Re: [PATCH] nvme: update keep alive interval when kato is modified

[Sagi Grimberg]

On 8/2/21 7:12 PM, sasaki tatsuya wrote:
> Currently the connection between host and NVMe-oF target gets
> disconnected by keep-alive timeout when a user connects to a target
> with a relatively large kato value and then sets the smaller kato
> with a set features command (e.g. connects with 60 seconds kato value
> and then sets 10 seconds kato value).
>
> The cause is that keep alive command interval on the host, which is
> defined as unsigned int kato in nvme_ctrl structure, does not follow
> the kato value changes.
>
> This patch updates the keep alive interval in the following steps when
> the kato is modified by a set features command: stops the keep alive
> work queue, then sets the kato as new timer value and re-start the queue.
>
> Signed-off-by: Tatsuya Sasaki <tatsuya6.sasaki@xxxxxxxxxx>
> ---
>   drivers/nvme/host/core.c  |  3 ++-
>   drivers/nvme/host/ioctl.c | 17 +++++++++++++++++
>   drivers/nvme/host/nvme.h  |  1 +
>   3 files changed, 20 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
> index dfd9dec0c1f6..89c52da15618 100644
> --- a/drivers/nvme/host/core.c
> +++ b/drivers/nvme/host/core.c
> @@ -1255,13 +1255,14 @@ static void nvme_keep_alive_work(struct work_struct *work)
>   	blk_execute_rq_nowait(NULL, rq, 0, nvme_keep_alive_end_io);
>   }
>   
> -static void nvme_start_keep_alive(struct nvme_ctrl *ctrl)
> +void nvme_start_keep_alive(struct nvme_ctrl *ctrl)
>   {
>   	if (unlikely(ctrl->kato == 0))
>   		return;
>   
>   	nvme_queue_keep_alive_work(ctrl);
>   }
> +EXPORT_SYMBOL_GPL(nvme_start_keep_alive);
>   
>   void nvme_stop_keep_alive(struct nvme_ctrl *ctrl)
>   {
> diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c
> index 305ddd415e45..0066728e77b2 100644
> --- a/drivers/nvme/host/ioctl.c
> +++ b/drivers/nvme/host/ioctl.c
> @@ -231,6 +231,23 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
>   			return -EFAULT;
>   	}
>   
> +	/*
> +	 * Keep alive commands interval on the host should be updated
> +	 * when KATO is modified by Set Features commands.
> +	 */

This is too specific. I'd move it to nvme_user_cmd_post similar to how
we handle passthru commands (with start/end that does stuff based on the
effects)

> +	if (!status && cmd.opcode == nvme_admin_set_features &&
> +	    (cmd.cdw10 & 0xFF) == NVME_FEAT_KATO) {
> +		/* ms -> s */
> +		unsigned int new_kato = DIV_ROUND_UP(cmd.cdw11, 1000);
> +

The section below needs to move to a core rountine:
nvme_update_keep_alive()

> +		dev_info(ctrl->device,
> +			 "keep alive commands interval on the host is updated from %u milliseconds to %u milliseconds\n",
> +			 ctrl->kato * 1000 / 2, new_kato * 1000 / 2);
> +		nvme_stop_keep_alive(ctrl);
> +		ctrl->kato = new_kato;
> +		nvme_start_keep_alive(ctrl);
> +	}
> +
>   	return status;
>   }
>