btrfs: Use ref_cnt for set_block_group_ro()

More than one code call set_block_group_ro() and restore rw in fail.

Old code use bool bit to save blockgroup's ro state, it can not
support parallel case(it is confirmd exist in my debug log).

This patch use ref count to store ro state, and rename
set_block_group_ro/set_block_group_rw
to
inc_block_group_ro/dec_block_group_ro.

Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index aac314e14188facff9ed95136eab647709352359..f57e6cae394b623ce6edefdabf5b75085f0b855e 100644 (file)
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1300,7 +1300,7 @@ struct btrfs_block_group_cache {
        /* for raid56, this is a full stripe, without parity */
        unsigned long full_stripe_len;
 
-       unsigned int ro:1;
+       unsigned int ro;
        unsigned int iref:1;
        unsigned int has_caching_ctl:1;
        unsigned int removed:1;
@@ -3495,9 +3495,9 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 void btrfs_block_rsv_release(struct btrfs_root *root,
                             struct btrfs_block_rsv *block_rsv,
                             u64 num_bytes);
-int btrfs_set_block_group_ro(struct btrfs_root *root,
+int btrfs_inc_block_group_ro(struct btrfs_root *root,
                             struct btrfs_block_group_cache *cache);
-void btrfs_set_block_group_rw(struct btrfs_root *root,
+void btrfs_dec_block_group_ro(struct btrfs_root *root,
                              struct btrfs_block_group_cache *cache);
 void btrfs_put_block_group_cache(struct btrfs_fs_info *info);
 u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo);

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 07204bf601edac25ad2757bbd7c85961eeabc56c..5cefa02b40a9e09b98895c74978de3cd4dda8eb9 100644 (file)
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -8723,14 +8723,13 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags)
        return flags;
 }
 
-static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force)
+static int inc_block_group_ro(struct btrfs_block_group_cache *cache, int force)
 {
        struct btrfs_space_info *sinfo = cache->space_info;
        u64 num_bytes;
        u64 min_allocable_bytes;
        int ret = -ENOSPC;
 
-
        /*
         * We need some metadata space and system metadata space for
         * allocating chunks in some corner cases until we force to set
@@ -8747,6 +8746,7 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force)
        spin_lock(&cache->lock);
 
        if (cache->ro) {
+               cache->ro++;
                ret = 0;
                goto out;
        }
@@ -8758,7 +8758,7 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force)
            sinfo->bytes_may_use + sinfo->bytes_readonly + num_bytes +
            min_allocable_bytes <= sinfo->total_bytes) {
                sinfo->bytes_readonly += num_bytes;
-               cache->ro = 1;
+               cache->ro++;
                list_add_tail(&cache->ro_list, &sinfo->ro_bgs);
                ret = 0;
        }
@@ -8768,7 +8768,7 @@ out:
        return ret;
 }
 
-int btrfs_set_block_group_ro(struct btrfs_root *root,
+int btrfs_inc_block_group_ro(struct btrfs_root *root,
                             struct btrfs_block_group_cache *cache)
 
 {
@@ -8776,8 +8776,6 @@ int btrfs_set_block_group_ro(struct btrfs_root *root,
        u64 alloc_flags;
        int ret;
 
-       BUG_ON(cache->ro);
-
 again:
        trans = btrfs_join_transaction(root);
        if (IS_ERR(trans))
@@ -8820,7 +8818,7 @@ again:
                        goto out;
        }
 
-       ret = set_block_group_ro(cache, 0);
+       ret = inc_block_group_ro(cache, 0);
        if (!ret)
                goto out;
        alloc_flags = get_alloc_profile(root, cache->space_info->flags);
@@ -8828,7 +8826,7 @@ again:
                             CHUNK_ALLOC_FORCE);
        if (ret < 0)
                goto out;
-       ret = set_block_group_ro(cache, 0);
+       ret = inc_block_group_ro(cache, 0);
 out:
        if (cache->flags & BTRFS_BLOCK_GROUP_SYSTEM) {
                alloc_flags = update_block_group_flags(root, cache->flags);
@@ -8891,7 +8889,7 @@ u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo)
        return free_bytes;
 }
 
-void btrfs_set_block_group_rw(struct btrfs_root *root,
+void btrfs_dec_block_group_ro(struct btrfs_root *root,
                              struct btrfs_block_group_cache *cache)
 {
        struct btrfs_space_info *sinfo = cache->space_info;
@@ -8901,11 +8899,13 @@ void btrfs_set_block_group_rw(struct btrfs_root *root,
 
        spin_lock(&sinfo->lock);
        spin_lock(&cache->lock);
-       num_bytes = cache->key.offset - cache->reserved - cache->pinned -
-                   cache->bytes_super - btrfs_block_group_used(&cache->item);
-       sinfo->bytes_readonly -= num_bytes;
-       cache->ro = 0;
-       list_del_init(&cache->ro_list);
+       if (!--cache->ro) {
+               num_bytes = cache->key.offset - cache->reserved -
+                           cache->pinned - cache->bytes_super -
+                           btrfs_block_group_used(&cache->item);
+               sinfo->bytes_readonly -= num_bytes;
+               list_del_init(&cache->ro_list);
+       }
        spin_unlock(&cache->lock);
        spin_unlock(&sinfo->lock);
 }
@@ -9421,7 +9421,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 
                set_avail_alloc_bits(root->fs_info, cache->flags);
                if (btrfs_chunk_readonly(root, cache->key.objectid)) {
-                       set_block_group_ro(cache, 1);
+                       inc_block_group_ro(cache, 1);
                } else if (btrfs_block_group_used(&cache->item) == 0) {
                        spin_lock(&info->unused_bgs_lock);
                        /* Should always be true but just in case. */
@@ -9449,11 +9449,11 @@ int btrfs_read_block_groups(struct btrfs_root *root)
                list_for_each_entry(cache,
                                &space_info->block_groups[BTRFS_RAID_RAID0],
                                list)
-                       set_block_group_ro(cache, 1);
+                       inc_block_group_ro(cache, 1);
                list_for_each_entry(cache,
                                &space_info->block_groups[BTRFS_RAID_SINGLE],
                                list)
-                       set_block_group_ro(cache, 1);
+                       inc_block_group_ro(cache, 1);
        }
 
        init_global_block_rsv(info);
@@ -9941,7 +9941,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
                spin_unlock(&block_group->lock);
 
                /* We don't want to force the issue, only flip if it's ok. */
-               ret = set_block_group_ro(block_group, 0);
+               ret = inc_block_group_ro(block_group, 0);
                up_write(&space_info->groups_sem);
                if (ret < 0) {
                        ret = 0;
@@ -9955,7 +9955,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
                /* 1 for btrfs_orphan_reserve_metadata() */
                trans = btrfs_start_transaction(root, 1);
                if (IS_ERR(trans)) {
-                       btrfs_set_block_group_rw(root, block_group);
+                       btrfs_dec_block_group_ro(root, block_group);
                        ret = PTR_ERR(trans);
                        goto next;
                }
@@ -9982,14 +9982,14 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
                                  EXTENT_DIRTY, GFP_NOFS);
                if (ret) {
                        mutex_unlock(&fs_info->unused_bg_unpin_mutex);
-                       btrfs_set_block_group_rw(root, block_group);
+                       btrfs_dec_block_group_ro(root, block_group);
                        goto end_trans;
                }
                ret = clear_extent_bits(&fs_info->freed_extents[1], start, end,
                                  EXTENT_DIRTY, GFP_NOFS);
                if (ret) {
                        mutex_unlock(&fs_info->unused_bg_unpin_mutex);
-                       btrfs_set_block_group_rw(root, block_group);
+                       btrfs_dec_block_group_ro(root, block_group);
                        goto end_trans;
                }
                mutex_unlock(&fs_info->unused_bg_unpin_mutex);

diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 88cbb5995667951192b94c2627f759a771eda9e3..52fe55ad11d66e5b8ad2922c8f26d08e2dc23055 100644 (file)
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -4215,14 +4215,12 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
        rc->block_group = btrfs_lookup_block_group(fs_info, group_start);
        BUG_ON(!rc->block_group);
 
-       if (!rc->block_group->ro) {
-               ret = btrfs_set_block_group_ro(extent_root, rc->block_group);
-               if (ret) {
-                       err = ret;
-                       goto out;
-               }
-               rw = 1;
+       ret = btrfs_inc_block_group_ro(extent_root, rc->block_group);
+       if (ret) {
+               err = ret;
+               goto out;
        }
+       rw = 1;
 
        path = btrfs_alloc_path();
        if (!path) {
@@ -4294,7 +4292,7 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
        WARN_ON(btrfs_block_group_used(&rc->block_group->item) > 0);
 out:
        if (err && rw)
-               btrfs_set_block_group_rw(extent_root, rc->block_group);
+               btrfs_dec_block_group_ro(extent_root, rc->block_group);
        iput(rc->data_inode);
        btrfs_put_block_group(rc->block_group);
        kfree(rc);