summaryrefslogtreecommitdiff
path: root/fs/btrfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/btrfs')
-rw-r--r--fs/btrfs/block-group.c18
-rw-r--r--fs/btrfs/compression.c2
-rw-r--r--fs/btrfs/discard.c51
-rw-r--r--fs/btrfs/disk-io.c28
-rw-r--r--fs/btrfs/extent_io.c7
-rw-r--r--fs/btrfs/extent_io.h2
-rw-r--r--fs/btrfs/fs.h1
-rw-r--r--fs/btrfs/inode.c7
-rw-r--r--fs/btrfs/scrub.c4
-rw-r--r--fs/btrfs/send.c6
-rw-r--r--fs/btrfs/super.c4
-rw-r--r--fs/btrfs/tree-checker.c2
-rw-r--r--fs/btrfs/volumes.c91
13 files changed, 89 insertions, 134 deletions
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index b96b23594334..b14faa1d57b8 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -1888,6 +1888,17 @@ void btrfs_reclaim_bgs_work(struct work_struct *work)
up_write(&space_info->groups_sem);
goto next;
}
+
+ /*
+ * Cache the zone_unusable value before turning the block group
+ * to read only. As soon as the block group is read only it's
+ * zone_unusable value gets moved to the block group's read-only
+ * bytes and isn't available for calculations anymore. We also
+ * cache it before unlocking the block group, to prevent races
+ * (reports from KCSAN and such tools) with tasks updating it.
+ */
+ zone_unusable = bg->zone_unusable;
+
spin_unlock(&bg->lock);
spin_unlock(&space_info->lock);
@@ -1904,13 +1915,6 @@ void btrfs_reclaim_bgs_work(struct work_struct *work)
goto next;
}
- /*
- * Cache the zone_unusable value before turning the block group
- * to read only. As soon as the blog group is read only it's
- * zone_unusable value gets moved to the block group's read-only
- * bytes and isn't available for calculations anymore.
- */
- zone_unusable = bg->zone_unusable;
ret = inc_block_group_ro(bg, 0);
up_write(&space_info->groups_sem);
if (ret < 0)
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 0c4d486c3048..18d2210dc724 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -606,7 +606,7 @@ void btrfs_submit_compressed_read(struct btrfs_bio *bbio)
free_extent_map(em);
cb->nr_folios = DIV_ROUND_UP(compressed_len, PAGE_SIZE);
- cb->compressed_folios = kcalloc(cb->nr_folios, sizeof(struct page *), GFP_NOFS);
+ cb->compressed_folios = kcalloc(cb->nr_folios, sizeof(struct folio *), GFP_NOFS);
if (!cb->compressed_folios) {
ret = BLK_STS_RESOURCE;
goto out_free_bio;
diff --git a/fs/btrfs/discard.c b/fs/btrfs/discard.c
index e815d165cccc..de23c4b3515e 100644
--- a/fs/btrfs/discard.c
+++ b/fs/btrfs/discard.c
@@ -94,8 +94,6 @@ static void __add_to_discard_list(struct btrfs_discard_ctl *discard_ctl,
struct btrfs_block_group *block_group)
{
lockdep_assert_held(&discard_ctl->lock);
- if (!btrfs_run_discard_work(discard_ctl))
- return;
if (list_empty(&block_group->discard_list) ||
block_group->discard_index == BTRFS_DISCARD_INDEX_UNUSED) {
@@ -118,6 +116,9 @@ static void add_to_discard_list(struct btrfs_discard_ctl *discard_ctl,
if (!btrfs_is_block_group_data_only(block_group))
return;
+ if (!btrfs_run_discard_work(discard_ctl))
+ return;
+
spin_lock(&discard_ctl->lock);
__add_to_discard_list(discard_ctl, block_group);
spin_unlock(&discard_ctl->lock);
@@ -167,13 +168,7 @@ static bool remove_from_discard_list(struct btrfs_discard_ctl *discard_ctl,
block_group->discard_eligible_time = 0;
queued = !list_empty(&block_group->discard_list);
list_del_init(&block_group->discard_list);
- /*
- * If the block group is currently running in the discard workfn, we
- * don't want to deref it, since it's still being used by the workfn.
- * The workfn will notice this case and deref the block group when it is
- * finished.
- */
- if (queued && !running)
+ if (queued)
btrfs_put_block_group(block_group);
spin_unlock(&discard_ctl->lock);
@@ -250,6 +245,18 @@ again:
block_group->used != 0) {
if (btrfs_is_block_group_data_only(block_group)) {
__add_to_discard_list(discard_ctl, block_group);
+ /*
+ * The block group must have been moved to other
+ * discard list even if discard was disabled in
+ * the meantime or a transaction abort happened,
+ * otherwise we can end up in an infinite loop,
+ * always jumping into the 'again' label and
+ * keep getting this block group over and over
+ * in case there are no other block groups in
+ * the discard lists.
+ */
+ ASSERT(block_group->discard_index !=
+ BTRFS_DISCARD_INDEX_UNUSED);
} else {
list_del_init(&block_group->discard_list);
btrfs_put_block_group(block_group);
@@ -260,9 +267,10 @@ again:
block_group->discard_cursor = block_group->start;
block_group->discard_state = BTRFS_DISCARD_EXTENTS;
}
- discard_ctl->block_group = block_group;
}
if (block_group) {
+ btrfs_get_block_group(block_group);
+ discard_ctl->block_group = block_group;
*discard_state = block_group->discard_state;
*discard_index = block_group->discard_index;
}
@@ -493,9 +501,20 @@ static void btrfs_discard_workfn(struct work_struct *work)
block_group = peek_discard_list(discard_ctl, &discard_state,
&discard_index, now);
- if (!block_group || !btrfs_run_discard_work(discard_ctl))
+ if (!block_group)
+ return;
+ if (!btrfs_run_discard_work(discard_ctl)) {
+ spin_lock(&discard_ctl->lock);
+ btrfs_put_block_group(block_group);
+ discard_ctl->block_group = NULL;
+ spin_unlock(&discard_ctl->lock);
return;
+ }
if (now < block_group->discard_eligible_time) {
+ spin_lock(&discard_ctl->lock);
+ btrfs_put_block_group(block_group);
+ discard_ctl->block_group = NULL;
+ spin_unlock(&discard_ctl->lock);
btrfs_discard_schedule_work(discard_ctl, false);
return;
}
@@ -547,15 +566,7 @@ static void btrfs_discard_workfn(struct work_struct *work)
spin_lock(&discard_ctl->lock);
discard_ctl->prev_discard = trimmed;
discard_ctl->prev_discard_time = now;
- /*
- * If the block group was removed from the discard list while it was
- * running in this workfn, then we didn't deref it, since this function
- * still owned that reference. But we set the discard_ctl->block_group
- * back to NULL, so we can use that condition to know that now we need
- * to deref the block_group.
- */
- if (discard_ctl->block_group == NULL)
- btrfs_put_block_group(block_group);
+ btrfs_put_block_group(block_group);
discard_ctl->block_group = NULL;
__btrfs_discard_schedule_work(discard_ctl, now, false);
spin_unlock(&discard_ctl->lock);
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index ca821e5966bd..cc8d73693701 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -4329,6 +4329,14 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info)
btrfs_cleanup_defrag_inodes(fs_info);
/*
+ * Handle the error fs first, as it will flush and wait for all ordered
+ * extents. This will generate delayed iputs, thus we want to handle
+ * it first.
+ */
+ if (unlikely(BTRFS_FS_ERROR(fs_info)))
+ btrfs_error_commit_super(fs_info);
+
+ /*
* Wait for any fixup workers to complete.
* If we don't wait for them here and they are still running by the time
* we call kthread_stop() against the cleaner kthread further below, we
@@ -4349,6 +4357,19 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info)
btrfs_flush_workqueue(fs_info->delalloc_workers);
/*
+ * We can have ordered extents getting their last reference dropped from
+ * the fs_info->workers queue because for async writes for data bios we
+ * queue a work for that queue, at btrfs_wq_submit_bio(), that runs
+ * run_one_async_done() which calls btrfs_bio_end_io() in case the bio
+ * has an error, and that later function can do the final
+ * btrfs_put_ordered_extent() on the ordered extent attached to the bio,
+ * which adds a delayed iput for the inode. So we must flush the queue
+ * so that we don't have delayed iputs after committing the current
+ * transaction below and stopping the cleaner and transaction kthreads.
+ */
+ btrfs_flush_workqueue(fs_info->workers);
+
+ /*
* When finishing a compressed write bio we schedule a work queue item
* to finish an ordered extent - btrfs_finish_compressed_write_work()
* calls btrfs_finish_ordered_extent() which in turns does a call to
@@ -4418,9 +4439,6 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info)
btrfs_err(fs_info, "commit super ret %d", ret);
}
- if (BTRFS_FS_ERROR(fs_info))
- btrfs_error_commit_super(fs_info);
-
kthread_stop(fs_info->transaction_kthread);
kthread_stop(fs_info->cleaner_kthread);
@@ -4543,10 +4561,6 @@ static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info)
/* cleanup FS via transaction */
btrfs_cleanup_transaction(fs_info);
- mutex_lock(&fs_info->cleaner_mutex);
- btrfs_run_delayed_iputs(fs_info);
- mutex_unlock(&fs_info->cleaner_mutex);
-
down_write(&fs_info->cleanup_work_sem);
up_write(&fs_info->cleanup_work_sem);
}
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index c021aae8875e..06922529f19d 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2842,10 +2842,10 @@ struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
return eb;
}
-#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
u64 start)
{
+#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
struct extent_buffer *eb, *exists = NULL;
int ret;
@@ -2881,8 +2881,11 @@ again:
free_eb:
btrfs_release_extent_buffer(eb);
return exists;
-}
+#else
+ /* Stub to avoid linker error when compiled with optimizations turned off. */
+ return NULL;
#endif
+}
static struct extent_buffer *grab_extent_buffer(struct btrfs_fs_info *fs_info,
struct folio *folio)
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 6c5328bfabc2..2aefc64cdd29 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -297,6 +297,8 @@ static inline int num_extent_pages(const struct extent_buffer *eb)
*/
static inline int num_extent_folios(const struct extent_buffer *eb)
{
+ if (!eb->folios[0])
+ return 0;
if (folio_order(eb->folios[0]))
return 1;
return num_extent_pages(eb);
diff --git a/fs/btrfs/fs.h b/fs/btrfs/fs.h
index b572d6b9730b..f46fba127caa 100644
--- a/fs/btrfs/fs.h
+++ b/fs/btrfs/fs.h
@@ -285,6 +285,7 @@ enum {
#define BTRFS_FEATURE_INCOMPAT_SAFE_CLEAR 0ULL
#define BTRFS_DEFAULT_COMMIT_INTERVAL (30)
+#define BTRFS_WARNING_COMMIT_INTERVAL (300)
#define BTRFS_DEFAULT_MAX_INLINE (2048)
struct btrfs_dev_replace {
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 3be6f8e8e157..a06fca7934d5 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1116,6 +1116,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
struct extent_state *cached = NULL;
struct extent_map *em;
int ret = 0;
+ bool free_pages = false;
u64 start = async_extent->start;
u64 end = async_extent->start + async_extent->ram_size - 1;
@@ -1136,7 +1137,10 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
}
if (async_extent->compress_type == BTRFS_COMPRESS_NONE) {
+ ASSERT(!async_extent->folios);
+ ASSERT(async_extent->nr_folios == 0);
submit_uncompressed_range(inode, async_extent, locked_folio);
+ free_pages = true;
goto done;
}
@@ -1152,6 +1156,7 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
* fall back to uncompressed.
*/
submit_uncompressed_range(inode, async_extent, locked_folio);
+ free_pages = true;
goto done;
}
@@ -1193,6 +1198,8 @@ static void submit_one_async_extent(struct async_chunk *async_chunk,
done:
if (async_chunk->blkcg_css)
kthread_associate_blkcg(NULL);
+ if (free_pages)
+ free_async_extent_pages(async_extent);
kfree(async_extent);
return;
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 531312efee8d..5d0060eb8ff4 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -1541,8 +1541,8 @@ static int scrub_find_fill_first_stripe(struct btrfs_block_group *bg,
u64 extent_gen;
int ret;
- if (unlikely(!extent_root)) {
- btrfs_err(fs_info, "no valid extent root for scrub");
+ if (unlikely(!extent_root || !csum_root)) {
+ btrfs_err(fs_info, "no valid extent or csum root for scrub");
return -EUCLEAN;
}
memset(stripe->sectors, 0, sizeof(struct scrub_sector_verification) *
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index f437138fefbc..bb8a0945b0fd 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -487,10 +487,8 @@ static int fs_path_ensure_buf(struct fs_path *p, int len)
if (p->buf_len >= len)
return 0;
- if (len > PATH_MAX) {
- WARN_ON(1);
- return -ENOMEM;
- }
+ if (WARN_ON(len > PATH_MAX))
+ return -ENAMETOOLONG;
path_len = p->end - p->start;
old_buf_len = p->buf_len;
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index a5f29ff3fbc2..8098f5cc9b7b 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -569,6 +569,10 @@ static int btrfs_parse_param(struct fs_context *fc, struct fs_parameter *param)
break;
case Opt_commit_interval:
ctx->commit_interval = result.uint_32;
+ if (ctx->commit_interval > BTRFS_WARNING_COMMIT_INTERVAL) {
+ btrfs_warn(NULL, "excessive commit interval %u, use with care",
+ ctx->commit_interval);
+ }
if (ctx->commit_interval == 0)
ctx->commit_interval = BTRFS_DEFAULT_COMMIT_INTERVAL;
break;
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 43979891f7c8..2b66a6130269 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -2235,7 +2235,7 @@ int btrfs_verify_level_key(struct extent_buffer *eb,
btrfs_err(fs_info,
"tree level mismatch detected, bytenr=%llu level expected=%u has=%u",
eb->start, check->level, found_level);
- return -EIO;
+ return -EUCLEAN;
}
if (!check->has_first_key)
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 3f8afbd1ebb5..b80110fe30fe 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -733,82 +733,6 @@ const u8 *btrfs_sb_fsid_ptr(const struct btrfs_super_block *sb)
return has_metadata_uuid ? sb->metadata_uuid : sb->fsid;
}
-/*
- * We can have very weird soft links passed in.
- * One example is "/proc/self/fd/<fd>", which can be a soft link to
- * a block device.
- *
- * But it's never a good idea to use those weird names.
- * Here we check if the path (not following symlinks) is a good one inside
- * "/dev/".
- */
-static bool is_good_dev_path(const char *dev_path)
-{
- struct path path = { .mnt = NULL, .dentry = NULL };
- char *path_buf = NULL;
- char *resolved_path;
- bool is_good = false;
- int ret;
-
- if (!dev_path)
- goto out;
-
- path_buf = kmalloc(PATH_MAX, GFP_KERNEL);
- if (!path_buf)
- goto out;
-
- /*
- * Do not follow soft link, just check if the original path is inside
- * "/dev/".
- */
- ret = kern_path(dev_path, 0, &path);
- if (ret)
- goto out;
- resolved_path = d_path(&path, path_buf, PATH_MAX);
- if (IS_ERR(resolved_path))
- goto out;
- if (strncmp(resolved_path, "/dev/", strlen("/dev/")))
- goto out;
- is_good = true;
-out:
- kfree(path_buf);
- path_put(&path);
- return is_good;
-}
-
-static int get_canonical_dev_path(const char *dev_path, char *canonical)
-{
- struct path path = { .mnt = NULL, .dentry = NULL };
- char *path_buf = NULL;
- char *resolved_path;
- int ret;
-
- if (!dev_path) {
- ret = -EINVAL;
- goto out;
- }
-
- path_buf = kmalloc(PATH_MAX, GFP_KERNEL);
- if (!path_buf) {
- ret = -ENOMEM;
- goto out;
- }
-
- ret = kern_path(dev_path, LOOKUP_FOLLOW, &path);
- if (ret)
- goto out;
- resolved_path = d_path(&path, path_buf, PATH_MAX);
- if (IS_ERR(resolved_path)) {
- ret = PTR_ERR(resolved_path);
- goto out;
- }
- ret = strscpy(canonical, resolved_path, PATH_MAX);
-out:
- kfree(path_buf);
- path_put(&path);
- return ret;
-}
-
static bool is_same_device(struct btrfs_device *device, const char *new_path)
{
struct path old = { .mnt = NULL, .dentry = NULL };
@@ -1513,23 +1437,12 @@ struct btrfs_device *btrfs_scan_one_device(const char *path, blk_mode_t flags,
bool new_device_added = false;
struct btrfs_device *device = NULL;
struct file *bdev_file;
- char *canonical_path = NULL;
u64 bytenr;
dev_t devt;
int ret;
lockdep_assert_held(&uuid_mutex);
- if (!is_good_dev_path(path)) {
- canonical_path = kmalloc(PATH_MAX, GFP_KERNEL);
- if (canonical_path) {
- ret = get_canonical_dev_path(path, canonical_path);
- if (ret < 0) {
- kfree(canonical_path);
- canonical_path = NULL;
- }
- }
- }
/*
* Avoid an exclusive open here, as the systemd-udev may initiate the
* device scan which may race with the user's mount or mkfs command,
@@ -1574,8 +1487,7 @@ struct btrfs_device *btrfs_scan_one_device(const char *path, blk_mode_t flags,
goto free_disk_super;
}
- device = device_list_add(canonical_path ? : path, disk_super,
- &new_device_added);
+ device = device_list_add(path, disk_super, &new_device_added);
if (!IS_ERR(device) && new_device_added)
btrfs_free_stale_devices(device->devt, device);
@@ -1584,7 +1496,6 @@ free_disk_super:
error_bdev_put:
fput(bdev_file);
- kfree(canonical_path);
return device;
}
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-02 17:59:02 +0300