@@ -795,7 +795,8 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
int i, j, rc = 0;
int timeout, optype;
struct mid_q_entry *midQ[MAX_COMPOUND];
- unsigned int credits = 0;
+ bool cancelled_mid[MAX_COMPOUND] = {false};
+ unsigned int credits[MAX_COMPOUND] = {0};
char *buf;
timeout = flags & CIFS_TIMEOUT_MASK;
@@ -813,13 +814,31 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
return -ENOENT;
/*
- * Ensure that we do not send more than 50 overlapping requests
- * to the same server. We may make this configurable later or
- * use ses->maxReq.
+ * Ensure we obtain 1 credit per request in the compound chain.
+ * It can be optimized further by waiting for all the credits
+ * at once but this can wait long enough if we don't have enough
+ * credits due to some heavy operations in progress or the server
+ * not granting us much, so a fallback to the current approach is
+ * needed anyway.
*/
- rc = wait_for_free_request(ses->server, timeout, optype);
- if (rc)
- return rc;
+ for (i = 0; i < num_rqst; i++) {
+ rc = wait_for_free_request(ses->server, timeout, optype);
+ if (rc) {
+ /*
+ * We haven't sent an SMB packet to the server yet but
+ * we already obtained credits for i requests in the
+ * compound chain - need to return those credits back
+ * for future use. Note that we need to call add_credits
+ * multiple times to match the way we obtained credits
+ * in the first place and to account for in flight
+ * requests correctly.
+ */
+ for (j = 0; j < i; j++)
+ add_credits(ses->server, 1, optype);
+ return rc;
+ }
+ credits[i] = 1;
+ }
/*
* Make sure that we sign in the same order that we send on this socket
@@ -835,8 +854,10 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
for (j = 0; j < i; j++)
cifs_delete_mid(midQ[j]);
mutex_unlock(&ses->server->srv_mutex);
+
/* Update # of requests on wire to server */
- add_credits(ses->server, 1, optype);
+ for (j = 0; j < num_rqst; j++)
+ add_credits(ses->server, credits[j], optype);
return PTR_ERR(midQ[i]);
}
@@ -883,17 +904,16 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
if (midQ[i]->mid_state == MID_REQUEST_SUBMITTED) {
midQ[i]->mid_flags |= MID_WAIT_CANCELLED;
midQ[i]->callback = DeleteMidQEntry;
- spin_unlock(&GlobalMid_Lock);
- add_credits(ses->server, 1, optype);
- return rc;
+ cancelled_mid[i] = true;
}
spin_unlock(&GlobalMid_Lock);
}
}
for (i = 0; i < num_rqst; i++)
- if (midQ[i]->resp_buf)
- credits += ses->server->ops->get_credits(midQ[i]);
+ if (!cancelled_mid[i] && midQ[i]->resp_buf
+ && (midQ[i]->mid_state == MID_RESPONSE_RECEIVED))
+ credits[i] = ses->server->ops->get_credits(midQ[i]);
for (i = 0; i < num_rqst; i++) {
if (rc < 0)
@@ -901,8 +921,9 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
rc = cifs_sync_mid_result(midQ[i], ses->server);
if (rc != 0) {
- add_credits(ses->server, credits, optype);
- return rc;
+ /* mark this mid as cancelled to not free it below */
+ cancelled_mid[i] = true;
+ goto out;
}
if (!midQ[i]->resp_buf ||
@@ -949,9 +970,11 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses,
* This is prevented above by using a noop callback that will not
* wake this thread except for the very last PDU.
*/
- for (i = 0; i < num_rqst; i++)
- cifs_delete_mid(midQ[i]);
- add_credits(ses->server, credits, optype);
+ for (i = 0; i < num_rqst; i++) {
+ if (!cancelled_mid[i])
+ cifs_delete_mid(midQ[i]);
+ add_credits(ses->server, credits[i], optype);
+ }
return rc;
}
In SMB3 protocol every part of the compound chain consumes credits individually, so we need to call wait_for_free_credits() for each of the PDUs in the chain. If an operation is interrupted, we must ensure we return all credits taken from the server structure back. Cc: <stable@vger.kernel.org> Signed-off-by: Pavel Shilovsky <pshilov@microsoft.com> --- fs/cifs/transport.c | 59 +++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 41 insertions(+), 18 deletions(-)