} else {
+		int ch = chan->chan_id - 8; /* ch8-->0 ch9-->1 ... ch11->3 */
+		val = dma_readl(pd, CTL3);
 
-	dma_writel(pd, CTL0, val);
+		if (pd_chan->dir == DMA_TO_DEVICE)
+			val |= 0x1 << (DMA_CTL0_BITS_PER_CH * ch +
+				       DMA_CTL0_DIR_SHIFT_BITS);
+		else
+			val &= ~(0x1 << (DMA_CTL0_BITS_PER_CH * ch +
+					 DMA_CTL0_DIR_SHIFT_BITS));
+
+		dma_writel(pd, CTL3, val);
+	}
 
 	dev_dbg(chan2dev(chan), "pdc_set_dir: chan %d -> %x\n",
 		chan->chan_id, val);
@@ -219,13 +237,26 @@ static void pdc_set_mode(struct dma_chan *chan, u32 mode)
 	struct pch_dma *pd = to_pd(chan->device);
 	u32 val;
 
-	val = dma_readl(pd, CTL0);
+	if (chan->chan_id < 8) {
+		val = dma_readl(pd, CTL0);
+
+		val &= ~(DMA_CTL0_MODE_MASK_BITS <<
+			(DMA_CTL0_BITS_PER_CH * chan->chan_id));
+		val |= mode << (DMA_CTL0_BITS_PER_CH * chan->chan_id);
 
-	val &= ~(DMA_CTL0_MODE_MASK_BITS <<
-		(DMA_CTL0_BITS_PER_CH * chan->chan_id));
-	val |= mode << (DMA_CTL0_BITS_PER_CH * chan->chan_id);
+		dma_writel(pd, CTL0, val);
+	} else {
+		int ch = chan->chan_id - 8; /* ch8-->0 ch9-->1 ... ch11->3 */
+
+		val = dma_readl(pd, CTL3);
+
+		val &= ~(DMA_CTL0_MODE_MASK_BITS <<
+			(DMA_CTL0_BITS_PER_CH * ch));
+		val |= mode << (DMA_CTL0_BITS_PER_CH * ch);
 
-	dma_writel(pd, CTL0, val);
+		dma_writel(pd, CTL3, val);
+
+	}
 
 	dev_dbg(chan2dev(chan), "pdc_set_mode: chan %d -> %x\n",
 		chan->chan_id, val);
@@ -251,9 +282,6 @@ static bool pdc_is_idle(struct pch_dma_chan *pd_chan)
 
 static void pdc_dostart(struct pch_dma_chan *pd_chan, struct pch_dma_desc* desc)
 {
-	struct pch_dma *pd = to_pd(pd_chan->chan.device);
-	u32 val;
-
 	if (!pdc_is_idle(pd_chan)) {
 		dev_err(chan2dev(&pd_chan->chan),
 			"BUG: Attempt to start non-idle channel\n");
@@ -279,10 +307,6 @@ static void pdc_dostart(struct pch_dma_chan *pd_chan, struct pch_dma_desc* desc)
 		channel_writel(pd_chan, NEXT, desc->txd.phys);
 		pdc_set_mode(&pd_chan->chan, DMA_CTL0_SG);
 	}
-
-	val = dma_readl(pd, CTL2);
-	val |= 1 << (DMA_CTL2_START_SHIFT_BITS + pd_chan->chan.chan_id);
-	dma_writel(pd, CTL2, val);
 }
 
 static void pdc_chain_complete(struct pch_dma_chan *pd_chan,
@@ -403,7 +427,7 @@ static struct pch_dma_desc *pdc_desc_get(struct pch_dma_chan *pd_chan)
 {
 	struct pch_dma_desc *desc, *_d;
 	struct pch_dma_desc *ret = NULL;
-	int i;
+	int i = 0;
 
 	spin_lock(&pd_chan->lock);
 	list_for_each_entry_safe(desc, _d, &pd_chan->free_list, desc_node) {
@@ -478,7 +502,6 @@ static int pd_alloc_chan_resources(struct dma_chan *chan)
 	spin_unlock_bh(&pd_chan->lock);
 
 	pdc_enable_irq(chan, 1);
-	pdc_set_dir(chan);
 
 	return pd_chan->descs_allocated;
 }
@@ -561,6 +584,9 @@ static struct dma_async_tx_descriptor *pd_prep_slave_sg(struct dma_chan *chan,
 	else
 		return NULL;
 
+	pd_chan->dir = direction;
+	pdc_set_dir(chan);
+
 	for_each_sg(sgl, sg, sg_len, i) {
 		desc = pdc_desc_get(pd_chan);
 
@@ -703,6 +729,7 @@ static void pch_dma_save_regs(struct pch_dma *pd)
 	pd->regs.dma_ctl0 = dma_readl(pd, CTL0);
 	pd->regs.dma_ctl1 = dma_readl(pd, CTL1);
 	pd->regs.dma_ctl2 = dma_readl(pd, CTL2);
+	pd->regs.dma_ctl3 = dma_readl(pd, CTL3);
 
 	list_for_each_entry_safe(chan, _c, &pd->dma.channels, device_node) {
 		pd_chan = to_pd_chan(chan);
@@ -725,6 +752,7 @@ static void pch_dma_restore_regs(struct pch_dma *pd)
 	dma_writel(pd, CTL0, pd->regs.dma_ctl0);
 	dma_writel(pd, CTL1, pd->regs.dma_ctl1);
 	dma_writel(pd, CTL2, pd->regs.dma_ctl2);
+	dma_writel(pd, CTL3, pd->regs.dma_ctl3);
 
 	list_for_each_entry_safe(chan, _c, &pd->dma.channels, device_node) {
 		pd_chan = to_pd_chan(chan);
@@ -850,8 +878,6 @@ static int __devinit pch_dma_probe(struct pci_dev *pdev,
 
 		pd_chan->membase = &regs->desc[i];
 
-		pd_chan->dir = (i % 2) ? DMA_FROM_DEVICE : DMA_TO_DEVICE;
-
 		spin_lock_init(&pd_chan->lock);
 
 		INIT_LIST_HEAD(&pd_chan->active_list);
@@ -929,13 +955,23 @@ static void __devexit pch_dma_remove(struct pci_dev *pdev)
 #define PCI_DEVICE_ID_ML7213_DMA1_8CH	0x8026
 #define PCI_DEVICE_ID_ML7213_DMA2_8CH	0x802B
 #define PCI_DEVICE_ID_ML7213_DMA3_4CH	0x8034
+#define PCI_DEVICE_ID_ML7213_DMA4_12CH	0x8032
+#define PCI_DEVICE_ID_ML7223_DMA1_4CH	0x800B
+#define PCI_DEVICE_ID_ML7223_DMA2_4CH	0x800E
+#define PCI_DEVICE_ID_ML7223_DMA3_4CH	0x8017
+#define PCI_DEVICE_ID_ML7223_DMA4_4CH	0x803B
 
-static const struct pci_device_id pch_dma_id_table[] = {
+DEFINE_PCI_DEVICE_TABLE(pch_dma_id_table) = {
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_EG20T_PCH_DMA_8CH), 8 },
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_EG20T_PCH_DMA_4CH), 4 },
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7213_DMA1_8CH), 8}, /* UART Video */
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7213_DMA2_8CH), 8}, /* PCMIF SPI */
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7213_DMA3_4CH), 4}, /* FPGA */
+	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7213_DMA4_12CH), 12}, /* I2S */
+	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7223_DMA1_4CH), 4}, /* UART */
+	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7223_DMA2_4CH), 4}, /* Video SPI */
+	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7223_DMA3_4CH), 4}, /* Security */
+	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7223_DMA4_4CH), 4}, /* FPGA */
 	{ 0, },
 };
 

drivers/dma/ppc4xx/adma.c

@@ -2313,7 +2313,7 @@ static struct dma_async_tx_descriptor *ppc440spe_adma_prep_dma_memcpy(
 	if (unlikely(!len))
 		return NULL;
 
-	BUG_ON(unlikely(len > PPC440SPE_ADMA_DMA_MAX_BYTE_COUNT));
+	BUG_ON(len > PPC440SPE_ADMA_DMA_MAX_BYTE_COUNT);
 
 	spin_lock_bh(&ppc440spe_chan->lock);
 
@@ -2354,7 +2354,7 @@ static struct dma_async_tx_descriptor *ppc440spe_adma_prep_dma_memset(
 	if (unlikely(!len))
 		return NULL;
 
-	BUG_ON(unlikely(len > PPC440SPE_ADMA_DMA_MAX_BYTE_COUNT));
+	BUG_ON(len > PPC440SPE_ADMA_DMA_MAX_BYTE_COUNT);
 
 	spin_lock_bh(&ppc440spe_chan->lock);
 
@@ -2397,7 +2397,7 @@ static struct dma_async_tx_descriptor *ppc440spe_adma_prep_dma_xor(
 				     dma_dest, dma_src, src_cnt));
 	if (unlikely(!len))
 		return NULL;
-	BUG_ON(unlikely(len > PPC440SPE_ADMA_XOR_MAX_BYTE_COUNT));
+	BUG_ON(len > PPC440SPE_ADMA_XOR_MAX_BYTE_COUNT);
 
 	dev_dbg(ppc440spe_chan->device->common.dev,
 		"ppc440spe adma%d: %s src_cnt: %d len: %u int_en: %d\n",
@@ -2887,7 +2887,7 @@ static struct dma_async_tx_descriptor *ppc440spe_adma_prep_dma_pq(
 	ADMA_LL_DBG(prep_dma_pq_dbg(ppc440spe_chan->device->id,
 				    dst, src, src_cnt));
 	BUG_ON(!len);
-	BUG_ON(unlikely(len > PPC440SPE_ADMA_XOR_MAX_BYTE_COUNT));
+	BUG_ON(len > PPC440SPE_ADMA_XOR_MAX_BYTE_COUNT);
 	BUG_ON(!src_cnt);
 
 	if (src_cnt == 1 && dst[1] == src[0]) {

drivers/dma/ste_dma40.c

@@ -1829,7 +1829,7 @@ d40_get_dev_addr(struct d40_chan *chan, enum dma_data_direction direction)
 {
 	struct stedma40_platform_data *plat = chan->base->plat_data;
 	struct stedma40_chan_cfg *cfg = &chan->dma_cfg;
-	dma_addr_t addr;
+	dma_addr_t addr = 0;
 
 	if (chan->runtime_addr)
 		return chan->runtime_addr;
@@ -2962,4 +2962,4 @@ static int __init stedma40_init(void)
 {
 	return platform_driver_probe(&d40_driver, d40_probe);
 }
-arch_initcall(stedma40_init);
+subsys_initcall(stedma40_init);

include/linux/dw_dmac.h

@@ -3,6 +3,7 @@
  * AVR32 systems.)
  *
  * Copyright (C) 2007 Atmel Corporation
+ * Copyright (C) 2010-2011 ST Microelectronics
  *
  * 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