Skip to contentSkip to conversation

nvmet-rdma: +1 to *queue_size from hsqsize/hrqsize

RFC PATCH New Jay Freyensee 2 participants 3 messages Aug 6, 2016 – Aug 8, 2016

Subsystems (2)
NVMe NVMe target
lore
Share & tools
lore

Jay Freyensee

author
#0 lore raw

RFC PATCH 1/1 nvmet-rdma: +1 to *queue_size from hsqsize/hrqsize

Now that the host will be sending hsqsize/hrqsize 0-based values, the target may need to be adjusted as well.

I am not 100% clear this is the only spot on the target. Lightly testing host+target I haven't noticed any difference with this change but my target went down so
I cannot do anything else for now so just make this an RFC...

Signed-off-by Jay Freyensee <james_p_freyensee at linux.intel.com>
---
 drivers/nvme/target/rdma.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/nvme/target/rdma.c +4 -4
index e06d504..e710c46 100644
@@ -1004,11 +1004,11 @@ nvmet_rdma_parse_cm_connect_req(struct rdma_conn_param *conn,
 	queue->host_qid = le16_to_cpu(req->qid);
 
 	/*
-	 * req->hsqsize corresponds to our recv queue size
-	 * req->hrqsize corresponds to our send queue size
+	 * req->hsqsize corresponds to our recv queue size, plus 1
+	 * req->hrqsize corresponds to our send queue size, plus 1
 	 */
-	queue->recv_queue_size = le16_to_cpu(req->hsqsize);
-	queue->send_queue_size = le16_to_cpu(req->hrqsize);
+	queue->recv_queue_size = le16_to_cpu(req->hsqsize)+1;
+	queue->send_queue_size = le16_to_cpu(req->hrqsize)+1;
 
 	if (!queue->host_qid && queue->recv_queue_size > NVMF_AQ_DEPTH)
 		return NVME_RDMA_CM_INVALID_HSQSIZE;
Signature · 1 line
2.7.4
#1 lore raw

RFC PATCH 1/1 nvmet-rdma: +1 to *queue_size from hsqsize/hrqsize

Hi Jay,

drivers/nvme/target/rdma.c nvmet_rdma_parse_cm_connect_req(struct rdma_conn_param *conn, · line 1011

Quoted 4 lines · Jay Freyensee
drivers/nvme/target/rdma.c +2 -2
-	queue->recv_queue_size = le16_to_cpu(req->hsqsize);
-	queue->send_queue_size = le16_to_cpu(req->hrqsize);
+	queue->recv_queue_size = le16_to_cpu(req->hsqsize)+1;
+	queue->send_queue_size = le16_to_cpu(req->hrqsize)+1;

I think the +1 needs to be inside the le16_to_cpu() in both cases.

-- Daniel

author
#2 lore raw

RFC PATCH 1/1 nvmet-rdma: +1 to *queue_size from hsqsize/hrqsize

Quoted 10 lines · Daniel

Hi Jay,

drivers/nvme/target/rdma.c nvmet_rdma_parse_cm_connect_req(struct rdma_conn_param *conn, · line 1011

Quoted 5 lines · Jay Freyensee
drivers/nvme/target/rdma.c +2 -2
-	queue->recv_queue_size = le16_to_cpu(req->hsqsize);
-	queue->send_queue_size = le16_to_cpu(req->hrqsize);
+	queue->recv_queue_size = le16_to_cpu(req->hsqsize)+1;
+	queue->send_queue_size = le16_to_cpu(req->hrqsize)+1;

I think the +1 needs to be inside the le16_to_cpu() in both cases.

Thanks.

This patch will be dependent on the "nvme-rdma: sqsize/hsqsize/hrqsize is 0-based val" patch, which reading the spec again I still think it's 0-based.

Quoted 2 lines

-- Daniel