[PATCH v23 06/20] nvme-tcp: Add DDP data-path

Aurelien Aptel aaptel at nvidia.com
Thu Mar 7 07:44:13 PST 2024


Sagi Grimberg <sagi at grimberg.me> writes:
>> +static void nvme_tcp_complete_request(struct request *rq,
>> +                                   __le16 status,
>> +                                   union nvme_result result,
>> +                                   __u16 command_id)
>> +{
>> +     struct nvme_tcp_request *req = blk_mq_rq_to_pdu(rq);
>> +
>> +     if (nvme_tcp_is_ddp_offloaded(req)) {
>> +             req->nvme_status = status;
>
> this can just be called req->status I think.

Since req->status already exists, we have checked whether it can be
safely used instead of adding nvme_status and it seems to be ok.

We will remove nvme_status.

>> +             req->result = result;
> I think it will be cleaner to always capture req->result and req->status
> regardless of ddp offload.

Sure, we will set status and result in the function before the offload
check:

static void nvme_tcp_complete_request(struct request *rq,
                                      __le16 status,
                                      union nvme_result result,
                                      __u16 command_id)
{
        struct nvme_tcp_request *req = blk_mq_rq_to_pdu(rq);

        req->status = status;
        req->result = result;

        if (nvme_tcp_is_ddp_offloaded(req)) {
                /* complete when teardown is confirmed to be done */
                nvme_tcp_teardown_ddp(req->queue, rq);
                return;
        }

        if (!nvme_try_complete_req(rq, status, result))
                nvme_complete_rq(rq);
}



More information about the Linux-nvme mailing list