Currently, we check the wb_err too early for directories, before all of
the unsafe child requests have been waited on. In order to fix that we
need to check the mapping->wb_err later nearer to the end of ceph_fsync.
We also have an overly-complex method for tracking errors after
blocklisting. The errors recorded in cleanup_session_requests go to a
completely separate field in the inode, but we end up reporting them the
same way we would for any other error (in fsync).
There's no real benefit to tracking these errors in two different
places, since the only reporting mechanism for them is in fsync, and
we'd need to advance them both every time.
Given that, we can just remove i_meta_err, and convert the places that
used it to instead just use mapping->wb_err instead. That also fixes
the original problem by ensuring that we do a check_and_advance of the
wb_err at the end of the fsync op.
Cc: stable@vger.kernel.org
URL: https://tracker.ceph.com/issues/52864
Reported-by: Patrick Donnelly <pdonnell@redhat.com>
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Xiubo Li <xiubli@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
int ceph_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
int ceph_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
- struct ceph_file_info *fi = file->private_data;
struct inode *inode = file->f_mapping->host;
struct ceph_inode_info *ci = ceph_inode(inode);
u64 flush_tid;
struct inode *inode = file->f_mapping->host;
struct ceph_inode_info *ci = ceph_inode(inode);
u64 flush_tid;
- if (errseq_check(&ci->i_meta_err, READ_ONCE(fi->meta_err))) {
- spin_lock(&file->f_lock);
- err = errseq_check_and_advance(&ci->i_meta_err,
- &fi->meta_err);
- spin_unlock(&file->f_lock);
- if (err < 0)
- ret = err;
- }
+ err = file_check_and_advance_wb_err(file);
+ if (err < 0)
+ ret = err;
out:
dout("fsync %p%s result=%d\n", inode, datasync ? " datasync" : "", ret);
return ret;
out:
dout("fsync %p%s result=%d\n", inode, datasync ? " datasync" : "", ret);
return ret;
spin_lock_init(&fi->rw_contexts_lock);
INIT_LIST_HEAD(&fi->rw_contexts);
spin_lock_init(&fi->rw_contexts_lock);
INIT_LIST_HEAD(&fi->rw_contexts);
- fi->meta_err = errseq_sample(&ci->i_meta_err);
fi->filp_gen = READ_ONCE(ceph_inode_to_client(inode)->filp_gen);
return 0;
fi->filp_gen = READ_ONCE(ceph_inode_to_client(inode)->filp_gen);
return 0;
ceph_fscache_inode_init(ci);
ceph_fscache_inode_init(ci);
{
struct ceph_mds_request *req;
struct rb_node *p;
{
struct ceph_mds_request *req;
struct rb_node *p;
- struct ceph_inode_info *ci;
dout("cleanup_session_requests mds%d\n", session->s_mds);
mutex_lock(&mdsc->mutex);
dout("cleanup_session_requests mds%d\n", session->s_mds);
mutex_lock(&mdsc->mutex);
struct ceph_mds_request, r_unsafe_item);
pr_warn_ratelimited(" dropping unsafe request %llu\n",
req->r_tid);
struct ceph_mds_request, r_unsafe_item);
pr_warn_ratelimited(" dropping unsafe request %llu\n",
req->r_tid);
- if (req->r_target_inode) {
- /* dropping unsafe change of inode's attributes */
- ci = ceph_inode(req->r_target_inode);
- errseq_set(&ci->i_meta_err, -EIO);
- }
- if (req->r_unsafe_dir) {
- /* dropping unsafe directory operation */
- ci = ceph_inode(req->r_unsafe_dir);
- errseq_set(&ci->i_meta_err, -EIO);
- }
+ if (req->r_target_inode)
+ mapping_set_error(req->r_target_inode->i_mapping, -EIO);
+ if (req->r_unsafe_dir)
+ mapping_set_error(req->r_unsafe_dir->i_mapping, -EIO);
__unregister_request(mdsc, req);
}
/* zero r_attempts, so kick_requests() will re-send requests */
__unregister_request(mdsc, req);
}
/* zero r_attempts, so kick_requests() will re-send requests */
spin_unlock(&mdsc->cap_dirty_lock);
if (dirty_dropped) {
spin_unlock(&mdsc->cap_dirty_lock);
if (dirty_dropped) {
- errseq_set(&ci->i_meta_err, -EIO);
+ mapping_set_error(inode->i_mapping, -EIO);
if (ci->i_wrbuffer_ref_head == 0 &&
ci->i_wr_ref == 0 &&
if (ci->i_wrbuffer_ref_head == 0 &&
ci->i_wr_ref == 0 &&
#ifdef CONFIG_CEPH_FSCACHE
struct fscache_cookie *fscache;
#endif
#ifdef CONFIG_CEPH_FSCACHE
struct fscache_cookie *fscache;
#endif
struct inode vfs_inode; /* at end */
};
struct inode vfs_inode; /* at end */
};
spinlock_t rw_contexts_lock;
struct list_head rw_contexts;
spinlock_t rw_contexts_lock;
struct list_head rw_contexts;
u32 filp_gen;
atomic_t num_locks;
};
u32 filp_gen;
atomic_t num_locks;
};