Merge branch 'ioat-md-accel-for-linus' of git://lost.foo-projects.org/~dwillia2/git/iop

* 'ioat-md-accel-for-linus' of git://lost.foo-projects.org/~dwillia2/git/iop: (28 commits)
  ioatdma: add the unisys "i/oat" pci vendor/device id
  ARM: Add drivers/dma to arch/arm/Kconfig
  iop3xx: surface the iop3xx DMA and AAU units to the iop-adma driver
  iop13xx: surface the iop13xx adma units to the iop-adma driver
  dmaengine: driver for the iop32x, iop33x, and iop13xx raid engines
  md: remove raid5 compute_block and compute_parity5
  md: handle_stripe5 - request io processing in raid5_run_ops
  md: handle_stripe5 - add request/completion logic for async expand ops
  md: handle_stripe5 - add request/completion logic for async read ops
  md: handle_stripe5 - add request/completion logic for async check ops
  md: handle_stripe5 - add request/completion logic for async compute ops
  md: handle_stripe5 - add request/completion logic for async write ops
  md: common infrastructure for running operations with raid5_run_ops
  md: raid5_run_ops - run stripe operations outside sh->lock
  raid5: replace custom debug PRINTKs with standard pr_debug
  raid5: refactor handle_stripe5 and handle_stripe6 (v3)
  async_tx: add the async_tx api
  xor: make 'xor_blocks' a library routine for use with async_tx
  dmaengine: make clients responsible for managing channels
  dmaengine: refactor dmaengine around dma_async_tx_descriptor
  ...

5 files changed, 439 insertions, 115 deletions

diff --git a/include/linux/async_tx.h b/include/linux/async_tx.h
new file mode 100644
index 00000000000..ff1255079fa
--- /dev/null
+++ b/include/linux/async_tx.h
@@ -0,0 +1,156 @@
+/*
+ * Copyright © 2006, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ */
+#ifndef _ASYNC_TX_H_
+#define _ASYNC_TX_H_
+#include <linux/dmaengine.h>
+#include <linux/spinlock.h>
+#include <linux/interrupt.h>
+
+/**
+ * dma_chan_ref - object used to manage dma channels received from the
+ *   dmaengine core.
+ * @chan - the channel being tracked
+ * @node - node for the channel to be placed on async_tx_master_list
+ * @rcu - for list_del_rcu
+ * @count - number of times this channel is listed in the pool
+ *	(for channels with multiple capabiities)
+ */
+struct dma_chan_ref {
+	struct dma_chan *chan;
+	struct list_head node;
+	struct rcu_head rcu;
+	atomic_t count;
+};
+
+/**
+ * async_tx_flags - modifiers for the async_* calls
+ * @ASYNC_TX_XOR_ZERO_DST: this flag must be used for xor operations where the
+ * the destination address is not a source.  The asynchronous case handles this
+ * implicitly, the synchronous case needs to zero the destination block.
+ * @ASYNC_TX_XOR_DROP_DST: this flag must be used if the destination address is
+ * also one of the source addresses.  In the synchronous case the destination
+ * address is an implied source, whereas the asynchronous case it must be listed
+ * as a source.  The destination address must be the first address in the source
+ * array.
+ * @ASYNC_TX_ASSUME_COHERENT: skip cache maintenance operations
+ * @ASYNC_TX_ACK: immediately ack the descriptor, precludes setting up a
+ * dependency chain
+ * @ASYNC_TX_DEP_ACK: ack the dependency descriptor.  Useful for chaining.
+ * @ASYNC_TX_KMAP_SRC: if the transaction is to be performed synchronously
+ * take an atomic mapping (KM_USER0) on the source page(s)
+ * @ASYNC_TX_KMAP_DST: if the transaction is to be performed synchronously
+ * take an atomic mapping (KM_USER0) on the dest page(s)
+ */
+enum async_tx_flags {
+	ASYNC_TX_XOR_ZERO_DST	 = (1 << 0),
+	ASYNC_TX_XOR_DROP_DST	 = (1 << 1),
+	ASYNC_TX_ASSUME_COHERENT = (1 << 2),
+	ASYNC_TX_ACK		 = (1 << 3),
+	ASYNC_TX_DEP_ACK	 = (1 << 4),
+	ASYNC_TX_KMAP_SRC	 = (1 << 5),
+	ASYNC_TX_KMAP_DST	 = (1 << 6),
+};
+
+#ifdef CONFIG_DMA_ENGINE
+void async_tx_issue_pending_all(void);
+enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx);
+void async_tx_run_dependencies(struct dma_async_tx_descriptor *tx);
+struct dma_chan *
+async_tx_find_channel(struct dma_async_tx_descriptor *depend_tx,
+	enum dma_transaction_type tx_type);
+#else
+static inline void async_tx_issue_pending_all(void)
+{
+	do { } while (0);
+}
+
+static inline enum dma_status
+dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
+{
+	return DMA_SUCCESS;
+}
+
+static inline void
+async_tx_run_dependencies(struct dma_async_tx_descriptor *tx,
+	struct dma_chan *host_chan)
+{
+	do { } while (0);
+}
+
+static inline struct dma_chan *
+async_tx_find_channel(struct dma_async_tx_descriptor *depend_tx,
+	enum dma_transaction_type tx_type)
+{
+	return NULL;
+}
+#endif
+
+/**
+ * async_tx_sync_epilog - actions to take if an operation is run synchronously
+ * @flags: async_tx flags
+ * @depend_tx: transaction depends on depend_tx
+ * @cb_fn: function to call when the transaction completes
+ * @cb_fn_param: parameter to pass to the callback routine
+ */
+static inline void
+async_tx_sync_epilog(unsigned long flags,
+	struct dma_async_tx_descriptor *depend_tx,
+	dma_async_tx_callback cb_fn, void *cb_fn_param)
+{
+	if (cb_fn)
+		cb_fn(cb_fn_param);
+
+	if (depend_tx && (flags & ASYNC_TX_DEP_ACK))
+		async_tx_ack(depend_tx);
+}
+
+void
+async_tx_submit(struct dma_chan *chan, struct dma_async_tx_descriptor *tx,
+	enum async_tx_flags flags, struct dma_async_tx_descriptor *depend_tx,
+	dma_async_tx_callback cb_fn, void *cb_fn_param);
+
+struct dma_async_tx_descriptor *
+async_xor(struct page *dest, struct page **src_list, unsigned int offset,
+	int src_cnt, size_t len, enum async_tx_flags flags,
+	struct dma_async_tx_descriptor *depend_tx,
+	dma_async_tx_callback cb_fn, void *cb_fn_param);
+
+struct dma_async_tx_descriptor *
+async_xor_zero_sum(struct page *dest, struct page **src_list,
+	unsigned int offset, int src_cnt, size_t len,
+	u32 *result, enum async_tx_flags flags,
+	struct dma_async_tx_descriptor *depend_tx,
+	dma_async_tx_callback cb_fn, void *cb_fn_param);
+
+struct dma_async_tx_descriptor *
+async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
+	unsigned int src_offset, size_t len, enum async_tx_flags flags,
+	struct dma_async_tx_descriptor *depend_tx,
+	dma_async_tx_callback cb_fn, void *cb_fn_param);
+
+struct dma_async_tx_descriptor *
+async_memset(struct page *dest, int val, unsigned int offset,
+	size_t len, enum async_tx_flags flags,
+	struct dma_async_tx_descriptor *depend_tx,
+	dma_async_tx_callback cb_fn, void *cb_fn_param);
+
+struct dma_async_tx_descriptor *
+async_trigger_callback(enum async_tx_flags flags,
+	struct dma_async_tx_descriptor *depend_tx,
+	dma_async_tx_callback cb_fn, void *cb_fn_param);
+#endif /* _ASYNC_TX_H_ */
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index c94d8f1d62e..a3b6035b6c8 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -21,29 +21,40 @@
 #ifndef DMAENGINE_H
 #define DMAENGINE_H
 
-#ifdef CONFIG_DMA_ENGINE
-
 #include <linux/device.h>
 #include <linux/uio.h>
 #include <linux/kref.h>
 #include <linux/completion.h>
 #include <linux/rcupdate.h>
+#include <linux/dma-mapping.h>
 
 /**
- * enum dma_event - resource PNP/power managment events
+ * enum dma_state - resource PNP/power managment state
  * @DMA_RESOURCE_SUSPEND: DMA device going into low power state
  * @DMA_RESOURCE_RESUME: DMA device returning to full power
- * @DMA_RESOURCE_ADDED: DMA device added to the system
+ * @DMA_RESOURCE_AVAILABLE: DMA device available to the system
  * @DMA_RESOURCE_REMOVED: DMA device removed from the system
  */
-enum dma_event {
+enum dma_state {
 	DMA_RESOURCE_SUSPEND,
 	DMA_RESOURCE_RESUME,
-	DMA_RESOURCE_ADDED,
+	DMA_RESOURCE_AVAILABLE,
 	DMA_RESOURCE_REMOVED,
 };
 
 /**
+ * enum dma_state_client - state of the channel in the client
+ * @DMA_ACK: client would like to use, or was using this channel
+ * @DMA_DUP: client has already seen this channel, or is not using this channel
+ * @DMA_NAK: client does not want to see any more channels
+ */
+enum dma_state_client {
+	DMA_ACK,
+	DMA_DUP,
+	DMA_NAK,
+};
+
+/**
  * typedef dma_cookie_t - an opaque DMA cookie
  *
  * if dma_cookie_t is >0 it's a DMA request cookie, <0 it's an error code
@@ -65,6 +76,31 @@ enum dma_status {
 };
 
 /**
+ * enum dma_transaction_type - DMA transaction types/indexes
+ */
+enum dma_transaction_type {
+	DMA_MEMCPY,
+	DMA_XOR,
+	DMA_PQ_XOR,
+	DMA_DUAL_XOR,
+	DMA_PQ_UPDATE,
+	DMA_ZERO_SUM,
+	DMA_PQ_ZERO_SUM,
+	DMA_MEMSET,
+	DMA_MEMCPY_CRC32C,
+	DMA_INTERRUPT,
+};
+
+/* last transaction type for creation of the capabilities mask */
+#define DMA_TX_TYPE_END (DMA_INTERRUPT + 1)
+
+/**
+ * dma_cap_mask_t - capabilities bitmap modeled after cpumask_t.
+ * See linux/cpumask.h
+ */
+typedef struct { DECLARE_BITMAP(bits, DMA_TX_TYPE_END); } dma_cap_mask_t;
+
+/**
  * struct dma_chan_percpu - the per-CPU part of struct dma_chan
  * @refcount: local_t used for open-coded "bigref" counting
  * @memcpy_count: transaction counter
@@ -80,7 +116,6 @@ struct dma_chan_percpu {
 
 /**
  * struct dma_chan - devices supply DMA channels, clients use them
- * @client: ptr to the client user of this chan, will be %NULL when unused
  * @device: ptr to the dma device who supplies this channel, always !%NULL
  * @cookie: last cookie value returned to client
  * @chan_id: channel ID for sysfs
@@ -88,12 +123,10 @@ struct dma_chan_percpu {
  * @refcount: kref, used in "bigref" slow-mode
  * @slow_ref: indicates that the DMA channel is free
  * @rcu: the DMA channel's RCU head
- * @client_node: used to add this to the client chan list
  * @device_node: used to add this to the device chan list
  * @local: per-cpu pointer to a struct dma_chan_percpu
  */
 struct dma_chan {
-	struct dma_client *client;
 	struct dma_device *device;
 	dma_cookie_t cookie;
 
@@ -105,11 +138,11 @@ struct dma_chan {
 	int slow_ref;
 	struct rcu_head rcu;
 
-	struct list_head client_node;
 	struct list_head device_node;
 	struct dma_chan_percpu *local;
 };
 
+
 void dma_chan_cleanup(struct kref *kref);
 
 static inline void dma_chan_get(struct dma_chan *chan)
@@ -134,169 +167,206 @@ static inline void dma_chan_put(struct dma_chan *chan)
 
 /*
  * typedef dma_event_callback - function pointer to a DMA event callback
+ * For each channel added to the system this routine is called for each client.
+ * If the client would like to use the channel it returns '1' to signal (ack)
+ * the dmaengine core to take out a reference on the channel and its
+ * corresponding device.  A client must not 'ack' an available channel more
+ * than once.  When a channel is removed all clients are notified.  If a client
+ * is using the channel it must 'ack' the removal.  A client must not 'ack' a
+ * removed channel more than once.
+ * @client - 'this' pointer for the client context
+ * @chan - channel to be acted upon
+ * @state - available or removed
  */
-typedef void (*dma_event_callback) (struct dma_client *client,
-		struct dma_chan *chan, enum dma_event event);
+struct dma_client;
+typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client,
+		struct dma_chan *chan, enum dma_state state);
 
 /**
  * struct dma_client - info on the entity making use of DMA services
  * @event_callback: func ptr to call when something happens
- * @chan_count: number of chans allocated
- * @chans_desired: number of chans requested. Can be +/- chan_count
- * @lock: protects access to the channels list
- * @channels: the list of DMA channels allocated
+ * @cap_mask: only return channels that satisfy the requested capabilities
+ *  a value of zero corresponds to any capability
  * @global_node: list_head for global dma_client_list
  */
 struct dma_client {
 	dma_event_callback	event_callback;
-	unsigned int		chan_count;
-	unsigned int		chans_desired;
-
-	spinlock_t		lock;
-	struct list_head	channels;
+	dma_cap_mask_t		cap_mask;
 	struct list_head	global_node;
 };
 
+typedef void (*dma_async_tx_callback)(void *dma_async_param);
+/**
+ * struct dma_async_tx_descriptor - async transaction descriptor
+ * ---dma generic offload fields---
+ * @cookie: tracking cookie for this transaction, set to -EBUSY if
+ *	this tx is sitting on a dependency list
+ * @ack: the descriptor can not be reused until the client acknowledges
+ *	receipt, i.e. has has a chance to establish any dependency chains
+ * @phys: physical address of the descriptor
+ * @tx_list: driver common field for operations that require multiple
+ *	descriptors
+ * @chan: target channel for this operation
+ * @tx_submit: set the prepared descriptor(s) to be executed by the engine
+ * @tx_set_dest: set a destination address in a hardware descriptor
+ * @tx_set_src: set a source address in a hardware descriptor
+ * @callback: routine to call after this operation is complete
+ * @callback_param: general parameter to pass to the callback routine
+ * ---async_tx api specific fields---
+ * @depend_list: at completion this list of transactions are submitted
+ * @depend_node: allow this transaction to be executed after another
+ *	transaction has completed, possibly on another channel
+ * @parent: pointer to the next level up in the dependency chain
+ * @lock: protect the dependency list
+ */
+struct dma_async_tx_descriptor {
+	dma_cookie_t cookie;
+	int ack;
+	dma_addr_t phys;
+	struct list_head tx_list;
+	struct dma_chan *chan;
+	dma_cookie_t (*tx_submit)(struct dma_async_tx_descriptor *tx);
+	void (*tx_set_dest)(dma_addr_t addr,
+		struct dma_async_tx_descriptor *tx, int index);
+	void (*tx_set_src)(dma_addr_t addr,
+		struct dma_async_tx_descriptor *tx, int index);
+	dma_async_tx_callback callback;
+	void *callback_param;
+	struct list_head depend_list;
+	struct list_head depend_node;
+	struct dma_async_tx_descriptor *parent;
+	spinlock_t lock;
+};
+
 /**
  * struct dma_device - info on the entity supplying DMA services
  * @chancnt: how many DMA channels are supported
  * @channels: the list of struct dma_chan
  * @global_node: list_head for global dma_device_list
+ * @cap_mask: one or more dma_capability flags
+ * @max_xor: maximum number of xor sources, 0 if no capability
  * @refcount: reference count
  * @done: IO completion struct
  * @dev_id: unique device ID
+ * @dev: struct device reference for dma mapping api
  * @device_alloc_chan_resources: allocate resources and return the
  *	number of allocated descriptors
  * @device_free_chan_resources: release DMA channel's resources
- * @device_memcpy_buf_to_buf: memcpy buf pointer to buf pointer
- * @device_memcpy_buf_to_pg: memcpy buf pointer to struct page
- * @device_memcpy_pg_to_pg: memcpy struct page/offset to struct page/offset
- * @device_memcpy_complete: poll the status of an IOAT DMA transaction
- * @device_memcpy_issue_pending: push appended descriptors to hardware
+ * @device_prep_dma_memcpy: prepares a memcpy operation
+ * @device_prep_dma_xor: prepares a xor operation
+ * @device_prep_dma_zero_sum: prepares a zero_sum operation
+ * @device_prep_dma_memset: prepares a memset operation
+ * @device_prep_dma_interrupt: prepares an end of chain interrupt operation
+ * @device_dependency_added: async_tx notifies the channel about new deps
+ * @device_issue_pending: push pending transactions to hardware
  */
 struct dma_device {
 
 	unsigned int chancnt;
 	struct list_head channels;
 	struct list_head global_node;
+	dma_cap_mask_t  cap_mask;
+	int max_xor;
 
 	struct kref refcount;
 	struct completion done;
 
 	int dev_id;
+	struct device *dev;
 
 	int (*device_alloc_chan_resources)(struct dma_chan *chan);
 	void (*device_free_chan_resources)(struct dma_chan *chan);
-	dma_cookie_t (*device_memcpy_buf_to_buf)(struct dma_chan *chan,
-			void *dest, void *src, size_t len);
-	dma_cookie_t (*device_memcpy_buf_to_pg)(struct dma_chan *chan,
-			struct page *page, unsigned int offset, void *kdata,
-			size_t len);
-	dma_cookie_t (*device_memcpy_pg_to_pg)(struct dma_chan *chan,
-			struct page *dest_pg, unsigned int dest_off,
-			struct page *src_pg, unsigned int src_off, size_t len);
-	enum dma_status (*device_memcpy_complete)(struct dma_chan *chan,
+
+	struct dma_async_tx_descriptor *(*device_prep_dma_memcpy)(
+		struct dma_chan *chan, size_t len, int int_en);
+	struct dma_async_tx_descriptor *(*device_prep_dma_xor)(
+		struct dma_chan *chan, unsigned int src_cnt, size_t len,
+		int int_en);
+	struct dma_async_tx_descriptor *(*device_prep_dma_zero_sum)(
+		struct dma_chan *chan, unsigned int src_cnt, size_t len,
+		u32 *result, int int_en);
+	struct dma_async_tx_descriptor *(*device_prep_dma_memset)(
+		struct dma_chan *chan, int value, size_t len, int int_en);
+	struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
+		struct dma_chan *chan);
+
+	void (*device_dependency_added)(struct dma_chan *chan);
+	enum dma_status (*device_is_tx_complete)(struct dma_chan *chan,
 			dma_cookie_t cookie, dma_cookie_t *last,
 			dma_cookie_t *used);
-	void (*device_memcpy_issue_pending)(struct dma_chan *chan);
+	void (*device_issue_pending)(struct dma_chan *chan);
 };
 
 /* --- public DMA engine API --- */
 
-struct dma_client *dma_async_client_register(dma_event_callback event_callback);
+void dma_async_client_register(struct dma_client *client);
 void dma_async_client_unregister(struct dma_client *client);
-void dma_async_client_chan_request(struct dma_client *client,
-		unsigned int number);
+void dma_async_client_chan_request(struct dma_client *client);
+dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan,
+	void *dest, void *src, size_t len);
+dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan,
+	struct page *page, unsigned int offset, void *kdata, size_t len);
+dma_cookie_t dma_async_memcpy_pg_to_pg(struct dma_chan *chan,
+	struct page *dest_pg, unsigned int dest_off, struct page *src_pg,
+	unsigned int src_off, size_t len);
+void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx,
+	struct dma_chan *chan);
 
-/**
- * dma_async_memcpy_buf_to_buf - offloaded copy between virtual addresses
- * @chan: DMA channel to offload copy to
- * @dest: destination address (virtual)
- * @src: source address (virtual)
- * @len: length
- *
- * Both @dest and @src must be mappable to a bus address according to the
- * DMA mapping API rules for streaming mappings.
- * Both @dest and @src must stay memory resident (kernel memory or locked
- * user space pages).
- */
-static inline dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan,
-	void *dest, void *src, size_t len)
+static inline void
+async_tx_ack(struct dma_async_tx_descriptor *tx)
 {
-	int cpu = get_cpu();
-	per_cpu_ptr(chan->local, cpu)->bytes_transferred += len;
-	per_cpu_ptr(chan->local, cpu)->memcpy_count++;
-	put_cpu();
-
-	return chan->device->device_memcpy_buf_to_buf(chan, dest, src, len);
+	tx->ack = 1;
 }
 
-/**
- * dma_async_memcpy_buf_to_pg - offloaded copy from address to page
- * @chan: DMA channel to offload copy to
- * @page: destination page
- * @offset: offset in page to copy to
- * @kdata: source address (virtual)
- * @len: length
- *
- * Both @page/@offset and @kdata must be mappable to a bus address according
- * to the DMA mapping API rules for streaming mappings.
- * Both @page/@offset and @kdata must stay memory resident (kernel memory or
- * locked user space pages)
- */
-static inline dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan,
-	struct page *page, unsigned int offset, void *kdata, size_t len)
+#define first_dma_cap(mask) __first_dma_cap(&(mask))
+static inline int __first_dma_cap(const dma_cap_mask_t *srcp)
 {
-	int cpu = get_cpu();
-	per_cpu_ptr(chan->local, cpu)->bytes_transferred += len;
-	per_cpu_ptr(chan->local, cpu)->memcpy_count++;
-	put_cpu();
+	return min_t(int, DMA_TX_TYPE_END,
+		find_first_bit(srcp->bits, DMA_TX_TYPE_END));
+}
 
-	return chan->device->device_memcpy_buf_to_pg(chan, page, offset,
-	                                             kdata, len);
+#define next_dma_cap(n, mask) __next_dma_cap((n), &(mask))
+static inline int __next_dma_cap(int n, const dma_cap_mask_t *srcp)
+{
+	return min_t(int, DMA_TX_TYPE_END,
+		find_next_bit(srcp->bits, DMA_TX_TYPE_END, n+1));
 }
 
-/**
- * dma_async_memcpy_pg_to_pg - offloaded copy from page to page
- * @chan: DMA channel to offload copy to
- * @dest_pg: destination page
- * @dest_off: offset in page to copy to
- * @src_pg: source page
- * @src_off: offset in page to copy from
- * @len: length
- *
- * Both @dest_page/@dest_off and @src_page/@src_off must be mappable to a bus
- * address according to the DMA mapping API rules for streaming mappings.
- * Both @dest_page/@dest_off and @src_page/@src_off must stay memory resident
- * (kernel memory or locked user space pages).
- */
-static inline dma_cookie_t dma_async_memcpy_pg_to_pg(struct dma_chan *chan,
-	struct page *dest_pg, unsigned int dest_off, struct page *src_pg,
-	unsigned int src_off, size_t len)
+#define dma_cap_set(tx, mask) __dma_cap_set((tx), &(mask))
+static inline void
+__dma_cap_set(enum dma_transaction_type tx_type, dma_cap_mask_t *dstp)
 {
-	int cpu = get_cpu();
-	per_cpu_ptr(chan->local, cpu)->bytes_transferred += len;
-	per_cpu_ptr(chan->local, cpu)->memcpy_count++;
-	put_cpu();
+	set_bit(tx_type, dstp->bits);
+}
 
-	return chan->device->device_memcpy_pg_to_pg(chan, dest_pg, dest_off,
-	                                            src_pg, src_off, len);
+#define dma_has_cap(tx, mask) __dma_has_cap((tx), &(mask))
+static inline int
+__dma_has_cap(enum dma_transaction_type tx_type, dma_cap_mask_t *srcp)
+{
+	return test_bit(tx_type, srcp->bits);
 }
 
+#define for_each_dma_cap_mask(cap, mask) \
+	for ((cap) = first_dma_cap(mask);	\
+		(cap) < DMA_TX_TYPE_END;	\
+		(cap) = next_dma_cap((cap), (mask)))
+
 /**
- * dma_async_memcpy_issue_pending - flush pending copies to HW
+ * dma_async_issue_pending - flush pending transactions to HW
  * @chan: target DMA channel
  *
  * This allows drivers to push copies to HW in batches,
  * reducing MMIO writes where possible.
  */
-static inline void dma_async_memcpy_issue_pending(struct dma_chan *chan)
+static inline void dma_async_issue_pending(struct dma_chan *chan)
 {
-	return chan->device->device_memcpy_issue_pending(chan);
+	return chan->device->device_issue_pending(chan);
 }
 
+#define dma_async_memcpy_issue_pending(chan) dma_async_issue_pending(chan)
+
 /**
- * dma_async_memcpy_complete - poll for transaction completion
+ * dma_async_is_tx_complete - poll for transaction completion
  * @chan: DMA channel
  * @cookie: transaction identifier to check status of
  * @last: returns last completed cookie, can be NULL
@@ -306,12 +376,15 @@ static inline void dma_async_memcpy_issue_pending(struct dma_chan *chan)
  * internal state and can be used with dma_async_is_complete() to check
  * the status of multiple cookies without re-checking hardware state.
  */
-static inline enum dma_status dma_async_memcpy_complete(struct dma_chan *chan,
+static inline enum dma_status dma_async_is_tx_complete(struct dma_chan *chan,
 	dma_cookie_t cookie, dma_cookie_t *last, dma_cookie_t *used)
 {
-	return chan->device->device_memcpy_complete(chan, cookie, last, used);
+	return chan->device->device_is_tx_complete(chan, cookie, last, used);
 }
 
+#define dma_async_memcpy_complete(chan, cookie, last, used)\
+	dma_async_is_tx_complete(chan, cookie, last, used)
+
 /**
  * dma_async_is_complete - test a cookie against chan state
  * @cookie: transaction identifier to test status of
@@ -334,6 +407,7 @@ static inline enum dma_status dma_async_is_complete(dma_cookie_t cookie,
 	return DMA_IN_PROGRESS;
 }
 
+enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie);
 
 /* --- DMA device --- */
 
@@ -362,5 +436,4 @@ dma_cookie_t dma_memcpy_pg_to_iovec(struct dma_chan *chan, struct iovec *iov,
 	struct dma_pinned_list *pinned_list, struct page *page,
 	unsigned int offset, size_t len);
 
-#endif /* CONFIG_DMA_ENGINE */
 #endif /* DMAENGINE_H */
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 9366182fffa..2c7add16953 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -479,6 +479,9 @@
 #define PCI_DEVICE_ID_IBM_ICOM_V2_ONE_PORT_RVX_ONE_PORT_MDM_PCIE 0x0361
 #define PCI_DEVICE_ID_IBM_ICOM_FOUR_PORT_MODEL	0x252
 
+#define PCI_VENDOR_ID_UNISYS		0x1018
+#define PCI_DEVICE_ID_UNISYS_DMA_DIRECTOR 0x001C
+
 #define PCI_VENDOR_ID_COMPEX2		0x101a /* pci.ids says "AT&T GIS (NCR)" */
 #define PCI_DEVICE_ID_COMPEX2_100VG	0x0005
 
diff --git a/include/linux/raid/raid5.h b/include/linux/raid/raid5.h
index d8286db60b9..93678f57ccb 100644
--- a/include/linux/raid/raid5.h
+++ b/include/linux/raid/raid5.h
@@ -116,13 +116,46 @@
  *  attach a request to an active stripe (add_stripe_bh())
  *     lockdev attach-buffer unlockdev
  *  handle a stripe (handle_stripe())
- *     lockstripe clrSTRIPE_HANDLE ... (lockdev check-buffers unlockdev) .. change-state .. record io needed unlockstripe schedule io
+ *     lockstripe clrSTRIPE_HANDLE ...
+ *		(lockdev check-buffers unlockdev) ..
+ *		change-state ..
+ *		record io/ops needed unlockstripe schedule io/ops
  *  release an active stripe (release_stripe())
  *     lockdev if (!--cnt) { if  STRIPE_HANDLE, add to handle_list else add to inactive-list } unlockdev
  *
  * The refcount counts each thread that have activated the stripe,
  * plus raid5d if it is handling it, plus one for each active request
- * on a cached buffer.
+ * on a cached buffer, and plus one if the stripe is undergoing stripe
+ * operations.
+ *
+ * Stripe operations are performed outside the stripe lock,
+ * the stripe operations are:
+ * -copying data between the stripe cache and user application buffers
+ * -computing blocks to save a disk access, or to recover a missing block
+ * -updating the parity on a write operation (reconstruct write and
+ *  read-modify-write)
+ * -checking parity correctness
+ * -running i/o to disk
+ * These operations are carried out by raid5_run_ops which uses the async_tx
+ * api to (optionally) offload operations to dedicated hardware engines.
+ * When requesting an operation handle_stripe sets the pending bit for the
+ * operation and increments the count.  raid5_run_ops is then run whenever
+ * the count is non-zero.
+ * There are some critical dependencies between the operations that prevent some
+ * from being requested while another is in flight.
+ * 1/ Parity check operations destroy the in cache version of the parity block,
+ *    so we prevent parity dependent operations like writes and compute_blocks
+ *    from starting while a check is in progress.  Some dma engines can perform
+ *    the check without damaging the parity block, in these cases the parity
+ *    block is re-marked up to date (assuming the check was successful) and is
+ *    not re-read from disk.
+ * 2/ When a write operation is requested we immediately lock the affected
+ *    blocks, and mark them as not up to date.  This causes new read requests
+ *    to be held off, as well as parity checks and compute block operations.
+ * 3/ Once a compute block operation has been requested handle_stripe treats
+ *    that block as if it is up to date.  raid5_run_ops guaruntees that any
+ *    operation that is dependent on the compute block result is initiated after
+ *    the compute block completes.
  */
 
