[PATCH 00/11] Critical bug fixes for RDMA/cxgb4

[PATCH 00/11] Critical bug fixes for RDMA/cxgb4

[Vipul Pandya]

Hi Roland,

This patch series fixes critical bugs for RDMA/cxgb4. It fixes bugs in following
areas:
- Aborts connection in error scenarios
- Logs only critical errors
- Holds the reference of the QP untill TID is released
- Avoids race condition in endpoint timeout  
- Fixes reconnect and version mismatch related bugs in MPAv2 related logic
- Avoids removing hwtid which was not inserted
- Addresses several sparse warnings

The patch-series is based on Roland's infiniband tree for-next branch.

Thanks,
Vipul Pandya

Vipul Pandya (11):
  RDMA/cxgb4: abort connections that receive unexpected streaming mode
    data
  RDMA/cxgb4: abort connections when moving to ERROR state.
  RDMA/cxgb4: Display streaming mode error only if detected in RTS.
  RDMA/cxgb4: keep qp referenced until TID released.
  RDMA/cxgb4: Always log async errors.
  RDMA/cxgb4: only log rx_data warnings if cpl status is non zero.
  RDMA/cxgb4: endpoint timeout race condition
  RDMA/cxgb4: don't reconnect on abort for mpa_rev 1
  RDMA/cxgb4: Don't wakeup threads for MPAv2
  RDMA/cxgb4: Insert hwtid in pass_accept_req instead in pass_establish
  RDMA/cxgb4: Address sparse warnings

 drivers/infiniband/hw/cxgb4/cm.c       |  170 +++++++++++++++++++-------------
 drivers/infiniband/hw/cxgb4/device.c   |    3 +-
 drivers/infiniband/hw/cxgb4/ev.c       |    8 +-
 drivers/infiniband/hw/cxgb4/iw_cxgb4.h |    2 +
 drivers/infiniband/hw/cxgb4/qp.c       |    1 +
 5 files changed, 113 insertions(+), 71 deletions(-)

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 01/11] RDMA/cxgb4: abort connections that receive unexpected streaming mode data

[Vipul Pandya]

This error means the rdma connection was knocked out of rdma mode probably due
to an error on the connection.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |   19 ++++++++++---------
 1 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index c13745c..9cab6a6 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1391,30 +1391,31 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb)
 	skb_pull(skb, sizeof(*hdr));
 	skb_trim(skb, dlen);
 
-	ep->rcv_seq += dlen;
-	BUG_ON(ep->rcv_seq != (ntohl(hdr->seq) + dlen));
-
 	/* update RX credits */
 	update_rx_credits(ep, dlen);
 
 	switch (state_read(&ep->com)) {
 	case MPA_REQ_SENT:
+		ep->rcv_seq += dlen;
 		process_mpa_reply(ep, skb);
 		break;
 	case MPA_REQ_WAIT:
+		ep->rcv_seq += dlen;
 		process_mpa_request(ep, skb);
 		break;
-	case MPA_REP_SENT:
-		break;
 	default:
 		pr_err("%s Unexpected streaming data." \
 		       " ep %p state %d tid %u status %d\n",
 		       __func__, ep, state_read(&ep->com), ep->hwtid, status);
 
-		/*
-		 * The ep will timeout and inform the ULP of the failure.
-		 * See ep_timeout().
-		 */
+		if (ep->com.qp) {
+			struct c4iw_qp_attributes attrs;
+
+			attrs.next_state = C4IW_QP_STATE_ERROR;
+			c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
+				       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
+		}
+		c4iw_ep_disconnect(ep, 1, GFP_KERNEL);
 		break;
 	}
 	return 0;
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 02/11] RDMA/cxgb4: abort connections when moving to ERROR state.

[Vipul Pandya]

If a FINI operation fails, then we need to ABORT instead
of CLOSE.  Also, if we ABORT due to unexpected STREAMING
data, then wake up anybody blocked in FINI...

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |    1 +
 drivers/infiniband/hw/cxgb4/qp.c |    1 +
 2 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 9cab6a6..3dce47a 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1438,6 +1438,7 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
 	mutex_lock(&ep->com.mutex);
 	switch (ep->com.state) {
 	case ABORTING:
+		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
 		__state_set(&ep->com, DEAD);
 		release = 1;
 		break;
diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index 05bfe53..17ba4f8 100644
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -1383,6 +1383,7 @@ err:
 	qhp->ep = NULL;
 	set_state(qhp, C4IW_QP_STATE_ERROR);
 	free = 1;
+	abort = 1;
 	wake_up(&qhp->wait);
 	BUG_ON(!ep);
 	flush_qp(qhp);
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 03/11] RDMA/cxgb4: Display streaming mode error only if detected in RTS.

[Vipul Pandya]

With later firmware, the chances of getting streaming mode data after
we exit RTS is likely, so we don't need to warn for it.  The only real
case where we don't expect it is when the QP is in RTS.

move QP to ERROR when streaming mode data received.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |   26 ++++++++++++++------------
 1 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 3dce47a..31d1fac 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1403,21 +1403,23 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb)
 		ep->rcv_seq += dlen;
 		process_mpa_request(ep, skb);
 		break;
-	default:
-		pr_err("%s Unexpected streaming data." \
-		       " ep %p state %d tid %u status %d\n",
-		       __func__, ep, state_read(&ep->com), ep->hwtid, status);
-
-		if (ep->com.qp) {
-			struct c4iw_qp_attributes attrs;
-
-			attrs.next_state = C4IW_QP_STATE_ERROR;
-			c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
-				       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
-		}
+	case FPDU_MODE: {
+		struct c4iw_qp_attributes attrs;
+		BUG_ON(!ep->com.qp);
+		if (ep->com.qp->attr.state == C4IW_QP_STATE_RTS)
+			pr_err("%s Unexpected streaming data." \
+			       " ep %p state %d tid %u status %d\n",
+			       __func__, ep, state_read(&ep->com),
+			       ep->hwtid, status);
+		attrs.next_state = C4IW_QP_STATE_ERROR;
+		c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
+			       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
 		c4iw_ep_disconnect(ep, 1, GFP_KERNEL);
 		break;
 	}
+	default:
+		break;
+	}
 	return 0;
 }
 
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 04/11] RDMA/cxgb4: keep qp referenced until TID released.

[Vipul Pandya]

The driver is currently releasing the last ref on the QP too early.
This can cause bus errors due to HW still fetching WRs from the hw queue.
The fix is to keep a qp ref until we release the HW TID.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c       |   20 ++++++++++++++++----
 drivers/infiniband/hw/cxgb4/iw_cxgb4.h |    1 +
 2 files changed, 17 insertions(+), 4 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 31d1fac..ebcdb3f 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -143,6 +143,18 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status);
 static LIST_HEAD(timeout_list);
 static spinlock_t timeout_lock;
 
+static void deref_qp(struct c4iw_ep *ep)
+{
+	c4iw_qp_rem_ref(&ep->com.qp->ibqp);
+	clear_bit(QP_REFERENCED, &ep->com.flags);
+}
+
+static void ref_qp(struct c4iw_ep *ep)
+{
+	set_bit(QP_REFERENCED, &ep->com.flags);
+	c4iw_qp_add_ref(&ep->com.qp->ibqp);
+}
+
 static void start_ep_timer(struct c4iw_ep *ep)
 {
 	PDBG("%s ep %p\n", __func__, ep);
@@ -271,6 +283,8 @@ void _c4iw_free_ep(struct kref *kref)
 
 	ep = container_of(kref, struct c4iw_ep, com.kref);
 	PDBG("%s ep %p state %s\n", __func__, ep, states[state_read(&ep->com)]);
+	if (test_bit(QP_REFERENCED, &ep->com.flags))
+		deref_qp(ep);
 	if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) {
 		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid);
 		dst_release(ep->dst);
@@ -863,7 +877,6 @@ static void close_complete_upcall(struct c4iw_ep *ep)
 		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
 		ep->com.cm_id->rem_ref(ep->com.cm_id);
 		ep->com.cm_id = NULL;
-		ep->com.qp = NULL;
 		set_bit(CLOSE_UPCALL, &ep->com.history);
 	}
 }
@@ -906,7 +919,6 @@ static void peer_abort_upcall(struct c4iw_ep *ep)
 		ep->com.cm_id->event_handler(ep->com.cm_id, &event);
 		ep->com.cm_id->rem_ref(ep->com.cm_id);
 		ep->com.cm_id = NULL;
-		ep->com.qp = NULL;
 		set_bit(ABORT_UPCALL, &ep->com.history);
 	}
 }
@@ -946,7 +958,6 @@ static void connect_reply_upcall(struct c4iw_ep *ep, int status)
 	if (status < 0) {
 		ep->com.cm_id->rem_ref(ep->com.cm_id);
 		ep->com.cm_id = NULL;
-		ep->com.qp = NULL;
 	}
 }
 
@@ -2434,6 +2445,7 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	cm_id->add_ref(cm_id);
 	ep->com.cm_id = cm_id;
 	ep->com.qp = qp;
+	ref_qp(ep);
 
 	/* bind QP to EP and move to RTS */
 	attrs.mpa_attr = ep->mpa_attr;
@@ -2464,7 +2476,6 @@ int c4iw_accept_cr(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	return 0;
 err1:
 	ep->com.cm_id = NULL;
-	ep->com.qp = NULL;
 	cm_id->rem_ref(cm_id);
 err:
 	c4iw_put_ep(&ep->com);
@@ -2505,6 +2516,7 @@ int c4iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	ep->com.cm_id = cm_id;
 	ep->com.qp = get_qhp(dev, conn_param->qpn);
 	BUG_ON(!ep->com.qp);
+	ref_qp(ep);
 	PDBG("%s qpn 0x%x qp %p cm_id %p\n", __func__, conn_param->qpn,
 	     ep->com.qp, cm_id);
 
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index 9c1644f..0aaaa0e 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -716,6 +716,7 @@ enum c4iw_ep_flags {
 	ABORT_REQ_IN_PROGRESS	= 1,
 	RELEASE_RESOURCES	= 2,
 	CLOSE_SENT		= 3,
+	QP_REFERENCED           = 5,
 };
 
 enum c4iw_ep_history {
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 05/11] RDMA/cxgb4: Always log async errors.

[Vipul Pandya]

Log AEs even if the QP isn't in RTS.  It is useful
information.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |    6 +++---
 drivers/infiniband/hw/cxgb4/ev.c |    8 +++++---
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index ebcdb3f..5989991 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1419,9 +1419,9 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb)
 		BUG_ON(!ep->com.qp);
 		if (ep->com.qp->attr.state == C4IW_QP_STATE_RTS)
 			pr_err("%s Unexpected streaming data." \
-			       " ep %p state %d tid %u status %d\n",
-			       __func__, ep, state_read(&ep->com),
-			       ep->hwtid, status);
+			       " qpid %u ep %p state %d tid %u status %d\n",
+			       __func__, ep->com.qp->wq.sq.qid, ep,
+			       state_read(&ep->com), ep->hwtid, status);
 		attrs.next_state = C4IW_QP_STATE_ERROR;
 		c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp,
 			       C4IW_QP_ATTR_NEXT_STATE, &attrs, 1);
diff --git a/drivers/infiniband/hw/cxgb4/ev.c b/drivers/infiniband/hw/cxgb4/ev.c
index cf2f6b4..1a840b2 100644
--- a/drivers/infiniband/hw/cxgb4/ev.c
+++ b/drivers/infiniband/hw/cxgb4/ev.c
@@ -46,9 +46,11 @@ static void post_qp_event(struct c4iw_dev *dev, struct c4iw_cq *chp,
 
 	if ((qhp->attr.state == C4IW_QP_STATE_ERROR) ||
 	    (qhp->attr.state == C4IW_QP_STATE_TERMINATE)) {
-		PDBG("%s AE received after RTS - "
-		     "qp state %d qpid 0x%x status 0x%x\n", __func__,
-		     qhp->attr.state, qhp->wq.sq.qid, CQE_STATUS(err_cqe));
+		pr_err("%s AE after RTS - qpid 0x%x opcode %d status 0x%x "\
+		       "type %d wrid.hi 0x%x wrid.lo 0x%x\n",
+		       __func__, CQE_QPID(err_cqe), CQE_OPCODE(err_cqe),
+		       CQE_STATUS(err_cqe), CQE_TYPE(err_cqe),
+		       CQE_WRID_HI(err_cqe), CQE_WRID_LOW(err_cqe));
 		return;
 	}
 
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 06/11] RDMA/cxgb4: only log rx_data warnings if cpl status is non zero.

[Vipul Pandya]

With newer firmware, we can get streaming data due to connection
errors before the driver moves the QP out of RTS.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 5989991..51ceb61 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1417,7 +1417,7 @@ static int rx_data(struct c4iw_dev *dev, struct sk_buff *skb)
 	case FPDU_MODE: {
 		struct c4iw_qp_attributes attrs;
 		BUG_ON(!ep->com.qp);
-		if (ep->com.qp->attr.state == C4IW_QP_STATE_RTS)
+		if (status)
 			pr_err("%s Unexpected streaming data." \
 			       " qpid %u ep %p state %d tid %u status %d\n",
 			       __func__, ep->com.qp->wq.sq.qid, ep,
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 07/11] RDMA/cxgb4: endpoint timeout race condition

[Vipul Pandya]

The endpoint timeout logic had a race that could cause an endpoint
object to be freed while it was still on the timedout list.  This
can happen if the timer is stopped after it had fired, but before
the timedout thread processed the endpoint timeout.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c       |   29 ++++++++++++++++-------------
 drivers/infiniband/hw/cxgb4/iw_cxgb4.h |    1 +
 2 files changed, 17 insertions(+), 13 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 51ceb61..ab5b4dd 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -159,10 +159,12 @@ static void start_ep_timer(struct c4iw_ep *ep)
 {
 	PDBG("%s ep %p\n", __func__, ep);
 	if (timer_pending(&ep->timer)) {
-		PDBG("%s stopped / restarted timer ep %p\n", __func__, ep);
-		del_timer_sync(&ep->timer);
-	} else
-		c4iw_get_ep(&ep->com);
+		pr_err("%s timer already started! ep %p\n",
+		       __func__, ep);
+		return;
+	}
+	clear_bit(TIMEOUT, &ep->com.flags);
+	c4iw_get_ep(&ep->com);
 	ep->timer.expires = jiffies + ep_timeout_secs * HZ;
 	ep->timer.data = (unsigned long)ep;
 	ep->timer.function = ep_timeout;
@@ -171,14 +173,10 @@ static void start_ep_timer(struct c4iw_ep *ep)
 
 static void stop_ep_timer(struct c4iw_ep *ep)
 {
-	PDBG("%s ep %p\n", __func__, ep);
-	if (!timer_pending(&ep->timer)) {
-		WARN(1, "%s timer stopped when its not running! "
-		       "ep %p state %u\n", __func__, ep, ep->com.state);
-		return;
-	}
+	PDBG("%s ep %p stopping\n", __func__, ep);
 	del_timer_sync(&ep->timer);
-	c4iw_put_ep(&ep->com);
+	if (!test_and_set_bit(TIMEOUT, &ep->com.flags))
+		c4iw_put_ep(&ep->com);
 }
 
 static int c4iw_l2t_send(struct c4iw_rdev *rdev, struct sk_buff *skb,
@@ -3191,11 +3189,16 @@ static DECLARE_WORK(skb_work, process_work);
 static void ep_timeout(unsigned long arg)
 {
 	struct c4iw_ep *ep = (struct c4iw_ep *)arg;
+	int kickit = 0;
 
 	spin_lock(&timeout_lock);
-	list_add_tail(&ep->entry, &timeout_list);
+	if (!test_and_set_bit(TIMEOUT, &ep->com.flags)) {
+		list_add_tail(&ep->entry, &timeout_list);
+		kickit = 1;
+	}
 	spin_unlock(&timeout_lock);
-	queue_work(workq, &skb_work);
+	if (kickit)
+		queue_work(workq, &skb_work);
 }
 
 /*
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index 0aaaa0e..94a3b3c 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -716,6 +716,7 @@ enum c4iw_ep_flags {
 	ABORT_REQ_IN_PROGRESS	= 1,
 	RELEASE_RESOURCES	= 2,
 	CLOSE_SENT		= 3,
+	TIMEOUT                 = 4,
 	QP_REFERENCED           = 5,
 };
 
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 08/11] RDMA/cxgb4: don't reconnect on abort for mpa_rev 1

[Vipul Pandya]

only reconnect if the endpoint wasn't freed.

peer_abort() should only attempt to reconnect if the endpoint wasn't freed.
Also remove hwtid from the debugfs idr.

Add missing check for peer2peer in MPAv2 code

use correct mpa version on reject.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |   11 +++++------
 1 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index ab5b4dd..88933af 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -284,10 +284,10 @@ void _c4iw_free_ep(struct kref *kref)
 	if (test_bit(QP_REFERENCED, &ep->com.flags))
 		deref_qp(ep);
 	if (test_bit(RELEASE_RESOURCES, &ep->com.flags)) {
+		remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid);
 		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid);
 		dst_release(ep->dst);
 		cxgb4_l2t_release(ep->l2t);
-		remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid);
 	}
 	kfree(ep);
 }
@@ -699,7 +699,7 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen)
 	memset(mpa, 0, sizeof(*mpa));
 	memcpy(mpa->key, MPA_KEY_REP, sizeof(mpa->key));
 	mpa->flags = MPA_REJECT;
-	mpa->revision = mpa_rev;
+	mpa->revision = ep->mpa_attr.version;
 	mpa->private_data_size = htons(plen);
 
 	if (ep->mpa_attr.version == 2 && ep->mpa_attr.enhanced_rdma_conn) {
@@ -2176,7 +2176,7 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb)
 		break;
 	case MPA_REQ_SENT:
 		stop_ep_timer(ep);
-		if (mpa_rev == 2 && ep->tried_with_mpa_v1)
+		if (mpa_rev == 1 || (mpa_rev == 2 && ep->tried_with_mpa_v1))
 			connect_reply_upcall(ep, -ECONNRESET);
 		else {
 			/*
@@ -2248,9 +2248,8 @@ static int peer_abort(struct c4iw_dev *dev, struct sk_buff *skb)
 out:
 	if (release)
 		release_ep_resources(ep);
-
-	/* retry with mpa-v1 */
-	if (ep && ep->retry_with_mpa_v1) {
+	else if (ep->retry_with_mpa_v1) {
+		remove_handle(ep->com.dev, &ep->com.dev->hwtid_idr, ep->hwtid);
 		cxgb4_remove_tid(ep->com.dev->rdev.lldi.tids, 0, ep->hwtid);
 		dst_release(ep->dst);
 		cxgb4_l2t_release(ep->l2t);
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 09/11] RDMA/cxgb4: Don't wakeup threads for MPAv2

[Vipul Pandya]

Don't wakeup threads blocked in rdma_init/rdma_fini if we are on
MPAv2, and want to retry connection with MPAv1.

Stop ep-timer on getting MPA version mismatch, before doing the
abort_connection - in process_mpa_request.

Take care to stop ep-timer in error paths for process_mpa_request.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |   12 +++++++++++-
 1 files changed, 11 insertions(+), 1 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 88933af..06c3a52 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1300,11 +1300,13 @@ static void process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
 	if (mpa->revision > mpa_rev) {
 		printk(KERN_ERR MOD "%s MPA version mismatch. Local = %d,"
 		       " Received = %d\n", __func__, mpa_rev, mpa->revision);
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
 
 	if (memcmp(mpa->key, MPA_KEY_REQ, sizeof(mpa->key))) {
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
@@ -1315,6 +1317,7 @@ static void process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
 	 * Fail if there's too much private data.
 	 */
 	if (plen > MPA_MAX_PRIVATE_DATA) {
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
@@ -1323,6 +1326,7 @@ static void process_mpa_request(struct c4iw_ep *ep, struct sk_buff *skb)
 	 * If plen does not account for pkt size
 	 */
 	if (ep->mpa_pkt_len > (sizeof(*mpa) + plen)) {
+		stop_ep_timer(ep);
 		abort_connection(ep, skb, GFP_KERNEL);
 		return;
 	}
@@ -3286,8 +3290,14 @@ static int peer_abort_intr(struct c4iw_dev *dev, struct sk_buff *skb)
 
 	/*
 	 * Wake up any threads in rdma_init() or rdma_fini().
+	 * However, if we are on MPAv2 and want to retry with MPAv1
+	 * then, don't wake up yet.
 	 */
-	c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
+	if (mpa_rev == 2 && !ep->tried_with_mpa_v1) {
+		if (ep->com.state != MPA_REQ_SENT)
+			c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
+	} else
+		c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET);
 	sched(dev, skb);
 	return 0;
 }
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 10/11] RDMA/cxgb4: Insert hwtid in pass_accept_req instead in pass_establish

[Vipul Pandya]

CPL_ABORT_REQ_RSS can come before TCP connection is established. In such case
peer_abort was trying to remove the hwtid which was not inserted. To avoid this
we insert the hwtid when we are sure that we are surely going to send passive
accept request.

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 06c3a52..37ea2fc 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -2010,6 +2010,7 @@ static int pass_accept_req(struct c4iw_dev *dev, struct sk_buff *skb)
 
 	init_timer(&child_ep->timer);
 	cxgb4_insert_tid(t, child_ep, hwtid);
+	insert_handle(dev, &dev->hwtid_idr, child_ep, child_ep->hwtid);
 	accept_cr(child_ep, peer_ip, skb, req);
 	set_bit(PASS_ACCEPT_REQ, &child_ep->com.history);
 	goto out;
@@ -2035,7 +2036,6 @@ static int pass_establish(struct c4iw_dev *dev, struct sk_buff *skb)
 	     ntohs(req->tcp_opt));
 
 	set_emss(ep, ntohs(req->tcp_opt));
-	insert_handle(dev, &dev->hwtid_idr, ep, ep->hwtid);
 
 	dst_confirm(ep->dst);
 	state_set(&ep->com, MPA_REQ_WAIT);
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 11/11] RDMA/cxgb4: Address sparse warnings

[Vipul Pandya]

It fixes following types of sparse warnings
- cast to pointer from integer of different size
- cast from pointer to integer of different size
- incorrect type in assignment (different base types)
- incorrect type in argument 1 (different base types)
- cast from restricted __be64
- cast from restricted __be32

Signed-off-by: Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org>
---
 drivers/infiniband/hw/cxgb4/cm.c     |   68 +++++++++++++++++++---------------
 drivers/infiniband/hw/cxgb4/device.c |    3 +-
 2 files changed, 40 insertions(+), 31 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 37ea2fc..dfca515 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1492,11 +1492,11 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 			V_FW_OFLD_CONNECTION_WR_ASTID(atid));
 	req->tcb.cplrxdataack_cplpassacceptrpl =
 			htons(F_FW_OFLD_CONNECTION_WR_CPLRXDATAACK);
-	req->tcb.tx_max = jiffies;
+	req->tcb.tx_max = (__force __be32) jiffies;
 	req->tcb.rcv_adv = htons(1);
 	cxgb4_best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx);
 	wscale = compute_wscale(rcv_win);
-	req->tcb.opt0 = TCAM_BYPASS(1) |
+	req->tcb.opt0 = (__force __be64) (TCAM_BYPASS(1) |
 		(nocong ? NO_CONG(1) : 0) |
 		KEEP_ALIVE(1) |
 		DELACK(1) |
@@ -1507,20 +1507,20 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 		SMAC_SEL(ep->smac_idx) |
 		DSCP(ep->tos) |
 		ULP_MODE(ULP_MODE_TCPDDP) |
-		RCV_BUFSIZ(rcv_win >> 10);
-	req->tcb.opt2 = PACE(1) |
+		RCV_BUFSIZ(rcv_win >> 10));
+	req->tcb.opt2 = (__force __be32) (PACE(1) |
 		TX_QUEUE(ep->com.dev->rdev.lldi.tx_modq[ep->tx_chan]) |
 		RX_CHANNEL(0) |
 		CCTRL_ECN(enable_ecn) |
-		RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);
+		RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid));
 	if (enable_tcp_timestamps)
-		req->tcb.opt2 |= TSTAMPS_EN(1);
+		req->tcb.opt2 |= (__force __be32) TSTAMPS_EN(1);
 	if (enable_tcp_sack)
-		req->tcb.opt2 |= SACK_EN(1);
+		req->tcb.opt2 |= (__force __be32) SACK_EN(1);
 	if (wscale && enable_tcp_window_scaling)
-		req->tcb.opt2 |= WND_SCALE_EN(1);
-	req->tcb.opt0 = cpu_to_be64(req->tcb.opt0);
-	req->tcb.opt2 = cpu_to_be32(req->tcb.opt2);
+		req->tcb.opt2 |= (__force __be32) WND_SCALE_EN(1);
+	req->tcb.opt0 = cpu_to_be64((__force u64) req->tcb.opt0);
+	req->tcb.opt2 = cpu_to_be32((__force u32) req->tcb.opt2);
 	set_wr_txq(skb, CPL_PRIORITY_CONTROL, ep->ctrlq_idx);
 	set_bit(ACT_OFLD_CONN, &ep->com.history);
 	c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
@@ -2773,7 +2773,8 @@ static void active_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb,
 	struct c4iw_ep *ep;
 	int atid = be32_to_cpu(req->tid);
 
-	ep = (struct c4iw_ep *)lookup_atid(dev->rdev.lldi.tids, req->tid);
+	ep = (struct c4iw_ep *)lookup_atid(dev->rdev.lldi.tids,
+					   (__force u32) req->tid);
 	if (!ep)
 		return;
 
@@ -2817,7 +2818,7 @@ static void passive_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb,
 	struct cpl_pass_accept_req *cpl;
 	int ret;
 
-	rpl_skb = (struct sk_buff *)cpu_to_be64(req->cookie);
+	rpl_skb = (__force struct sk_buff *)cpu_to_be64(req->cookie);
 	BUG_ON(!rpl_skb);
 	if (req->retval) {
 		PDBG("%s passive open failure %d\n", __func__, req->retval);
@@ -2828,7 +2829,8 @@ static void passive_ofld_conn_reply(struct c4iw_dev *dev, struct sk_buff *skb,
 	} else {
 		cpl = (struct cpl_pass_accept_req *)cplhdr(rpl_skb);
 		OPCODE_TID(cpl) = htonl(MK_OPCODE_TID(CPL_PASS_ACCEPT_REQ,
-						      htonl(req->tid)));
+					(__force u32) htonl(
+					(__force u32) req->tid)));
 		ret = pass_accept_req(dev, rpl_skb);
 		if (!ret)
 			kfree_skb(rpl_skb);
@@ -2874,10 +2876,10 @@ static void build_cpl_pass_accept_req(struct sk_buff *skb, int stid , u8 tos)
 	struct tcp_options_received tmp_opt;
 
 	/* Store values from cpl_rx_pkt in temporary location. */
-	vlantag = cpl->vlan;
-	len = cpl->len;
-	l2info  = cpl->l2info;
-	hdr_len = cpl->hdr_len;
+	vlantag = (__force u16) cpl->vlan;
+	len = (__force u16) cpl->len;
+	l2info  = (__force u32) cpl->l2info;
+	hdr_len = (__force u16) cpl->hdr_len;
 	intf = cpl->iff;
 
 	__skb_pull(skb, sizeof(*req) + sizeof(struct rss_header));
@@ -2888,19 +2890,24 @@ static void build_cpl_pass_accept_req(struct sk_buff *skb, int stid , u8 tos)
 	 */
 	memset(&tmp_opt, 0, sizeof(tmp_opt));
 	tcp_clear_options(&tmp_opt);
-	tcp_parse_options(skb, &tmp_opt, 0, 0, NULL);
+	tcp_parse_options(skb, &tmp_opt, NULL, 0, NULL);
 
 	req = (struct cpl_pass_accept_req *)__skb_push(skb, sizeof(*req));
 	memset(req, 0, sizeof(*req));
 	req->l2info = cpu_to_be16(V_SYN_INTF(intf) |
-			 V_SYN_MAC_IDX(G_RX_MACIDX(htonl(l2info))) |
+			 V_SYN_MAC_IDX(G_RX_MACIDX(
+			 (__force int) htonl(l2info))) |
 			 F_SYN_XACT_MATCH);
-	req->hdr_len = cpu_to_be32(V_SYN_RX_CHAN(G_RX_CHAN(htonl(l2info))) |
-				V_TCP_HDR_LEN(G_RX_TCPHDR_LEN(htons(hdr_len))) |
-				V_IP_HDR_LEN(G_RX_IPHDR_LEN(htons(hdr_len))) |
-				V_ETH_HDR_LEN(G_RX_ETHHDR_LEN(htonl(l2info))));
-	req->vlan = vlantag;
-	req->len = len;
+	req->hdr_len = cpu_to_be32(V_SYN_RX_CHAN(G_RX_CHAN(
+					(__force int) htonl(l2info))) |
+				   V_TCP_HDR_LEN(G_RX_TCPHDR_LEN(
+					(__force int) htons(hdr_len))) |
+				   V_IP_HDR_LEN(G_RX_IPHDR_LEN(
+					(__force int) htons(hdr_len))) |
+				   V_ETH_HDR_LEN(G_RX_ETHHDR_LEN(
+					(__force int) htonl(l2info))));
+	req->vlan = (__force __be16) vlantag;
+	req->len = (__force __be16) len;
 	req->tos_stid = cpu_to_be32(PASS_OPEN_TID(stid) |
 				    PASS_OPEN_TOS(tos));
 	req->tcpopt.mss = htons(tmp_opt.mss_clamp);
@@ -2929,7 +2936,7 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb,
 	req->op_compl = htonl(V_WR_OP(FW_OFLD_CONNECTION_WR) | FW_WR_COMPL(1));
 	req->len16_pkd = htonl(FW_WR_LEN16(DIV_ROUND_UP(sizeof(*req), 16)));
 	req->le.version_cpl = htonl(F_FW_OFLD_CONNECTION_WR_CPL);
-	req->le.filter = filter;
+	req->le.filter = (__force __be32) filter;
 	req->le.lport = lport;
 	req->le.pport = rport;
 	req->le.u.ipv4.lip = laddr;
@@ -2955,7 +2962,7 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb,
 	 * TP will ignore any value > 0 for MSS index.
 	 */
 	req->tcb.opt0 = cpu_to_be64(V_MSS_IDX(0xF));
-	req->cookie = cpu_to_be64((u64)skb);
+	req->cookie = (__force __u64) cpu_to_be64((u64)skb);
 
 	set_wr_txq(req_skb, CPL_PRIORITY_CONTROL, port_id);
 	cxgb4_ofld_send(dev->rdev.lldi.ports[0], req_skb);
@@ -3005,7 +3012,8 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb)
 	/*
 	 * Calculate the server tid from filter hit index from cpl_rx_pkt.
 	 */
-	stid = cpu_to_be32(rss->hash_val) - dev->rdev.lldi.tids->sftid_base
+	stid = (__force int) cpu_to_be32((__force u32) rss->hash_val)
+					  - dev->rdev.lldi.tids->sftid_base
 					  + dev->rdev.lldi.tids->nstids;
 
 	lep = (struct c4iw_ep *)lookup_stid(dev->rdev.lldi.tids, stid);
@@ -3066,10 +3074,10 @@ static int rx_pkt(struct c4iw_dev *dev, struct sk_buff *skb)
 
 	step = dev->rdev.lldi.nrxq / dev->rdev.lldi.nchan;
 	rss_qid = dev->rdev.lldi.rxq_ids[pi->port_id * step];
-	window = htons(tcph->window);
+	window = (__force u16) htons((__force u16)tcph->window);
 
