[PATCH net-next 00/10] s390/qeth: updates 2018-07-11

[PATCH net-next 00/10] s390/qeth: updates 2018-07-11

[Julian Wiedmann]

Hi Dave,

please apply this first batch of qeth patches for net-next. It brings the
usual cleanups, and some performance improvements to the transmit paths.

Thanks,
Julian


Julian Wiedmann (10):
  s390/qeth: various buffer management cleanups
  s390/qeth: remove unused buffer->aob pointer
  s390/qeth: fine-tune RX modesetting
  s390/qeth: clean up Output Queue selection
  s390/qeth: consolidate ccwgroup driver definition
  s390/qeth: clean up exported symbols
  s390/qeth: increase GSO max size for eligible L3 devices
  s390/qeth: add a L3 xmit wrapper
  s390/qeth: speed-up L3 IQD xmit
  s390/qeth: speed-up IPv4 OSA xmit

 drivers/s390/net/qeth_core.h      |  22 ++-
 drivers/s390/net/qeth_core_main.c | 131 +++++---------
 drivers/s390/net/qeth_core_mpc.h  |   2 +
 drivers/s390/net/qeth_l2.h        |   5 +
 drivers/s390/net/qeth_l2_main.c   |  20 +--
 drivers/s390/net/qeth_l3_main.c   | 360 ++++++++++++++++++++++----------------
 6 files changed, 281 insertions(+), 259 deletions(-)

-- 
2.16.4

[PATCH net-next 01/10] s390/qeth: various buffer management cleanups

[Julian Wiedmann]

Use the new qeth_scrub_qdio_buffer() helper, remove an extra parameter
from qeth_clear_output_buffer(), init the bufstates.user field just once
(in qeth_flush_buffers()) and remove some noisy trace messages.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_core_main.c | 66 ++++++++++++---------------------------
 1 file changed, 20 insertions(+), 46 deletions(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index d01ac29fd986..916c17d93ddb 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -1267,8 +1267,7 @@ static void qeth_release_skbs(struct qeth_qdio_out_buffer *buf)
 }
 
 static void qeth_clear_output_buffer(struct qeth_qdio_out_q *queue,
-		struct qeth_qdio_out_buffer *buf,
-		enum qeth_qdio_buffer_states newbufstate)
+				     struct qeth_qdio_out_buffer *buf)
 {
 	int i;
 
@@ -1276,23 +1275,19 @@ static void qeth_clear_output_buffer(struct qeth_qdio_out_q *queue,
 	if (buf->buffer->element[0].sflags & SBAL_SFLAGS0_PCI_REQ)
 		atomic_dec(&queue->set_pci_flags_count);
 
-	if (newbufstate == QETH_QDIO_BUF_EMPTY) {
-		qeth_release_skbs(buf);
-	}
+	qeth_release_skbs(buf);
+
 	for (i = 0; i < QETH_MAX_BUFFER_ELEMENTS(queue->card); ++i) {
 		if (buf->buffer->element[i].addr && buf->is_header[i])
 			kmem_cache_free(qeth_core_header_cache,
 				buf->buffer->element[i].addr);
 		buf->is_header[i] = 0;
-		buf->buffer->element[i].length = 0;
-		buf->buffer->element[i].addr = NULL;
-		buf->buffer->element[i].eflags = 0;
-		buf->buffer->element[i].sflags = 0;
 	}
-	buf->buffer->element[15].eflags = 0;
-	buf->buffer->element[15].sflags = 0;
+
+	qeth_scrub_qdio_buffer(buf->buffer,
+			       QETH_MAX_BUFFER_ELEMENTS(queue->card));
 	buf->next_element_to_fill = 0;
-	atomic_set(&buf->state, newbufstate);
+	atomic_set(&buf->state, QETH_QDIO_BUF_EMPTY);
 }
 
 static void qeth_clear_outq_buffers(struct qeth_qdio_out_q *q, int free)
@@ -1303,7 +1298,7 @@ static void qeth_clear_outq_buffers(struct qeth_qdio_out_q *q, int free)
 		if (!q->bufs[j])
 			continue;
 		qeth_cleanup_handled_pending(q, j, 1);
-		qeth_clear_output_buffer(q, q->bufs[j], QETH_QDIO_BUF_EMPTY);
+		qeth_clear_output_buffer(q, q->bufs[j]);
 		if (free) {
 			kmem_cache_free(qeth_qdio_outbuf_cache, q->bufs[j]);
 			q->bufs[j] = NULL;
@@ -2473,15 +2468,12 @@ static int qeth_ulp_setup(struct qeth_card *card)
 
 static int qeth_init_qdio_out_buf(struct qeth_qdio_out_q *q, int bidx)
 {
-	int rc;
 	struct qeth_qdio_out_buffer *newbuf;
 
-	rc = 0;
 	newbuf = kmem_cache_zalloc(qeth_qdio_outbuf_cache, GFP_ATOMIC);
-	if (!newbuf) {
-		rc = -ENOMEM;
-		goto out;
-	}
+	if (!newbuf)
+		return -ENOMEM;
+
 	newbuf->buffer = q->qdio_bufs[bidx];
 	skb_queue_head_init(&newbuf->skb_list);
 	lockdep_set_class(&newbuf->skb_list.lock, &qdio_out_skb_queue_key);
@@ -2490,15 +2482,7 @@ static int qeth_init_qdio_out_buf(struct qeth_qdio_out_q *q, int bidx)
 	newbuf->next_pending = q->bufs[bidx];
 	atomic_set(&newbuf->state, QETH_QDIO_BUF_EMPTY);
 	q->bufs[bidx] = newbuf;
-	if (q->bufstates) {
-		q->bufstates[bidx].user = newbuf;
-		QETH_CARD_TEXT_(q->card, 2, "nbs%d", bidx);
-		QETH_CARD_TEXT_(q->card, 2, "%lx", (long) newbuf);
-		QETH_CARD_TEXT_(q->card, 2, "%lx",
-				(long) newbuf->next_pending);
-	}
-out:
-	return rc;
+	return 0;
 }
 
 static void qeth_free_qdio_out_buf(struct qeth_qdio_out_q *q)
@@ -2908,8 +2892,7 @@ int qeth_init_qdio_queues(struct qeth_card *card)
 				   QDIO_MAX_BUFFERS_PER_Q);
 		for (j = 0; j < QDIO_MAX_BUFFERS_PER_Q; ++j) {
 			qeth_clear_output_buffer(card->qdio.out_qs[i],
-					card->qdio.out_qs[i]->bufs[j],
-					QETH_QDIO_BUF_EMPTY);
+						 card->qdio.out_qs[i]->bufs[j]);
 		}
 		card->qdio.out_qs[i]->card = card;
 		card->qdio.out_qs[i]->next_buf_to_fill = 0;
@@ -3634,10 +3617,10 @@ int qeth_configure_cq(struct qeth_card *card, enum qeth_cq cq)
 }
 EXPORT_SYMBOL_GPL(qeth_configure_cq);
 
-
-static void qeth_qdio_cq_handler(struct qeth_card *card,
-		unsigned int qdio_err,
-		unsigned int queue, int first_element, int count) {
+static void qeth_qdio_cq_handler(struct qeth_card *card, unsigned int qdio_err,
+				 unsigned int queue, int first_element,
+				 int count)
+{
 	struct qeth_qdio_q *cq = card->qdio.c_q;
 	int i;
 	int rc;
@@ -3663,25 +3646,17 @@ static void qeth_qdio_cq_handler(struct qeth_card *card,
 	for (i = first_element; i < first_element + count; ++i) {
 		int bidx = i % QDIO_MAX_BUFFERS_PER_Q;
 		struct qdio_buffer *buffer = cq->qdio_bufs[bidx];
-		int e;
+		int e = 0;
 
