OpenSDV
/
original_kernel
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

nvme-fc: correct csn initialization and increments on error 

This patch fixes a long-standing bug that initialized the FC-NVME
cmnd iu CSN value to 1. Early FC-NVME specs had the connection starting
with CSN=1. By the time the spec reached approval, the language had
changed to state a connection should start with CSN=0.  This patch
corrects the initialization value for FC-NVME connections.

Additionally, in reviewing the transport, the CSN value is assigned to
the new IU early in the start routine. It's possible that a later dma
map request may fail, causing the command to never be sent to the
controller.  Change the location of the assignment so that it is
immediately prior to calling the lldd. Add a comment block to explain
the impacts if the lldd were to additionally fail sending the command.

Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Ewan D. Milne <emilne@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
This commit is contained in:
James Smart 2019-04-08 11:15:19 -07:00 committed by Christoph Hellwig
parent a3761c3c91
commit 67f471b6ed
1 changed files with 15 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
drivers/nvme/host/fc.c
View File

@ -1845,7 +1845,7 @@ nvme_fc_init_queue(struct nvme_fc_ctrl *ctrl, int idx)
memset(queue, 0, sizeof(*queue));
queue->ctrl = ctrl;
queue->qnum = idx;
atomic_set(&queue->csn, 1);
atomic_set(&queue->csn, 0);
queue->dev = ctrl->dev;
if (idx > 0)
@ -1887,7 +1887,7 @@ nvme_fc_free_queue(struct nvme_fc_queue *queue)
*/
queue->connection_id = 0;
atomic_set(&queue->csn, 1);
atomic_set(&queue->csn, 0);
}
static void
@ -2183,7 +2183,6 @@ nvme_fc_start_fcp_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_queue *queue,
{
struct nvme_fc_cmd_iu *cmdiu = &op->cmd_iu;
struct nvme_command *sqe = &cmdiu->sqe;
u32 csn;
int ret, opstate;
/*
@ -2198,8 +2197,6 @@ nvme_fc_start_fcp_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_queue *queue,
/* format the FC-NVME CMD IU and fcp_req */
cmdiu->connection_id = cpu_to_be64(queue->connection_id);
csn = atomic_inc_return(&queue->csn);
cmdiu->csn = cpu_to_be32(csn);
cmdiu->data_len = cpu_to_be32(data_len);
switch (io_dir) {
case NVMEFC_FCP_WRITE:
@ -2257,11 +2254,24 @@ nvme_fc_start_fcp_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_queue *queue,
if (!(op->flags & FCOP_FLAGS_AEN))
blk_mq_start_request(op->rq);
cmdiu->csn = cpu_to_be32(atomic_inc_return(&queue->csn));
ret = ctrl->lport->ops->fcp_io(&ctrl->lport->localport,
&ctrl->rport->remoteport,
queue->lldd_handle, &op->fcp_req);
if (ret) {
/*
* If the lld fails to send the command is there an issue with
* the csn value? If the command that fails is the Connect,
* no - as the connection won't be live. If it is a command
* post-connect, it's possible a gap in csn may be created.
* Does this matter? As Linux initiators don't send fused
* commands, no. The gap would exist, but as there's nothing
* that depends on csn order to be delivered on the target
* side, it shouldn't hurt. It would be difficult for a
* target to even detect the csn gap as it has no idea when the
* cmd with the csn was supposed to arrive.
*/
opstate = atomic_xchg(&op->state, FCPOP_STATE_COMPLETE);
__nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate);