linux-kernel - Re: [PATCH RESEND v5 3/7] blk-mq: add a new queue sysfs attribute async

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: <4f3ed928-2fef-469d-8bfc-76041267b1c6@acm.org>
Date: Mon, 17 Nov 2025 15:45:32 -0800
From: Bart Van Assche <bvanassche@....org>
To: Yu Kuai <yukuai@...as.com>, axboe@...nel.dk, linux-block@...r.kernel.org,
 linux-kernel@...r.kernel.org
Cc: nilay@...ux.ibm.com
Subject: Re: [PATCH RESEND v5 3/7] blk-mq: add a new queue sysfs attribute
 async_depth

On 11/15/25 7:52 PM, Yu Kuai wrote:
> +static void blk_mq_limit_depth(struct blk_mq_alloc_data *data)
> +{
> +	struct elevator_mq_ops *ops;
> +
> +	/* If elevator is none, don't limit requests */

Nobody calls an I/O scheduler an elevator anymore. So I propose to
change "elevator is none" into "no I/O scheduler has been configured".

> +	/*
> +	 * Flush/passthrough requests are special and go directly to the
> +	 * dispatch list, they don't have limit.
> +	 */
"they don't have a limit" -> "are not subject to the async_depth limit".

> +	/*
> +	 * By default, sync requests have no limit, and async requests is
> +	 * limited to async_depth.
> +	 */

"is" -> "are"

 > -		/*
 > -		 * All requests use scheduler tags when an I/O scheduler is
 > -		 * enabled for the queue.
 > -		 */
 > -		data->rq_flags |= RQF_SCHED_TAGS;

Why has the above comment been removed? I think that it is useful.

Additionally, please split this patch into two patches: a first patch
that introduces the function blk_mq_limit_depth() and a second patch
with the remaining changes from this patch.

Thanks,

Bart.