Merge tag 'md-3.7' of git://neil.brown.name/md

Pull md updates from NeilBrown:
 - "discard" support, some dm-raid improvements and other assorted bits
   and pieces.

* tag 'md-3.7' of git://neil.brown.name/md: (29 commits)
  md: refine reporting of resync/reshape delays.
  md/raid5: be careful not to resize_stripes too big.
  md: make sure manual changes to recovery checkpoint are saved.
  md/raid10: use correct limit variable
  md: writing to sync_action should clear the read-auto state.
  Subject: [PATCH] md:change resync_mismatches to atomic64_t to avoid races
  md/raid5: make sure to_read and to_write never go negative.
  md: When RAID5 is dirty, force reconstruct-write instead of read-modify-write.
  md/raid5: protect debug message against NULL derefernce.
  md/raid5: add some missing locking in handle_failed_stripe.
  MD: raid5 avoid unnecessary zero page for trim
  MD: raid5 trim support
  md/bitmap:Don't use IS_ERR to judge alloc_page().
  md/raid1: Don't release reference to device while handling read error.
  raid: replace list_for_each_continue_rcu with new interface
  add further __init annotations to crypto/xor.c
  DM RAID: Fix for "sync" directive ineffectiveness
  DM RAID: Fix comparison of index and quantity for "rebuild" parameter
  DM RAID: Add rebuild capability for RAID10
  DM RAID: Move 'rebuild' checking code to its own function
  ...

1 files changed, 106 insertions, 39 deletions

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 95c88012a3b9..9ab768acfb62 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -674,7 +674,18 @@ static struct md_rdev * find_rdev_nr(struct mddev *mddev, int nr)
 	return NULL;
 }
 
-static struct md_rdev * find_rdev(struct mddev * mddev, dev_t dev)
+static struct md_rdev *find_rdev_nr_rcu(struct mddev *mddev, int nr)
+{
+	struct md_rdev *rdev;
+
+	rdev_for_each_rcu(rdev, mddev)
+		if (rdev->desc_nr == nr)
+			return rdev;
+
+	return NULL;
+}
+
+static struct md_rdev *find_rdev(struct mddev *mddev, dev_t dev)
 {
 	struct md_rdev *rdev;
 
@@ -685,6 +696,17 @@ static struct md_rdev * find_rdev(struct mddev * mddev, dev_t dev)
 	return NULL;
 }
 
+static struct md_rdev *find_rdev_rcu(struct mddev *mddev, dev_t dev)
+{
+	struct md_rdev *rdev;
+
+	rdev_for_each_rcu(rdev, mddev)
+		if (rdev->bdev->bd_dev == dev)
+			return rdev;
+
+	return NULL;
+}
+
 static struct md_personality *find_pers(int level, char *clevel)
 {
 	struct md_personality *pers;
@@ -2022,8 +2044,14 @@ EXPORT_SYMBOL(md_integrity_register);
 /* Disable data integrity if non-capable/non-matching disk is being added */
 void md_integrity_add_rdev(struct md_rdev *rdev, struct mddev *mddev)
 {
-	struct blk_integrity *bi_rdev = bdev_get_integrity(rdev->bdev);
-	struct blk_integrity *bi_mddev = blk_get_integrity(mddev->gendisk);
+	struct blk_integrity *bi_rdev;
+	struct blk_integrity *bi_mddev;
+
+	if (!mddev->gendisk)
+		return;
+
+	bi_rdev = bdev_get_integrity(rdev->bdev);
+	bi_mddev = blk_get_integrity(mddev->gendisk);
 
 	if (!bi_mddev) /* nothing to do */
 		return;
@@ -3754,6 +3782,8 @@ resync_start_store(struct mddev *mddev, const char *buf, size_t len)
 		return -EINVAL;
 
 	mddev->recovery_cp = n;
+	if (mddev->pers)
+		set_bit(MD_CHANGE_CLEAN, &mddev->flags);
 	return len;
 }
 static struct md_sysfs_entry md_resync_start =
@@ -4231,6 +4261,13 @@ action_store(struct mddev *mddev, const char *page, size_t len)
 		set_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
 		set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
 	}
+	if (mddev->ro == 2) {
+		/* A write to sync_action is enough to justify
+		 * canceling read-auto mode
+		 */
+		mddev->ro = 0;
+		md_wakeup_thread(mddev->sync_thread);
+	}
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
 	sysfs_notify_dirent_safe(mddev->sysfs_action);
@@ -4241,7 +4278,8 @@ static ssize_t
 mismatch_cnt_show(struct mddev *mddev, char *page)
 {
 	return sprintf(page, "%llu\n",
-		       (unsigned long long) mddev->resync_mismatches);
+		       (unsigned long long)
+		       atomic64_read(&mddev->resync_mismatches));
 }
 
 static struct md_sysfs_entry md_scan_mode =
@@ -4362,6 +4400,10 @@ sync_completed_show(struct mddev *mddev, char *page)
 	if (!test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		return sprintf(page, "none\n");
 
+	if (mddev->curr_resync == 1 ||
+	    mddev->curr_resync == 2)
+		return sprintf(page, "delayed\n");
+
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ||
 	    test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
 		max_sectors = mddev->resync_max_sectors;
@@ -5207,7 +5249,7 @@ static void md_clean(struct mddev *mddev)
 	mddev->new_layout = 0;
 	mddev->new_chunk_sectors = 0;
 	mddev->curr_resync = 0;
-	mddev->resync_mismatches = 0;
+	atomic64_set(&mddev->resync_mismatches, 0);
 	mddev->suspend_lo = mddev->suspend_hi = 0;
 	mddev->sync_speed_min = mddev->sync_speed_max = 0;
 	mddev->recovery = 0;
@@ -5509,8 +5551,9 @@ static int get_array_info(struct mddev * mddev, void __user * arg)
 	int nr,working,insync,failed,spare;
 	struct md_rdev *rdev;
 
-	nr=working=insync=failed=spare=0;
-	rdev_for_each(rdev, mddev) {
+	nr = working = insync = failed = spare = 0;
+	rcu_read_lock();
+	rdev_for_each_rcu(rdev, mddev) {
 		nr++;
 		if (test_bit(Faulty, &rdev->flags))
 			failed++;
@@ -5522,6 +5565,7 @@ static int get_array_info(struct mddev * mddev, void __user * arg)
 				spare++;
 		}
 	}
+	rcu_read_unlock();
 
 	info.major_version = mddev->major_version;
 	info.minor_version = mddev->minor_version;
@@ -5605,7 +5649,8 @@ static int get_disk_info(struct mddev * mddev, void __user * arg)
 	if (copy_from_user(&info, arg, sizeof(info)))
 		return -EFAULT;
 
-	rdev = find_rdev_nr(mddev, info.number);
+	rcu_read_lock();
+	rdev = find_rdev_nr_rcu(mddev, info.number);
 	if (rdev) {
 		info.major = MAJOR(rdev->bdev->bd_dev);
 		info.minor = MINOR(rdev->bdev->bd_dev);
@@ -5624,6 +5669,7 @@ static int get_disk_info(struct mddev * mddev, void __user * arg)
 		info.raid_disk = -1;
 		info.state = (1<<MD_DISK_REMOVED);
 	}
+	rcu_read_unlock();
 
 	if (copy_to_user(arg, &info, sizeof(info)))
 		return -EFAULT;
@@ -6232,18 +6278,22 @@ static int update_array_info(struct mddev *mddev, mdu_array_info_t *info)
 static int set_disk_faulty(struct mddev *mddev, dev_t dev)
 {
 	struct md_rdev *rdev;
+	int err = 0;
 
 	if (mddev->pers == NULL)
 		return -ENODEV;
 
-	rdev = find_rdev(mddev, dev);
+	rcu_read_lock();
+	rdev = find_rdev_rcu(mddev, dev);
 	if (!rdev)
-		return -ENODEV;
-
-	md_error(mddev, rdev);
-	if (!test_bit(Faulty, &rdev->flags))
-		return -EBUSY;
-	return 0;
+		err =  -ENODEV;
+	else {
+		md_error(mddev, rdev);
+		if (!test_bit(Faulty, &rdev->flags))
+			err = -EBUSY;
+	}
+	rcu_read_unlock();
+	return err;
 }
 
 /*
@@ -6315,6 +6365,27 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode,
 		goto abort;
 	}
 
+	/* Some actions do not requires the mutex */
+	switch (cmd) {
+	case GET_ARRAY_INFO:
+		if (!mddev->raid_disks && !mddev->external)
+			err = -ENODEV;
+		else
+			err = get_array_info(mddev, argp);
+		goto abort;
+
+	case GET_DISK_INFO:
+		if (!mddev->raid_disks && !mddev->external)
+			err = -ENODEV;
+		else
+			err = get_disk_info(mddev, argp);
+		goto abort;
+
+	case SET_DISK_FAULTY:
+		err = set_disk_faulty(mddev, new_decode_dev(arg));
+		goto abort;
+	}
+
 	err = mddev_lock(mddev);
 	if (err) {
 		printk(KERN_INFO 
@@ -6387,18 +6458,10 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode,
 	 */
 	switch (cmd)
 	{
-		case GET_ARRAY_INFO:
-			err = get_array_info(mddev, argp);
-			goto done_unlock;
-
 		case GET_BITMAP_FILE:
 			err = get_bitmap_file(mddev, argp);
 			goto done_unlock;
 
-		case GET_DISK_INFO:
-			err = get_disk_info(mddev, argp);
-			goto done_unlock;
-
 		case RESTART_ARRAY_RW:
 			err = restart_array(mddev);
 			goto done_unlock;
@@ -6480,10 +6543,6 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode,
 			err = hot_add_disk(mddev, new_decode_dev(arg));
 			goto done_unlock;
 
-		case SET_DISK_FAULTY:
-			err = set_disk_faulty(mddev, new_decode_dev(arg));
-			goto done_unlock;
-
 		case RUN_ARRAY:
 			err = do_md_run(mddev);
 			goto done_unlock;
@@ -6641,7 +6700,7 @@ static int md_thread(void * arg)
 
 		clear_bit(THREAD_WAKEUP, &thread->flags);
 		if (!kthread_should_stop())
-			thread->run(thread->mddev);
+			thread->run(thread);
 	}
 
 	return 0;
@@ -6656,8 +6715,8 @@ void md_wakeup_thread(struct md_thread *thread)
 	}
 }
 
-struct md_thread *md_register_thread(void (*run) (struct mddev *), struct mddev *mddev,
-				 const char *name)
+struct md_thread *md_register_thread(void (*run) (struct md_thread *),
+		struct mddev *mddev, const char *name)
 {
 	struct md_thread *thread;
 
@@ -6752,7 +6811,11 @@ static void status_resync(struct seq_file *seq, struct mddev * mddev)
 	int scale;
 	unsigned int per_milli;
 
-	resync = mddev->curr_resync - atomic_read(&mddev->recovery_active);
+	if (mddev->curr_resync <= 3)
+		resync = 0;
+	else
+		resync = mddev->curr_resync
+			- atomic_read(&mddev->recovery_active);
 
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ||
 	    test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
@@ -6978,7 +7041,7 @@ static int md_seq_show(struct seq_file *seq, void *v)
 				if (mddev->curr_resync > 2) {
 					status_resync(seq, mddev);
 					seq_printf(seq, "\n      ");
-				} else if (mddev->curr_resync == 1 || mddev->curr_resync == 2)
+				} else if (mddev->curr_resync >= 1)
 					seq_printf(seq, "\tresync=DELAYED\n      ");
 				else if (mddev->recovery_cp < MaxSector)
 					seq_printf(seq, "\tresync=PENDING\n      ");
@@ -7206,8 +7269,9 @@ EXPORT_SYMBOL_GPL(md_allow_write);
 
 #define SYNC_MARKS	10
 #define	SYNC_MARK_STEP	(3*HZ)
-void md_do_sync(struct mddev *mddev)
+void md_do_sync(struct md_thread *thread)
 {
+	struct mddev *mddev = thread->mddev;
 	struct mddev *mddev2;
 	unsigned int currspeed = 0,
 		 window;
@@ -7311,7 +7375,7 @@ void md_do_sync(struct mddev *mddev)
 		 * which defaults to physical size, but can be virtual size
 		 */
 		max_sectors = mddev->resync_max_sectors;
-		mddev->resync_mismatches = 0;
+		atomic64_set(&mddev->resync_mismatches, 0);
 		/* we don't use the checkpoint if there's a bitmap */
 		if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
 			j = mddev->resync_min;
@@ -7367,8 +7431,11 @@ void md_do_sync(struct mddev *mddev)
 		       "md: resuming %s of %s from checkpoint.\n",
 		       desc, mdname(mddev));
 		mddev->curr_resync = j;
-	}
+	} else
+		mddev->curr_resync = 3; /* no longer delayed */
 	mddev->curr_resync_completed = j;
+	sysfs_notify(&mddev->kobj, NULL, "sync_completed");
+	md_new_event(mddev);
 
 	blk_start_plug(&plug);
 	while (j < max_sectors) {
@@ -7421,7 +7488,8 @@ void md_do_sync(struct mddev *mddev)
 			break;
 
 		j += sectors;
-		if (j>1) mddev->curr_resync = j;
+		if (j > 2)
+			mddev->curr_resync = j;
 		mddev->curr_mark_cnt = io_sectors;
 		if (last_check == 0)
 			/* this is the earliest that rebuild will be
@@ -7543,8 +7611,6 @@ static int remove_and_add_spares(struct mddev *mddev)
 	int spares = 0;
 	int removed = 0;
 
-	mddev->curr_resync_completed = 0;
-
 	rdev_for_each(rdev, mddev)
 		if (rdev->raid_disk >= 0 &&
 		    !test_bit(Blocked, &rdev->flags) &&
@@ -7739,6 +7805,7 @@ void md_check_recovery(struct mddev *mddev)
 		/* Set RUNNING before clearing NEEDED to avoid
 		 * any transients in the value of "sync_action".
 		 */
+		mddev->curr_resync_completed = 0;
 		set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
 		/* Clear some bits that don't mean anything, but
 		 * might be left set
@@ -7752,7 +7819,7 @@ void md_check_recovery(struct mddev *mddev)
 		/* no recovery is running.
 		 * remove any failed drives, then
 		 * add spares if possible.
-		 * Spare are also removed and re-added, to allow
+		 * Spares are also removed and re-added, to allow
 		 * the personality to fail the re-add.
 		 */