libceph: fix preallocation check in get_reply()

The check that makes sure that we have enough memory allocated to read
in the entire header of the message in question is currently busted.
It compares front_len of the incoming message with iov_len field of
ceph_msg::front structure, which is used primarily to indicate the
amount of data already read in, and not the size of the allocated
buffer.  Under certain conditions (e.g. a short read from a socket
followed by that socket's shutdown and owning ceph_connection reset)
this results in a warning similar to

[85688.975866] libceph: get_reply front 198 > preallocated 122 (4#0)

and, through another bug, leads to forever hung tasks and forced
reboots.  Fix this by comparing front_len with front_alloc_len field of
struct ceph_msg, which stores the actual size of the buffer.

Fixes: http://tracker.ceph.com/issues/5425

Signed-off-by: Ilya Dryomov <ilya.dryomov@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
This commit is contained in:
Ilya Dryomov 2014-01-09 20:08:21 +02:00
parent 3f0a4ac55f
commit f2be82b005
2 changed files with 3 additions and 4 deletions

View File

@ -3130,7 +3130,6 @@ struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags,
INIT_LIST_HEAD(&m->data);
/* front */
m->front_alloc_len = front_len;
if (front_len) {
if (front_len > PAGE_CACHE_SIZE) {
m->front.iov_base = __vmalloc(front_len, flags,
@ -3147,7 +3146,7 @@ struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags,
} else {
m->front.iov_base = NULL;
}
m->front.iov_len = front_len;
m->front_alloc_len = m->front.iov_len = front_len;
dout("ceph_msg_new %p front %d\n", m, front_len);
return m;

View File

@ -2522,9 +2522,9 @@ static struct ceph_msg *get_reply(struct ceph_connection *con,
req->r_reply, req->r_reply->con);
ceph_msg_revoke_incoming(req->r_reply);
if (front_len > req->r_reply->front.iov_len) {
if (front_len > req->r_reply->front_alloc_len) {
pr_warning("get_reply front %d > preallocated %d (%u#%llu)\n",
front_len, (int)req->r_reply->front.iov_len,
front_len, req->r_reply->front_alloc_len,
(unsigned int)con->peer_name.type,
le64_to_cpu(con->peer_name.num));
m = ceph_msg_new(CEPH_MSG_OSD_OPREPLY, front_len, GFP_NOFS,