* [PATCH] nvmet-rdma: remove NVMET_RDMA_REQ_INVALIDATE_RKEY flag
@ 2024-03-20 9:20 Guixin Liu
2024-03-20 10:45 ` Sagi Grimberg
0 siblings, 1 reply; 3+ messages in thread
From: Guixin Liu @ 2024-03-20 9:20 UTC (permalink / raw)
To: hch, sagi, kch; +Cc: linux-nvme
We can simply use invalidate_rkey to check instead of adding a flag.
Signed-off-by: Guixin Liu <kanie@linux.alibaba.com>
---
drivers/nvme/target/rdma.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
index f2bb9d95ecf4..d0bc72c75471 100644
--- a/drivers/nvme/target/rdma.c
+++ b/drivers/nvme/target/rdma.c
@@ -53,7 +53,6 @@ struct nvmet_rdma_cmd {
enum {
NVMET_RDMA_REQ_INLINE_DATA = (1 << 0),
- NVMET_RDMA_REQ_INVALIDATE_RKEY = (1 << 1),
};
struct nvmet_rdma_rsp {
@@ -722,7 +721,7 @@ static void nvmet_rdma_queue_response(struct nvmet_req *req)
struct rdma_cm_id *cm_id = rsp->queue->cm_id;
struct ib_send_wr *first_wr;
- if (rsp->flags & NVMET_RDMA_REQ_INVALIDATE_RKEY) {
+ if (rsp->invalidate_rkey) {
rsp->send_wr.opcode = IB_WR_SEND_WITH_INV;
rsp->send_wr.ex.invalidate_rkey = rsp->invalidate_rkey;
} else {
@@ -905,10 +904,8 @@ static u16 nvmet_rdma_map_sgl_keyed(struct nvmet_rdma_rsp *rsp,
goto error_out;
rsp->n_rdma += ret;
- if (invalidate) {
+ if (invalidate)
rsp->invalidate_rkey = key;
- rsp->flags |= NVMET_RDMA_REQ_INVALIDATE_RKEY;
- }
return 0;
--
2.43.0
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] nvmet-rdma: remove NVMET_RDMA_REQ_INVALIDATE_RKEY flag
2024-03-20 9:20 [PATCH] nvmet-rdma: remove NVMET_RDMA_REQ_INVALIDATE_RKEY flag Guixin Liu
@ 2024-03-20 10:45 ` Sagi Grimberg
2024-03-20 10:56 ` Guixin Liu
0 siblings, 1 reply; 3+ messages in thread
From: Sagi Grimberg @ 2024-03-20 10:45 UTC (permalink / raw)
To: Guixin Liu, hch, kch; +Cc: linux-nvme
On 20/03/2024 11:20, Guixin Liu wrote:
> We can simply use invalidate_rkey to check instead of adding a flag.
>
> Signed-off-by: Guixin Liu <kanie@linux.alibaba.com>
> ---
> drivers/nvme/target/rdma.c | 7 ++-----
> 1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
> index f2bb9d95ecf4..d0bc72c75471 100644
> --- a/drivers/nvme/target/rdma.c
> +++ b/drivers/nvme/target/rdma.c
> @@ -53,7 +53,6 @@ struct nvmet_rdma_cmd {
>
> enum {
> NVMET_RDMA_REQ_INLINE_DATA = (1 << 0),
> - NVMET_RDMA_REQ_INVALIDATE_RKEY = (1 << 1),
> };
>
> struct nvmet_rdma_rsp {
> @@ -722,7 +721,7 @@ static void nvmet_rdma_queue_response(struct nvmet_req *req)
> struct rdma_cm_id *cm_id = rsp->queue->cm_id;
> struct ib_send_wr *first_wr;
>
> - if (rsp->flags & NVMET_RDMA_REQ_INVALIDATE_RKEY) {
> + if (rsp->invalidate_rkey) {
> rsp->send_wr.opcode = IB_WR_SEND_WITH_INV;
> rsp->send_wr.ex.invalidate_rkey = rsp->invalidate_rkey;
> } else {
> @@ -905,10 +904,8 @@ static u16 nvmet_rdma_map_sgl_keyed(struct nvmet_rdma_rsp *rsp,
> goto error_out;
> rsp->n_rdma += ret;
>
> - if (invalidate) {
> + if (invalidate)
> rsp->invalidate_rkey = key;
> - rsp->flags |= NVMET_RDMA_REQ_INVALIDATE_RKEY;
> - }
>
> return 0;
>
Who clears invalidate_rkey between executions?
I think you need to clear right after nvmet_rdma_get_rsp()
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] nvmet-rdma: remove NVMET_RDMA_REQ_INVALIDATE_RKEY flag
2024-03-20 10:45 ` Sagi Grimberg
@ 2024-03-20 10:56 ` Guixin Liu
0 siblings, 0 replies; 3+ messages in thread
From: Guixin Liu @ 2024-03-20 10:56 UTC (permalink / raw)
To: Sagi Grimberg, hch, kch; +Cc: linux-nvme
>
> Who clears invalidate_rkey between executions?
> I think you need to clear right after nvmet_rdma_get_rsp()
You are right, I need clear invalidate_rkey to 0 after nvmet_rdma_get_rsp().
Thanks,
Guixin Liu
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2024-03-20 10:57 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-03-20 9:20 [PATCH] nvmet-rdma: remove NVMET_RDMA_REQ_INVALIDATE_RKEY flag Guixin Liu
2024-03-20 10:45 ` Sagi Grimberg
2024-03-20 10:56 ` Guixin Liu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox