[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20240719091959.17163-3-jacky_gam_2001@163.com>
Date: Fri, 19 Jul 2024 17:19:59 +0800
From: Ping Gan <jacky_gam_2001@....com>
To: hare@...e.de,
hch@....de,
sagi@...mberg.me,
kch@...dia.com,
linux-nvme@...ts.infradead.org,
linux-kernel@...r.kernel.org
Cc: ping.gan@...l.com,
Ping Gan <jacky_gam_2001@....com>
Subject: [PATCH v3 2/2] nvmet-rdma: use unbound_wq for nvmet-rdma by default
To use unbounded workqueue to handle RDMA's IO of CQ by default, which
was discussed in below session.
https://lore.kernel.org/lkml/20240719084953.8050-1-jacky_gam_2001@163.com/
Signed-off-by: Ping Gan <jacky_gam_2001@....com>
---
drivers/nvme/target/rdma.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
index 1eff8ca6a5f1..6a54217e3bc5 100644
--- a/drivers/nvme/target/rdma.c
+++ b/drivers/nvme/target/rdma.c
@@ -1260,7 +1260,7 @@ static int nvmet_rdma_create_queue_ib(struct nvmet_rdma_queue *queue)
nr_cqe = queue->recv_queue_size + 2 * queue->send_queue_size;
queue->cq = ib_cq_pool_get(ndev->device, nr_cqe + 1,
- queue->comp_vector, IB_POLL_WORKQUEUE);
+ queue->comp_vector, IB_POLL_UNBOUND_WORKQUEUE);
if (IS_ERR(queue->cq)) {
ret = PTR_ERR(queue->cq);
pr_err("failed to create CQ cqe= %d ret= %d\n",
--
2.26.2
Powered by blists - more mailing lists