UBUNTU: SAUCE: (no-up) ubuntu: dm-raid45

ExternalDriver: dm-raid45
Description: This software extends device-mapper by RAID4 and RAID5 mappings.
Url: http://people.redhat.com/~heinzm/sw/dm/dm-raid45/
Version: 2009.04.24 (2.6.30-rc3)
Signed-off-by: Manoj Iyer <manoj.iyer@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

ubuntu/Kconfig

@@ -10,6 +10,7 @@ menu "Ubuntu Supplied Third-Party Device Drivers"
 ##
 ##
 ##
+source "ubuntu/dm-raid4-5/Kconfig"
 ##
 ##
 ##

ubuntu/Makefile

@@ -12,6 +12,7 @@
 ##
 ##
 ##
+obj-$(CONFIG_DM_RAID45)		+= dm-raid4-5/
 ##
 ##
 ##

ubuntu/dm-raid4-5/BOM

+Downloaded from:	http://people.redhat.com/~heinzm/sw/dm/dm-raid45/
+Current Version:	2009.04.24 (2.6.30-rc3)
+Comments:		All of the patches to dmraid1/dm-log, etc are upstream.

ubuntu/dm-raid4-5/Kconfig

+config DM_RAID45
+	tristate "RAID 4/5 target (EXPERIMENTAL)"
+	depends on BLK_DEV_DM && XOR_BLOCKS && EXPERIMENTAL
+	default m
+	---help---
+	A target that supports RAID4 and RAID5 mappings.

ubuntu/dm-raid4-5/Makefile

+EXTRA_CFLAGS += -I$(srctree)/drivers/md
+
+obj-$(CONFIG_DM_RAID45) := dm-raid45.o
+dm-raid45-objs := dm-raid4-5.o dm-memcache.o dm-region-hash.o dm-message.o

ubuntu/dm-raid4-5/dm-memcache.c

+/*
+ * Copyright (C) 2006-2008 Red Hat, Inc. All rights reserved.
+ *
+ * Module Author: Heinz Mauelshagen <heinzm@redhat.com>
+ *
+ * Device-mapper memory object handling:
+ *
+ * o allocate/free total_pages in a per client page pool.
+ *
+ * o allocate/free memory objects with chunks (1..n) of
+ *   pages_per_chunk pages hanging off.
+ *
+ * This file is released under the GPL.
+ */
+
+#define	DM_MEM_CACHE_VERSION	"0.2"
+
+#include "dm.h"
+#include "dm-memcache.h"
+#include <linux/dm-io.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+
+struct dm_mem_cache_client {
+	spinlock_t lock;
+	mempool_t *objs_pool;
+	struct page_list *free_list;
+	unsigned objects;
+	unsigned chunks;
+	unsigned pages_per_chunk;
+	unsigned free_pages;
+	unsigned total_pages;
+};
+
+/*
+ * Free pages and page_list elements of client.
+ */
+static void free_cache_pages(struct page_list *list)
+{
+	while (list) {
+		struct page_list *pl = list;
+
+		list = pl->next;
+		BUG_ON(!pl->page);
+		__free_page(pl->page);
+		kfree(pl);
+	}
+}
+
+/*
+ * Alloc number of pages and page_list elements as required by client.
+ */
+static struct page_list *alloc_cache_pages(unsigned pages)
+{
+	struct page_list *pl, *ret = NULL;
+	struct page *page;
+
+	while (pages--) {
+		page = alloc_page(GFP_NOIO);
+		if (!page)
+			goto err;
+
+		pl = kmalloc(sizeof(*pl), GFP_NOIO);
+		if (!pl) {
+			__free_page(page);
+			goto err;
+		}
+
+		pl->page = page;
+		pl->next = ret;
+		ret = pl;
+	}
+
+	return ret;
+
+err:
+	free_cache_pages(ret);
+	return NULL;
+}
+
+/*
+ * Allocate page_list elements from the pool to chunks of the memory object.
+ */
+static void alloc_chunks(struct dm_mem_cache_client *cl,
+			 struct dm_mem_cache_object *obj)
+{
+	unsigned chunks = cl->chunks;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	local_irq_disable();
+	while (chunks--) {
+		unsigned p = cl->pages_per_chunk;
+
+		obj[chunks].pl = NULL;
+
+		while (p--) {
+			struct page_list *pl;
+
+			/* Take next element from free list */
+			spin_lock(&cl->lock);
+			pl = cl->free_list;
+			BUG_ON(!pl);
+			cl->free_list = pl->next;
+			spin_unlock(&cl->lock);
+
+			pl->next = obj[chunks].pl;
+			obj[chunks].pl = pl;
+		}
+	}
+
+	local_irq_restore(flags);
+}
+
+/*
+ * Free page_list elements putting them back onto free list
+ */
+static void free_chunks(struct dm_mem_cache_client *cl,
+			struct dm_mem_cache_object *obj)
+{
+	unsigned chunks = cl->chunks;
+	unsigned long flags;
+	struct page_list *next, *pl;
+
+	local_irq_save(flags);
+	local_irq_disable();
+	while (chunks--) {
+		for (pl = obj[chunks].pl; pl; pl = next) {
+			next = pl->next;
+
+			spin_lock(&cl->lock);
+			pl->next = cl->free_list;
+			cl->free_list = pl;
+			cl->free_pages++;
+			spin_unlock(&cl->lock);
+		}
+	}
+
+	local_irq_restore(flags);
+}
+
+/*
+ * Create/destroy dm memory cache client resources.
+ */
+struct dm_mem_cache_client *
+dm_mem_cache_client_create(unsigned objects, unsigned chunks,
+			   unsigned pages_per_chunk)
+{
+	unsigned total_pages = objects * chunks * pages_per_chunk;
+	struct dm_mem_cache_client *client;
+
+	BUG_ON(!total_pages);
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
+	if (!client)
+		return ERR_PTR(-ENOMEM);
+
+	client->objs_pool = mempool_create_kmalloc_pool(objects,
+				chunks * sizeof(struct dm_mem_cache_object));
+	if (!client->objs_pool)
+		goto err;
+
+	client->free_list = alloc_cache_pages(total_pages);
+	if (!client->free_list)
+		goto err1;
+
+	spin_lock_init(&client->lock);
+	client->objects = objects;
+	client->chunks = chunks;
+	client->pages_per_chunk = pages_per_chunk;
+	client->free_pages = client->total_pages = total_pages;
+	return client;
+
+err1:
+	mempool_destroy(client->objs_pool);
+err:
+	kfree(client);
+	return ERR_PTR(-ENOMEM);
+}
+EXPORT_SYMBOL(dm_mem_cache_client_create);
+
+void dm_mem_cache_client_destroy(struct dm_mem_cache_client *cl)
+{
+	BUG_ON(cl->free_pages != cl->total_pages);
+	free_cache_pages(cl->free_list);
+	mempool_destroy(cl->objs_pool);
+	kfree(cl);
+}
+EXPORT_SYMBOL(dm_mem_cache_client_destroy);
+
+/*
+ * Grow a clients cache by an amount of pages.
+ *
+ * Don't call from interrupt context!
+ */
+int dm_mem_cache_grow(struct dm_mem_cache_client *cl, unsigned objects)
+{
+	unsigned pages = objects * cl->chunks * cl->pages_per_chunk;
+	struct page_list *pl, *last;
+
+	BUG_ON(!pages);
+	pl = alloc_cache_pages(pages);
+	if (!pl)
+		return -ENOMEM;
+
+	last = pl;
+	while (last->next)
+		last = last->next;
+
+	spin_lock_irq(&cl->lock);
+	last->next = cl->free_list;
+	cl->free_list = pl;
+	cl->free_pages += pages;
+	cl->total_pages += pages;
+	cl->objects++;
+	spin_unlock_irq(&cl->lock);
+
+	mempool_resize(cl->objs_pool, cl->objects, GFP_NOIO);
+	return 0;
+}
+EXPORT_SYMBOL(dm_mem_cache_grow);
+
+/* Shrink a clients cache by an amount of pages */
+int dm_mem_cache_shrink(struct dm_mem_cache_client *cl, unsigned objects)
+{
+	int r;
+	unsigned pages = objects * cl->chunks * cl->pages_per_chunk, p = pages;
+	unsigned long flags;
+	struct page_list *last = NULL, *pl, *pos;
+
+	BUG_ON(!pages);
+
+	spin_lock_irqsave(&cl->lock, flags);
+	pl = pos = cl->free_list;
+	while (p-- && pos->next) {
+		last = pos;
+		pos = pos->next;
+	}
+
+	if (++p)
+		r = -ENOMEM;
+	else {
+		r = 0;
+		cl->free_list = pos;
+		cl->free_pages -= pages;
+		cl->total_pages -= pages;
+		cl->objects--;
+		last->next = NULL;
+	}
+	spin_unlock_irqrestore(&cl->lock, flags);
+
+	if (!r) {
+		free_cache_pages(pl);
+		mempool_resize(cl->objs_pool, cl->objects, GFP_NOIO);
+	}
+
+	return r;
+}
+EXPORT_SYMBOL(dm_mem_cache_shrink);
+
+/*
+ * Allocate/free a memory object
+ *
+ * Can be called from interrupt context
+ */
+struct dm_mem_cache_object *dm_mem_cache_alloc(struct dm_mem_cache_client *cl)
+{
+	int r = 0;
+	unsigned pages = cl->chunks * cl->pages_per_chunk;
+	unsigned long flags;
+	struct dm_mem_cache_object *obj;
+
+	obj = mempool_alloc(cl->objs_pool, GFP_NOIO);
+	if (!obj)
+		return ERR_PTR(-ENOMEM);
+
+	spin_lock_irqsave(&cl->lock, flags);
+	if (pages > cl->free_pages)
+		r = -ENOMEM;
+	else
+		cl->free_pages -= pages;
+	spin_unlock_irqrestore(&cl->lock, flags);
+
+	if (r) {
+		mempool_free(obj, cl->objs_pool);
+		return ERR_PTR(r);
+	}
+
+	alloc_chunks(cl, obj);
+	return obj;
+}
+EXPORT_SYMBOL(dm_mem_cache_alloc);
+
+void dm_mem_cache_free(struct dm_mem_cache_client *cl,
+		       struct dm_mem_cache_object *obj)
+{
+	free_chunks(cl, obj);
+	mempool_free(obj, cl->objs_pool);
+}
+EXPORT_SYMBOL(dm_mem_cache_free);
+
+MODULE_DESCRIPTION(DM_NAME " dm memory cache");
+MODULE_AUTHOR("Heinz Mauelshagen <hjm@redhat.com>");
+MODULE_LICENSE("GPL");

ubuntu/dm-raid4-5/dm-memcache.h

+/*
+ * Copyright (C) 2006-2008 Red Hat, Inc. All rights reserved.
+ *
+ * Module Author: Heinz Mauelshagen <Mauelshagen@RedHat.com>
+ *
+ * Device-mapper memory object handling:
+ *
+ * o allocate/free total_pages in a per client page pool.
+ *
+ * o allocate/free memory objects with chunks (1..n) of
+ *   pages_per_chunk pages hanging off.
+ *
+ * This file is released under the GPL.
+ */
+
+#ifndef _DM_MEM_CACHE_H
+#define _DM_MEM_CACHE_H
+
+#define	DM_MEM_CACHE_H_VERSION	"0.1"
+
+#include "dm.h"
+#include <linux/dm-io.h>
+
+static inline struct page_list *pl_elem(struct page_list *pl, unsigned p)
+{
+	while (pl && p--)
+		pl = pl->next;
+
+	return pl;
+}
+
+struct dm_mem_cache_object {
+	struct page_list *pl; /* Dynamically allocated array */
+	void *private;	      /* Caller context reference */
+};
+
+struct dm_mem_cache_client;
+
+/*
+ * Create/destroy dm memory cache client resources.
+ *
+ * On creation, a number of @objects with @chunks of
+ * @pages_per_chunk pages will be allocated.
+ */
+struct dm_mem_cache_client *
+dm_mem_cache_client_create(unsigned objects, unsigned chunks,
+			   unsigned pages_per_chunk);
+void dm_mem_cache_client_destroy(struct dm_mem_cache_client *client);
+
+/*
+ * Grow/shrink a dm memory cache client resources
+ * by @objetcs amount of objects.
+ */
+int dm_mem_cache_grow(struct dm_mem_cache_client *client, unsigned objects);
+int dm_mem_cache_shrink(struct dm_mem_cache_client *client, unsigned objects);
+
+/*
+ * Allocate/free a memory object
+ *
+ * On allocation one object with an amount of chunks and
+ * an amount of pages per chunk will be returned on success.
+ */
+struct dm_mem_cache_object *
+dm_mem_cache_alloc(struct dm_mem_cache_client *client);
+void dm_mem_cache_free(struct dm_mem_cache_client *client,
+		       struct dm_mem_cache_object *object);
+
+#endif

ubuntu/dm-raid4-5/dm-message.c

+/*
+ * Copyright (C) 2007,2008 Red Hat Inc. All rights reserved.
+ *
+ * Module Author: Heinz Mauelshagen <heinzm@redhat.com>
+ *
+ * General device-mapper message interface argument parser.
+ *
+ * This file is released under the GPL.
+ *
+ * device-mapper message parser.
+ *
+ */
+
+#include "dm.h"
+#include "dm-message.h"
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#define DM_MSG_PREFIX	"dm_message"
+
+/* Basename of a path. */
+static inline char *
+basename(char *s)
+{
+	char *p = strrchr(s, '/');
+
+	return p ? p + 1 : s;
+}
+
+/* Get an argument depending on type. */
+static void
+message_arguments(struct dm_msg *msg, int argc, char **argv)
+{
+
+	if (argc) {
+		int i;
+		struct dm_message_argument *args = msg->spec->args;
+
+		for (i = 0; i < args->num_args; i++) {
+			int r;
+			unsigned long **ptr = args->ptr;
+			enum dm_message_argument_type type = args->types[i];
+
+			switch (type) {
+			case dm_msg_base_t:
+				((char **) ptr)[i] = basename(argv[i]);
+				break;
+
+			case dm_msg_str_t:
+				((char **) ptr)[i] = argv[i];
+				break;
+
+			case dm_msg_int_t:
+				r = sscanf(argv[i], "%d", ((int **) ptr)[i]);
+				goto check;
+
+			case dm_msg_uint_t:
+				r = sscanf(argv[i], "%u",
+					   ((unsigned **) ptr)[i]);
+				goto check;
+
+			case dm_msg_uint64_t:
+				r = sscanf(argv[i], "%llu",
+					   ((unsigned long long **) ptr)[i]);
+
+check:
+				if (r != 1) {
+					set_bit(dm_msg_ret_undef, &msg->ret);
+					set_bit(dm_msg_ret_arg, &msg->ret);
+				}
+			}
+		}
+	}
+}
+
+/* Parse message options. */
+static void
+message_options_parse(struct dm_msg *msg, int argc, char **argv)
+{
+	int hit = 0;
+	unsigned long *action;
+	size_t l1 = strlen(*argv), l_hit = 0;
+	struct dm_message_option *o = msg->spec->options;
+	char **option, **option_end = o->options + o->num_options;
+
+	for (option = o->options, action = o->actions;
+	     option < option_end; option++, action++) {
+		size_t l2 = strlen(*option);
+
+		if (!strnicmp(*argv, *option, min(l1, l2))) {
+			hit++;
+			l_hit = l2;
+			set_bit(*action, &msg->action);
+		}
+	}
+
+	/* Assume error. */
+	msg->ret = 0;
+	set_bit(dm_msg_ret_option, &msg->ret);
+	if (!hit || l1 > l_hit)
+		set_bit(dm_msg_ret_undef, &msg->ret);	/* Undefined option. */
+	else if (hit > 1)
+		set_bit(dm_msg_ret_ambiguous, &msg->ret); /* Ambiguous option.*/
+	else {
+		clear_bit(dm_msg_ret_option, &msg->ret); /* Option OK. */
+		message_arguments(msg, --argc, ++argv);
+	}
+}
+
+static inline void
+print_ret(const char *caller, unsigned long ret)
+{
+	struct {
+		unsigned long err;
+		const char *err_str;
+	} static err_msg[] = {
+		{ dm_msg_ret_ambiguous, "message ambiguous" },
+		{ dm_msg_ret_inval, "message invalid" },
+		{ dm_msg_ret_undef, "message undefined" },
+		{ dm_msg_ret_arg, "message argument" },
+		{ dm_msg_ret_argcount, "message argument count" },
+		{ dm_msg_ret_option, "option" },
+	}, *e = ARRAY_END(err_msg);
+
+	while (e-- > err_msg) {
+		if (test_bit(e->err, &ret))
+			DMERR("%s %s", caller, e->err_str);
+	}
+}
+
+/* Parse a message action. */
+int
+dm_message_parse(const char *caller, struct dm_msg *msg, void *context,
+		 int argc, char **argv)
+{
+	int hit = 0;
+	size_t l1, l_hit = 0;
+	struct dm_msg_spec *s, *s_hit = NULL,
+			   *s_end = msg->specs + msg->num_specs;
+
+	if (argc < 2)
+		return -EINVAL;
+
+	l1 = strlen(*argv);
+	for (s = msg->specs; s < s_end; s++) {
+		size_t l2 = strlen(s->cmd);
+
+		if (!strnicmp(*argv, s->cmd, min(l1, l2))) {
+			hit++;
+			l_hit = l2;
+			s_hit = s;
+		}
+	}
+
+	msg->ret = 0;
+	if (!hit || l1 > l_hit)	/* No hit or message string too long. */
+		set_bit(dm_msg_ret_undef, &msg->ret);
+	else if (hit > 1)	/* Ambiguous message. */
+		set_bit(dm_msg_ret_ambiguous, &msg->ret);
+	else if (argc - 2 != s_hit->args->num_args) {
+		set_bit(dm_msg_ret_undef, &msg->ret);
+		set_bit(dm_msg_ret_argcount, &msg->ret);
+	}
+
+	if (msg->ret)
+		goto bad;
+
+	msg->action = 0;
+	msg->spec = s_hit;
+	set_bit(s_hit->action, &msg->action);
+	message_options_parse(msg, --argc, ++argv);
+
+	if (!msg->ret)
+		return msg->spec->f(msg, context);
+
+bad:
+	print_ret(caller, msg->ret);
+	return -EINVAL;
+}
+EXPORT_SYMBOL(dm_message_parse);
+
+MODULE_DESCRIPTION(DM_NAME " device-mapper target message parser");
+MODULE_AUTHOR("Heinz Mauelshagen <hjm@redhat.com>");
+MODULE_LICENSE("GPL");

ubuntu/dm-raid4-5/dm-message.h

+/*
+ * Copyright (C) 2007,2008 Red Hat, Inc. All rights reserved.
+ *
+ * Module Author: Heinz Mauelshagen <Mauelshagen@RedHat.de>
+ *
+ * General device-mapper message interface argument parser.
+ *
+ * This file is released under the GPL.
+ *
+ */
+
+#ifndef DM_MESSAGE_H
+#define DM_MESSAGE_H
+
+/* Factor out to dm.h. */
+/* Reference to array end. */
+#define ARRAY_END(a)    ((a) + ARRAY_SIZE(a))
+
+/* Message return bits. */
+enum dm_message_return {
+	dm_msg_ret_ambiguous,		/* Action ambiguous. */
+	dm_msg_ret_inval,		/* Action invalid. */
+	dm_msg_ret_undef,		/* Action undefined. */
+
+	dm_msg_ret_option,		/* Option error. */
+	dm_msg_ret_arg,			/* Argument error. */
+	dm_msg_ret_argcount,		/* Argument count error. */
+};
+
+/* Message argument type conversions. */
+enum dm_message_argument_type {
+	dm_msg_base_t,		/* Basename string. */
+	dm_msg_str_t,		/* String. */
+	dm_msg_int_t,		/* Signed int. */
+	dm_msg_uint_t,		/* Unsigned int. */
+	dm_msg_uint64_t,	/* Unsigned int 64. */
+};
+
+/* A message option. */
+struct dm_message_option {
+	unsigned num_options;
+	char **options;
+	unsigned long *actions;
+};
+
+/* Message arguments and types. */
+struct dm_message_argument {
+	unsigned num_args;
+	unsigned long **ptr;
+	enum dm_message_argument_type types[];
+};
+
+/* Client message. */
+struct dm_msg {
+	unsigned long action;		/* Identified action. */
+	unsigned long ret;		/* Return bits. */
+	unsigned num_specs;		/* # of sepcifications listed. */
+	struct dm_msg_spec *specs;	/* Specification list. */
+	struct dm_msg_spec *spec;	/* Specification selected. */
+};
+
+/* Secification of the message. */
+struct dm_msg_spec {
+	const char *cmd;	/* Name of the command (i.e. 'bandwidth'). */
+	unsigned long action;
+	struct dm_message_option *options;
+	struct dm_message_argument *args;
+	unsigned long parm;	/* Parameter to pass through to callback. */
+	/* Function to process for action. */
+	int (*f) (struct dm_msg *msg, void *context);
+};
+
+/* Parameter access macros. */
+#define	DM_MSG_PARM(msg) ((msg)->spec->parm)
+
+#define	DM_MSG_STR_ARGS(msg, idx) ((char *) *(msg)->spec->args->ptr[idx])
+#define	DM_MSG_INT_ARGS(msg, idx) ((int) *(msg)->spec->args->ptr[idx])
+#define	DM_MSG_UINT_ARGS(msg, idx) ((unsigned) DM_MSG_INT_ARG(msg, idx))
+#define	DM_MSG_UINT64_ARGS(msg, idx) ((uint64_t)  *(msg)->spec->args->ptr[idx])
+
+#define	DM_MSG_STR_ARG(msg)	DM_MSG_STR_ARGS(msg, 0)
+#define	DM_MSG_INT_ARG(msg)	DM_MSG_INT_ARGS(msg, 0)
+#define	DM_MSG_UINT_ARG(msg)	DM_MSG_UINT_ARGS(msg, 0)
+#define	DM_MSG_UINT64_ARG(msg)	DM_MSG_UINT64_ARGS(msg, 0)
+
+
+/* Parse a message and its options and optionally call a function back. */
+int dm_message_parse(const char *caller, struct dm_msg *msg, void *context,
+		     int argc, char **argv);
+
+#endif

ubuntu/dm-raid4-5/dm-raid4-5.h

+/*
+ * Copyright (C) 2006  Red Hat GmbH
+ *
+ * Module Author: Heinz Mauelshagen (Mauelshagen@RedHat.com)
+ *
+ * This file is released under the GPL.
+ *
+ */
+
+#ifndef _DM_RAID45_H
+#define _DM_RAID45_H
+
+/* Factor out to dm.h! */
+#define	STR_LEN(ptr, str) ptr, str, strlen(ptr)
+
+enum lock_type { RAID45_EX, RAID45_SHARED };
+
+struct dmraid45_locking_type {
+        /* Request a lock on a stripe. */
+        void* (*lock)(sector_t key, enum lock_type type);
+
+        /* Release a lock on a stripe. */
+        void (*unlock)(void *lock_handle);
+
+};
+
+#endif

ubuntu/dm-raid4-5/dm-raid45.h

+/*
+ * Copyright (C) 2006-2008 Red Hat, Inc. All rights reserved.
+ *
+ * Module Author: Heinz Mauelshagen (Mauelshagen@RedHat.com)
+ *
+ * Locking definitions for the device-mapper RAID45 target.
+ *
+ * This file is released under the GPL.
+ *
+ */
+
+#ifndef _DM_RAID45_H
+#define _DM_RAID45_H
+
+/* Factor out to dm.h! */
+#define	STR_LEN(ptr, str)	(ptr), (str), strlen((ptr))
+
+enum dm_lock_type { DM_RAID45_EX, DM_RAID45_SHARED };
+
+struct dm_raid45_locking_type {
+	/* Request a lock on a stripe. */
+	void* (*lock)(sector_t key, enum dm_lock_type type);
+
+	/* Release a lock on a stripe. */
+	void (*unlock)(void *lock_handle);
+};
+
+#endif

ubuntu/dm-raid4-5/dm-region-hash.c

+/*
+ * Copyright (C) 2003 Sistina Software Limited.
+ * Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved.
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/dm-dirty-log.h>
+#include "dm-region-hash.h"
+
+#include <linux/ctype.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+
+#include "dm.h"
+
+#define	DM_MSG_PREFIX	"region hash"
+
+/*-----------------------------------------------------------------
+ * Region hash
+ *
+ * The mirror splits itself up into discrete regions.  Each
+ * region can be in one of three states: clean, dirty,
+ * nosync.  There is no need to put clean regions in the hash.
+ *
+ * In addition to being present in the hash table a region _may_
+ * be present on one of three lists.
+ *
+ *   clean_regions: Regions on this list have no io pending to
+ *   them, they are in sync, we are no longer interested in them,
+ *   they are dull.  dm_rh_update_states() will remove them from the
+ *   hash table.
+ *
+ *   quiesced_regions: These regions have been spun down, ready
+ *   for recovery.  rh_recovery_start() will remove regions from
+ *   this list and hand them to kmirrord, which will schedule the
+ *   recovery io with kcopyd.
+ *
+ *   recovered_regions: Regions that kcopyd has successfully
+ *   recovered.  dm_rh_update_states() will now schedule any delayed
+ *   io, up the recovery_count, and remove the region from the
+ *   hash.
+ *
+ * There are 2 locks:
+ *   A rw spin lock 'hash_lock' protects just the hash table,
+ *   this is never held in write mode from interrupt context,
+ *   which I believe means that we only have to disable irqs when
+ *   doing a write lock.
+ *
+ *   An ordinary spin lock 'region_lock' that protects the three
+ *   lists in the region_hash, with the 'state', 'list' and
+ *   'delayed_bios' fields of the regions.  This is used from irq
+ *   context, so all other uses will have to suspend local irqs.
+ *---------------------------------------------------------------*/
+struct dm_region_hash {
+	uint32_t region_size;
+	unsigned region_shift;
+
+	/* holds persistent region state */
+	struct dm_dirty_log *log;
+
+	/* hash table */
+	rwlock_t hash_lock;
+	mempool_t *region_pool;
+	unsigned mask;
+	unsigned nr_buckets;
+	unsigned prime;
+	unsigned shift;
+	struct list_head *buckets;
+
+	unsigned max_recovery; /* Max # of regions to recover in parallel */
+
+	spinlock_t region_lock;
+	atomic_t recovery_in_flight;
+	struct semaphore recovery_count;
+	struct list_head clean_regions;
+	struct list_head quiesced_regions;
+	struct list_head recovered_regions;
+	struct list_head failed_recovered_regions;
+
+	void *context;
+	sector_t target_begin;
+
+	/* Callback function to schedule bios writes */
+	void (*dispatch_bios)(void *context, struct bio_list *bios);
+
+	/* Callback function to wakeup callers worker thread. */
+	void (*wakeup_workers)(void *context);
+
+	/* Callback function to wakeup callers recovery waiters. */
+	void (*wakeup_all_recovery_waiters)(void *context);
+};
+
+struct dm_region {
+	struct dm_region_hash *rh;	/* FIXME: can we get rid of this ? */
+	region_t key;
+	int state;
+
+	struct list_head hash_list;
+	struct list_head list;
+
+	atomic_t pending;
+	struct bio_list delayed_bios;
+};
+
+/*
+ * Conversion fns
+ */
+region_t dm_rh_sector_to_region(struct dm_region_hash *rh, sector_t sector)
+{
+	return sector >> rh->region_shift;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_sector_to_region);
+
+sector_t dm_rh_region_to_sector(struct dm_region_hash *rh, region_t region)
+{
+	return region << rh->region_shift;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_region_to_sector);
+
+region_t dm_rh_bio_to_region(struct dm_region_hash *rh, struct bio *bio)
+{
+	return dm_rh_sector_to_region(rh, bio->bi_sector - rh->target_begin);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_bio_to_region);
+
+void *dm_rh_region_context(struct dm_region *reg)
+{
+	return reg->rh->context;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_region_context);
+
+region_t dm_rh_get_region_key(struct dm_region *reg)
+{
+	return reg->key;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_get_region_key);
+
+sector_t dm_rh_get_region_size(struct dm_region_hash *rh)
+{
+	return rh->region_size;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_get_region_size);
+
+/*
+ * FIXME: shall we pass in a structure instead of all these args to
+ * dm_region_hash_create()????
+ */
+#define RH_HASH_MULT 2654435387U
+#define RH_HASH_SHIFT 12
+
+#define MIN_REGIONS 64
+struct dm_region_hash *dm_region_hash_create(
+		void *context, void (*dispatch_bios)(void *context,
+						     struct bio_list *bios),
+		void (*wakeup_workers)(void *context),
+		void (*wakeup_all_recovery_waiters)(void *context),
+		sector_t target_begin, unsigned max_recovery,
+		struct dm_dirty_log *log, uint32_t region_size,
+		region_t nr_regions)
+{
+	struct dm_region_hash *rh;
+	unsigned nr_buckets, max_buckets;
+	size_t i;
+
+	/*
+	 * Calculate a suitable number of buckets for our hash
+	 * table.
+	 */
+	max_buckets = nr_regions >> 6;
+	for (nr_buckets = 128u; nr_buckets < max_buckets; nr_buckets <<= 1)
+		;
+	nr_buckets >>= 1;
+
+	rh = kmalloc(sizeof(*rh), GFP_KERNEL);
+	if (!rh) {
+		DMERR("unable to allocate region hash memory");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	rh->context = context;
+	rh->dispatch_bios = dispatch_bios;
+	rh->wakeup_workers = wakeup_workers;
+	rh->wakeup_all_recovery_waiters = wakeup_all_recovery_waiters;
+	rh->target_begin = target_begin;
+	rh->max_recovery = max_recovery;
+	rh->log = log;
+	rh->region_size = region_size;
+	rh->region_shift = ffs(region_size) - 1;
+	rwlock_init(&rh->hash_lock);
+	rh->mask = nr_buckets - 1;
+	rh->nr_buckets = nr_buckets;
+
+	rh->shift = RH_HASH_SHIFT;
+	rh->prime = RH_HASH_MULT;
+
+	rh->buckets = vmalloc(nr_buckets * sizeof(*rh->buckets));
+	if (!rh->buckets) {
+		DMERR("unable to allocate region hash bucket memory");
+		kfree(rh);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	for (i = 0; i < nr_buckets; i++)
+		INIT_LIST_HEAD(rh->buckets + i);
+
+	spin_lock_init(&rh->region_lock);
+	sema_init(&rh->recovery_count, 0);
+	atomic_set(&rh->recovery_in_flight, 0);
+	INIT_LIST_HEAD(&rh->clean_regions);
+	INIT_LIST_HEAD(&rh->quiesced_regions);
+	INIT_LIST_HEAD(&rh->recovered_regions);
+	INIT_LIST_HEAD(&rh->failed_recovered_regions);
+
+	rh->region_pool = mempool_create_kmalloc_pool(MIN_REGIONS,
+						      sizeof(struct dm_region));
+	if (!rh->region_pool) {
+		vfree(rh->buckets);
+		kfree(rh);
+		rh = ERR_PTR(-ENOMEM);
+	}
+
+	return rh;
+}
+// EXPORT_SYMBOL_GPL(dm_region_hash_create);
+
+void dm_region_hash_destroy(struct dm_region_hash *rh)
+{
+	unsigned h;
+	struct dm_region *reg, *nreg;
+
+	BUG_ON(!list_empty(&rh->quiesced_regions));
+	for (h = 0; h < rh->nr_buckets; h++) {
+		list_for_each_entry_safe(reg, nreg, rh->buckets + h,
+					 hash_list) {
+			BUG_ON(atomic_read(&reg->pending));
+			mempool_free(reg, rh->region_pool);
+		}
+	}
+
+	if (rh->log)
+		dm_dirty_log_destroy(rh->log);
+
+	if (rh->region_pool)
+		mempool_destroy(rh->region_pool);
+
+	vfree(rh->buckets);
+	kfree(rh);
+}
+// EXPORT_SYMBOL_GPL(dm_region_hash_destroy);
+
+struct dm_dirty_log *dm_rh_dirty_log(struct dm_region_hash *rh)
+{
+	return rh->log;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_dirty_log);
+
+static unsigned rh_hash(struct dm_region_hash *rh, region_t region)
+{
+	return (unsigned) ((region * rh->prime) >> rh->shift) & rh->mask;
+}
+
+static struct dm_region *__rh_lookup(struct dm_region_hash *rh, region_t region)
+{
+	struct dm_region *reg;
+	struct list_head *bucket = rh->buckets + rh_hash(rh, region);
+
+	list_for_each_entry(reg, bucket, hash_list)
+		if (reg->key == region)
+			return reg;
+
+	return NULL;
+}
+
+static void __rh_insert(struct dm_region_hash *rh, struct dm_region *reg)
+{
+	list_add(&reg->hash_list, rh->buckets + rh_hash(rh, reg->key));
+}
+
+static struct dm_region *__rh_alloc(struct dm_region_hash *rh, region_t region)
+{
+	struct dm_region *reg, *nreg;
+
+	nreg = mempool_alloc(rh->region_pool, GFP_ATOMIC);
+	if (unlikely(!nreg))
+		nreg = kmalloc(sizeof(*nreg), GFP_NOIO | __GFP_NOFAIL);
+
+	nreg->state = rh->log->type->in_sync(rh->log, region, 1) ?
+		      DM_RH_CLEAN : DM_RH_NOSYNC;
+	nreg->rh = rh;
+	nreg->key = region;
+	INIT_LIST_HEAD(&nreg->list);
+	atomic_set(&nreg->pending, 0);
+	bio_list_init(&nreg->delayed_bios);
+
+	write_lock_irq(&rh->hash_lock);
+	reg = __rh_lookup(rh, region);
+	if (reg)
+		/* We lost the race. */
+		mempool_free(nreg, rh->region_pool);
+	else {
+		__rh_insert(rh, nreg);
+		if (nreg->state == DM_RH_CLEAN) {
+			spin_lock(&rh->region_lock);
+			list_add(&nreg->list, &rh->clean_regions);
+			spin_unlock(&rh->region_lock);
+		}
+
+		reg = nreg;
+	}
+	write_unlock_irq(&rh->hash_lock);
+
+	return reg;
+}
+
+static struct dm_region *__rh_find(struct dm_region_hash *rh, region_t region)
+{
+	struct dm_region *reg;
+
+	reg = __rh_lookup(rh, region);
+	if (!reg) {
+		read_unlock(&rh->hash_lock);
+		reg = __rh_alloc(rh, region);
+		read_lock(&rh->hash_lock);
+	}
+
+	return reg;
+}
+
+int dm_rh_get_state(struct dm_region_hash *rh, region_t region, int may_block)
+{
+	int r;
+	struct dm_region *reg;
+
+	read_lock(&rh->hash_lock);
+	reg = __rh_lookup(rh, region);
+	read_unlock(&rh->hash_lock);
+
+	if (reg)
+		return reg->state;
+
+	/*
+	 * The region wasn't in the hash, so we fall back to the
+	 * dirty log.
+	 */
+	r = rh->log->type->in_sync(rh->log, region, may_block);
+
+	/*
+	 * Any error from the dirty log (eg. -EWOULDBLOCK) gets
+	 * taken as a DM_RH_NOSYNC
+	 */
+	return r == 1 ? DM_RH_CLEAN : DM_RH_NOSYNC;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_get_state);
+
+static void complete_resync_work(struct dm_region *reg, int success)
+{
+	struct dm_region_hash *rh = reg->rh;
+
+	rh->log->type->set_region_sync(rh->log, reg->key, success);
+
+	/*
+	 * Dispatch the bios before we call 'wake_up_all'.
+	 * This is important because if we are suspending,
+	 * we want to know that recovery is complete and
+	 * the work queue is flushed.  If we wake_up_all
+	 * before we dispatch_bios (queue bios and call wake()),
+	 * then we risk suspending before the work queue
+	 * has been properly flushed.
+	 */
+	rh->dispatch_bios(rh->context, &reg->delayed_bios);
+	if (atomic_dec_and_test(&rh->recovery_in_flight))
+		rh->wakeup_all_recovery_waiters(rh->context);
+	up(&rh->recovery_count);
+}
+
+/* dm_rh_mark_nosync
+ * @ms
+ * @bio
+ * @done
+ * @error
+ *
+ * The bio was written on some mirror(s) but failed on other mirror(s).
+ * We can successfully endio the bio but should avoid the region being
+ * marked clean by setting the state DM_RH_NOSYNC.
+ *
+ * This function is _not_ safe in interrupt context!
+ */
+void dm_rh_mark_nosync(struct dm_region_hash *rh,
+		       struct bio *bio, unsigned done, int error)
+{
+	unsigned long flags;
+	struct dm_dirty_log *log = rh->log;
+	struct dm_region *reg;
+	region_t region = dm_rh_bio_to_region(rh, bio);
+	int recovering = 0;
+
+	/* We must inform the log that the sync count has changed. */
+	log->type->set_region_sync(log, region, 0);
+
+	read_lock(&rh->hash_lock);
+	reg = __rh_find(rh, region);
+	read_unlock(&rh->hash_lock);
+
+	/* region hash entry should exist because write was in-flight */
+	BUG_ON(!reg);
+	BUG_ON(!list_empty(&reg->list));
+
+	spin_lock_irqsave(&rh->region_lock, flags);
+	/*
+	 * Possible cases:
+	 *   1) DM_RH_DIRTY
+	 *   2) DM_RH_NOSYNC: was dirty, other preceeding writes failed
+	 *   3) DM_RH_RECOVERING: flushing pending writes
+	 * Either case, the region should have not been connected to list.
+	 */
+	recovering = (reg->state == DM_RH_RECOVERING);
+	reg->state = DM_RH_NOSYNC;
+	BUG_ON(!list_empty(&reg->list));
+	spin_unlock_irqrestore(&rh->region_lock, flags);
+
+	bio_endio(bio, error);
+	if (recovering)
+		complete_resync_work(reg, 0);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_mark_nosync);
+
+void dm_rh_update_states(struct dm_region_hash *rh, int errors_handled)
+{
+	struct dm_region *reg, *next;
+
+	LIST_HEAD(clean);
+	LIST_HEAD(recovered);
+	LIST_HEAD(failed_recovered);
+
+	/*
+	 * Quickly grab the lists.
+	 */
+	write_lock_irq(&rh->hash_lock);
+	spin_lock(&rh->region_lock);
+	if (!list_empty(&rh->clean_regions)) {
+		list_splice_init(&rh->clean_regions, &clean);
+
+		list_for_each_entry(reg, &clean, list)
+			list_del(&reg->hash_list);
+	}
+
+	if (!list_empty(&rh->recovered_regions)) {
+		list_splice_init(&rh->recovered_regions, &recovered);
+
+		list_for_each_entry(reg, &recovered, list)
+			list_del(&reg->hash_list);
+	}
+
+	if (!list_empty(&rh->failed_recovered_regions)) {
+		list_splice_init(&rh->failed_recovered_regions,
+				 &failed_recovered);
+
+		list_for_each_entry(reg, &failed_recovered, list)
+			list_del(&reg->hash_list);
+	}
+
+	spin_unlock(&rh->region_lock);
+	write_unlock_irq(&rh->hash_lock);
+
+	/*
+	 * All the regions on the recovered and clean lists have
+	 * now been pulled out of the system, so no need to do
+	 * any more locking.
+	 */
+	list_for_each_entry_safe(reg, next, &recovered, list) {
+		rh->log->type->clear_region(rh->log, reg->key);
+		complete_resync_work(reg, 1);
+		mempool_free(reg, rh->region_pool);
+	}
+
+	list_for_each_entry_safe(reg, next, &failed_recovered, list) {
+		complete_resync_work(reg, errors_handled ? 0 : 1);
+		mempool_free(reg, rh->region_pool);
+	}
+
+	list_for_each_entry_safe(reg, next, &clean, list) {
+		rh->log->type->clear_region(rh->log, reg->key);
+		mempool_free(reg, rh->region_pool);
+	}
+
+	rh->log->type->flush(rh->log);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_update_states);
+
+void dm_rh_inc(struct dm_region_hash *rh, region_t region)
+{
+	struct dm_region *reg;
+
+	read_lock(&rh->hash_lock);
+	reg = __rh_find(rh, region);
+
+	spin_lock_irq(&rh->region_lock);
+	atomic_inc(&reg->pending);
+
+	if (reg->state == DM_RH_CLEAN) {
+		reg->state = DM_RH_DIRTY;
+		list_del_init(&reg->list);	/* take off the clean list */
+		spin_unlock_irq(&rh->region_lock);
+
+		rh->log->type->mark_region(rh->log, reg->key);
+	} else
+		spin_unlock_irq(&rh->region_lock);
+
+
+	read_unlock(&rh->hash_lock);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_inc);
+
+void dm_rh_inc_pending(struct dm_region_hash *rh, struct bio_list *bios)
+{
+	struct bio *bio;
+
+	for (bio = bios->head; bio; bio = bio->bi_next)
+		dm_rh_inc(rh, dm_rh_bio_to_region(rh, bio));
+}
+// EXPORT_SYMBOL_GPL(dm_rh_inc_pending);
+
+void dm_rh_dec(struct dm_region_hash *rh, region_t region)
+{
+	unsigned long flags;
+	struct dm_region *reg;
+	int should_wake = 0;
+
+	read_lock(&rh->hash_lock);
+	reg = __rh_lookup(rh, region);
+	read_unlock(&rh->hash_lock);
+
+	spin_lock_irqsave(&rh->region_lock, flags);
+	if (atomic_dec_and_test(&reg->pending)) {
+		/*
+		 * There is no pending I/O for this region.
+		 * We can move the region to corresponding list for next action.
+		 * At this point, the region is not yet connected to any list.
+		 *
+		 * If the state is DM_RH_NOSYNC, the region should be kept off
+		 * from clean list.
+		 * The hash entry for DM_RH_NOSYNC will remain in memory
+		 * until the region is recovered or the map is reloaded.
+		 */
+
+		/* do nothing for DM_RH_NOSYNC */
+		if (reg->state == DM_RH_RECOVERING) {
+			list_add_tail(&reg->list, &rh->quiesced_regions);
+		} else if (reg->state == DM_RH_DIRTY) {
+			reg->state = DM_RH_CLEAN;
+			list_add(&reg->list, &rh->clean_regions);
+		}
+		should_wake = 1;
+	}
+	spin_unlock_irqrestore(&rh->region_lock, flags);
+
+	if (should_wake)
+		rh->wakeup_workers(rh->context);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_dec);
+
+/*
+ * Starts quiescing a region in preparation for recovery.
+ */
+static int __rh_recovery_prepare(struct dm_region_hash *rh)
+{
+	int r;
+	region_t region;
+	struct dm_region *reg;
+
+	/*
+	 * Ask the dirty log what's next.
+	 */
+	r = rh->log->type->get_resync_work(rh->log, &region);
+	if (r <= 0)
+		return r;
+
+	/*
+	 * Get this region, and start it quiescing by setting the
+	 * recovering flag.
+	 */
+	read_lock(&rh->hash_lock);
+	reg = __rh_find(rh, region);
+	read_unlock(&rh->hash_lock);
+
+	spin_lock_irq(&rh->region_lock);
+	reg->state = DM_RH_RECOVERING;
+
+	/* Already quiesced ? */
+	if (atomic_read(&reg->pending))
+		list_del_init(&reg->list);
+	else
+		list_move(&reg->list, &rh->quiesced_regions);
+
+	spin_unlock_irq(&rh->region_lock);
+
+	return 1;
+}
+
+void dm_rh_recovery_prepare(struct dm_region_hash *rh)
+{
+	/* Extra reference to avoid race with dm_rh_stop_recovery */
+	atomic_inc(&rh->recovery_in_flight);
+
+	while (!down_trylock(&rh->recovery_count)) {
+		atomic_inc(&rh->recovery_in_flight);
+		if (__rh_recovery_prepare(rh) <= 0) {
+			atomic_dec(&rh->recovery_in_flight);
+			up(&rh->recovery_count);
+			break;
+		}
+	}
+
+	/* Drop the extra reference */
+	if (atomic_dec_and_test(&rh->recovery_in_flight))
+		rh->wakeup_all_recovery_waiters(rh->context);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_recovery_prepare);
+
+/*
+ * Returns any quiesced regions.
+ */
+struct dm_region *dm_rh_recovery_start(struct dm_region_hash *rh)
+{
+	struct dm_region *reg = NULL;
+
+	spin_lock_irq(&rh->region_lock);
+	if (!list_empty(&rh->quiesced_regions)) {
+		reg = list_entry(rh->quiesced_regions.next,
+				 struct dm_region, list);
+		list_del_init(&reg->list);  /* remove from the quiesced list */
+	}
+	spin_unlock_irq(&rh->region_lock);
+
+	return reg;
+}
+// EXPORT_SYMBOL_GPL(dm_rh_recovery_start);
+
+void dm_rh_recovery_end(struct dm_region *reg, int success)
+{
+	struct dm_region_hash *rh = reg->rh;
+
+	spin_lock_irq(&rh->region_lock);
+	if (success)
+		list_add(&reg->list, &reg->rh->recovered_regions);
+	else {
+		reg->state = DM_RH_NOSYNC;
+		list_add(&reg->list, &reg->rh->failed_recovered_regions);
+	}
+	spin_unlock_irq(&rh->region_lock);
+
+	rh->wakeup_workers(rh->context);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_recovery_end);
+
+/* Return recovery in flight count. */
+int dm_rh_recovery_in_flight(struct dm_region_hash *rh)
+{
+	return atomic_read(&rh->recovery_in_flight);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_recovery_in_flight);
+
+int dm_rh_flush(struct dm_region_hash *rh)
+{
+	return rh->log->type->flush(rh->log);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_flush);
+
+void dm_rh_delay(struct dm_region_hash *rh, struct bio *bio)
+{
+	struct dm_region *reg;
+
+	read_lock(&rh->hash_lock);
+	reg = __rh_find(rh, dm_rh_bio_to_region(rh, bio));
+	bio_list_add(&reg->delayed_bios, bio);
+	read_unlock(&rh->hash_lock);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_delay);
+
+void dm_rh_delay_by_region(struct dm_region_hash *rh,
+			   struct bio *bio, region_t region)
+{
+	struct dm_region *reg;
+
+	/* FIXME: locking. */
+	read_lock(&rh->hash_lock);
+	reg = __rh_find(rh, region);
+	bio_list_add(&reg->delayed_bios, bio);
+	read_unlock(&rh->hash_lock);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_delay_by_region);
+
+void dm_rh_stop_recovery(struct dm_region_hash *rh)
+{
+	int i;
+
+	/* wait for any recovering regions */
+	for (i = 0; i < rh->max_recovery; i++)
+		down(&rh->recovery_count);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_stop_recovery);
+
+void dm_rh_start_recovery(struct dm_region_hash *rh)
+{
+	int i;
+
+	for (i = 0; i < rh->max_recovery; i++)
+		up(&rh->recovery_count);
+
+	rh->wakeup_workers(rh->context);
+}
+// EXPORT_SYMBOL_GPL(dm_rh_start_recovery);
+
+MODULE_DESCRIPTION(DM_NAME " region hash");
+MODULE_AUTHOR("Joe Thornber/Heinz Mauelshagen <dm-devel@redhat.com>");
+MODULE_LICENSE("GPL");

ubuntu/dm-raid4-5/dm-region-hash.h

+/*
+ * Copyright (C) 2003 Sistina Software Limited.
+ * Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved.
+ *
+ * Device-Mapper dirty region hash interface.
+ *
+ * This file is released under the GPL.
+ */
+
+#ifndef DM_REGION_HASH_H
+#define DM_REGION_HASH_H
+
+#include <linux/dm-dirty-log.h>
+
+/*-----------------------------------------------------------------
+ * Region hash
+ *----------------------------------------------------------------*/
+struct dm_region_hash;
+struct dm_region;
+
+/*
+ * States a region can have.
+ */
+enum dm_rh_region_states {
+	DM_RH_CLEAN	 = 0x01,	/* No writes in flight. */
+	DM_RH_DIRTY	 = 0x02,	/* Writes in flight. */
+	DM_RH_NOSYNC	 = 0x04,	/* Out of sync. */
+	DM_RH_RECOVERING = 0x08,	/* Under resynchronization. */
+};
+
+/*
+ * Region hash create/destroy.
+ */
+struct bio_list;
+struct dm_region_hash *dm_region_hash_create(
+		void *context, void (*dispatch_bios)(void *context,
+						     struct bio_list *bios),
+		void (*wakeup_workers)(void *context),
+		void (*wakeup_all_recovery_waiters)(void *context),
+		sector_t target_begin, unsigned max_recovery,
+		struct dm_dirty_log *log, uint32_t region_size,
+		region_t nr_regions);
+void dm_region_hash_destroy(struct dm_region_hash *rh);
+
+struct dm_dirty_log *dm_rh_dirty_log(struct dm_region_hash *rh);
+
+/*
+ * Conversion functions.
+ */
+region_t dm_rh_bio_to_region(struct dm_region_hash *rh, struct bio *bio);
+sector_t dm_rh_region_to_sector(struct dm_region_hash *rh, region_t region);
+region_t dm_rh_sector_to_region(struct dm_region_hash *rh, sector_t sector);
+void *dm_rh_region_context(struct dm_region *reg);
+
+/*
+ * Get region size and key (ie. number of the region).
+ */
+sector_t dm_rh_get_region_size(struct dm_region_hash *rh);
+region_t dm_rh_get_region_key(struct dm_region *reg);
+
+/*
+ * Get/set/update region state (and dirty log).
+ *
+ */
+int dm_rh_get_state(struct dm_region_hash *rh, region_t region, int may_block);
+void dm_rh_set_state(struct dm_region_hash *rh, region_t region,
+		     enum dm_rh_region_states state, int may_block);
+
+/* Non-zero errors_handled leaves the state of the region NOSYNC */
+void dm_rh_update_states(struct dm_region_hash *rh, int errors_handled);
+
+/* Flush the region hash and dirty log. */
+int dm_rh_flush(struct dm_region_hash *rh);
+
+/* Inc/dec pending count on regions. */
+void dm_rh_inc(struct dm_region_hash *rh, region_t region);
+void dm_rh_inc_pending(struct dm_region_hash *rh, struct bio_list *bios);
+void dm_rh_dec(struct dm_region_hash *rh, region_t region);
+
+/* Delay bios on regions. */
+void dm_rh_delay(struct dm_region_hash *rh, struct bio *bio);
+void dm_rh_delay_by_region(struct dm_region_hash *rh, struct bio *bio,
+			   region_t region);
+
+void dm_rh_mark_nosync(struct dm_region_hash *rh,
+		       struct bio *bio, unsigned done, int error);
+
+/*
+ * Region recovery control.
+ */
+
+/* Prepare some regions for recovery by starting to quiesce them. */
+void dm_rh_recovery_prepare(struct dm_region_hash *rh);
+
+/* Try fetching a quiesced region for recovery. */
+struct dm_region *dm_rh_recovery_start(struct dm_region_hash *rh);
+
+/* Report recovery end on a region. */
+void dm_rh_recovery_end(struct dm_region *reg, int error);
+
+/* Returns number of regions with recovery work outstanding. */
+int dm_rh_recovery_in_flight(struct dm_region_hash *rh);
+
+/* Start/stop recovery. */
+void dm_rh_start_recovery(struct dm_region_hash *rh);
+void dm_rh_stop_recovery(struct dm_region_hash *rh);
+
+#endif /* DM_REGION_HASH_H */