netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: davem@davemloft.net
Cc: netdev@vger.kernel.org, Stephen Hemminger <sthemmin@microsoft.com>
Subject: [PATCH net-next 09/15] netvsc: optimize netvsc_send_pkt
Date: Wed,  3 May 2017 16:01:11 -0700	[thread overview]
Message-ID: <20170503230117.20070-10-sthemmin@microsoft.com> (raw)
In-Reply-To: <20170503230117.20070-1-sthemmin@microsoft.com>

Hand optimize netvsc_send_pkt by adding likely/unlikely.
Also don't print pointer in warning message, instead dump info.

Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
---
 drivers/net/hyperv/netvsc.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c
index 59ca5fd6797d..d9bd1a2db4db 100644
--- a/drivers/net/hyperv/netvsc.c
+++ b/drivers/net/hyperv/netvsc.c
@@ -751,9 +751,9 @@ static u32 netvsc_copy_to_send_buf(struct netvsc_device *net_device,
 	return msg_size;
 }
 
-static inline int netvsc_send_pkt(
+static int netvsc_send_pkt(
 	struct hv_device *device,
-	struct hv_netvsc_packet *packet,
+	const struct hv_netvsc_packet *packet,
 	struct netvsc_device *net_device,
 	struct hv_page_buffer **pb,
 	struct sk_buff *skb)
@@ -766,7 +766,6 @@ static inline int netvsc_send_pkt(
 	struct netdev_queue *txq = netdev_get_tx_queue(ndev, packet->q_idx);
 	u64 req_id;
 	int ret;
-	struct hv_page_buffer *pgbuf;
 	u32 ring_avail = hv_ringbuf_avail_percent(out_channel);
 
 	nvmsg.hdr.msg_type = NVSP_MSG1_TYPE_SEND_RNDIS_PKT;
@@ -788,10 +787,12 @@ static inline int netvsc_send_pkt(
 
 	req_id = (ulong)skb;
 
-	if (out_channel->rescind)
+	if (unlikely(out_channel->rescind))
 		return -ENODEV;
 
 	if (packet->page_buf_cnt) {
+		struct hv_page_buffer *pgbuf;
+
 		pgbuf = packet->cp_partial ? (*pb) +
 			packet->rmsg_pgcnt : (*pb);
 		ret = vmbus_sendpacket_pagebuffer_ctl(out_channel,
@@ -809,20 +810,23 @@ static inline int netvsc_send_pkt(
 					   VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
 	}
 
-	if (ret == 0) {
-		atomic_inc_return(&nvchan->queue_sends);
+	if (likely(ret == 0)) {
+		atomic_inc(&nvchan->queue_sends);
 
 		if (ring_avail < RING_AVAIL_PERCENT_LOWATER)
 			netif_tx_stop_queue(txq);
-	} else if (ret == -EAGAIN) {
+	} else if (likely(ret == -EAGAIN)) {
 		netif_tx_stop_queue(txq);
 		if (atomic_read(&nvchan->queue_sends) < 1) {
 			netif_tx_wake_queue(txq);
 			ret = -ENOSPC;
 		}
 	} else {
-		netdev_err(ndev, "Unable to send packet %p ret %d\n",
-			   packet, ret);
+		if (net_ratelimit())
+			netdev_warn(ndev,
+				    "Unable to send packet qid %u index %d len %u (%d)\n",
+				    packet->q_idx, packet->send_buf_index,
+				    packet->total_data_buflen, ret);
 	}
 
 	return ret;
-- 
2.11.0

  parent reply	other threads:[~2017-05-03 23:01 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-03 23:01 [PATCH net-next 00/15] netvsc: misc patches Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 01/15] vmbus: simplify hv_ringbuffer_read Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 02/15] vmbus: fix unnecessary signal events as result of NAPI Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 03/15] netvsc: make sure napi enabled before vmbus_open Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 04/15] netvsc: don't reacquire rtnl on device removal Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 05/15] netvsc: optimize avail percent calculation Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 06/15] netvsc: prefetch the first incoming ring element Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 07/15] netvsc: convert ring_size to unsigned Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 08/15] netvsc: allow overriding send/recv buffer size Stephen Hemminger
2017-05-03 23:01 ` Stephen Hemminger [this message]
2017-05-03 23:01 ` [PATCH net-next 10/15] netvsc: replace modulus with mask for alignment Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 11/15] netvsc: reduce unnecessary memset Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 12/15] netvsc: size receive completion ring based on receive area Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 13/15] netvsc: convert open count from atomic to refcount Stephen Hemminger
2017-05-03 23:01 ` [PATCH net-next 14/15] netvsc: optimize receive completions Stephen Hemminger
2017-05-03 23:01 ` [PATCH 15/15] netvsc: use vzalloc_node for receive completion data Stephen Hemminger
2017-05-03 23:35 ` [PATCH net-next 00/15] netvsc: misc patches David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170503230117.20070-10-sthemmin@microsoft.com \
    --to=stephen@networkplumber.org \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.org \
    --cc=sthemmin@microsoft.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).