Btrfs: use seqlock to protect fs_info->avail_{data, metadata, system}_alloc_bits

There is no lock to protect
  fs_info->avail_{data, metadata, system}_alloc_bits,
it may introduce some problem, such as the wrong profile
information, so we add a seqlock to protect them.

Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>

4 files changed, 49 insertions, 32 deletions

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 50def99f5379..5f5c30aaef36 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1483,6 +1483,8 @@ struct btrfs_fs_info {
         struct rb_root defrag_inodes;
         atomic_t defrag_running;
 
+        /* Used to protect avail_{data, metadata, system}_alloc_bits */
+        seqlock_t profiles_lock;
         /*
          * these three are in extended format (availability of single
          * chunks is denoted by BTRFS_AVAIL_ALLOC_BIT_SINGLE bit, other
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 11f6dbcb1191..00b6742fdde7 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -2040,6 +2040,7 @@ int open_ctree(struct super_block *sb,
         spin_lock_init(&fs_info->tree_mod_seq_lock);
         rwlock_init(&fs_info->tree_mod_log_lock);
         mutex_init(&fs_info->reloc_mutex);
+        seqlock_init(&fs_info->profiles_lock);
 
         init_completion(&fs_info->kobj_unregister);
         INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 115d1646bf50..faff98f720de 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3227,12 +3227,14 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
         u64 extra_flags = chunk_to_extended(flags) &
                                 BTRFS_EXTENDED_PROFILE_MASK;
 
+        write_seqlock(&fs_info->profiles_lock);
         if (flags & BTRFS_BLOCK_GROUP_DATA)
                 fs_info->avail_data_alloc_bits |= extra_flags;
         if (flags & BTRFS_BLOCK_GROUP_METADATA)
                 fs_info->avail_metadata_alloc_bits |= extra_flags;
         if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
                 fs_info->avail_system_alloc_bits |= extra_flags;
+        write_sequnlock(&fs_info->profiles_lock);
 }
 
 /*
@@ -3324,12 +3326,18 @@ u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags)
 
 static u64 get_alloc_profile(struct btrfs_root *root, u64 flags)
 {
-        if (flags & BTRFS_BLOCK_GROUP_DATA)
-                flags |= root->fs_info->avail_data_alloc_bits;
-        else if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
-                flags |= root->fs_info->avail_system_alloc_bits;
-        else if (flags & BTRFS_BLOCK_GROUP_METADATA)
-                flags |= root->fs_info->avail_metadata_alloc_bits;
+        unsigned seq;
+
+        do {
+                seq = read_seqbegin(&root->fs_info->profiles_lock);
+
+                if (flags & BTRFS_BLOCK_GROUP_DATA)
+                        flags |= root->fs_info->avail_data_alloc_bits;
+                else if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
+                        flags |= root->fs_info->avail_system_alloc_bits;
+                else if (flags & BTRFS_BLOCK_GROUP_METADATA)
+                        flags |= root->fs_info->avail_metadata_alloc_bits;
+        } while (read_seqretry(&root->fs_info->profiles_lock, seq));
 
         return btrfs_reduce_alloc_profile(root, flags);
 }
@@ -7967,12 +7975,14 @@ static void clear_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags)
         u64 extra_flags = chunk_to_extended(flags) &
                                 BTRFS_EXTENDED_PROFILE_MASK;
 
+        write_seqlock(&fs_info->profiles_lock);
         if (flags & BTRFS_BLOCK_GROUP_DATA)
                 fs_info->avail_data_alloc_bits &= ~extra_flags;
         if (flags & BTRFS_BLOCK_GROUP_METADATA)
                 fs_info->avail_metadata_alloc_bits &= ~extra_flags;
         if (flags & BTRFS_BLOCK_GROUP_SYSTEM)
                 fs_info->avail_system_alloc_bits &= ~extra_flags;
+        write_sequnlock(&fs_info->profiles_lock);
 }
 
 int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 1ce581d58d87..8c9ea4cd66bb 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1430,14 +1430,19 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
         u64 devid;
         u64 num_devices;
         u8 *dev_uuid;
+        unsigned seq;
         int ret = 0;
         bool clear_super = false;
 
         mutex_lock(&uuid_mutex);
 
