nvme-tcp: Do not reset transport on data digest errors
PATCH v2 1/1 nvme-tcp: Do not reset transport on data digest errors
The spec says
7.4.6.1 Digest Error handling
When a host detects a data digest error in a C2HData PDU, that host shall continue processing C2HData PDUs associated with the command and when the command processing has completed, if a successful status was returned by the controller, the host shall fail the command with a non-fatal transport error.
Currently the transport is reseted when a data digest error is detected. To fix this, keep track of the final status in the queue object and use it when completing the request.
---
The status member placed so that it fills up a hole in struct nvme_tcp_request:
struct nvme_tcp_request {
struct nvme_request req; /* 0 32 */
void * pdu; /* 32 8 */
struct nvme_tcp_queue * queue; /* 40 8 */
u32 data_len; /* 48 4 */
u32 pdu_len; /* 52 4 */
u32 pdu_sent; /* 56 4 */
u16 ttag; /* 60 2 */
u16 status; /* 62 2 */ /* --- cacheline 1 boundary (64 bytes) --- */
struct list_head entry; /* 64 16 */
struct llist_node lentry; /* 80 8 */
__le32 ddgst; /* 88 4 */
/* XXX 4 bytes hole, try to pack */
struct bio * curr_bio; /* 96 8 */
struct iov_iter iter; /* 104 40 */
/* --- cacheline 2 boundary (128 bytes) was 16 bytes ago --- */
size_t offset; /* 144 8 */
size_t data_sent; /* 152 8 */
enum nvme_tcp_send_state state; /* 160 4 */
/* size: 168, cachelines: 3, members: 16 */ /* sum members: 160, holes: 1, sum holes: 4 */ /* padding: 4 */
/* last cacheline: 40 bytes */
};
v1:
- https://lore.kernel.org/linux-nvme/[email protected]/
- moved 'status' from nvme_tcp_queue to nvme_tcp_request.
drivers/nvme/host/tcp.c | 25 +++++++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-)
drivers/nvme/host/tcp.c +21 -4
@@ -45,6 +45,7 @@ struct nvme_tcp_request {u32 pdu_len;u32 pdu_sent;u16 ttag; +u16 status;struct list_head entry;struct llist_node lentry;__le32 ddgst;
@@ -485,7 +486,9 @@ static void nvme_tcp_error_recovery(struct nvme_ctrl *ctrl)static int nvme_tcp_process_nvme_cqe (struct nvme_tcp_queue *queue,struct nvme_completion *cqe) { +struct nvme_tcp_request *req;struct request *rq; +u16 status; rq =nvme_find_rq (nvme_tcp_tagset (queue), cqe->command_id);if (!rq) {
@@ -496,7 +499,12 @@ static int nvme_tcp_process_nvme_cqe(struct nvme_tcp_queue *queue,return -EINVAL ; } -if (!nvme_try_complete_req (rq, cqe->status, cqe->result)) + req =blk_mq_rq_to_pdu (rq); + status = req->status; +if (status ==NVME_SC_SUCCESS ) + status = cqe->status; + +if (!nvme_try_complete_req (rq, status, cqe->result))nvme_complete_rq (rq); queue->nr_cqe++;
@@ -506,6 +514,7 @@ static int nvme_tcp_process_nvme_cqe(struct nvme_tcp_queue *queue,static int nvme_tcp_handle_c2h_data (struct nvme_tcp_queue *queue,struct nvme_tcp_data_pdu *pdu) { +struct nvme_tcp_request *req;struct request *rq; rq =nvme_find_rq (nvme_tcp_tagset (queue), pdu->command_id);
@@ -534,6 +543,8 @@ static int nvme_tcp_handle_c2h_data(struct nvme_tcp_queue *queue,return -EPROTO ; } + req =blk_mq_rq_to_pdu (rq); + req->status =NVME_SC_SUCCESS ;return 0 ; }
@@ -758,7 +769,7 @@ static int nvme_tcp_recv_data(struct nvme_tcp_queue *queue, struct sk_buff *skb, queue->ddgst_remaining =NVME_TCP_DIGEST_LENGTH ; }else {if (pdu->hdr.flags &NVME_TCP_F_DATA_SUCCESS ) { -nvme_tcp_end_request (rq,NVME_SC_SUCCESS ); +nvme_tcp_end_request (rq, req->status); queue->nr_cqe++; }nvme_tcp_init_recv_ctx (queue);