 	/* Calcuate filter portion for LE region. */
-	filter = cpu_to_be32(select_ntuple(dev, dst, e));
+	filter = (__force unsigned int) cpu_to_be32(select_ntuple(dev, dst, e));
 
 	/*
 	 * Synthesize the cpl_pass_accept_req. We have everything except the
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index ba11c76..dc6adb2 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -797,7 +797,8 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp,
 		       "RSS %#llx, FL %#llx, len %u\n",
 		       pci_name(ctx->lldi.pdev), gl->va,
 		       (unsigned long long)be64_to_cpu(*rsp),
-		       (unsigned long long)be64_to_cpu(*(u64 *)gl->va),
+		       (unsigned long long)be64_to_cpu(
+		       *(__force __be64 *)gl->va),
 		       gl->tot_len);
 
 		return 0;
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 00/11] Critical bug fixes for RDMA/cxgb4

[Steve Wise]

Reviewed-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 00/11] Critical bug fixes for RDMA/cxgb4

[Steve Wise]

And:

Acked-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>

Not sure which one I should be using :)


On 1/7/2013 9:44 AM, Steve Wise wrote:
> Reviewed-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
> -- 
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 00/11] Critical bug fixes for RDMA/cxgb4

[Roland Dreier]

On Mon, Jan 7, 2013 at 5:11 AM, Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org> wrote:
> This patch series fixes critical bugs for RDMA/cxgb4. It fixes bugs in following
> areas:
> - Aborts connection in error scenarios
> - Logs only critical errors
> - Holds the reference of the QP untill TID is released
> - Avoids race condition in endpoint timeout
> - Fixes reconnect and version mismatch related bugs in MPAv2 related logic
> - Avoids removing hwtid which was not inserted
> - Addresses several sparse warnings

How can sparse warnings be a critical bug?

Which of these patches are really needed for 3.8?

 - R.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 00/11] Critical bug fixes for RDMA/cxgb4

[Vipul Pandya]

On 08-01-2013 06:03, Roland Dreier wrote:
> On Mon, Jan 7, 2013 at 5:11 AM, Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org> wrote:
>> This patch series fixes critical bugs for RDMA/cxgb4. It fixes bugs in following
>> areas:
>> - Aborts connection in error scenarios
>> - Logs only critical errors
>> - Holds the reference of the QP untill TID is released
>> - Avoids race condition in endpoint timeout
>> - Fixes reconnect and version mismatch related bugs in MPAv2 related logic
>> - Avoids removing hwtid which was not inserted
>> - Addresses several sparse warnings
> 
> How can sparse warnings be a critical bug?
> 
> Which of these patches are really needed for 3.8?
> 
>  - R.
> 

Hi Roland,

We would like first 10 patches to go in 3.8 since they are fixing kernel
panics, race condition, warning trace etc. I think sparse warnings patch
can be left out.

Should I send the series again with only first 10 patches in it?

Thanks,
Vipul
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 00/11] Critical bug fixes for RDMA/cxgb4

[Vipul Pandya]

On 08-01-2013 11:43, Vipul Pandya wrote:
> 
> 
> On 08-01-2013 06:03, Roland Dreier wrote:
>> On Mon, Jan 7, 2013 at 5:11 AM, Vipul Pandya <vipul-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org> wrote:
>>> This patch series fixes critical bugs for RDMA/cxgb4. It fixes bugs in following
>>> areas:
>>> - Aborts connection in error scenarios
>>> - Logs only critical errors
>>> - Holds the reference of the QP untill TID is released
>>> - Avoids race condition in endpoint timeout
>>> - Fixes reconnect and version mismatch related bugs in MPAv2 related logic
>>> - Avoids removing hwtid which was not inserted
>>> - Addresses several sparse warnings
>>
>> How can sparse warnings be a critical bug?
>>
>> Which of these patches are really needed for 3.8?
>>
>>  - R.
>>
> 
> Hi Roland,
> 
> We would like first 10 patches to go in 3.8 since they are fixing kernel
> panics, race condition, warning trace etc. I think sparse warnings patch
> can be left out.
> 
> Should I send the series again with only first 10 patches in it?
> 
> Thanks,
> Vipul
> 

Hi Roland,

Any update on this?

Thanks,
Vipul
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html