summaryrefslogtreecommitdiff
path: root/fs/btrfs
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2022-08-19 23:33:48 +0300
committerLinus Torvalds <torvalds@linux-foundation.org>2022-08-19 23:33:48 +0300
commit42c54d5491ed7b9fe89a499224494277a33b23df (patch)
treec26ad94e0ae342c5c9720a68db7b9829a1f72582 /fs/btrfs
parenta3a78b6332ba8e1949541150efcc8051ceeb2762 (diff)
parent899b7f69f244e539ea5df1b4d756046337de44a5 (diff)
downloadlinux-42c54d5491ed7b9fe89a499224494277a33b23df.tar.xz
Merge tag 'for-6.0-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux
Pull btrfs fixes from David Sterba: "A few short fixes and a lockdep warning fix (needs moving some code): - tree-log replay fixes: - fix error handling when looking up extent refs - fix warning when setting inode number of links - relocation fixes: - reset block group read-only status when relocation fails - unset control structure if transaction fails when starting to process a block group - add lockdep annotations to fix a warning during relocation where blocks temporarily belong to another tree and can lead to reversed dependencies - tree-checker verifies that extent items don't overlap" * tag 'for-6.0-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: btrfs: tree-checker: check for overlapping extent items btrfs: fix warning during log replay when bumping inode link count btrfs: fix lost error handling when looking up extended ref on log replay btrfs: fix lockdep splat with reloc root extent buffers btrfs: move lockdep class helpers to locking.c btrfs: unset reloc control if transaction commit fails in prepare_to_relocate() btrfs: reset RO counter on block group if we fail to relocate
Diffstat (limited to 'fs/btrfs')
-rw-r--r--fs/btrfs/block-group.c4
-rw-r--r--fs/btrfs/ctree.c3
-rw-r--r--fs/btrfs/ctree.h2
-rw-r--r--fs/btrfs/disk-io.c82
-rw-r--r--fs/btrfs/disk-io.h10
-rw-r--r--fs/btrfs/extent-tree.c18
-rw-r--r--fs/btrfs/extent_io.c11
-rw-r--r--fs/btrfs/locking.c91
-rw-r--r--fs/btrfs/locking.h14
-rw-r--r--fs/btrfs/relocation.c9
-rw-r--r--fs/btrfs/tree-checker.c25
-rw-r--r--fs/btrfs/tree-log.c8
12 files changed, 176 insertions, 101 deletions
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index c3aecfb0a71d..993aca2f1e18 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -1640,9 +1640,11 @@ void btrfs_reclaim_bgs_work(struct work_struct *work)
div64_u64(zone_unusable * 100, bg->length));
trace_btrfs_reclaim_block_group(bg);
ret = btrfs_relocate_chunk(fs_info, bg->start);
- if (ret)
+ if (ret) {
+ btrfs_dec_block_group_ro(bg);
btrfs_err(fs_info, "error relocating chunk %llu",
bg->start);
+ }
next:
btrfs_put_block_group(bg);
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 6e556031a8f3..ebfa35fe1c38 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2075,6 +2075,9 @@ cow_done:
if (!p->skip_locking) {
level = btrfs_header_level(b);
+
+ btrfs_maybe_reset_lockdep_class(root, b);
+
if (level <= write_lock_level) {
btrfs_tree_lock(b);
p->locks[level] = BTRFS_WRITE_LOCK;
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 4db85b9dc7ed..4edb4bfb2166 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1173,6 +1173,8 @@ enum {
BTRFS_ROOT_ORPHAN_CLEANUP,
/* This root has a drop operation that was started previously. */
BTRFS_ROOT_UNFINISHED_DROP,
+ /* This reloc root needs to have its buffers lockdep class reset. */
+ BTRFS_ROOT_RESET_LOCKDEP_CLASS,
};
static inline void btrfs_wake_unfinished_drop(struct btrfs_fs_info *fs_info)
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 4c3166f3c725..820b1f1e6b67 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -87,88 +87,6 @@ struct async_submit_bio {
};
/*
- * Lockdep class keys for extent_buffer->lock's in this root. For a given
- * eb, the lockdep key is determined by the btrfs_root it belongs to and
- * the level the eb occupies in the tree.
- *
- * Different roots are used for different purposes and may nest inside each
- * other and they require separate keysets. As lockdep keys should be
- * static, assign keysets according to the purpose of the root as indicated
- * by btrfs_root->root_key.objectid. This ensures that all special purpose
- * roots have separate keysets.
- *
- * Lock-nesting across peer nodes is always done with the immediate parent
- * node locked thus preventing deadlock. As lockdep doesn't know this, use
- * subclass to avoid triggering lockdep warning in such cases.
- *
- * The key is set by the readpage_end_io_hook after the buffer has passed
- * csum validation but before the pages are unlocked. It is also set by
- * btrfs_init_new_buffer on freshly allocated blocks.
- *
- * We also add a check to make sure the highest level of the tree is the
- * same as our lockdep setup here. If BTRFS_MAX_LEVEL changes, this code
- * needs update as well.
- */
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-# if BTRFS_MAX_LEVEL != 8
-# error
-# endif
-
-#define DEFINE_LEVEL(stem, level) \
- .names[level] = "btrfs-" stem "-0" #level,
-
-#define DEFINE_NAME(stem) \
- DEFINE_LEVEL(stem, 0) \
- DEFINE_LEVEL(stem, 1) \
- DEFINE_LEVEL(stem, 2) \
- DEFINE_LEVEL(stem, 3) \
- DEFINE_LEVEL(stem, 4) \
- DEFINE_LEVEL(stem, 5) \
- DEFINE_LEVEL(stem, 6) \
- DEFINE_LEVEL(stem, 7)
-
-static struct btrfs_lockdep_keyset {
- u64 id; /* root objectid */
- /* Longest entry: btrfs-free-space-00 */
- char names[BTRFS_MAX_LEVEL][20];
- struct lock_class_key keys[BTRFS_MAX_LEVEL];
-} btrfs_lockdep_keysets[] = {
- { .id = BTRFS_ROOT_TREE_OBJECTID, DEFINE_NAME("root") },
- { .id = BTRFS_EXTENT_TREE_OBJECTID, DEFINE_NAME("extent") },
- { .id = BTRFS_CHUNK_TREE_OBJECTID, DEFINE_NAME("chunk") },
- { .id = BTRFS_DEV_TREE_OBJECTID, DEFINE_NAME("dev") },
- { .id = BTRFS_CSUM_TREE_OBJECTID, DEFINE_NAME("csum") },
- { .id = BTRFS_QUOTA_TREE_OBJECTID, DEFINE_NAME("quota") },
- { .id = BTRFS_TREE_LOG_OBJECTID, DEFINE_NAME("log") },
- { .id = BTRFS_TREE_RELOC_OBJECTID, DEFINE_NAME("treloc") },
- { .id = BTRFS_DATA_RELOC_TREE_OBJECTID, DEFINE_NAME("dreloc") },
- { .id = BTRFS_UUID_TREE_OBJECTID, DEFINE_NAME("uuid") },
- { .id = BTRFS_FREE_SPACE_TREE_OBJECTID, DEFINE_NAME("free-space") },
- { .id = 0, DEFINE_NAME("tree") },
-};
-
-#undef DEFINE_LEVEL
-#undef DEFINE_NAME
-
-void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb,
- int level)
-{
- struct btrfs_lockdep_keyset *ks;
-
- BUG_ON(level >= ARRAY_SIZE(ks->keys));
-
- /* find the matching keyset, id 0 is the default entry */
- for (ks = btrfs_lockdep_keysets; ks->id; ks++)
- if (ks->id == objectid)
- break;
-
- lockdep_set_class_and_name(&eb->lock,
- &ks->keys[level], ks->names[level]);
-}
-
-#endif
-
-/*
* Compute the csum of a btree block and store the result to provided buffer.
*/
static void csum_tree_block(struct extent_buffer *buf, u8 *result)
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 8993b428e09c..47ad8e0a2d33 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -137,14 +137,4 @@ int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags);
int btrfs_get_free_objectid(struct btrfs_root *root, u64 *objectid);
int btrfs_init_root_free_objectid(struct btrfs_root *root);
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-void btrfs_set_buffer_lockdep_class(u64 objectid,
- struct extent_buffer *eb, int level);
-#else
-static inline void btrfs_set_buffer_lockdep_class(u64 objectid,
- struct extent_buffer *eb, int level)
-{
-}
-#endif
-
#endif
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index ea3ec1e761e8..ab944d1f94ef 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -4867,6 +4867,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct extent_buffer *buf;
+ u64 lockdep_owner = owner;
buf = btrfs_find_create_tree_block(fs_info, bytenr, owner, level);
if (IS_ERR(buf))
@@ -4886,11 +4887,26 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
}
/*
+ * The reloc trees are just snapshots, so we need them to appear to be
+ * just like any other fs tree WRT lockdep.
+ *
+ * The exception however is in replace_path() in relocation, where we
+ * hold the lock on the original fs root and then search for the reloc
+ * root. At that point we need to make sure any reloc root buffers are
+ * set to the BTRFS_TREE_RELOC_OBJECTID lockdep class in order to make
+ * lockdep happy.
+ */
+ if (lockdep_owner == BTRFS_TREE_RELOC_OBJECTID &&
+ !test_bit(BTRFS_ROOT_RESET_LOCKDEP_CLASS, &root->state))
+ lockdep_owner = BTRFS_FS_TREE_OBJECTID;
+
+ /*
* This needs to stay, because we could allocate a freed block from an
* old tree into a new tree, so we need to make sure this new block is
* set to the appropriate level and owner.
*/
- btrfs_set_buffer_lockdep_class(owner, buf, level);
+ btrfs_set_buffer_lockdep_class(lockdep_owner, buf, level);
+
__btrfs_tree_lock(buf, nest);
btrfs_clean_tree_block(buf);
clear_bit(EXTENT_BUFFER_STALE, &buf->bflags);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index bfae67c593c5..eed81a7e36a4 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -6140,6 +6140,7 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
struct extent_buffer *exists = NULL;
struct page *p;
struct address_space *mapping = fs_info->btree_inode->i_mapping;
+ u64 lockdep_owner = owner_root;
int uptodate = 1;
int ret;
@@ -6164,7 +6165,15 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
eb = __alloc_extent_buffer(fs_info, start, len);
if (!eb)
return ERR_PTR(-ENOMEM);
- btrfs_set_buffer_lockdep_class(owner_root, eb, level);
+
+ /*
+ * The reloc trees are just snapshots, so we need them to appear to be
+ * just like any other fs tree WRT lockdep.
+ */
+ if (lockdep_owner == BTRFS_TREE_RELOC_OBJECTID)
+ lockdep_owner = BTRFS_FS_TREE_OBJECTID;
+
+ btrfs_set_buffer_lockdep_class(lockdep_owner, eb, level);
num_pages = num_extent_pages(eb);
for (i = 0; i < num_pages; i++, index++) {
diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c
index 33461b4f9c8b..9063072b399b 100644
--- a/fs/btrfs/locking.c
+++ b/fs/btrfs/locking.c
@@ -14,6 +14,93 @@
#include "locking.h"
/*
+ * Lockdep class keys for extent_buffer->lock's in this root. For a given
+ * eb, the lockdep key is determined by the btrfs_root it belongs to and
+ * the level the eb occupies in the tree.
+ *
+ * Different roots are used for different purposes and may nest inside each
+ * other and they require separate keysets. As lockdep keys should be
+ * static, assign keysets according to the purpose of the root as indicated
+ * by btrfs_root->root_key.objectid. This ensures that all special purpose
+ * roots have separate keysets.
+ *
+ * Lock-nesting across peer nodes is always done with the immediate parent
+ * node locked thus preventing deadlock. As lockdep doesn't know this, use
+ * subclass to avoid triggering lockdep warning in such cases.
+ *
+ * The key is set by the readpage_end_io_hook after the buffer has passed
+ * csum validation but before the pages are unlocked. It is also set by
+ * btrfs_init_new_buffer on freshly allocated blocks.
+ *
+ * We also add a check to make sure the highest level of the tree is the
+ * same as our lockdep setup here. If BTRFS_MAX_LEVEL changes, this code
+ * needs update as well.