-        all_avail = root->fs_info->avail_data_alloc_bits |
-                root->fs_info->avail_system_alloc_bits |
-                root->fs_info->avail_metadata_alloc_bits;
+        do {
+                seq = read_seqbegin(&root->fs_info->profiles_lock);
+
+                all_avail = root->fs_info->avail_data_alloc_bits |
+                            root->fs_info->avail_system_alloc_bits |
+                            root->fs_info->avail_metadata_alloc_bits;
+        } while (read_seqretry(&root->fs_info->profiles_lock, seq));
 
         num_devices = root->fs_info->fs_devices->num_devices;
         btrfs_dev_replace_lock(&root->fs_info->dev_replace);
@@ -3043,6 +3048,7 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
         int mixed = 0;
         int ret;
         u64 num_devices;
+        unsigned seq;
 
         if (btrfs_fs_closing(fs_info) ||
             atomic_read(&fs_info->balance_pause_req) ||
@@ -3126,22 +3132,26 @@ int btrfs_balance(struct btrfs_balance_control *bctl,
         /* allow to reduce meta or sys integrity only if force set */
         allowed = BTRFS_BLOCK_GROUP_DUP | BTRFS_BLOCK_GROUP_RAID1 |
                         BTRFS_BLOCK_GROUP_RAID10;
-        if (((bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
-             (fs_info->avail_system_alloc_bits & allowed) &&
-             !(bctl->sys.target & allowed)) ||
-            ((bctl->meta.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
-             (fs_info->avail_metadata_alloc_bits & allowed) &&
-             !(bctl->meta.target & allowed))) {
-                if (bctl->flags & BTRFS_BALANCE_FORCE) {
-                        printk(KERN_INFO "btrfs: force reducing metadata "
-                               "integrity\n");
-                } else {
-                        printk(KERN_ERR "btrfs: balance will reduce metadata "
-                               "integrity, use force if you want this\n");
-                        ret = -EINVAL;
-                        goto out;
+        do {
+                seq = read_seqbegin(&fs_info->profiles_lock);
+
+                if (((bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
+                     (fs_info->avail_system_alloc_bits & allowed) &&
+                     !(bctl->sys.target & allowed)) ||
+                    ((bctl->meta.flags & BTRFS_BALANCE_ARGS_CONVERT) &&
+                     (fs_info->avail_metadata_alloc_bits & allowed) &&
+                     !(bctl->meta.target & allowed))) {
+                        if (bctl->flags & BTRFS_BALANCE_FORCE) {
+                                printk(KERN_INFO "btrfs: force reducing metadata "
+                                       "integrity\n");
+                        } else {
+                                printk(KERN_ERR "btrfs: balance will reduce metadata "
+                                       "integrity, use force if you want this\n");
+                                ret = -EINVAL;
+                                goto out;
+                        }
                 }
-        }
+        } while (read_seqretry(&fs_info->profiles_lock, seq));
 
         if (bctl->sys.flags & BTRFS_BALANCE_ARGS_CONVERT) {
                 int num_tolerated_disk_barrier_failures;
@@ -3980,10 +3990,7 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans,
         if (ret)
                 return ret;
 
-        alloc_profile = BTRFS_BLOCK_GROUP_METADATA |
-                                fs_info->avail_metadata_alloc_bits;
-        alloc_profile = btrfs_reduce_alloc_profile(root, alloc_profile);
-
+        alloc_profile = btrfs_get_alloc_profile(extent_root, 0);
         ret = __btrfs_alloc_chunk(trans, extent_root, &map, &chunk_size,
                                   &stripe_size, chunk_offset, alloc_profile);
         if (ret)
@@ -3991,10 +3998,7 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans,
 
         sys_chunk_offset = chunk_offset + chunk_size;
 
-        alloc_profile = BTRFS_BLOCK_GROUP_SYSTEM |
-                                fs_info->avail_system_alloc_bits;
-        alloc_profile = btrfs_reduce_alloc_profile(root, alloc_profile);
-
+        alloc_profile = btrfs_get_alloc_profile(fs_info->chunk_root, 0);
         ret = __btrfs_alloc_chunk(trans, extent_root, &sys_map,
                                   &sys_chunk_size, &sys_stripe_size,
                                   sys_chunk_offset, alloc_profile);