linux-kernel - Re: [PATCH v20 02/12] Add infrastructure for copy offload in block and request layer.

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-next>] [day] [month] [year] [list]

Date: Thu, 30 May 2024 10:11:15 -0700
From: Bart Van Assche <bvanassche@....org>
To: Nitesh Shetty <nj.shetty@...sung.com>
Cc: Damien Le Moal <dlemoal@...nel.org>, Jens Axboe <axboe@...nel.dk>,
 Jonathan Corbet <corbet@....net>, Alasdair Kergon <agk@...hat.com>,
 Mike Snitzer <snitzer@...nel.org>, Mikulas Patocka <mpatocka@...hat.com>,
 Keith Busch <kbusch@...nel.org>, Christoph Hellwig <hch@....de>,
 Sagi Grimberg <sagi@...mberg.me>, Chaitanya Kulkarni <kch@...dia.com>,
 Alexander Viro <viro@...iv.linux.org.uk>,
 Christian Brauner <brauner@...nel.org>, Jan Kara <jack@...e.cz>,
 martin.petersen@...cle.com, david@...morbit.com, hare@...e.de,
 damien.lemoal@...nsource.wdc.com, anuj20.g@...sung.com, joshi.k@...sung.com,
 nitheshshetty@...il.com, gost.dev@...sung.com, linux-block@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-doc@...r.kernel.org,
 dm-devel@...ts.linux.dev, linux-nvme@...ts.infradead.org,
 linux-fsdevel@...r.kernel.org
Subject: Re: [PATCH v20 02/12] Add infrastructure for copy offload in block
 and request layer.

On 5/30/24 00:16, Nitesh Shetty wrote:
> +static inline bool blk_copy_offload_attempt_combine(struct request_queue *q,
> +                         struct bio *bio)
> +{
> +    struct blk_plug *plug = current->plug;
> +    struct request *rq;
> +
> +    if (!plug || rq_list_empty(plug->mq_list))
> +        return false;
> +
> +    rq_list_for_each(&plug->mq_list, rq) {
> +        if (rq->q == q) {
> +            if (!blk_copy_offload_combine(rq, bio))
> +                return true;
> +            break;
> +        }
> +
> +        /*
> +         * Only keep iterating plug list for combines if we have multiple
> +         * queues
> +         */
> +        if (!plug->multiple_queues)
> +            break;
> +    }
> +    return false;
> +}

This new approach has the following two disadvantages:
* Without plug, REQ_OP_COPY_SRC and REQ_OP_COPY_DST are not combined. These two
   operation types are the only operation types for which not using a plug causes
   an I/O failure.
* A loop is required to combine the REQ_OP_COPY_SRC and REQ_OP_COPY_DST operations.

Please switch to the approach Hannes suggested, namely bio chaining. Chaining
REQ_OP_COPY_SRC and REQ_OP_COPY_DST bios before these are submitted eliminates the
two disadvantages mentioned above.

Thanks,

Bart.