+ */
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+#if BTRFS_MAX_LEVEL != 8
+#error
+#endif
+
+#define DEFINE_LEVEL(stem, level) \
+ .names[level] = "btrfs-" stem "-0" #level,
+
+#define DEFINE_NAME(stem) \
+ DEFINE_LEVEL(stem, 0) \
+ DEFINE_LEVEL(stem, 1) \
+ DEFINE_LEVEL(stem, 2) \
+ DEFINE_LEVEL(stem, 3) \
+ DEFINE_LEVEL(stem, 4) \
+ DEFINE_LEVEL(stem, 5) \
+ DEFINE_LEVEL(stem, 6) \
+ DEFINE_LEVEL(stem, 7)
+
+static struct btrfs_lockdep_keyset {
+ u64 id; /* root objectid */
+ /* Longest entry: btrfs-free-space-00 */
+ char names[BTRFS_MAX_LEVEL][20];
+ struct lock_class_key keys[BTRFS_MAX_LEVEL];
+} btrfs_lockdep_keysets[] = {
+ { .id = BTRFS_ROOT_TREE_OBJECTID, DEFINE_NAME("root") },
+ { .id = BTRFS_EXTENT_TREE_OBJECTID, DEFINE_NAME("extent") },
+ { .id = BTRFS_CHUNK_TREE_OBJECTID, DEFINE_NAME("chunk") },
+ { .id = BTRFS_DEV_TREE_OBJECTID, DEFINE_NAME("dev") },
+ { .id = BTRFS_CSUM_TREE_OBJECTID, DEFINE_NAME("csum") },
+ { .id = BTRFS_QUOTA_TREE_OBJECTID, DEFINE_NAME("quota") },
+ { .id = BTRFS_TREE_LOG_OBJECTID, DEFINE_NAME("log") },
+ { .id = BTRFS_TREE_RELOC_OBJECTID, DEFINE_NAME("treloc") },
+ { .id = BTRFS_DATA_RELOC_TREE_OBJECTID, DEFINE_NAME("dreloc") },
+ { .id = BTRFS_UUID_TREE_OBJECTID, DEFINE_NAME("uuid") },
+ { .id = BTRFS_FREE_SPACE_TREE_OBJECTID, DEFINE_NAME("free-space") },
+ { .id = 0, DEFINE_NAME("tree") },
+};
+
+#undef DEFINE_LEVEL
+#undef DEFINE_NAME
+
+void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb, int level)
+{
+ struct btrfs_lockdep_keyset *ks;
+
+ BUG_ON(level >= ARRAY_SIZE(ks->keys));
+
+ /* Find the matching keyset, id 0 is the default entry */
+ for (ks = btrfs_lockdep_keysets; ks->id; ks++)
+ if (ks->id == objectid)
+ break;
+
+ lockdep_set_class_and_name(&eb->lock, &ks->keys[level], ks->names[level]);
+}
+
+void btrfs_maybe_reset_lockdep_class(struct btrfs_root *root, struct extent_buffer *eb)
+{
+ if (test_bit(BTRFS_ROOT_RESET_LOCKDEP_CLASS, &root->state))
+ btrfs_set_buffer_lockdep_class(root->root_key.objectid,
+ eb, btrfs_header_level(eb));
+}
+
+#endif
+
+/*
* Extent buffer locking
* =====================
*
@@ -164,6 +251,8 @@ struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root)
while (1) {
eb = btrfs_root_node(root);
+
+ btrfs_maybe_reset_lockdep_class(root, eb);
btrfs_tree_lock(eb);
if (eb == root->node)
break;
@@ -185,6 +274,8 @@ struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
while (1) {
eb = btrfs_root_node(root);
+
+ btrfs_maybe_reset_lockdep_class(root, eb);
btrfs_tree_read_lock(eb);
if (eb == root->node)
break;
diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h
index bbc45534ae9a..ab268be09bb5 100644
--- a/fs/btrfs/locking.h
+++ b/fs/btrfs/locking.h
@@ -131,4 +131,18 @@ void btrfs_drew_write_unlock(struct btrfs_drew_lock *lock);
void btrfs_drew_read_lock(struct btrfs_drew_lock *lock);
void btrfs_drew_read_unlock(struct btrfs_drew_lock *lock);
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb, int level);
+void btrfs_maybe_reset_lockdep_class(struct btrfs_root *root, struct extent_buffer *eb);
+#else
+static inline void btrfs_set_buffer_lockdep_class(u64 objectid,
+ struct extent_buffer *eb, int level)
+{
+}
+static inline void btrfs_maybe_reset_lockdep_class(struct btrfs_root *root,
