[ofa-general] Re: [PATCH for-2.6.22] IB/ipoib: fix performance regression on Mellanox

[Michael S. Tsirkin]

> Quoting Roland Dreier <rdreier at cisco.com>:
> Subject: Re: [PATCH for-2.6.22] IB/ipoib: fix performance regression on Mellanox
> 
> seems like this leaves rx_drain_qp in the data structure and also in
> the comment in ipoib.h...

Right, add this on top of it.

> not sure if there are any other remnants of
> the previous approach that should be cleaned up.

Hopefully not - compiler'd notice any uses of rx_drain_qp, and
that really is the only change.

----->

Remove unused rx_drain_qp.

Signed-off-by: Michael S. Tsirkin <mst at dev.mellanox.co.il>

---

diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index 158759e..285c143 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -156,7 +156,7 @@ struct ipoib_cm_data {
  * - and then invoke a Destroy QP or Reset QP.
  *
  * We use the second option and wait for a completion on the
- * rx_drain_qp before destroying QPs attached to our SRQ.
+ * same CQ before destroying QPs attached to our SRQ.
  */
 
 enum ipoib_cm_state {
@@ -199,7 +199,6 @@ struct ipoib_cm_dev_priv {
 	struct ib_srq  	       *srq;
 	struct ipoib_cm_rx_buf *srq_ring;
 	struct ib_cm_id        *id;
-	struct ib_qp           *rx_drain_qp;   /* generates WR described in 10.3.1 */
 	struct list_head        passive_ids;   /* state: LIVE */
 	struct list_head        rx_error_list; /* state: ERROR */
 	struct list_head        rx_flush_list; /* state: FLUSH, drain not started */

-- 
MST