From: Julian Wiedmann <jwi@linux.ibm.com>
Subject: s390/qeth: don't take queue lock in send_packet_fast()
Patch-mainline: v4.15-rc1
Git-commit: 7c2e9ba373264b29a35d6df06cc5fb125d189b90
References: FATE#326350, LTC#169511, bsc#1113509

Summary:     qeth: performance improvements
Description: This adds recent functional and performance improvements for the
             qeth network driver.
             Primarily this brings Scatter-Gather support for HiperSockets,
             reduced CPU consumption in the L3 IPv4 transmit path for OSA,
             improved Promiscuous Mode performance due to IFF_UNICAST_FLT,
             support for Scatter-Gather on z/VM virtual NICs, and
             support for delayed GRO flushing.

             For sanity & stability reasons, this effectively constitutes a
             backport of the qeth device driver from 4.19 mainline.
             

Upstream-Description:

             s390/qeth: don't take queue lock in send_packet_fast()

             Locking the output queue prior to TX is needed on OSA devices,
             to synchronize against a packing flush from the TX completion code
             (via qeth_check_outbound_queue()).
             But send_packet_fast() is only used for IQDs, which don't do packing.
             So remove the locking, and apply some easy cleanups.

             Signed-off-by: Julian Wiedmann <jwi@linux.vnet.ibm.com>
             Reviewed-by: Ursula Braun <ubraun@linux.vnet.ibm.com>
             Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Acked-by: Petr Tesarik <ptesarik@suse.com>
---
 drivers/s390/net/qeth_core.h      |    3 +--
 drivers/s390/net/qeth_core_main.c |   22 +++++-----------------
 drivers/s390/net/qeth_l2_main.c   |    2 +-
 drivers/s390/net/qeth_l3_main.c   |    4 ++--
 4 files changed, 9 insertions(+), 22 deletions(-)

--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -978,8 +978,7 @@ int qeth_get_priority_queue(struct qeth_
 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 *);
-int qeth_do_send_packet_fast(struct qeth_card *card,
-			     struct qeth_qdio_out_q *queue, struct sk_buff *skb,
+int qeth_do_send_packet_fast(struct qeth_qdio_out_q *queue, struct sk_buff *skb,
 			     struct qeth_hdr *hdr, unsigned int offset,
 			     unsigned int hd_len);
 int qeth_do_send_packet(struct qeth_card *card, struct qeth_qdio_out_q *queue,
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -4063,35 +4063,23 @@ static int qeth_fill_buffer(struct qeth_
 	return flush_cnt;
 }
 
-int qeth_do_send_packet_fast(struct qeth_card *card,
-			     struct qeth_qdio_out_q *queue, struct sk_buff *skb,
+int qeth_do_send_packet_fast(struct qeth_qdio_out_q *queue, struct sk_buff *skb,
 			     struct qeth_hdr *hdr, unsigned int offset,
 			     unsigned int hd_len)
 {
-	struct qeth_qdio_out_buffer *buffer;
-	int index;
+	int index = queue->next_buf_to_fill;
+	struct qeth_qdio_out_buffer *buffer = queue->bufs[index];
 
-	/* spin until we get the queue ... */
-	while (atomic_cmpxchg(&queue->state, QETH_OUT_Q_UNLOCKED,
-			      QETH_OUT_Q_LOCKED) != QETH_OUT_Q_UNLOCKED);
-	/* ... now we've got the queue */
-	index = queue->next_buf_to_fill;
-	buffer = queue->bufs[queue->next_buf_to_fill];
 	/*
 	 * check if buffer is empty to make sure that we do not 'overtake'
 	 * ourselves and try to fill a buffer that is already primed
 	 */
 	if (atomic_read(&buffer->state) != QETH_QDIO_BUF_EMPTY)
-		goto out;
-	queue->next_buf_to_fill = (queue->next_buf_to_fill + 1) %
-					  QDIO_MAX_BUFFERS_PER_Q;
-	atomic_set(&queue->state, QETH_OUT_Q_UNLOCKED);
+		return -EBUSY;
+	queue->next_buf_to_fill = (index + 1) % QDIO_MAX_BUFFERS_PER_Q;
 	qeth_fill_buffer(queue, buffer, skb, hdr, offset, hd_len);
 	qeth_flush_buffers(queue, index, 1);
 	return 0;
-out:
-	atomic_set(&queue->state, QETH_OUT_Q_UNLOCKED);
-	return -EBUSY;
 }
 EXPORT_SYMBOL_GPL(qeth_do_send_packet_fast);
 
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -734,7 +734,7 @@ static int qeth_l2_xmit_iqd(struct qeth_
 		rc = -E2BIG;
 		goto out;
 	}
-	rc = qeth_do_send_packet_fast(card, queue, skb, hdr, data_offset,
+	rc = qeth_do_send_packet_fast(queue, skb, hdr, data_offset,
 				      sizeof(*hdr) + data_offset);
 out:
 	if (rc)
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -2793,8 +2793,8 @@ static netdev_tx_t qeth_l3_hard_start_xm
 		rc = qeth_do_send_packet(card, queue, new_skb, hdr, hd_len,
 					 hd_len, elements);
 	} else
-		rc = qeth_do_send_packet_fast(card, queue, new_skb, hdr,
-					      data_offset, hd_len);
+		rc = qeth_do_send_packet_fast(queue, new_skb, hdr, data_offset,
+					      hd_len);
 
 	if (!rc) {
 		card->stats.tx_packets++;