Merge rsync://rsync.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6

Conflicts:

	drivers/usb/input/hid.h

10 files changed, 130 insertions, 47 deletions

diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 08a40f4e4f6..a1086ee8ccc 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -20,6 +20,7 @@
 #include <asm/atomic.h>
 #include <linux/scatterlist.h>
 #include <asm/page.h>
+#include <asm/unaligned.h>
 
 #include "dm.h"
 
@@ -85,7 +86,10 @@ struct crypt_config {
 	 */
 	struct crypt_iv_operations *iv_gen_ops;
 	char *iv_mode;
-	struct crypto_cipher *iv_gen_private;
+	union {
+		struct crypto_cipher *essiv_tfm;
+		int benbi_shift;
+	} iv_gen_private;
 	sector_t iv_offset;
 	unsigned int iv_size;
 
@@ -101,7 +105,7 @@ struct crypt_config {
 #define MIN_POOL_PAGES 32
 #define MIN_BIO_PAGES  8
 
-static kmem_cache_t *_crypt_io_pool;
+static struct kmem_cache *_crypt_io_pool;
 
 /*
  * Different IV generation algorithms:
@@ -113,6 +117,9 @@ static kmem_cache_t *_crypt_io_pool;
  *        encrypted with the bulk cipher using a salt as key. The salt
  *        should be derived from the bulk cipher's key via hashing.
  *
+ * benbi: the 64-bit "big-endian 'narrow block'-count", starting at 1
+ *        (needed for LRW-32-AES and possible other narrow block modes)
+ *
  * plumb: unimplemented, see:
  * http://article.gmane.org/gmane.linux.kernel.device-mapper.dm-crypt/454
  */
@@ -191,21 +198,61 @@ static int crypt_iv_essiv_ctr(struct crypt_config *cc, struct dm_target *ti,
 	}
 	kfree(salt);
 
-	cc->iv_gen_private = essiv_tfm;
+	cc->iv_gen_private.essiv_tfm = essiv_tfm;
 	return 0;
 }
 
 static void crypt_iv_essiv_dtr(struct crypt_config *cc)
 {
-	crypto_free_cipher(cc->iv_gen_private);
-	cc->iv_gen_private = NULL;
+	crypto_free_cipher(cc->iv_gen_private.essiv_tfm);
+	cc->iv_gen_private.essiv_tfm = NULL;
 }
 
 static int crypt_iv_essiv_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
 {
 	memset(iv, 0, cc->iv_size);
 	*(u64 *)iv = cpu_to_le64(sector);
-	crypto_cipher_encrypt_one(cc->iv_gen_private, iv, iv);
+	crypto_cipher_encrypt_one(cc->iv_gen_private.essiv_tfm, iv, iv);
+	return 0;
+}
+
+static int crypt_iv_benbi_ctr(struct crypt_config *cc, struct dm_target *ti,
+			      const char *opts)
+{
+	unsigned int bs = crypto_blkcipher_blocksize(cc->tfm);
+	int log = long_log2(bs);
+
+	/* we need to calculate how far we must shift the sector count
+	 * to get the cipher block count, we use this shift in _gen */
+
+	if (1 << log != bs) {
+		ti->error = "cypher blocksize is not a power of 2";
+		return -EINVAL;
+	}
+
+	if (log > 9) {
+		ti->error = "cypher blocksize is > 512";
+		return -EINVAL;
+	}
+
+	cc->iv_gen_private.benbi_shift = 9 - log;
+
+	return 0;
+}
+
+static void crypt_iv_benbi_dtr(struct crypt_config *cc)
+{
+}
+
+static int crypt_iv_benbi_gen(struct crypt_config *cc, u8 *iv, sector_t sector)
+{
+	__be64 val;
+
+	memset(iv, 0, cc->iv_size - sizeof(u64)); /* rest is cleared below */
+
+	val = cpu_to_be64(((u64)sector << cc->iv_gen_private.benbi_shift) + 1);
+	put_unaligned(val, (__be64 *)(iv + cc->iv_size - sizeof(u64)));
+
 	return 0;
 }
 
@@ -219,13 +266,18 @@ static struct crypt_iv_operations crypt_iv_essiv_ops = {
 	.generator = crypt_iv_essiv_gen
 };
 
+static struct crypt_iv_operations crypt_iv_benbi_ops = {
+	.ctr	   = crypt_iv_benbi_ctr,
+	.dtr	   = crypt_iv_benbi_dtr,
+	.generator = crypt_iv_benbi_gen
+};
 
 static int
 crypt_convert_scatterlist(struct crypt_config *cc, struct scatterlist *out,
                           struct scatterlist *in, unsigned int length,
                           int write, sector_t sector)
 {
-	u8 iv[cc->iv_size];
+	u8 iv[cc->iv_size] __attribute__ ((aligned(__alignof__(u64))));
 	struct blkcipher_desc desc = {
 		.tfm = cc->tfm,
 		.info = iv,
@@ -458,11 +510,11 @@ static void dec_pending(struct crypt_io *io, int error)
  * interrupt context.
  */
 static struct workqueue_struct *_kcryptd_workqueue;
-static void kcryptd_do_work(void *data);
+static void kcryptd_do_work(struct work_struct *work);
 
 static void kcryptd_queue_io(struct crypt_io *io)
 {
-	INIT_WORK(&io->work, kcryptd_do_work, io);
+	INIT_WORK(&io->work, kcryptd_do_work);
 	queue_work(_kcryptd_workqueue, &io->work);
 }
 
@@ -618,9 +670,9 @@ static void process_read_endio(struct crypt_io *io)
 	dec_pending(io, crypt_convert(cc, &ctx));
 }
 
-static void kcryptd_do_work(void *data)
+static void kcryptd_do_work(struct work_struct *work)
 {
-	struct crypt_io *io = data;
+	struct crypt_io *io = container_of(work, struct crypt_io, work);
 
 	if (io->post_process)
 		process_read_endio(io);
@@ -768,7 +820,7 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	cc->tfm = tfm;
 
 	/*
-	 * Choose ivmode. Valid modes: "plain", "essiv:<esshash>".
+	 * Choose ivmode. Valid modes: "plain", "essiv:<esshash>", "benbi".
 	 * See comments at iv code
 	 */
 
@@ -778,6 +830,8 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		cc->iv_gen_ops = &crypt_iv_plain_ops;
 	else if (strcmp(ivmode, "essiv") == 0)
 		cc->iv_gen_ops = &crypt_iv_essiv_ops;
+	else if (strcmp(ivmode, "benbi") == 0)
+		cc->iv_gen_ops = &crypt_iv_benbi_ops;
 	else {
 		ti->error = "Invalid IV mode";
 		goto bad2;
diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index d13bb15a8a0..4510ad8f971 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -606,9 +606,14 @@ static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param)
 		return __get_name_cell(param->name);
 
 	md = dm_get_md(huge_decode_dev(param->dev));
-	if (md)
-		mdptr = dm_get_mdptr(md);
+	if (!md)
+		goto out;
 
+	mdptr = dm_get_mdptr(md);
+	if (!mdptr)
+		dm_put(md);
+
+out:
 	return mdptr;
 }
 
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index d754e0bc6e9..cf8bf052138 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -101,11 +101,11 @@ typedef int (*action_fn) (struct pgpath *pgpath);
 
 #define MIN_IOS 256	/* Mempool size */
 
-static kmem_cache_t *_mpio_cache;
+static struct kmem_cache *_mpio_cache;
 
 struct workqueue_struct *kmultipathd;
-static void process_queued_ios(void *data);
-static void trigger_event(void *data);
+static void process_queued_ios(struct work_struct *work);
+static void trigger_event(struct work_struct *work);
 
 
 /*-----------------------------------------------
@@ -173,8 +173,8 @@ static struct multipath *alloc_multipath(struct dm_target *ti)
 		INIT_LIST_HEAD(&m->priority_groups);
 		spin_lock_init(&m->lock);
 		m->queue_io = 1;
-		INIT_WORK(&m->process_queued_ios, process_queued_ios, m);
-		INIT_WORK(&m->trigger_event, trigger_event, m);
+		INIT_WORK(&m->process_queued_ios, process_queued_ios);
+		INIT_WORK(&m->trigger_event, trigger_event);
 		m->mpio_pool = mempool_create_slab_pool(MIN_IOS, _mpio_cache);
 		if (!m->mpio_pool) {
 			kfree(m);
@@ -379,9 +379,10 @@ static void dispatch_queued_ios(struct multipath *m)
 	}
 }
 
-static void process_queued_ios(void *data)
+static void process_queued_ios(struct work_struct *work)
 {
-	struct multipath *m = (struct multipath *) data;
+	struct multipath *m =
+		container_of(work, struct multipath, process_queued_ios);
 	struct hw_handler *hwh = &m->hw_handler;
 	struct pgpath *pgpath = NULL;
 	unsigned init_required = 0, must_queue = 1;
@@ -421,9 +422,10 @@ out:
  * An event is triggered whenever a path is taken out of use.
  * Includes path failure and PG bypass.
  */
-static void trigger_event(void *data)
+static void trigger_event(struct work_struct *work)
 {
-	struct multipath *m = (struct multipath *) data;
+	struct multipath *m =
+		container_of(work, struct multipath, trigger_event);
 
 	dm_table_event(m->ti->table);
 }
diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c
index 659224cb7c5..fc8cbb168e3 100644
--- a/drivers/md/dm-raid1.c
+++ b/drivers/md/dm-raid1.c
@@ -24,6 +24,7 @@
 
 static struct workqueue_struct *_kmirrord_wq;
 static struct work_struct _kmirrord_work;
+static DECLARE_WAIT_QUEUE_HEAD(_kmirrord_recovery_stopped);
 
 static inline void wake(void)
 {
@@ -83,6 +84,7 @@ struct region_hash {
 	struct list_head *buckets;
 
 	spinlock_t region_lock;
+	atomic_t recovery_in_flight;
 	struct semaphore recovery_count;
 	struct list_head clean_regions;
 	struct list_head quiesced_regions;
@@ -191,6 +193,7 @@ static int rh_init(struct region_hash *rh, struct mirror_set *ms,
 
 	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);
@@ -382,6 +385,8 @@ static void rh_update_states(struct region_hash *rh)
 		rh->log->type->clear_region(rh->log, reg->key);
 		rh->log->type->complete_resync_work(rh->log, reg->key, 1);
 		dispatch_bios(rh->ms, &reg->delayed_bios);
+		if (atomic_dec_and_test(&rh->recovery_in_flight))
+			wake_up_all(&_kmirrord_recovery_stopped);
 		up(&rh->recovery_count);
 		mempool_free(reg, rh->region_pool);
 	}
@@ -502,11 +507,21 @@ static int __rh_recovery_prepare(struct region_hash *rh)
 
 static void rh_recovery_prepare(struct region_hash *rh)
 {
-	while (!down_trylock(&rh->recovery_count))
+	/* Extra reference to avoid race with 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))
+		wake_up_all(&_kmirrord_recovery_stopped);
 }
 
 /*
@@ -868,7 +883,7 @@ static void do_mirror(struct mirror_set *ms)
 	do_writes(ms, &writes);
 }
 
-static void do_work(void *ignored)
+static void do_work(struct work_struct *ignored)
 {
 	struct mirror_set *ms;
 
@@ -1177,6 +1192,11 @@ static void mirror_postsuspend(struct dm_target *ti)
 	struct dirty_log *log = ms->rh.log;
 
 	rh_stop_recovery(&ms->rh);
+
+	/* Wait for all I/O we generated to complete */
+	wait_event(_kmirrord_recovery_stopped,
+		   !atomic_read(&ms->rh.recovery_in_flight));
+
 	if (log->type->suspend && log->type->suspend(log))
 		/* FIXME: need better error handling */
 		DMWARN("log suspend failed");
@@ -1249,7 +1269,7 @@ static int __init dm_mirror_init(void)
 		dm_dirty_log_exit();
 		return r;
 	}
-	INIT_WORK(&_kmirrord_work, do_work, NULL);
+	INIT_WORK(&_kmirrord_work, do_work);
 
 	r = dm_register_target(&mirror_target);
 	if (r < 0) {
diff --git a/drivers/md/dm-round-robin.c b/drivers/md/dm-round-robin.c
index c5a16c55012..6f9fcd4db9b 100644
--- a/drivers/md/dm-round-robin.c
+++ b/drivers/md/dm-round-robin.c
@@ -136,7 +136,7 @@ static int rr_add_path(struct path_selector *ps, struct path *path,
 
 	path->pscontext = pi;
 
-	list_add(&pi->list, &s->valid_paths);
+	list_add_tail(&pi->list, &s->valid_paths);
 
 	return 0;
 }
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 5281e009407..b0ce2ce8227 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -40,7 +40,7 @@
 #define SNAPSHOT_PAGES 256
 
 struct workqueue_struct *ksnapd;
-static void flush_queued_bios(void *data);
+static void flush_queued_bios(struct work_struct *work);
 
 struct pending_exception {
 	struct exception e;
@@ -88,8 +88,8 @@ struct pending_exception {
  * Hash table mapping origin volumes to lists of snapshots and
  * a lock to protect it
  */
-static kmem_cache_t *exception_cache;
-static kmem_cache_t *pending_cache;
+static struct kmem_cache *exception_cache;
+static struct kmem_cache *pending_cache;
 static mempool_t *pending_pool;
 
 /*
@@ -228,7 +228,7 @@ static int init_exception_table(struct exception_table *et, uint32_t size)
 	return 0;
 }
 
-static void exit_exception_table(struct exception_table *et, kmem_cache_t *mem)
+static void exit_exception_table(struct exception_table *et, struct kmem_cache *mem)
 {
 	struct list_head *slot;
 	struct exception *ex, *next;
@@ -528,7 +528,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	}
 
 	bio_list_init(&s->queued_bios);
-	INIT_WORK(&s->queued_bios_work, flush_queued_bios, s);
+	INIT_WORK(&s->queued_bios_work, flush_queued_bios);
 
 	/* Add snapshot to the list of snapshots for this origin */
 	/* Exceptions aren't triggered till snapshot_resume() is called */
@@ -603,9 +603,10 @@ static void flush_bios(struct bio *bio)
 	}
 }
 
-static void flush_queued_bios(void *data)
+static void flush_queued_bios(struct work_struct *work)
 {
-	struct dm_snapshot *s = (struct dm_snapshot *) data;
+	struct dm_snapshot *s =
+		container_of(work, struct dm_snapshot, queued_bios_work);
 	struct bio *queued_bios;
 	unsigned long flags;
 
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index b5764a86c8b..7ec1b112a6d 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -121,8 +121,8 @@ struct mapped_device {
 };
 
 #define MIN_IOS 256
-static kmem_cache_t *_io_cache;
-static kmem_cache_t *_tio_cache;
+static struct kmem_cache *_io_cache;
+static struct kmem_cache *_tio_cache;
 
 static int __init local_init(void)
 {
@@ -1285,7 +1285,7 @@ int dm_suspend(struct mapped_device *md, int do_lockfs)
 	down(&md->suspend_lock);
 
 	if (dm_suspended(md))
-		goto out;
+		goto out_unlock;
 
 	map = dm_get_table(md);
 
@@ -1361,6 +1361,8 @@ out:
 	}
 
 	dm_table_put(map);
+
+out_unlock:
 	up(&md->suspend_lock);
 	return r;
 }
diff --git a/drivers/md/kcopyd.c b/drivers/md/kcopyd.c
index f1db6eff485..b46f6c575f7 100644
--- a/drivers/md/kcopyd.c
+++ b/drivers/md/kcopyd.c
@@ -203,7 +203,7 @@ struct kcopyd_job {
 /* FIXME: this should scale with the number of pages */
 #define MIN_JOBS 512
 
-static kmem_cache_t *_job_cache;
+static struct kmem_cache *_job_cache;
 static mempool_t *_job_pool;
 
 /*
@@ -417,7 +417,7 @@ static int process_jobs(struct list_head *jobs, int (*fn) (struct kcopyd_job *))
 /*
  * kcopyd does this every time it's woken up.
  */
-static void do_work(void *ignored)
+static void do_work(struct work_struct *ignored)
 {
 	/*
 	 * The order that these are called is *very* important.
@@ -628,7 +628,7 @@ static int kcopyd_init(void)
 	}
 
 	kcopyd_clients++;
-	INIT_WORK(&_kcopyd_work, do_work, NULL);
+	INIT_WORK(&_kcopyd_work, do_work);
 	mutex_unlock(&kcopyd_init_lock);
 	return 0;
 }
diff --git a/drivers/md/md.c b/drivers/md/md.c
index d1113560440..6c4345bde07 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -39,10 +39,10 @@
 #include <linux/raid/bitmap.h>
 #include <linux/sysctl.h>
 #include <linux/buffer_head.h> /* for invalidate_bdev */
-#include <linux/suspend.h>
 #include <linux/poll.h>
 #include <linux/mutex.h>
 #include <linux/ctype.h>
+#include <linux/freezer.h>
 
 #include <linux/init.h>
 
@@ -3200,7 +3200,7 @@ static int do_md_run(mddev_t * mddev)
 
 	mddev->changed = 1;
 	md_new_event(mddev);
-	kobject_uevent(&mddev->gendisk->kobj, KOBJ_ONLINE);
+	kobject_uevent(&mddev->gendisk->kobj, KOBJ_CHANGE);
 	return 0;
 }
 
@@ -3314,7 +3314,6 @@ static int do_md_stop(mddev_t * mddev, int mode)
 
 			module_put(mddev->pers->owner);
 			mddev->pers = NULL;
-			kobject_uevent(&mddev->gendisk->kobj, KOBJ_OFFLINE);
 			if (mddev->ro)
 				mddev->ro = 0;
 		}
@@ -4487,6 +4486,7 @@ static int md_thread(void * arg)
 	 * many dirty RAID5 blocks.
 	 */
 
+	current->flags |= PF_NOFREEZE;
 	allow_signal(SIGKILL);
 	while (!kthread_should_stop()) {
 
@@ -4503,7 +4503,6 @@ static int md_thread(void * arg)
 			 test_bit(THREAD_WAKEUP, &thread->flags)
 			 || kthread_should_stop(),
 			 thread->timeout);
-		try_to_freeze();
 
 		clear_bit(THREAD_WAKEUP, &thread->flags);
 
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index e14f4578072..52914d5cec7 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -348,7 +348,7 @@ static int grow_one_stripe(raid5_conf_t *conf)
 
 static int grow_stripes(raid5_conf_t *conf, int num)
 {
-	kmem_cache_t *sc;
+	struct kmem_cache *sc;
 	int devs = conf->raid_disks;
 
 	sprintf(conf->cache_name[0], "raid5/%s", mdname(conf->mddev));
@@ -397,7 +397,7 @@ static int resize_stripes(raid5_conf_t *conf, int newsize)
 	LIST_HEAD(newstripes);
 	struct disk_info *ndisks;
 	int err = 0;
-	kmem_cache_t *sc;
+	struct kmem_cache *sc;
 	int i;
 
 	if (newsize <= conf->pool_size)
@@ -3659,7 +3659,7 @@ static void end_reshape(raid5_conf_t *conf)
 		bdev = bdget_disk(conf->mddev->gendisk, 0);
 		if (bdev) {
 			mutex_lock(&bdev->bd_inode->i_mutex);
-			i_size_write(bdev->bd_inode, conf->mddev->array_size << 10);
+			i_size_write(bdev->bd_inode, (loff_t)conf->mddev->array_size << 10);
 			mutex_unlock(&bdev->bd_inode->i_mutex);
 			bdput(bdev);
 		}