@@ -788,18 +799,24 @@ static int nvme_tcp_recv_ddgst(struct nvme_tcp_queue *queue,return 0 ;if (queue->recv_ddgst != queue->exp_ddgst) { +struct request *rq =nvme_cid_to_rq (nvme_tcp_tagset (queue), + pdu->command_id); +struct nvme_tcp_request *req =blk_mq_rq_to_pdu (rq); + + req->status =NVME_SC_DATA_XFER_ERROR ; +dev_err (queue->ctrl->ctrl.device,"data digest error: recv %#x expected %#x\n" ,le32_to_cpu (queue->recv_ddgst),le32_to_cpu (queue->exp_ddgst)); -return -EIO ; }if (pdu->hdr.flags &NVME_TCP_F_DATA_SUCCESS ) {struct request *rq =nvme_cid_to_rq (nvme_tcp_tagset (queue), pdu->command_id); +struct nvme_tcp_request *req =blk_mq_rq_to_pdu (rq); -nvme_tcp_end_request (rq,NVME_SC_SUCCESS ); +nvme_tcp_end_request (rq, req->status); queue->nr_cqe++; }
Signature · 1 line
2.29.2
Quoted 101 lines · Daniel Wagner
The spec says
7.4.6.1 Digest Error handling
When a host detects a data digest error in a C2HData PDU, that host shall continue processing C2HData PDUs associated with the command and when the command processing has completed, if a successful status was returned by the controller, the host shall fail the command with a non-fatal transport error.
Currently the transport is reseted when a data digest error is detected. To fix this, keep track of the final status in the queue object and use it when completing the request.
Signed-off-by Daniel Wagner <[email protected]>---
The status member placed so that it fills up a hole in struct
nvme_tcp_request:struct
nvme_tcp_request{struct nvme_request req; /* 0 32 */ void * pdu; /* 32 8 */ struct nvme_tcp_queue * queue; /* 40 8 */ u32 data_len; /* 48 4 */ u32 pdu_len; /* 52 4 */ u32 pdu_sent; /* 56 4 */ u16 ttag; /* 60 2 */ u16 status; /* 62 2 *//* --- cacheline 1 boundary (64 bytes) --- */
structlist_headentry; /* 64 16 */
structllist_nodelentry; /* 80 8 */
__le32ddgst; /* 88 4 *//* XXX 4 bytes hole, try to pack */
struct bio *
curr_bio; /* 96 8 */
structiov_iteriter; /* 104 40 */
/* --- cacheline 2 boundary (128 bytes) was 16 bytes ago --- */
size_toffset; /* 144 8 */
size_tdata_sent; /* 152 8 */
enumnvme_tcp_send_statestate; /* 160 4 *//* size: 168, cachelines: 3, members: 16 */ /* sum members: 160, holes: 1, sum holes: 4 */ /* padding: 4 */
/* last cacheline: 40 bytes */
};v1:
- https://lore.kernel.org/linux-nvme/[email protected]/
- moved 'status' fromnvme_tcp_queuetonvme_tcp_request.drivers/nvme/host/tcp.c | 25 +++++++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-)drivers/nvme/host/tcp.c +10
-1@@ -45,6 +45,7 @@ struct nvme_tcp_request {u32 pdu_len;u32 pdu_sent;u16 ttag; +u16 status;struct list_head entry;struct llist_node lentry;__le32 ddgst;@@ -485,7 +486,9 @@ static void nvme_tcp_error_recovery(struct nvme_ctrl *ctrl)static int nvme_tcp_process_nvme_cqe (struct nvme_tcp_queue *queue,struct nvme_completion *cqe) { +struct nvme_tcp_request *req;struct request *rq; +u16 status; rq =nvme_find_rq (nvme_tcp_tagset (queue), cqe->command_id);if (!rq) {@@ -496,7 +499,12 @@ static int nvme_tcp_process_nvme_cqe(struct nvme_tcp_queue *queue,return -EINVAL ; } -if (!nvme_try_complete_req (rq, cqe->status, cqe->result)) + req =blk_mq_rq_to_pdu (rq); + status = req->status; +if (status ==NVME_SC_SUCCESS ) + status = cqe->status; + +if (!nvme_try_complete_req (rq, status, cqe->result))nvme_complete_rq (rq); queue->nr_cqe++;@@ -506,6 +514,7 @@ static int nvme_tcp_process_nvme_cqe(struct nvme_tcp_queue *queue,static int nvme_tcp_handle_c2h_data (struct nvme_tcp_queue *queue,struct nvme_tcp_data_pdu *pdu) { +struct nvme_tcp_request *req;struct request *rq; rq =nvme_find_rq (nvme_tcp_tagset (queue), pdu->command_id);
drivers/nvme/host/tcp.c static int nvme_tcp_handle_c2h_data(struct nvme_tcp_queue *queue, · line 550
Quoted 8 lines
drivers/nvme/host/tcp.c +2
-0@@ -534,6 +543,8 @@ static int nvme_tcp_handle_c2h_data(struct nvme_tcp_queue *queue,return -EPROTO ; } + req =blk_mq_rq_to_pdu (rq); + req->status =NVME_SC_SUCCESS ;return 0 ; }
Can't you move the initialisation to nvme_tcp_setup_cmd_pdu()? That'll save to to have to reference the tcp request here ...