-		e = 0;
 		while ((e < QDIO_MAX_ELEMENTS_PER_BUFFER) &&
 		       buffer->element[e].addr) {
 			unsigned long phys_aob_addr;
 
 			phys_aob_addr = (unsigned long) buffer->element[e].addr;
 			qeth_qdio_handle_aob(card, phys_aob_addr);
-			buffer->element[e].addr = NULL;
-			buffer->element[e].eflags = 0;
-			buffer->element[e].sflags = 0;
-			buffer->element[e].length = 0;
-
 			++e;
 		}
-
-		buffer->element[15].eflags = 0;
-		buffer->element[15].sflags = 0;
+		qeth_scrub_qdio_buffer(buffer, QDIO_MAX_ELEMENTS_PER_BUFFER);
 	}
 	rc = do_QDIO(CARD_DDEV(card), QDIO_FLAG_SYNC_INPUT, queue,
 		    card->qdio.c_q->next_buf_to_init,
@@ -3782,8 +3757,7 @@ static void qeth_qdio_output_handler(struct ccw_device *ccwdev,
 				qeth_notify_skbs(queue, buffer, n);
 			}
 
-			qeth_clear_output_buffer(queue, buffer,
-						QETH_QDIO_BUF_EMPTY);
+			qeth_clear_output_buffer(queue, buffer);
 		}
 		qeth_cleanup_handled_pending(queue, bidx, 0);
 	}
-- 
2.16.4

[PATCH net-next 02/10] s390/qeth: remove unused buffer->aob pointer

[Julian Wiedmann]

Except for tracing, the pointer is not used.

At the same time, accessing it from qeth_qdio_output_handler() is racy:
whenever qeth_qdio_cq_handler() gets control, its call to
qeth_qdio_handle_aob() frees the AOB.

So the AOB pointer that qeth_qdio_output_handler() stores into 'buffer'
can go stale at any time, and trigger a use-after-free.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_core.h      | 1 -
 drivers/s390/net/qeth_core_main.c | 7 -------
 2 files changed, 8 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index a246a618f9a4..0ca6ea319d88 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -465,7 +465,6 @@ struct qeth_qdio_out_buffer {
 	struct sk_buff_head skb_list;
 	int is_header[QDIO_MAX_ELEMENTS_PER_BUFFER];
 
-	struct qaob *aob;
 	struct qeth_qdio_out_q *q;
 	struct qeth_qdio_out_buffer *next_pending;
 };
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 916c17d93ddb..cfe68e3bfe7a 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -473,7 +473,6 @@ static void qeth_cleanup_handled_pending(struct qeth_qdio_out_q *q, int bidx,
 	if (forced_cleanup && (atomic_read(&(q->bufs[bidx]->state)) ==
 					QETH_QDIO_BUF_HANDLED_DELAYED)) {
 		/* for recovery situations */
-		q->bufs[bidx]->aob = q->bufstates[bidx].aob;
 		qeth_init_qdio_out_buf(q, bidx);
 		QETH_CARD_TEXT(q->card, 2, "clprecov");
 	}
@@ -510,7 +509,6 @@ static void qeth_qdio_handle_aob(struct qeth_card *card,
 	}
 	qeth_notify_skbs(buffer->q, buffer, notification);
 
-	buffer->aob = NULL;
 	/* Free dangling allocations. The attached skbs are handled by
 	 * qeth_cleanup_handled_pending().
 	 */
@@ -2478,7 +2476,6 @@ static int qeth_init_qdio_out_buf(struct qeth_qdio_out_q *q, int bidx)
 	skb_queue_head_init(&newbuf->skb_list);
 	lockdep_set_class(&newbuf->skb_list.lock, &qdio_out_skb_queue_key);
 	newbuf->q = q;
-	newbuf->aob = NULL;
 	newbuf->next_pending = q->bufs[bidx];
 	atomic_set(&newbuf->state, QETH_QDIO_BUF_EMPTY);
 	q->bufs[bidx] = newbuf;
@@ -3735,11 +3732,7 @@ static void qeth_qdio_output_handler(struct ccw_device *ccwdev,
 				qeth_notify_skbs(queue, buffer,
 						 TX_NOTIFY_PENDING);
 			}
-			buffer->aob = queue->bufstates[bidx].aob;
 			QETH_CARD_TEXT_(queue->card, 5, "pel%d", bidx);
-			QETH_CARD_TEXT(queue->card, 5, "aob");
-			QETH_CARD_TEXT_(queue->card, 5, "%lx",
-					virt_to_phys(buffer->aob));
 
 			/* prepare the queue slot for re-use: */
 			qeth_scrub_qdio_buffer(buffer->buffer,
-- 
2.16.4

[PATCH net-next 03/10] s390/qeth: fine-tune RX modesetting

[Julian Wiedmann]

Changing a device's address lists (or its promisc mode) already triggers
an RX modeset, there's no need to do it manually from the L2 driver's
ndo_vlan_rx_kill_vid() hook.

Also when setting a device online, dev_open() already calls
dev_set_rx_mode(). So a manual modeset is only necessary from the
recovery path.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_l2_main.c | 5 +----
 drivers/s390/net/qeth_l3_main.c | 7 ++++---
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 2487f0aeb165..730ab51fbac5 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -26,7 +26,6 @@
 
 static int qeth_l2_set_offline(struct ccwgroup_device *);
 static int qeth_l2_stop(struct net_device *);
-static void qeth_l2_set_rx_mode(struct net_device *);
 static void qeth_bridgeport_query_support(struct qeth_card *card);
 static void qeth_bridge_state_change(struct qeth_card *card,
 					struct qeth_ipa_cmd *cmd);
@@ -344,7 +343,6 @@ static int qeth_l2_vlan_rx_kill_vid(struct net_device *dev,
 		rc = qeth_l2_send_setdelvlan(card, vid, IPA_CMD_DELVLAN);
 		kfree(tmpid);
 	}
-	qeth_l2_set_rx_mode(card->dev);
 	return rc;
 }
 
@@ -1125,13 +1123,12 @@ static int __qeth_l2_set_online(struct ccwgroup_device *gdev, int recovery_mode)
 		if (recovery_mode &&
 		    card->info.type != QETH_CARD_TYPE_OSN) {
 			__qeth_l2_open(card->dev);
+			qeth_l2_set_rx_mode(card->dev);
 		} else {
 			rtnl_lock();
 			dev_open(card->dev);
 			rtnl_unlock();
 		}
-		/* this also sets saved unicast addresses */
-		qeth_l2_set_rx_mode(card->dev);
 	}
 	/* let user_space know that device is online */
 	kobject_uevent(&gdev->dev.kobj, KOBJ_CHANGE);
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 5905dc63e256..a54881cdda82 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2666,11 +2666,12 @@ static int __qeth_l3_set_online(struct ccwgroup_device *gdev, int recovery_mode)
 	qeth_enable_hw_features(card->dev);
 	if (recover_flag == CARD_STATE_RECOVER) {
 		rtnl_lock();
-		if (recovery_mode)
+		if (recovery_mode) {
 			__qeth_l3_open(card->dev);
-		else
+			qeth_l3_set_rx_mode(card->dev);
+		} else {
 			dev_open(card->dev);
-		qeth_l3_set_rx_mode(card->dev);
+		}
 		rtnl_unlock();
 	}
 	qeth_trace_features(card);
-- 
2.16.4

[PATCH net-next 04/10] s390/qeth: clean up Output Queue selection

[Julian Wiedmann]

Consolidate duplicated code, fix the misuse of RTN_UNSPEC and simplify
the handling of non-unicast traffic on IQD devices.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_core.h      | 15 +++++++++++++--
 drivers/s390/net/qeth_core_main.c |  8 +-------
 drivers/s390/net/qeth_core_mpc.h  |  2 ++
 drivers/s390/net/qeth_l2_main.c   | 13 ++++---------
 drivers/s390/net/qeth_l3_main.c   | 16 +++++++---------
 5 files changed, 27 insertions(+), 27 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 0ca6ea319d88..082c06d6380b 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -661,7 +661,6 @@ struct qeth_card_info {
 	int portno;
 	enum qeth_card_types type;
 	enum qeth_link_types link_type;
-	int is_multicast_different;
 	int initial_mtu;
 	int max_mtu;
 	int broadcast_capable;
@@ -934,6 +933,19 @@ static inline int qeth_send_simple_setassparms_v6(struct qeth_card *card,
 						 data, QETH_PROT_IPV6);
 }
 
+int qeth_get_priority_queue(struct qeth_card *card, struct sk_buff *skb,
+			    int ipv);
+static inline struct qeth_qdio_out_q *qeth_get_tx_queue(struct qeth_card *card,
+							struct sk_buff *skb,
+							int ipv, int cast_type)
+{
+	if (IS_IQD(card) && cast_type != RTN_UNICAST)
+		return card->qdio.out_qs[card->qdio.no_out_queues - 1];
+	if (!card->qdio.do_prio_queueing)
+		return card->qdio.out_qs[card->qdio.default_out_queue];
+	return card->qdio.out_qs[qeth_get_priority_queue(card, skb, ipv)];
+}
+
 extern struct qeth_discipline qeth_l2_discipline;
 extern struct qeth_discipline qeth_l3_discipline;
 extern const struct attribute_group *qeth_generic_attr_groups[];
@@ -1001,7 +1013,6 @@ int qeth_bridgeport_query_ports(struct qeth_card *card,
 	enum qeth_sbp_roles *role, enum qeth_sbp_states *state);
 int qeth_bridgeport_setrole(struct qeth_card *card, enum qeth_sbp_roles role);
 int qeth_bridgeport_an_set(struct qeth_card *card, int enable);
-int qeth_get_priority_queue(struct qeth_card *, struct sk_buff *, int, int);
 int qeth_get_elements_no(struct qeth_card *card, struct sk_buff *skb,
 			 int extra_elems, int data_offset);
 int qeth_get_elements_for_frags(struct sk_buff *);
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index cfe68e3bfe7a..e8b18a9e07d7 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -1537,8 +1537,6 @@ static void qeth_determine_card_type(struct qeth_card *card)
 	card->qdio.default_out_queue = QETH_DEFAULT_QUEUE;
 	card->info.type = CARD_RDEV(card)->id.driver_info;
 	card->qdio.no_out_queues = QETH_MAX_QUEUES;
-	if (card->info.type == QETH_CARD_TYPE_IQD)
-		card->info.is_multicast_different = 0x0103;
 	qeth_update_from_chp_desc(card);
 }
 
@@ -3777,15 +3775,11 @@ static inline int qeth_cut_iqd_prio(struct qeth_card *card, int queue_num)
  * Note: Function assumes that we have 4 outbound queues.
  */
 int qeth_get_priority_queue(struct qeth_card *card, struct sk_buff *skb,
-			int ipv, int cast_type)
+			    int ipv)
 {
 	__be16 *tci;
 	u8 tos;
 
-	if (cast_type && card->info.is_multicast_different)
-		return card->info.is_multicast_different &
-			(card->qdio.no_out_queues - 1);
-
 	switch (card->qdio.do_prio_queueing) {
 	case QETH_PRIO_Q_ING_TOS:
 	case QETH_PRIO_Q_ING_PREC:
diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h
index 878e62f35169..54c35224262a 100644
--- a/drivers/s390/net/qeth_core_mpc.h
+++ b/drivers/s390/net/qeth_core_mpc.h
@@ -64,6 +64,8 @@ enum qeth_card_types {
 	QETH_CARD_TYPE_OSX     = 2,
 };
 
+#define IS_IQD(card)	((card)->info.type == QETH_CARD_TYPE_IQD)
+
 #define QETH_MPC_DIFINFO_LEN_INDICATES_LINK_TYPE 0x18
 /* only the first two bytes are looked at in qeth_get_cardname_short */
 enum qeth_link_types {
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 730ab51fbac5..5910fd524872 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -185,12 +185,12 @@ static void qeth_l2_del_all_macs(struct qeth_card *card)
 static int qeth_l2_get_cast_type(struct qeth_card *card, struct sk_buff *skb)
 {
 	if (card->info.type == QETH_CARD_TYPE_OSN)
-		return RTN_UNSPEC;
+		return RTN_UNICAST;
 	if (is_broadcast_ether_addr(skb->data))
 		return RTN_BROADCAST;
 	if (is_multicast_ether_addr(skb->data))
 		return RTN_MULTICAST;
-	return RTN_UNSPEC;
+	return RTN_UNICAST;
 }
 
 static void qeth_l2_fill_header(struct qeth_hdr *hdr, struct sk_buff *skb,
@@ -768,18 +768,13 @@ static netdev_tx_t qeth_l2_hard_start_xmit(struct sk_buff *skb,
 	int tx_bytes = skb->len;
 	int rc;
 
-	if (card->qdio.do_prio_queueing || (cast_type &&
-					card->info.is_multicast_different))
-		queue = card->qdio.out_qs[qeth_get_priority_queue(card, skb,
-					ipv, cast_type)];
-	else
-		queue = card->qdio.out_qs[card->qdio.default_out_queue];
-
 	if ((card->state != CARD_STATE_UP) || !card->lan_online) {
 		card->stats.tx_carrier_errors++;
 		goto tx_drop;
 	}
 
+	queue = qeth_get_tx_queue(card, skb, ipv, cast_type);
+
 	if (card->options.performance_stats) {
 		card->perf_stats.outbound_cnt++;
 		card->perf_stats.outbound_start_time = qeth_get_micros();
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index a54881cdda82..7b4de46bf6e3 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1978,17 +1978,17 @@ static int qeth_l3_get_cast_type(struct sk_buff *skb)
 		    (cast_type == RTN_MULTICAST) ||
 		    (cast_type == RTN_ANYCAST))
 			return cast_type;
-		return RTN_UNSPEC;
+		return RTN_UNICAST;
 	}
 	rcu_read_unlock();
 
 	/* no neighbour (eg AF_PACKET), fall back to target's IP address ... */
 	if (be16_to_cpu(skb->protocol) == ETH_P_IPV6)
 		return ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr) ?
-				RTN_MULTICAST : RTN_UNSPEC;
+				RTN_MULTICAST : RTN_UNICAST;
 	else if (be16_to_cpu(skb->protocol) == ETH_P_IP)
 		return ipv4_is_multicast(ip_hdr(skb)->daddr) ?
-				RTN_MULTICAST : RTN_UNSPEC;
+				RTN_MULTICAST : RTN_UNICAST;
 
 	/* ... and MAC address */
 	if (ether_addr_equal_64bits(eth_hdr(skb)->h_dest, skb->dev->broadcast))
@@ -1997,7 +1997,7 @@ static int qeth_l3_get_cast_type(struct sk_buff *skb)
 		return RTN_MULTICAST;
 
 	/* default to unicast */
-	return RTN_UNSPEC;
+	return RTN_UNICAST;
 }
 
 static void qeth_l3_fill_af_iucv_hdr(struct qeth_card *card,
@@ -2168,11 +2168,7 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 	struct sk_buff *new_skb = NULL;
 	int ipv = qeth_get_ip_version(skb);
 	int cast_type = qeth_l3_get_cast_type(skb);
-	struct qeth_qdio_out_q *queue =
-		card->qdio.out_qs[card->qdio.do_prio_queueing
-			|| (cast_type && card->info.is_multicast_different) ?
-			qeth_get_priority_queue(card, skb, ipv, cast_type) :
-			card->qdio.default_out_queue];
+	struct qeth_qdio_out_q *queue;
 	int tx_bytes = skb->len;
 	unsigned int hd_len = 0;
 	bool use_tso;
@@ -2195,6 +2191,8 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 	    (card->info.broadcast_capable == 0))
 		goto tx_drop;
 
+	queue = qeth_get_tx_queue(card, skb, ipv, cast_type);
+
 	if (card->options.performance_stats) {
 		card->perf_stats.outbound_cnt++;
 		card->perf_stats.outbound_start_time = qeth_get_micros();
-- 
2.16.4

[PATCH net-next 05/10] s390/qeth: consolidate ccwgroup driver definition

[Julian Wiedmann]

Reshuffle the code a bit so that everything is in one place.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_core_main.c | 42 +++++++++++++++++++--------------------
 1 file changed, 21 insertions(+), 21 deletions(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index e8b18a9e07d7..4ba40a64907a 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -5848,31 +5848,13 @@ static int qeth_core_restore(struct ccwgroup_device *gdev)
 	return 0;
 }
 
-static struct ccwgroup_driver qeth_core_ccwgroup_driver = {
-	.driver = {
-		.owner = THIS_MODULE,
-		.name = "qeth",
-	},
-	.ccw_driver = &qeth_ccw_driver,
-	.setup = qeth_core_probe_device,
-	.remove = qeth_core_remove_device,
-	.set_online = qeth_core_set_online,
-	.set_offline = qeth_core_set_offline,
-	.shutdown = qeth_core_shutdown,
-	.prepare = NULL,
-	.complete = NULL,
-	.freeze = qeth_core_freeze,
-	.thaw = qeth_core_thaw,
-	.restore = qeth_core_restore,
-};
-
 static ssize_t group_store(struct device_driver *ddrv, const char *buf,
 			   size_t count)
 {
 	int err;
 
-	err = ccwgroup_create_dev(qeth_core_root_dev,
-				  &qeth_core_ccwgroup_driver, 3, buf);
+	err = ccwgroup_create_dev(qeth_core_root_dev, to_ccwgroupdrv(ddrv), 3,
+				  buf);
 
 	return err ? err : count;
 }
@@ -5890,6 +5872,25 @@ static const struct attribute_group *qeth_drv_attr_groups[] = {
 	NULL,
 };
 
+static struct ccwgroup_driver qeth_core_ccwgroup_driver = {
+	.driver = {
+		.groups = qeth_drv_attr_groups,
+		.owner = THIS_MODULE,
+		.name = "qeth",
+	},
+	.ccw_driver = &qeth_ccw_driver,
+	.setup = qeth_core_probe_device,
+	.remove = qeth_core_remove_device,
+	.set_online = qeth_core_set_online,
+	.set_offline = qeth_core_set_offline,
+	.shutdown = qeth_core_shutdown,
+	.prepare = NULL,
+	.complete = NULL,
+	.freeze = qeth_core_freeze,
+	.thaw = qeth_core_thaw,
+	.restore = qeth_core_restore,
+};
+
 int qeth_do_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 {
 	struct qeth_card *card = dev->ml_priv;
@@ -6581,7 +6582,6 @@ static int __init qeth_core_init(void)
 	rc = ccw_driver_register(&qeth_ccw_driver);
 	if (rc)
 		goto ccw_err;
-	qeth_core_ccwgroup_driver.driver.groups = qeth_drv_attr_groups;
 	rc = ccwgroup_driver_register(&qeth_core_ccwgroup_driver);
 	if (rc)
 		goto ccwgroup_err;
-- 
2.16.4

[PATCH net-next 06/10] s390/qeth: clean up exported symbols

[Julian Wiedmann]

Remove some redundant EXPORTs. While at it, also move some L2-only
prototypes into the proper header file.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_core.h      | 6 ------
 drivers/s390/net/qeth_core_main.c | 8 +++-----
 drivers/s390/net/qeth_l2.h        | 5 +++++
 drivers/s390/net/qeth_l2_main.c   | 2 --
 4 files changed, 8 insertions(+), 13 deletions(-)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 082c06d6380b..a932aac62d0e 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -983,7 +983,6 @@ int qeth_send_ipa_cmd(struct qeth_card *, struct qeth_cmd_buffer *,
 		  void *);
 struct qeth_cmd_buffer *qeth_get_ipacmd_buffer(struct qeth_card *,
 			enum qeth_ipa_cmds, enum qeth_prot_versions);
-int qeth_query_setadapterparms(struct qeth_card *);
 struct sk_buff *qeth_core_get_next_skb(struct qeth_card *,
 		struct qeth_qdio_buffer *, struct qdio_buffer_element **, int *,
 		struct qeth_hdr **);
@@ -1009,10 +1008,6 @@ int qeth_query_switch_attributes(struct qeth_card *card,
 int qeth_send_control_data(struct qeth_card *, int, struct qeth_cmd_buffer *,
 	int (*reply_cb)(struct qeth_card *, struct qeth_reply*, unsigned long),
 	void *reply_param);
-int qeth_bridgeport_query_ports(struct qeth_card *card,
-	enum qeth_sbp_roles *role, enum qeth_sbp_states *state);
-int qeth_bridgeport_setrole(struct qeth_card *card, enum qeth_sbp_roles role);
-int qeth_bridgeport_an_set(struct qeth_card *card, int enable);
 int qeth_get_elements_no(struct qeth_card *card, struct sk_buff *skb,
 			 int extra_elems, int data_offset);
 int qeth_get_elements_for_frags(struct sk_buff *);
@@ -1036,7 +1031,6 @@ int qeth_set_access_ctrl_online(struct qeth_card *card, int fallback);
 int qeth_hdr_chk_and_bounce(struct sk_buff *, struct qeth_hdr **, int);
 int qeth_configure_cq(struct qeth_card *, enum qeth_cq);
 int qeth_hw_trap(struct qeth_card *, enum qeth_diags_trap_action);
-int qeth_query_ipassists(struct qeth_card *, enum qeth_prot_versions prot);
 void qeth_trace_features(struct qeth_card *);
 void qeth_close_dev(struct qeth_card *);
 int qeth_send_setassparms(struct qeth_card *, struct qeth_cmd_buffer *, __u16,
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 4ba40a64907a..d80972b9bfc7 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -3054,7 +3054,7 @@ static struct qeth_cmd_buffer *qeth_get_adapter_cmd(struct qeth_card *card,
 	return iob;
 }
 
-int qeth_query_setadapterparms(struct qeth_card *card)
+static int qeth_query_setadapterparms(struct qeth_card *card)
 {
 	int rc;
 	struct qeth_cmd_buffer *iob;
@@ -3067,7 +3067,6 @@ int qeth_query_setadapterparms(struct qeth_card *card)
 	rc = qeth_send_ipa_cmd(card, iob, qeth_query_setadapterparms_cb, NULL);
 	return rc;
 }
-EXPORT_SYMBOL_GPL(qeth_query_setadapterparms);
 
 static int qeth_query_ipassists_cb(struct qeth_card *card,
 		struct qeth_reply *reply, unsigned long data)
@@ -3107,7 +3106,8 @@ static int qeth_query_ipassists_cb(struct qeth_card *card,
 	return 0;
 }
 
-int qeth_query_ipassists(struct qeth_card *card, enum qeth_prot_versions prot)
+static int qeth_query_ipassists(struct qeth_card *card,
+				enum qeth_prot_versions prot)
 {
 	int rc;
 	struct qeth_cmd_buffer *iob;
@@ -3119,7 +3119,6 @@ int qeth_query_ipassists(struct qeth_card *card, enum qeth_prot_versions prot)
 	rc = qeth_send_ipa_cmd(card, iob, qeth_query_ipassists_cb, NULL);
 	return rc;
 }
-EXPORT_SYMBOL_GPL(qeth_query_ipassists);
 
 static int qeth_query_switch_attributes_cb(struct qeth_card *card,
 				struct qeth_reply *reply, unsigned long data)
@@ -3158,7 +3157,6 @@ int qeth_query_switch_attributes(struct qeth_card *card,
 	return qeth_send_ipa_cmd(card, iob,
 				qeth_query_switch_attributes_cb, sw_info);
 }
-EXPORT_SYMBOL_GPL(qeth_query_switch_attributes);
 
 static int qeth_query_setdiagass_cb(struct qeth_card *card,
 		struct qeth_reply *reply, unsigned long data)
diff --git a/drivers/s390/net/qeth_l2.h b/drivers/s390/net/qeth_l2.h
index f2130051ca11..ddc615b431a8 100644
--- a/drivers/s390/net/qeth_l2.h
+++ b/drivers/s390/net/qeth_l2.h
@@ -14,6 +14,11 @@ extern const struct attribute_group *qeth_l2_attr_groups[];
 int qeth_l2_create_device_attributes(struct device *);
 void qeth_l2_remove_device_attributes(struct device *);
 void qeth_l2_setup_bridgeport_attrs(struct qeth_card *card);
+int qeth_bridgeport_query_ports(struct qeth_card *card,
+				enum qeth_sbp_roles *role,
+				enum qeth_sbp_states *state);
+int qeth_bridgeport_setrole(struct qeth_card *card, enum qeth_sbp_roles role);
+int qeth_bridgeport_an_set(struct qeth_card *card, int enable);
 
 int qeth_l2_vnicc_set_state(struct qeth_card *card, u32 vnicc, bool state);
 int qeth_l2_vnicc_get_state(struct qeth_card *card, u32 vnicc, bool *state);
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 5910fd524872..8ac243de7a9e 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -1869,7 +1869,6 @@ int qeth_bridgeport_query_ports(struct qeth_card *card,
 		return rc;
 	return qeth_bridgeport_makerc(card, &cbctl, IPA_SBP_QUERY_BRIDGE_PORTS);
 }
-EXPORT_SYMBOL_GPL(qeth_bridgeport_query_ports);
 
 static int qeth_bridgeport_set_cb(struct qeth_card *card,
 	struct qeth_reply *reply, unsigned long data)
@@ -2017,7 +2016,6 @@ int qeth_bridgeport_an_set(struct qeth_card *card, int enable)
 		rc = qdio_pnso_brinfo(schid, 0, &response, NULL, NULL);
 	return qeth_anset_makerc(card, rc, response);
 }
-EXPORT_SYMBOL_GPL(qeth_bridgeport_an_set);
 
 static bool qeth_bridgeport_is_in_use(struct qeth_card *card)
 {
-- 
2.16.4

[PATCH net-next 07/10] s390/qeth: increase GSO max size for eligible L3 devices

[Julian Wiedmann]

When a L3 device doesn't offer TSO, allow the stack to build full-size
GSO skbs.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_l3_main.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 7b4de46bf6e3..d07803df460b 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2536,8 +2536,9 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 				NETIF_F_HW_VLAN_CTAG_RX |
 				NETIF_F_HW_VLAN_CTAG_FILTER;
 	netif_keep_dst(card->dev);
-	netif_set_gso_max_size(card->dev, (QETH_MAX_BUFFER_ELEMENTS(card) - 1) *
-					  PAGE_SIZE);
+	if (card->dev->hw_features & NETIF_F_TSO)
+		netif_set_gso_max_size(card->dev,
+				       PAGE_SIZE * (QETH_MAX_BUFFER_ELEMENTS(card) - 1));
 
 	SET_NETDEV_DEV(card->dev, &card->gdev->dev);
 	netif_napi_add(card->dev, &card->napi, qeth_poll, QETH_NAPI_WEIGHT);
-- 
2.16.4

[PATCH net-next 08/10] s390/qeth: add a L3 xmit wrapper

[Julian Wiedmann]

In preparation for future work, move the high-level xmit work into a
separate wrapper. This matches the L2 xmit code.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_l3_main.c | 122 +++++++++++++++++++++-------------------
 1 file changed, 65 insertions(+), 57 deletions(-)

diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index d07803df460b..b5a54cc07017 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2156,48 +2156,21 @@ static int qeth_l3_get_elements_no_tso(struct qeth_card *card,
 	return elements;
 }
 
-static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
-					   struct net_device *dev)
+static int qeth_l3_xmit(struct qeth_card *card, struct sk_buff *skb,
+			struct qeth_qdio_out_q *queue, int ipv, int cast_type)
 {
 	int rc;
 	__be16 *tag;
 	struct qeth_hdr *hdr = NULL;
 	int hdr_elements = 0;
 	int elements;
-	struct qeth_card *card = dev->ml_priv;
 	struct sk_buff *new_skb = NULL;
-	int ipv = qeth_get_ip_version(skb);
-	int cast_type = qeth_l3_get_cast_type(skb);
-	struct qeth_qdio_out_q *queue;
 	int tx_bytes = skb->len;
 	unsigned int hd_len = 0;
 	bool use_tso;
 	int data_offset = -1;
 	unsigned int nr_frags;
 
-	if (((card->info.type == QETH_CARD_TYPE_IQD) &&
-	     (((card->options.cq != QETH_CQ_ENABLED) && !ipv) ||
-	      ((card->options.cq == QETH_CQ_ENABLED) &&
-	       (be16_to_cpu(skb->protocol) != ETH_P_AF_IUCV)))) ||
-	    card->options.sniffer)
-			goto tx_drop;
-
-	if ((card->state != CARD_STATE_UP) || !card->lan_online) {
-		card->stats.tx_carrier_errors++;
-		goto tx_drop;
-	}
-
-	if ((cast_type == RTN_BROADCAST) &&
-	    (card->info.broadcast_capable == 0))
-		goto tx_drop;
-
-	queue = qeth_get_tx_queue(card, skb, ipv, cast_type);
-
-	if (card->options.performance_stats) {
-		card->perf_stats.outbound_cnt++;
-		card->perf_stats.outbound_start_time = qeth_get_micros();
-	}
-
 	/* Ignore segment size from skb_is_gso(), 1 page is always used. */
 	use_tso = skb_is_gso(skb) &&
 		  (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4);
@@ -2208,14 +2181,14 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 		hd_len = sizeof(*hdr);
 		hdr = kmem_cache_alloc(qeth_core_header_cache, GFP_ATOMIC);
 		if (!hdr)
-			goto tx_drop;
+			return -ENOMEM;
 		hdr_elements++;
 	} else {
 		/* create a clone with writeable headroom */
 		new_skb = skb_realloc_headroom(skb, sizeof(struct qeth_hdr_tso)
 					+ VLAN_HLEN);
 		if (!new_skb)
-			goto tx_drop;
+			return -ENOMEM;
 
 		if (ipv == 4) {
 			skb_pull(new_skb, ETH_HLEN);
@@ -2234,24 +2207,22 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 		}
 	}
 
-	netif_stop_queue(dev);
-
 	/* fix hardware limitation: as long as we do not have sbal
 	 * chaining we can not send long frag lists
 	 */
 	if ((card->info.type != QETH_CARD_TYPE_IQD) &&
 	    ((use_tso && !qeth_l3_get_elements_no_tso(card, new_skb, 1)) ||
 	     (!use_tso && !qeth_get_elements_no(card, new_skb, 0, 0)))) {
-		int lin_rc = skb_linearize(new_skb);
+		rc = skb_linearize(new_skb);
 
 		if (card->options.performance_stats) {
-			if (lin_rc)
+			if (rc)
 				card->perf_stats.tx_linfail++;
 			else
 				card->perf_stats.tx_lin++;
 		}
-		if (lin_rc)
-			goto tx_drop;
+		if (rc)
+			goto out;
 	}
 	nr_frags = skb_shinfo(new_skb)->nr_frags;
 
@@ -2290,9 +2261,8 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 		   qeth_get_elements_no(card, new_skb, hdr_elements,
 					(data_offset > 0) ? data_offset : 0);
 	if (!elements) {
-		if (data_offset >= 0)
-			kmem_cache_free(qeth_core_header_cache, hdr);
-		goto tx_drop;
+		rc = -E2BIG;
+		goto out;
 	}
 	elements += hdr_elements;
 
@@ -2306,17 +2276,18 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 			len = sizeof(struct qeth_hdr_layer3);
 		}
 
-		if (qeth_hdr_chk_and_bounce(new_skb, &hdr, len))
-			goto tx_drop;
+		if (qeth_hdr_chk_and_bounce(new_skb, &hdr, len)) {
+			rc = -EINVAL;
+			goto out;
+		}
 		rc = qeth_do_send_packet(card, queue, new_skb, hdr, hd_len,
 					 hd_len, elements);
 	} else
 		rc = qeth_do_send_packet_fast(queue, new_skb, hdr, data_offset,
 					      hd_len);
 
+out:
 	if (!rc) {
-		card->stats.tx_packets++;
-		card->stats.tx_bytes += tx_bytes;
 		if (new_skb != skb)
 			dev_kfree_skb_any(skb);
 		if (card->options.performance_stats) {
@@ -2330,30 +2301,67 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 				card->perf_stats.sg_frags_sent += nr_frags + 1;
 			}
 		}
-		rc = NETDEV_TX_OK;
 	} else {
+		if (new_skb != skb)
+			dev_kfree_skb_any(new_skb);
 		if (data_offset >= 0)
 			kmem_cache_free(qeth_core_header_cache, hdr);
+	}
+	return rc;
+}
 
-		if (rc == -EBUSY) {
-			if (new_skb != skb)
-				dev_kfree_skb_any(new_skb);
-			return NETDEV_TX_BUSY;
-		} else
+static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
+					   struct net_device *dev)
+{
+	int cast_type = qeth_l3_get_cast_type(skb);
+	struct qeth_card *card = dev->ml_priv;
+	int ipv = qeth_get_ip_version(skb);
+	struct qeth_qdio_out_q *queue;
+	int tx_bytes = skb->len;
+	int rc;
+
+	if (IS_IQD(card)) {
+		if (card->options.sniffer)
+			goto tx_drop;
+		if ((card->options.cq != QETH_CQ_ENABLED && !ipv) ||
+		    (card->options.cq == QETH_CQ_ENABLED &&
+		     skb->protocol != htons(ETH_P_AF_IUCV)))
 			goto tx_drop;
 	}
 
-	netif_wake_queue(dev);
-	if (card->options.performance_stats)
-		card->perf_stats.outbound_time += qeth_get_micros() -
-			card->perf_stats.outbound_start_time;
-	return rc;
+	if (card->state != CARD_STATE_UP || !card->lan_online) {
+		card->stats.tx_carrier_errors++;
+		goto tx_drop;
+	}
+
+	if (cast_type == RTN_BROADCAST && !card->info.broadcast_capable)
+		goto tx_drop;
+
+	queue = qeth_get_tx_queue(card, skb, ipv, cast_type);
+
+	if (card->options.performance_stats) {
+		card->perf_stats.outbound_cnt++;
+		card->perf_stats.outbound_start_time = qeth_get_micros();
+	}
+	netif_stop_queue(dev);
+
+	rc = qeth_l3_xmit(card, skb, queue, ipv, cast_type);
+
+	if (!rc) {
+		card->stats.tx_packets++;
+		card->stats.tx_bytes += tx_bytes;
+		if (card->options.performance_stats)
+			card->perf_stats.outbound_time += qeth_get_micros() -
+				card->perf_stats.outbound_start_time;
+		netif_wake_queue(dev);
+		return NETDEV_TX_OK;
+	} else if (rc == -EBUSY) {
+		return NETDEV_TX_BUSY;
+	} /* else fall through */
 
 tx_drop:
 	card->stats.tx_dropped++;
 	card->stats.tx_errors++;
-	if ((new_skb != skb) && new_skb)
-		dev_kfree_skb_any(new_skb);
 	dev_kfree_skb_any(skb);
 	netif_wake_queue(dev);
 	return NETDEV_TX_OK;
-- 
2.16.4

[PATCH net-next 09/10] s390/qeth: speed-up L3 IQD xmit

[Julian Wiedmann]

This implements a new xmit path for L3 HiperSockets, which carves the
HW header from skb headroom instead of allocating it from the hdr cache.
It also adds NETIF_F_SG support.

The delta in qeth_l3_xmit() is all just removal of IQD-specific code and
some minor consolidation.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_l3_main.c | 211 ++++++++++++++++++++++++----------------
 1 file changed, 128 insertions(+), 83 deletions(-)

diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index b5a54cc07017..0863ffa9ab59 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2000,19 +2000,18 @@ static int qeth_l3_get_cast_type(struct sk_buff *skb)
 	return RTN_UNICAST;
 }
 
-static void qeth_l3_fill_af_iucv_hdr(struct qeth_card *card,
-		struct qeth_hdr *hdr, struct sk_buff *skb)
+static void qeth_l3_fill_af_iucv_hdr(struct qeth_hdr *hdr, struct sk_buff *skb,
+				     unsigned int data_len)
 {
 	char daddr[16];
 	struct af_iucv_trans_hdr *iucv_hdr;
 
 	memset(hdr, 0, sizeof(struct qeth_hdr));
 	hdr->hdr.l3.id = QETH_HEADER_TYPE_LAYER3;
-	hdr->hdr.l3.ext_flags = 0;
-	hdr->hdr.l3.length = skb->len - ETH_HLEN;
+	hdr->hdr.l3.length = data_len;
 	hdr->hdr.l3.flags = QETH_HDR_IPV6 | QETH_CAST_UNICAST;
 
-	iucv_hdr = (struct af_iucv_trans_hdr *) (skb->data + ETH_HLEN);
+	iucv_hdr = (struct af_iucv_trans_hdr *)(skb_mac_header(skb) + ETH_HLEN);
 	memset(daddr, 0, sizeof(daddr));
 	daddr[0] = 0xfe;
 	daddr[1] = 0x80;
@@ -2156,63 +2155,122 @@ static int qeth_l3_get_elements_no_tso(struct qeth_card *card,
 	return elements;
 }
 
+static int qeth_l3_xmit_offload(struct qeth_card *card, struct sk_buff *skb,
+				struct qeth_qdio_out_q *queue, int ipv,
+				int cast_type)
+{
+	const unsigned int hw_hdr_len = sizeof(struct qeth_hdr);
+	unsigned int frame_len, nr_frags;
+	unsigned char eth_hdr[ETH_HLEN];
+	unsigned int hdr_elements = 0;
+	struct qeth_hdr *hdr = NULL;
+	unsigned int hd_len = 0;
+	int push_len, rc;
+
+	/* compress skb to fit into one IO buffer: */
+	if (!qeth_get_elements_no(card, skb, 0, 0)) {
+		rc = skb_linearize(skb);
+
+		if (card->options.performance_stats) {
+			if (rc)
+				card->perf_stats.tx_linfail++;
+			else
+				card->perf_stats.tx_lin++;
+		}
+		if (rc)
+			return rc;
+	}
+
+	/* re-use the L2 header area for the HW header: */
+	rc = skb_cow_head(skb, hw_hdr_len - ETH_HLEN);
+	if (rc)
+		return rc;
+	skb_copy_from_linear_data(skb, eth_hdr, ETH_HLEN);
+	skb_pull(skb, ETH_HLEN);
+	frame_len = skb->len;
+	nr_frags = skb_shinfo(skb)->nr_frags;
+
+	push_len = qeth_push_hdr(skb, &hdr, hw_hdr_len);
+	if (push_len < 0)
+		return push_len;
+	if (!push_len) {
+		/* hdr was added discontiguous from skb->data */
+		hd_len = hw_hdr_len;
+		hdr_elements = 1;
+	}
+
+	if (!qeth_get_elements_no(card, skb, hdr_elements, 0)) {
+		rc = -E2BIG;
+		goto out;
+	}
+
+	if (skb->protocol == htons(ETH_P_AF_IUCV))
+		qeth_l3_fill_af_iucv_hdr(hdr, skb, frame_len);
+	else
+		qeth_l3_fill_header(card, hdr, skb, ipv, cast_type, frame_len);
+
+	rc = qeth_do_send_packet_fast(queue, skb, hdr, 0, hd_len);
+out:
+	if (!rc) {
+		if (card->options.performance_stats && nr_frags) {
+			card->perf_stats.sg_skbs_sent++;
+			/* nr_frags + skb->data */
+			card->perf_stats.sg_frags_sent += nr_frags + 1;
+		}
+	} else {
+		if (!push_len)
+			kmem_cache_free(qeth_core_header_cache, hdr);
+		if (rc == -EBUSY) {
+			/* roll back to ETH header */
+			skb_pull(skb, push_len);
+			skb_push(skb, ETH_HLEN);
+			skb_copy_to_linear_data(skb, eth_hdr, ETH_HLEN);
+		}
+	}
+	return rc;
+}
+
 static int qeth_l3_xmit(struct qeth_card *card, struct sk_buff *skb,
 			struct qeth_qdio_out_q *queue, int ipv, int cast_type)
 {
-	int rc;
+	unsigned int hd_len, nr_frags;
+	int elements, len, rc;
 	__be16 *tag;
 	struct qeth_hdr *hdr = NULL;
 	int hdr_elements = 0;
-	int elements;
 	struct sk_buff *new_skb = NULL;
 	int tx_bytes = skb->len;
-	unsigned int hd_len = 0;
 	bool use_tso;
-	int data_offset = -1;
-	unsigned int nr_frags;
 
 	/* Ignore segment size from skb_is_gso(), 1 page is always used. */
 	use_tso = skb_is_gso(skb) &&
 		  (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4);
 
-	if (card->info.type == QETH_CARD_TYPE_IQD) {
-		new_skb = skb;
-		data_offset = ETH_HLEN;
-		hd_len = sizeof(*hdr);
-		hdr = kmem_cache_alloc(qeth_core_header_cache, GFP_ATOMIC);
-		if (!hdr)
-			return -ENOMEM;
-		hdr_elements++;
-	} else {
-		/* create a clone with writeable headroom */
-		new_skb = skb_realloc_headroom(skb, sizeof(struct qeth_hdr_tso)
-					+ VLAN_HLEN);
-		if (!new_skb)
-			return -ENOMEM;
-
-		if (ipv == 4) {
-			skb_pull(new_skb, ETH_HLEN);
-		}
+	/* create a clone with writeable headroom */
+	new_skb = skb_realloc_headroom(skb, sizeof(struct qeth_hdr_tso) +
+					    VLAN_HLEN);
+	if (!new_skb)
+		return -ENOMEM;
 
-		if (ipv != 4 && skb_vlan_tag_present(new_skb)) {
-			skb_push(new_skb, VLAN_HLEN);
-			skb_copy_to_linear_data(new_skb, new_skb->data + 4, 4);
-			skb_copy_to_linear_data_offset(new_skb, 4,
-				new_skb->data + 8, 4);
-			skb_copy_to_linear_data_offset(new_skb, 8,
-				new_skb->data + 12, 4);
-			tag = (__be16 *)(new_skb->data + 12);
-			*tag = cpu_to_be16(ETH_P_8021Q);
-			*(tag + 1) = cpu_to_be16(skb_vlan_tag_get(new_skb));
-		}
+	if (ipv == 4) {
+		skb_pull(new_skb, ETH_HLEN);
+	} else if (skb_vlan_tag_present(new_skb)) {
+		skb_push(new_skb, VLAN_HLEN);
+		skb_copy_to_linear_data(new_skb, new_skb->data + 4, 4);
+		skb_copy_to_linear_data_offset(new_skb, 4,
+					       new_skb->data + 8, 4);
+		skb_copy_to_linear_data_offset(new_skb, 8,
+					       new_skb->data + 12, 4);
+		tag = (__be16 *)(new_skb->data + 12);
+		*tag = cpu_to_be16(ETH_P_8021Q);
+		*(tag + 1) = cpu_to_be16(skb_vlan_tag_get(new_skb));
 	}
 
 	/* fix hardware limitation: as long as we do not have sbal
 	 * chaining we can not send long frag lists
 	 */
-	if ((card->info.type != QETH_CARD_TYPE_IQD) &&
-	    ((use_tso && !qeth_l3_get_elements_no_tso(card, new_skb, 1)) ||
-	     (!use_tso && !qeth_get_elements_no(card, new_skb, 0, 0)))) {
+	if ((use_tso && !qeth_l3_get_elements_no_tso(card, new_skb, 1)) ||
+	    (!use_tso && !qeth_get_elements_no(card, new_skb, 0, 0))) {
 		rc = skb_linearize(new_skb);
 
 		if (card->options.performance_stats) {
@@ -2234,20 +2292,9 @@ static int qeth_l3_xmit(struct qeth_card *card, struct sk_buff *skb,
 		qeth_tso_fill_header(card, hdr, new_skb);
 		hdr_elements++;
 	} else {
-		if (data_offset < 0) {
-			hdr = skb_push(new_skb, sizeof(struct qeth_hdr));
-			qeth_l3_fill_header(card, hdr, new_skb, ipv, cast_type,
-					    new_skb->len -
-					    sizeof(struct qeth_hdr));
-		} else {
-			if (be16_to_cpu(new_skb->protocol) == ETH_P_AF_IUCV)
-				qeth_l3_fill_af_iucv_hdr(card, hdr, new_skb);
-			else {
-				qeth_l3_fill_header(card, hdr, new_skb, ipv,
-						    cast_type,
-						    new_skb->len - data_offset);
-			}
-		}
+		hdr = skb_push(new_skb, sizeof(struct qeth_hdr));
+		qeth_l3_fill_header(card, hdr, new_skb, ipv, cast_type,
+				    new_skb->len - sizeof(struct qeth_hdr));
 
 		if (new_skb->ip_summed == CHECKSUM_PARTIAL) {
 			qeth_tx_csum(new_skb, &hdr->hdr.l3.ext_flags, ipv);
@@ -2258,34 +2305,28 @@ static int qeth_l3_xmit(struct qeth_card *card, struct sk_buff *skb,
 
 	elements = use_tso ?
 		   qeth_l3_get_elements_no_tso(card, new_skb, hdr_elements) :
-		   qeth_get_elements_no(card, new_skb, hdr_elements,
-					(data_offset > 0) ? data_offset : 0);
+		   qeth_get_elements_no(card, new_skb, hdr_elements, 0);
 	if (!elements) {
 		rc = -E2BIG;
 		goto out;
 	}
 	elements += hdr_elements;
 
-	if (card->info.type != QETH_CARD_TYPE_IQD) {
-		int len;
-		if (use_tso) {
-			hd_len = sizeof(struct qeth_hdr_tso) +
-				 ip_hdrlen(new_skb) + tcp_hdrlen(new_skb);
-			len = hd_len;
-		} else {
-			len = sizeof(struct qeth_hdr_layer3);
-		}
-
-		if (qeth_hdr_chk_and_bounce(new_skb, &hdr, len)) {
-			rc = -EINVAL;
-			goto out;
-		}
-		rc = qeth_do_send_packet(card, queue, new_skb, hdr, hd_len,
-					 hd_len, elements);
-	} else
-		rc = qeth_do_send_packet_fast(queue, new_skb, hdr, data_offset,
-					      hd_len);
+	if (use_tso) {
+		hd_len = sizeof(struct qeth_hdr_tso) +
+			 ip_hdrlen(new_skb) + tcp_hdrlen(new_skb);
+		len = hd_len;
+	} else {
+		hd_len = 0;
+		len = sizeof(struct qeth_hdr_layer3);
+	}
 