 struct stripe_head {
@@ -136,15 +169,46 @@ struct stripe_head {
 	spinlock_t		lock;
 	int			bm_seq;	/* sequence number for bitmap flushes */
 	int			disks;			/* disks in stripe */
+	/* stripe_operations
+	 * @pending - pending ops flags (set for request->issue->complete)
+	 * @ack - submitted ops flags (set for issue->complete)
+	 * @complete - completed ops flags (set for complete)
+	 * @target - STRIPE_OP_COMPUTE_BLK target
+	 * @count - raid5_runs_ops is set to run when this is non-zero
+	 */
+	struct stripe_operations {
+		unsigned long	   pending;
+		unsigned long	   ack;
+		unsigned long	   complete;
+		int		   target;
+		int		   count;
+		u32		   zero_sum_result;
+	} ops;
 	struct r5dev {
 		struct bio	req;
 		struct bio_vec	vec;
 		struct page	*page;
-		struct bio	*toread, *towrite, *written;
+		struct bio	*toread, *read, *towrite, *written;
 		sector_t	sector;			/* sector of this page */
 		unsigned long	flags;
 	} dev[1]; /* allocated with extra space depending of RAID geometry */
 };
+
+/* stripe_head_state - collects and tracks the dynamic state of a stripe_head
+ *     for handle_stripe.  It is only valid under spin_lock(sh->lock);
+ */
+struct stripe_head_state {
+	int syncing, expanding, expanded;
+	int locked, uptodate, to_read, to_write, failed, written;
+	int to_fill, compute, req_compute, non_overwrite;
+	int failed_num;
+};
+
+/* r6_state - extra state data only relevant to r6 */
+struct r6_state {
+	int p_failed, q_failed, qd_idx, failed_num[2];
+};
+
 /* Flags */
 #define	R5_UPTODATE	0	/* page contains current data */
 #define	R5_LOCKED	1	/* IO has been submitted on "req" */
@@ -158,6 +222,15 @@ struct stripe_head {
 #define	R5_ReWrite	9	/* have tried to over-write the readerror */
 
 #define	R5_Expanded	10	/* This block now has post-expand data */
+#define	R5_Wantcompute	11 /* compute_block in progress treat as
+				    * uptodate
+				    */
+#define	R5_Wantfill	12 /* dev->toread contains a bio that needs
+				    * filling
+				    */
+#define	R5_Wantprexor	13 /* distinguish blocks ready for rmw from
+				    * other "towrites"
+				    */
 /*
  * Write method
  */
@@ -180,6 +253,24 @@ struct stripe_head {
 #define	STRIPE_EXPAND_SOURCE	10
 #define	STRIPE_EXPAND_READY	11
 /*
+ * Operations flags (in issue order)
+ */
+#define STRIPE_OP_BIOFILL	0
+#define STRIPE_OP_COMPUTE_BLK	1
+#define STRIPE_OP_PREXOR	2
+#define STRIPE_OP_BIODRAIN	3
+#define STRIPE_OP_POSTXOR	4
+#define STRIPE_OP_CHECK	5
+#define STRIPE_OP_IO		6
+
+/* modifiers to the base operations
+ * STRIPE_OP_MOD_REPAIR_PD - compute the parity block and write it back
+ * STRIPE_OP_MOD_DMA_CHECK - parity is not corrupted by the check
+ */
+#define STRIPE_OP_MOD_REPAIR_PD 7
+#define STRIPE_OP_MOD_DMA_CHECK 8
+
+/*
  * Plugging:
  *
  * To improve write throughput, we need to delay the handling of some
diff --git a/include/linux/raid/xor.h b/include/linux/raid/xor.h
index f0d67cbdea4..3e120587ead 100644
--- a/include/linux/raid/xor.h
+++ b/include/linux/raid/xor.h
@@ -3,9 +3,10 @@
 
 #include <linux/raid/md.h>
 
-#define MAX_XOR_BLOCKS 5
+#define MAX_XOR_BLOCKS 4
 
-extern void xor_block(unsigned int count, unsigned int bytes, void **ptr);
+extern void xor_blocks(unsigned int count, unsigned int bytes,
+	void *dest, void **srcs);
 
 struct xor_block_template {
         struct xor_block_template *next;