1 files changed, 71 insertions, 8 deletions

diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index 564c92638b20..9f2290509aca 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -431,6 +431,35 @@ leave_no_lock:
         return ret;
 }
 
+/*
+ * blocked until all flighting bios are finished.
+ */
+static void btrfs_rm_dev_replace_blocked(struct btrfs_fs_info *fs_info)
+{
+        s64 writers;
+        DEFINE_WAIT(wait);
+
+        set_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state);
+        do {
+                prepare_to_wait(&fs_info->replace_wait, &wait,
+                                TASK_UNINTERRUPTIBLE);
+                writers = percpu_counter_sum(&fs_info->bio_counter);
+                if (writers)
+                        schedule();
+                finish_wait(&fs_info->replace_wait, &wait);
+        } while (writers);
+}
+
+/*
+ * we have removed target device, it is safe to allow new bios request.
+ */
+static void btrfs_rm_dev_replace_unblocked(struct btrfs_fs_info *fs_info)
+{
+        clear_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state);
+        if (waitqueue_active(&fs_info->replace_wait))
+                wake_up(&fs_info->replace_wait);
+}
+
 static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
                                        int scrub_ret)
 {
@@ -458,17 +487,11 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
         src_device = dev_replace->srcdev;
         btrfs_dev_replace_unlock(dev_replace);
 
-        /* replace old device with new one in mapping tree */
-        if (!scrub_ret)
-                btrfs_dev_replace_update_device_in_mapping_tree(fs_info,
-                                                                src_device,
-                                                                tgt_device);
-
         /*
          * flush all outstanding I/O and inode extent mappings before the
          * copy operation is declared as being finished
          */
-        ret = btrfs_start_delalloc_roots(root->fs_info, 0);
+        ret = btrfs_start_delalloc_roots(root->fs_info, 0, -1);
         if (ret) {
                 mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
                 return ret;
@@ -484,6 +507,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
         WARN_ON(ret);
 
         /* keep away write_all_supers() during the finishing procedure */
+        mutex_lock(&root->fs_info->chunk_mutex);
         mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
         btrfs_dev_replace_lock(dev_replace);
         dev_replace->replace_state =
@@ -494,7 +518,12 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
         dev_replace->time_stopped = get_seconds();
         dev_replace->item_needs_writeback = 1;
 
-        if (scrub_ret) {
+        /* replace old device with new one in mapping tree */
+        if (!scrub_ret) {
+                btrfs_dev_replace_update_device_in_mapping_tree(fs_info,
+                                                                src_device,
+                                                                tgt_device);
+        } else {
                 printk_in_rcu(KERN_ERR
                               "BTRFS: btrfs_scrub_dev(%s, %llu, %s) failed %d\n",
                               src_device->missing ? "<missing disk>" :
@@ -503,6 +532,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
                               rcu_str_deref(tgt_device->name), scrub_ret);
                 btrfs_dev_replace_unlock(dev_replace);
                 mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
+                mutex_unlock(&root->fs_info->chunk_mutex);
                 if (tgt_device)
                         btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
                 mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
@@ -532,8 +562,12 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
                 fs_info->fs_devices->latest_bdev = tgt_device->bdev;
         list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list);
 
+        btrfs_rm_dev_replace_blocked(fs_info);
+
         btrfs_rm_dev_replace_srcdev(fs_info, src_device);
 
+        btrfs_rm_dev_replace_unblocked(fs_info);
+
         /*
          * this is again a consistent state where no dev_replace procedure
          * is running, the target device is part of the filesystem, the
@@ -543,6 +577,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
          */
         btrfs_dev_replace_unlock(dev_replace);
         mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
+        mutex_unlock(&root->fs_info->chunk_mutex);
 
         /* write back the superblocks */
         trans = btrfs_start_transaction(root, 0);
@@ -862,3 +897,31 @@ void btrfs_dev_replace_unlock(struct btrfs_dev_replace *dev_replace)
                 mutex_unlock(&dev_replace->lock_management_lock);
         }
 }
+
+void btrfs_bio_counter_inc_noblocked(struct btrfs_fs_info *fs_info)
+{
+        percpu_counter_inc(&fs_info->bio_counter);
+}
+
+void btrfs_bio_counter_dec(struct btrfs_fs_info *fs_info)
+{
+        percpu_counter_dec(&fs_info->bio_counter);
+
+        if (waitqueue_active(&fs_info->replace_wait))
+                wake_up(&fs_info->replace_wait);
+}
+
+void btrfs_bio_counter_inc_blocked(struct btrfs_fs_info *fs_info)
+{
+        DEFINE_WAIT(wait);
+again:
+        percpu_counter_inc(&fs_info->bio_counter);
+        if (test_bit(BTRFS_FS_STATE_DEV_REPLACING, &fs_info->fs_state)) {
+                btrfs_bio_counter_dec(fs_info);
+                wait_event(fs_info->replace_wait,
+                           !test_bit(BTRFS_FS_STATE_DEV_REPLACING,
+                                     &fs_info->fs_state));
+                goto again;
+        }
+
+}