linux-kernel - Re: [PATCH v2 1/3] nvme: add capability to connect to an admin controller

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <d19a0776-5c3c-46f8-82ad-a1a83d5af884@suse.de>
Date: Thu, 3 Jul 2025 10:55:00 +0200
From: Hannes Reinecke <hare@...e.de>
To: Kamaljit Singh <kamaljit.singh1@....com>, kbusch@...nel.org,
 axboe@...nel.dk, hch@....de, sagi@...mberg.me,
 linux-nvme@...ts.infradead.org, linux-kernel@...r.kernel.org
Cc: cassel@...nel.org, dlemoal@...nel.org
Subject: Re: [PATCH v2 1/3] nvme: add capability to connect to an admin
 controller

On 7/2/25 02:58, Kamaljit Singh wrote:
> Suggested-by: Niklas Cassel <cassel@...nel.org>
> 
> Add capability to connect to an administrative controller by
> preventing ioq creation for admin-controllers.
> 
> * Add helper nvme_admin_ctrl() to check for an administrative controller
> * Add helper nvme_override_prohibited_io_queues() to override
>    queue_count in nvme_ctrl
> 
> Reference: NVMe Base rev 2.2, sec 3.1.3.4, fig 28.
> 
> Signed-off-by: Kamaljit Singh <kamaljit.singh1@....com>
> ---
>   drivers/nvme/host/core.c | 21 +++++++++++++++++++++
>   drivers/nvme/host/nvme.h |  1 +
>   drivers/nvme/host/rdma.c |  2 ++
>   drivers/nvme/host/tcp.c  |  2 ++
>   4 files changed, 26 insertions(+)
> 
> diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
> index e533d791955d..a1155fb8d5be 100644
> --- a/drivers/nvme/host/core.c
> +++ b/drivers/nvme/host/core.c
> @@ -3149,6 +3149,22 @@ static inline bool nvme_discovery_ctrl(struct nvme_ctrl *ctrl)
>   	return ctrl->opts && ctrl->opts->discovery_nqn;
>   }
>   
> +static inline bool nvme_admin_ctrl(struct nvme_ctrl *ctrl)
> +{
> +	return (ctrl->cntrltype == NVME_CTRL_ADMIN);
> +}
> +
> +/*
> + * An admin controller has one admin queue, but no I/O queues.
> + * Override queue_count so it only creates an admin queue.
> + */
> +void nvme_override_prohibited_io_queues(struct nvme_ctrl *ctrl)
> +{
> +	if (nvme_admin_ctrl(ctrl))
> +		ctrl->queue_count = 1;
> +}
> +EXPORT_SYMBOL_GPL(nvme_override_prohibited_io_queues);
> +
>   static bool nvme_validate_cntlid(struct nvme_subsystem *subsys,
>   		struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
>   {
> @@ -3215,6 +3231,11 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
>   		kfree(subsys);
>   		return -EINVAL;
>   	}
> +	if (nvme_admin_ctrl(ctrl))
> +		dev_info(ctrl->device,
> +			"Subsystem %s is an administrative controller",
> +			subsys->subnqn);
> +

Bzzt. A subsystem is a subsystem, a controller is a controller.
Better issue a message when connecting the controller.

>   	nvme_mpath_default_iopolicy(subsys);
>   
>   	subsys->dev.class = &nvme_subsys_class;
> diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
> index 7df2ea21851f..5e07ba634e97 100644
> --- a/drivers/nvme/host/nvme.h
> +++ b/drivers/nvme/host/nvme.h
> @@ -1212,6 +1212,7 @@ struct nvme_ctrl *nvme_ctrl_from_file(struct file *file);
>   struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid);
>   bool nvme_get_ns(struct nvme_ns *ns);
>   void nvme_put_ns(struct nvme_ns *ns);
> +void nvme_override_prohibited_io_queues(struct nvme_ctrl *ctrl);
>   
>   static inline bool nvme_multi_css(struct nvme_ctrl *ctrl)
>   {
> diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
> index 9bd3646568d0..7d50a2f31c3a 100644
> --- a/drivers/nvme/host/rdma.c
> +++ b/drivers/nvme/host/rdma.c
> @@ -1026,6 +1026,8 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
>   		goto destroy_admin;
>   	}
>   
> +	nvme_override_prohibited_io_queues(&ctrl->ctrl);
> +
>   	if (ctrl->ctrl.opts->queue_size > ctrl->ctrl.sqsize + 1) {
>   		dev_warn(ctrl->ctrl.device,
>   			"queue_size %zu > ctrl sqsize %u, clamping down\n",
> diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
> index d924008c3949..bfb52a487c45 100644
> --- a/drivers/nvme/host/tcp.c
> +++ b/drivers/nvme/host/tcp.c
> @@ -2381,6 +2381,8 @@ static int nvme_tcp_setup_ctrl(struct nvme_ctrl *ctrl, bool new)
>   		goto destroy_admin;
>   	}
>   
> +	nvme_override_prohibited_io_queues(ctrl);
> +
>   	if (opts->queue_size > ctrl->sqsize + 1)
>   		dev_warn(ctrl->device,
>   			"queue_size %zu > ctrl sqsize %u, clamping down\n",

And that is a bit convoluted.

Why not add a check in 'nvme_set_queue_count' and reduce the number of
queues to '1' there?

(Then you also have a place to put your message about the admin
controller ...)

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                  Kernel Storage Architect
hare@...e.de                                +49 911 74053 688
SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg
HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich