nvme-tcp: try to send request in queue_rq context

Today, nvme-tcp automatically schedules a send request
to a workqueue context, which is 1 more than we'd need
in case the socket buffer is wide open.

However, because we have async send activity (as a result
of r2t, or write_space callbacks), we need to synchronize
sends from possibly multiple contexts (ideally all running
on the same cpu though).

Thus, we only try to send directly from queue_rq in cases:
1. the send_list is empty
2. we can send it synchronously (i.e. not from the RX path)
3. we run on the same cpu as the queue->io_cpu to avoid
   contention on the send operation.

Proposed-by: Mark Wunderlich <mark.wunderlich@intel.com>
Signed-off-by: Mark Wunderlich <mark.wunderlich@intel.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Sagi Grimberg 2020-05-01 14:25:45 -07:00 committed by Jens Axboe
parent 72e5d757c6
commit db5ad6b7f8

View File

@ -76,6 +76,7 @@ struct nvme_tcp_queue {
int io_cpu; int io_cpu;
spinlock_t lock; spinlock_t lock;
struct mutex send_mutex;
struct list_head send_list; struct list_head send_list;
/* recv state */ /* recv state */
@ -132,6 +133,7 @@ static DEFINE_MUTEX(nvme_tcp_ctrl_mutex);
static struct workqueue_struct *nvme_tcp_wq; static struct workqueue_struct *nvme_tcp_wq;
static struct blk_mq_ops nvme_tcp_mq_ops; static struct blk_mq_ops nvme_tcp_mq_ops;
static struct blk_mq_ops nvme_tcp_admin_mq_ops; static struct blk_mq_ops nvme_tcp_admin_mq_ops;
static int nvme_tcp_try_send(struct nvme_tcp_queue *queue);
static inline struct nvme_tcp_ctrl *to_tcp_ctrl(struct nvme_ctrl *ctrl) static inline struct nvme_tcp_ctrl *to_tcp_ctrl(struct nvme_ctrl *ctrl)
{ {
@ -258,15 +260,29 @@ static inline void nvme_tcp_advance_req(struct nvme_tcp_request *req,
} }
} }
static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req) static inline void nvme_tcp_queue_request(struct nvme_tcp_request *req,
bool sync)
{ {
struct nvme_tcp_queue *queue = req->queue; struct nvme_tcp_queue *queue = req->queue;
bool empty;
spin_lock(&queue->lock); spin_lock(&queue->lock);
empty = list_empty(&queue->send_list) && !queue->request;
list_add_tail(&req->entry, &queue->send_list); list_add_tail(&req->entry, &queue->send_list);
spin_unlock(&queue->lock); spin_unlock(&queue->lock);
queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work); /*
* if we're the first on the send_list and we can try to send
* directly, otherwise queue io_work. Also, only do that if we
* are on the same cpu, so we don't introduce contention.
*/
if (queue->io_cpu == smp_processor_id() &&
sync && empty && mutex_trylock(&queue->send_mutex)) {
nvme_tcp_try_send(queue);
mutex_unlock(&queue->send_mutex);
} else {
queue_work_on(queue->io_cpu, nvme_tcp_wq, &queue->io_work);
}
} }
static inline struct nvme_tcp_request * static inline struct nvme_tcp_request *
@ -579,7 +595,7 @@ static int nvme_tcp_handle_r2t(struct nvme_tcp_queue *queue,
req->state = NVME_TCP_SEND_H2C_PDU; req->state = NVME_TCP_SEND_H2C_PDU;
req->offset = 0; req->offset = 0;
nvme_tcp_queue_request(req); nvme_tcp_queue_request(req, false);
return 0; return 0;
} }
@ -1065,11 +1081,14 @@ static void nvme_tcp_io_work(struct work_struct *w)
bool pending = false; bool pending = false;
int result; int result;
result = nvme_tcp_try_send(queue); if (mutex_trylock(&queue->send_mutex)) {
if (result > 0) result = nvme_tcp_try_send(queue);
pending = true; mutex_unlock(&queue->send_mutex);
else if (unlikely(result < 0)) if (result > 0)
break; pending = true;
else if (unlikely(result < 0))
break;
}
result = nvme_tcp_try_recv(queue); result = nvme_tcp_try_recv(queue);
if (result > 0) if (result > 0)
@ -1321,6 +1340,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
queue->ctrl = ctrl; queue->ctrl = ctrl;
INIT_LIST_HEAD(&queue->send_list); INIT_LIST_HEAD(&queue->send_list);
spin_lock_init(&queue->lock); spin_lock_init(&queue->lock);
mutex_init(&queue->send_mutex);
INIT_WORK(&queue->io_work, nvme_tcp_io_work); INIT_WORK(&queue->io_work, nvme_tcp_io_work);
queue->queue_size = queue_size; queue->queue_size = queue_size;
@ -1545,6 +1565,7 @@ static struct blk_mq_tag_set *nvme_tcp_alloc_tagset(struct nvme_ctrl *nctrl,
set->queue_depth = NVME_AQ_MQ_TAG_DEPTH; set->queue_depth = NVME_AQ_MQ_TAG_DEPTH;
set->reserved_tags = 2; /* connect + keep-alive */ set->reserved_tags = 2; /* connect + keep-alive */
set->numa_node = NUMA_NO_NODE; set->numa_node = NUMA_NO_NODE;
set->flags = BLK_MQ_F_BLOCKING;
set->cmd_size = sizeof(struct nvme_tcp_request); set->cmd_size = sizeof(struct nvme_tcp_request);
set->driver_data = ctrl; set->driver_data = ctrl;
set->nr_hw_queues = 1; set->nr_hw_queues = 1;
@ -1556,7 +1577,7 @@ static struct blk_mq_tag_set *nvme_tcp_alloc_tagset(struct nvme_ctrl *nctrl,
set->queue_depth = nctrl->sqsize + 1; set->queue_depth = nctrl->sqsize + 1;
set->reserved_tags = 1; /* fabric connect */ set->reserved_tags = 1; /* fabric connect */
set->numa_node = NUMA_NO_NODE; set->numa_node = NUMA_NO_NODE;
set->flags = BLK_MQ_F_SHOULD_MERGE; set->flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING;
set->cmd_size = sizeof(struct nvme_tcp_request); set->cmd_size = sizeof(struct nvme_tcp_request);
set->driver_data = ctrl; set->driver_data = ctrl;
set->nr_hw_queues = nctrl->queue_count - 1; set->nr_hw_queues = nctrl->queue_count - 1;
@ -2115,7 +2136,7 @@ static void nvme_tcp_submit_async_event(struct nvme_ctrl *arg)
ctrl->async_req.curr_bio = NULL; ctrl->async_req.curr_bio = NULL;
ctrl->async_req.data_len = 0; ctrl->async_req.data_len = 0;
nvme_tcp_queue_request(&ctrl->async_req); nvme_tcp_queue_request(&ctrl->async_req, true);
} }
static enum blk_eh_timer_return static enum blk_eh_timer_return
@ -2246,7 +2267,7 @@ static blk_status_t nvme_tcp_queue_rq(struct blk_mq_hw_ctx *hctx,
blk_mq_start_request(rq); blk_mq_start_request(rq);
nvme_tcp_queue_request(req); nvme_tcp_queue_request(req, true);
return BLK_STS_OK; return BLK_STS_OK;
} }