Quoted 27 lines
drivers/nvme/host/tcp.c +7
-2@@ -758,7 +769,7 @@ static int nvme_tcp_recv_data(struct nvme_tcp_queue *queue, struct sk_buff *skb, queue->ddgst_remaining =NVME_TCP_DIGEST_LENGTH ; }else {if (pdu->hdr.flags &NVME_TCP_F_DATA_SUCCESS ) { -nvme_tcp_end_request (rq,NVME_SC_SUCCESS ); +nvme_tcp_end_request (rq, req->status); queue->nr_cqe++; }nvme_tcp_init_recv_ctx (queue);@@ -788,18 +799,24 @@ static int nvme_tcp_recv_ddgst(struct nvme_tcp_queue *queue,return 0 ;if (queue->recv_ddgst != queue->exp_ddgst) { +struct request *rq =nvme_cid_to_rq (nvme_tcp_tagset (queue), + pdu->command_id); +struct nvme_tcp_request *req =blk_mq_rq_to_pdu (rq); + + req->status =NVME_SC_DATA_XFER_ERROR ; +dev_err (queue->ctrl->ctrl.device,"data digest error: recv %#x expected %#x\n" ,le32_to_cpu (queue->recv_ddgst),le32_to_cpu (queue->exp_ddgst)); -return -EIO ; }
drivers/nvme/host/tcp.c static int nvme_tcp_recv_ddgst(struct nvme_tcp_queue *queue, · line 822
Quoted 10 lines
drivers/nvme/host/tcp.c +2
-1if (pdu->hdr.flags &NVME_TCP_F_DATA_SUCCESS ) {struct request *rq =nvme_cid_to_rq (nvme_tcp_tagset (queue), pdu->command_id); +struct nvme_tcp_request *req =blk_mq_rq_to_pdu (rq); -nvme_tcp_end_request (rq,NVME_SC_SUCCESS ); +nvme_tcp_end_request (rq, req->status); queue->nr_cqe++; }
Otherwise:
Cheers,
Hannes
Signature · 4 lines
Dr. Hannes Reinecke Kernel Storage Architect [email protected] +49 911 74053 688 SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg HRB 36809 (AG Nürnberg), Geschäftsführer: Felix Imendörffer
Quoted 9 lines · Hannes Reinecke
Quoted 9 lines
drivers/nvme/host/tcp.c +1
-0@@ -506,6 +514,7 @@ static int nvme_tcp_process_nvme_cqe(struct nvme_tcp_queue *queue,static int nvme_tcp_handle_c2h_data (struct nvme_tcp_queue *queue,struct nvme_tcp_data_pdu *pdu) { +struct nvme_tcp_request *req;struct request *rq; rq =nvme_find_rq (nvme_tcp_tagset (queue), pdu->command_id);@@ -534,6 +543,8 @@ static int nvme_tcp_handle_c2h_data(struct nvme_tcp_queue *queue,return -EPROTO ;
Quoted 8 lines
drivers/nvme/host/tcp.c static int nvme_tcp_handle_c2h_data(struct nvme_tcp_queue *queue, · line 549
Quoted 5 lines
drivers/nvme/host/tcp.c +2
-0} + req =blk_mq_rq_to_pdu (rq); + req->status =NVME_SC_SUCCESS ;return 0 ; }Can't you move the initialisation to
nvme_tcp_setup_cmd_pdu()? That'll save to to have to reference the tcp request here ...
Yes, that's a why better place to initialize it. Let me spin a new version.