+ struct extent_buffer *eb)
+{
+}
+#endif
+
#endif
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index a6dc827e75af..45c02aba2492 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1326,7 +1326,9 @@ again:
btrfs_release_path(path);
path->lowest_level = level;
+ set_bit(BTRFS_ROOT_RESET_LOCKDEP_CLASS, &src->state);
ret = btrfs_search_slot(trans, src, &key, path, 0, 1);
+ clear_bit(BTRFS_ROOT_RESET_LOCKDEP_CLASS, &src->state);
path->lowest_level = 0;
if (ret) {
if (ret > 0)
@@ -3573,7 +3575,12 @@ int prepare_to_relocate(struct reloc_control *rc)
*/
return PTR_ERR(trans);
}
- return btrfs_commit_transaction(trans);
+
+ ret = btrfs_commit_transaction(trans);
+ if (ret)
+ unset_reloc_control(rc);
+
+ return ret;
}
static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 9e0e0ae2288c..43f905ab0a18 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -1233,7 +1233,8 @@ static void extent_err(const struct extent_buffer *eb, int slot,
}
static int check_extent_item(struct extent_buffer *leaf,
- struct btrfs_key *key, int slot)
+ struct btrfs_key *key, int slot,
+ struct btrfs_key *prev_key)
{
struct btrfs_fs_info *fs_info = leaf->fs_info;
struct btrfs_extent_item *ei;
@@ -1453,6 +1454,26 @@ static int check_extent_item(struct extent_buffer *leaf,
total_refs, inline_refs);
return -EUCLEAN;
}
+
+ if ((prev_key->type == BTRFS_EXTENT_ITEM_KEY) ||
+ (prev_key->type == BTRFS_METADATA_ITEM_KEY)) {
+ u64 prev_end = prev_key->objectid;
+
+ if (prev_key->type == BTRFS_METADATA_ITEM_KEY)
+ prev_end += fs_info->nodesize;
+ else
+ prev_end += prev_key->offset;
+
+ if (unlikely(prev_end > key->objectid)) {
+ extent_err(leaf, slot,
+ "previous extent [%llu %u %llu] overlaps current extent [%llu %u %llu]",
+ prev_key->objectid, prev_key->type,
+ prev_key->offset, key->objectid, key->type,
+ key->offset);
+ return -EUCLEAN;
+ }
+ }
+
return 0;
}
@@ -1621,7 +1642,7 @@ static int check_leaf_item(struct extent_buffer *leaf,
break;
case BTRFS_EXTENT_ITEM_KEY:
case BTRFS_METADATA_ITEM_KEY:
- ret = check_extent_item(leaf, key, slot);
+ ret = check_extent_item(leaf, key, slot, prev_key);
break;
case BTRFS_TREE_BLOCK_REF_KEY:
case BTRFS_SHARED_DATA_REF_KEY:
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index dcf75a8daa20..9205c4a5ca81 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -1146,7 +1146,9 @@ again:
extref = btrfs_lookup_inode_extref(NULL, root, path, name, namelen,
inode_objectid, parent_objectid, 0,
0);
- if (!IS_ERR_OR_NULL(extref)) {
+ if (IS_ERR(extref)) {
+ return PTR_ERR(extref);
+ } else if (extref) {
u32 item_size;
u32 cur_offset = 0;
unsigned long base;
@@ -1457,7 +1459,7 @@ static int add_link(struct btrfs_trans_handle *trans,
* on the inode will not free it. We will fixup the link count later.
*/
if (other_inode->i_nlink == 0)
- inc_nlink(other_inode);
+ set_nlink(other_inode, 1);
add_link:
ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode),
name, namelen, 0, ref_index);
@@ -1600,7 +1602,7 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
* free it. We will fixup the link count later.
*/
if (!ret && inode->i_nlink == 0)
- inc_nlink(inode);
+ set_nlink(inode, 1);
}
if (ret < 0)
goto out;