+	if (qeth_hdr_chk_and_bounce(new_skb, &hdr, len)) {
+		rc = -EINVAL;
+		goto out;
+	}
+	rc = qeth_do_send_packet(card, queue, new_skb, hdr, hd_len, hd_len,
+				 elements);
 out:
 	if (!rc) {
 		if (new_skb != skb)
@@ -2304,8 +2345,6 @@ static int qeth_l3_xmit(struct qeth_card *card, struct sk_buff *skb,
 	} else {
 		if (new_skb != skb)
 			dev_kfree_skb_any(new_skb);
-		if (data_offset >= 0)
-			kmem_cache_free(qeth_core_header_cache, hdr);
 	}
 	return rc;
 }
@@ -2345,7 +2384,10 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 	}
 	netif_stop_queue(dev);
 
-	rc = qeth_l3_xmit(card, skb, queue, ipv, cast_type);
+	if (IS_IQD(card))
+		rc = qeth_l3_xmit_offload(card, skb, queue, ipv, cast_type);
+	else
+		rc = qeth_l3_xmit(card, skb, queue, ipv, cast_type);
 
 	if (!rc) {
 		card->stats.tx_packets++;
@@ -2503,9 +2545,6 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 		if (!(card->info.unique_id & UNIQUE_ID_NOT_BY_CARD))
 			card->dev->dev_id = card->info.unique_id & 0xffff;
 
-		card->dev->hw_features |= NETIF_F_SG;
-		card->dev->vlan_features |= NETIF_F_SG;
-
 		if (!card->info.guestlan) {
 			card->dev->features |= NETIF_F_SG;
 			card->dev->hw_features |= NETIF_F_TSO |
@@ -2524,7 +2563,10 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 		if (!card->dev)
 			return -ENODEV;
 		card->dev->flags |= IFF_NOARP;
+		card->dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 		card->dev->netdev_ops = &qeth_l3_netdev_ops;
+		card->dev->needed_headroom = sizeof(struct qeth_hdr) - ETH_HLEN;
+
 		rc = qeth_l3_iqd_read_initial_mac(card);
 		if (rc)
 			return rc;
@@ -2543,6 +2585,9 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 	card->dev->features |=	NETIF_F_HW_VLAN_CTAG_TX |
 				NETIF_F_HW_VLAN_CTAG_RX |
 				NETIF_F_HW_VLAN_CTAG_FILTER;
+	card->dev->hw_features |= NETIF_F_SG;
+	card->dev->vlan_features |= NETIF_F_SG;
+
 	netif_keep_dst(card->dev);
 	if (card->dev->hw_features & NETIF_F_TSO)
 		netif_set_gso_max_size(card->dev,
-- 
2.16.4

[PATCH net-next 10/10] s390/qeth: speed-up IPv4 OSA xmit

[Julian Wiedmann]

Move the xmit of offload-eligible (ie IPv4) traffic on OSA over to the
new, copy-free path.
As with L2, we'll need to preserve the skb_orphan() behaviour of the
old code path until TX completion is sufficiently fast.

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
---
 drivers/s390/net/qeth_l3_main.c | 33 +++++++++++++++++++++------------
 1 file changed, 21 insertions(+), 12 deletions(-)

diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 0863ffa9ab59..062f62b49294 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2050,6 +2050,12 @@ static void qeth_l3_fill_header(struct qeth_card *card, struct qeth_hdr *hdr,
 		hdr->hdr.l3.vlan_id = skb_vlan_tag_get(skb);
 	}
 
+	if (!skb_is_gso(skb) && skb->ip_summed == CHECKSUM_PARTIAL) {
+		qeth_tx_csum(skb, &hdr->hdr.l3.ext_flags, ipv);
+		if (card->options.performance_stats)
+			card->perf_stats.tx_csum++;
+	}
+
 	/* OSA only: */
 	if (!ipv) {
 		hdr->hdr.l3.flags = QETH_HDR_PASSTHRU;
@@ -2164,8 +2170,8 @@ static int qeth_l3_xmit_offload(struct qeth_card *card, struct sk_buff *skb,
 	unsigned char eth_hdr[ETH_HLEN];
 	unsigned int hdr_elements = 0;
 	struct qeth_hdr *hdr = NULL;
+	int elements, push_len, rc;
 	unsigned int hd_len = 0;
-	int push_len, rc;
 
 	/* compress skb to fit into one IO buffer: */
 	if (!qeth_get_elements_no(card, skb, 0, 0)) {
@@ -2199,17 +2205,26 @@ static int qeth_l3_xmit_offload(struct qeth_card *card, struct sk_buff *skb,
 		hdr_elements = 1;
 	}
 
-	if (!qeth_get_elements_no(card, skb, hdr_elements, 0)) {
+	elements = qeth_get_elements_no(card, skb, hdr_elements, 0);
+	if (!elements) {
 		rc = -E2BIG;
 		goto out;
 	}
+	elements += hdr_elements;
 
 	if (skb->protocol == htons(ETH_P_AF_IUCV))
 		qeth_l3_fill_af_iucv_hdr(hdr, skb, frame_len);
 	else
 		qeth_l3_fill_header(card, hdr, skb, ipv, cast_type, frame_len);
 
-	rc = qeth_do_send_packet_fast(queue, skb, hdr, 0, hd_len);
+	if (IS_IQD(card)) {
+		rc = qeth_do_send_packet_fast(queue, skb, hdr, 0, hd_len);
+	} else {
+		/* TODO: drop skb_orphan() once TX completion is fast enough */
+		skb_orphan(skb);
+		rc = qeth_do_send_packet(card, queue, skb, hdr, 0, hd_len,
+					 elements);
+	}
 out:
 	if (!rc) {
 		if (card->options.performance_stats && nr_frags) {
@@ -2295,12 +2310,6 @@ static int qeth_l3_xmit(struct qeth_card *card, struct sk_buff *skb,
 		hdr = skb_push(new_skb, sizeof(struct qeth_hdr));
 		qeth_l3_fill_header(card, hdr, new_skb, ipv, cast_type,
 				    new_skb->len - sizeof(struct qeth_hdr));
-
-		if (new_skb->ip_summed == CHECKSUM_PARTIAL) {
-			qeth_tx_csum(new_skb, &hdr->hdr.l3.ext_flags, ipv);
-			if (card->options.performance_stats)
-				card->perf_stats.tx_csum++;
-		}
 	}
 
 	elements = use_tso ?
@@ -2384,7 +2393,7 @@ static netdev_tx_t qeth_l3_hard_start_xmit(struct sk_buff *skb,
 	}
 	netif_stop_queue(dev);
 
-	if (IS_IQD(card))
+	if (IS_IQD(card) || (!skb_is_gso(skb) && ipv == 4))
 		rc = qeth_l3_xmit_offload(card, skb, queue, ipv, cast_type);
 	else
 		rc = qeth_l3_xmit(card, skb, queue, ipv, cast_type);
@@ -2563,9 +2572,7 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 		if (!card->dev)
 			return -ENODEV;
 		card->dev->flags |= IFF_NOARP;
-		card->dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 		card->dev->netdev_ops = &qeth_l3_netdev_ops;
-		card->dev->needed_headroom = sizeof(struct qeth_hdr) - ETH_HLEN;
 
 		rc = qeth_l3_iqd_read_initial_mac(card);
 		if (rc)
@@ -2582,6 +2589,8 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 	card->dev->max_mtu = ETH_MAX_MTU;
 	card->dev->dev_port = card->info.portno;
 	card->dev->ethtool_ops = &qeth_l3_ethtool_ops;
+	card->dev->priv_flags &= ~IFF_TX_SKB_SHARING;
+	card->dev->needed_headroom = sizeof(struct qeth_hdr) - ETH_HLEN;
 	card->dev->features |=	NETIF_F_HW_VLAN_CTAG_TX |
 				NETIF_F_HW_VLAN_CTAG_RX |
 				NETIF_F_HW_VLAN_CTAG_FILTER;
-- 
2.16.4

Re: [PATCH net-next 00/10] s390/qeth: updates 2018-07-11

[David Miller]

From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Wed, 11 Jul 2018 17:42:37 +0200

> please apply this first batch of qeth patches for net-next. It brings the
> usual cleanups, and some performance improvements to the transmit paths.

Series applied, thank you.