On 11/16/21 11:15 PM, Christoph Hellwig wrote:
> On Tue, Nov 16, 2021 at 08:38:05PM -0700, Jens Axboe wrote:
>> /**
>> * nvme_submit_cmd() - Copy a command into a queue and ring the doorbell
>> * @nvmeq: The queue to use
>> @@ -511,10 +520,7 @@ static void nvme_submit_cmd(struct nvme_queue *nvmeq, struct nvme_command *cmd,
>> bool write_sq)
>> {
>> spin_lock(&nvmeq->sq_lock);
>> - memcpy(nvmeq->sq_cmds + (nvmeq->sq_tail << nvmeq->sqes),
>> - cmd, sizeof(*cmd));
>> - if (++nvmeq->sq_tail == nvmeq->q_depth)
>> - nvmeq->sq_tail = 0;
>> + nvme_copy_cmd(nvmeq, cmd);
>> nvme_write_sq_db(nvmeq, write_sq);
>> spin_unlock(&nvmeq->sq_lock);
>
> Given that nvme_submit_cmd only has two callers, I'd be tempted to just
> open code in the callers rather than creating a deep callchain here.
It's inlined, but if you prefer duplicating the code, then I can just
drop this patch.
--
Jens Axboe
