Fix a regression was introduced in commit
6df5a128f0fd ("IB/iser:
Suppress scsi command send completions").
The sig_count was wrongly set to be static variable, thus it is
possible that we won't reach to (sig_count % ISER_SIGNAL_BATCH) == 0
condition (due to races) and the send queue will be overflowed.
Instead keep sig_count per connection. We don't need it to be atomic
as we are safe under the iscsi session frwd_lock taken by libiscsi on
the queuecommand path.
Fixes:
6df5a128f0fd ("IB/iser: Suppress scsi command send completions")
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
* @cma_id: rdma_cm connection maneger handle
* @qp: Connection Queue-pair
* @post_recv_buf_count: post receive counter
+ * @sig_count: send work request signal count
* @rx_wr: receive work request for batch posts
* @device: reference to iser device
* @comp: iser completion context
struct rdma_cm_id *cma_id;
struct ib_qp *qp;
int post_recv_buf_count;
+ u8 sig_count;
struct ib_recv_wr rx_wr[ISER_MIN_POSTED_RX];
struct iser_device *device;
struct iser_comp *comp;
return 0;
}
-static inline bool iser_signal_comp(int sig_count)
+static inline bool iser_signal_comp(u8 sig_count)
{
return ((sig_count % ISER_SIGNAL_CMD_COUNT) == 0);
}
struct iscsi_scsi_req *hdr = (struct iscsi_scsi_req *)task->hdr;
struct scsi_cmnd *sc = task->sc;
struct iser_tx_desc *tx_desc = &iser_task->desc;
- static unsigned sig_count;
+ u8 sig_count = ++iser_conn->ib_conn.sig_count;
edtl = ntohl(hdr->data_length);
iser_task->status = ISER_TASK_STATUS_STARTED;
err = iser_post_send(&iser_conn->ib_conn, tx_desc,
- iser_signal_comp(++sig_count));
+ iser_signal_comp(sig_count));
if (!err)
return 0;