[PATCH v3] ceph: request Fw caps before updating the mtime in ceph_write_iter

[PATCH v3] ceph: request Fw caps before updating the mtime in ceph_write_iter

[Jeff Layton]

The current code will update the mtime and then try to get caps to
handle the write. If we end up having to request caps from the MDS, then
the mtime in the cap grant will clobber the updated mtime and it'll be
lost.

This is most noticable when two clients are alternately writing to the
same file. Fw caps are continually being granted and revoked, and the
mtime ends up stuck because the updated mtimes are always being
overwritten with the old one.

Fix this by changing the order of operations in ceph_write_iter to get
the caps before updating the times. Also, make sure we check the pool
full conditions before even getting any caps or uninlining.

URL: https://tracker.ceph.com/issues/46574
Cc: "Yan, Zheng" <ukernel@gmail.com>
Cc: Luis Henriques <lhenriques@suse.de>
Cc: Xiubo Li <xiubli@redhat.com>
Reported-by: Jozef Kováč <kovac@firma.zoznam.sk>
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
 fs/ceph/file.c | 32 +++++++++++++++++---------------
 1 file changed, 17 insertions(+), 15 deletions(-)

v3: remove privs before getting caps. That can generate a SETATTR
    request, which could deadlock with us holding Fw caps

diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index d1755ac1d964..3daebfaec8c6 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1722,32 +1722,26 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		goto out;
 	}
 
-	err = file_remove_privs(file);
-	if (err)
+	down_read(&osdc->lock);
+	map_flags = osdc->osdmap->flags;
+	pool_flags = ceph_pg_pool_flags(osdc->osdmap, ci->i_layout.pool_id);
+	up_read(&osdc->lock);
+	if ((map_flags & CEPH_OSDMAP_FULL) ||
+	    (pool_flags & CEPH_POOL_FLAG_FULL)) {
+		err = -ENOSPC;
 		goto out;
+	}
 
-	err = file_update_time(file);
+	err = file_remove_privs(file);
 	if (err)
 		goto out;
 
-	inode_inc_iversion_raw(inode);
-
 	if (ci->i_inline_version != CEPH_INLINE_NONE) {
 		err = ceph_uninline_data(file, NULL);
 		if (err < 0)
 			goto out;
 	}
 
-	down_read(&osdc->lock);
-	map_flags = osdc->osdmap->flags;
-	pool_flags = ceph_pg_pool_flags(osdc->osdmap, ci->i_layout.pool_id);
-	up_read(&osdc->lock);
-	if ((map_flags & CEPH_OSDMAP_FULL) ||
-	    (pool_flags & CEPH_POOL_FLAG_FULL)) {
-		err = -ENOSPC;
-		goto out;
-	}
-
 	dout("aio_write %p %llx.%llx %llu~%zd getting caps. i_size %llu\n",
 	     inode, ceph_vinop(inode), pos, count, i_size_read(inode));
 	if (fi->fmode & CEPH_FILE_MODE_LAZY)
@@ -1759,6 +1753,12 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	if (err < 0)
 		goto out;
 
+	err = file_update_time(file);
+	if (err)
+		goto out_caps;
+
+	inode_inc_iversion_raw(inode);
+
 	dout("aio_write %p %llx.%llx %llu~%zd got cap refs on %s\n",
 	     inode, ceph_vinop(inode), pos, count, ceph_cap_string(got));
 
@@ -1842,6 +1842,8 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	}
 
 	goto out_unlocked;
+out_caps:
+	ceph_put_cap_refs(ci, got);
 out:
 	if (direct_lock)
 		ceph_end_io_direct(inode);
-- 
2.31.1