[PATCH 2/2] nvmet-rdma: implement get_queue_size controller op
From: Max Gurtovoy <mgurtovoy@nvidia.com>
Date: 2021-09-21 19:05:43
Subsystem:
nvm express target driver, the rest · Maintainers:
Christoph Hellwig, Sagi Grimberg, Chaitanya Kulkarni, Linus Torvalds
Limit the maximal queue size for RDMA controllers. Today, the target reports a limit of 1024 and this limit isn't valid for some of the RDMA based controllers. For now, limit RDMA transport to 128 entries (the default queue depth configured for Linux NVMeoF host fabric drivers). Future general solution should use RDMA/core API to calculate this size according to device capabilities and number of WRs needed per NVMe IO request. Reported-by: Mark Ruijter <redacted> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> --- drivers/nvme/target/rdma.c | 8 ++++++++ 1 file changed, 8 insertions(+)
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
index 891174ccd44b..16342c5f31e8 100644
--- a/drivers/nvme/target/rdma.c
+++ b/drivers/nvme/target/rdma.c@@ -36,6 +36,8 @@ #define NVMET_RDMA_MAX_MDTS 8 #define NVMET_RDMA_MAX_METADATA_MDTS 5 +#define NVMET_RDMA_QUEUE_SIZE 128 + struct nvmet_rdma_srq; struct nvmet_rdma_cmd {
@@ -1975,6 +1977,11 @@ static u8 nvmet_rdma_get_mdts(const struct nvmet_ctrl *ctrl) return NVMET_RDMA_MAX_MDTS; } +static u16 nvmet_rdma_get_queue_size(const struct nvmet_ctrl *ctrl) +{ + return NVMET_RDMA_QUEUE_SIZE; +} + static const struct nvmet_fabrics_ops nvmet_rdma_ops = { .owner = THIS_MODULE, .type = NVMF_TRTYPE_RDMA,
@@ -1986,6 +1993,7 @@ static const struct nvmet_fabrics_ops nvmet_rdma_ops = { .delete_ctrl = nvmet_rdma_delete_ctrl, .disc_traddr = nvmet_rdma_disc_port_addr, .get_mdts = nvmet_rdma_get_mdts, + .get_queue_size = nvmet_rdma_get_queue_size, }; static void nvmet_rdma_remove_one(struct ib_device *ib_device, void *client_data)
--
2.18.1
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme