From: Sagi Grimberg <sagi@grimberg.me>
Date: Thu, 18 Jun 2020 17:30:24 -0700
Subject: [PATCH] nvme-tcp: optimize network stack with setting msg flags
Git-commit: 122e5b9f3d370ae11e1502d14ff5c7ea9b144a76
Patch-mainline: v5.9-rc1
References: bsc#1179519
according to batch size
If we have a long list of request to send, signal the network stack
that more is coming (MSG_MORE). If we have nothing else, signal MSG_EOR.
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Tested-by: Mark Wunderlich <mark.wunderlich@intel.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Hannes Reinecke <hare@suse.com>
---
drivers/nvme/host/tcp.c | 20 +++++++++++++++++---
1 file changed, 17 insertions(+), 3 deletions(-)
diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 2d3962c164a4..b2e73e19ef01 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -79,6 +79,7 @@ struct nvme_tcp_queue {
struct mutex send_mutex;
struct llist_head req_list;
struct list_head send_list;
+ bool more_requests;
/* recv state */
void *pdu;
@@ -277,7 +278,9 @@ static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req,
*/
if (queue->io_cpu == smp_processor_id() &&
sync && empty && mutex_trylock(&queue->send_mutex)) {
+ queue->more_requests = !last;
nvme_tcp_try_send(queue);
+ queue->more_requests = false;
mutex_unlock(&queue->send_mutex);
} else if (last) {
queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work);
@@ -877,6 +880,12 @@ static void nvme_tcp_state_change(struct sock *sk)
read_unlock(&sk->sk_callback_lock);
}
+static inline bool nvme_tcp_queue_more(struct nvme_tcp_queue *queue)
+{
+ return !list_empty(&queue->send_list) ||
+ !llist_empty(&queue->req_list) || queue->more_requests;
+}
+
static inline void nvme_tcp_done_send_req(struct nvme_tcp_queue *queue)
{
queue->request = NULL;
@@ -898,7 +907,7 @@ static int nvme_tcp_try_send_data(struct nvme_tcp_request *req)
bool last = nvme_tcp_pdu_last_send(req, len);
int ret, flags = MSG_DONTWAIT;
- if (last && !queue->data_digest)
+ if (last && !queue->data_digest && !nvme_tcp_queue_more(queue))
flags |= MSG_EOR;
else
flags |= MSG_MORE | MSG_SENDPAGE_NOTLAST;
@@ -945,7 +954,7 @@ static int nvme_tcp_try_send_cmd_pdu(struct nvme_tcp_request *req)
int flags = MSG_DONTWAIT;
int ret;
- if (inline_data)
+ if (inline_data || nvme_tcp_queue_more(queue))
flags |= MSG_MORE | MSG_SENDPAGE_NOTLAST;
else
flags |= MSG_EOR;
@@ -1010,12 +1019,17 @@ static int nvme_tcp_try_send_ddgst(struct nvme_tcp_request *req)
{
struct nvme_tcp_queue *queue = req->queue;
int ret;
- struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_EOR };
+ struct msghdr msg = { .msg_flags = MSG_DONTWAIT };
struct kvec iov = {
.iov_base = &req->ddgst + req->offset,
.iov_len = NVME_TCP_DIGEST_LENGTH - req->offset
};
+ if (nvme_tcp_queue_more(queue))
+ msg.msg_flags |= MSG_MORE;
+ else
+ msg.msg_flags |= MSG_EOR;
+
ret = kernel_sendmsg(queue->sock, &msg, &iov, 1, iov.iov_len);
if (unlikely(ret <= 0))
return ret;
--
2.16.4