linux-kernel - Re: [RFC PATCH] fs: block_dev: compute nr_vecs hint for improving writeback bvecs allocation

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: <20210106084548.GA3845805@T590>
Date:   Wed, 6 Jan 2021 16:45:48 +0800
From:   Ming Lei <ming.lei@...hat.com>
To:     Christoph Hellwig <hch@....de>
Cc:     Jens Axboe <axboe@...nel.dk>, linux-kernel@...r.kernel.org,
        linux-block@...r.kernel.org,
        Alexander Viro <viro@...iv.linux.org.uk>,
        "Darrick J . Wong" <darrick.wong@...cle.com>,
        linux-xfs@...r.kernel.org, linux-fsdevel@...r.kernel.org
Subject: Re: [RFC PATCH] fs: block_dev: compute nr_vecs hint for improving
 writeback bvecs allocation

On Tue, Jan 05, 2021 at 07:39:38PM +0100, Christoph Hellwig wrote:
> At least for iomap I think this is the wrong approach.  Between the
> iomap and writeback_control we know the maximum size of the writeback
> request and can just use that.

I think writeback_control can tell us nothing about max pages in single
bio:

- wbc->nr_to_write controls how many pages to writeback, this pages
  usually don't belong to same bio. Also this number is often much
  bigger than BIO_MAX_PAGES.

- wbc->range_start/range_end is similar too, which is often much more
  bigger than BIO_MAX_PAGES.

Also page/blocks_in_page can be mapped to different extent too, which is
only available when wpc->ops->map_blocks() is returned, which looks not
different with mpage_writepages(), in which bio is allocated with
BIO_MAX_PAGES vecs too.

Or you mean we can use iomap->length for this purpose? But iomap->length
still is still too big in case of xfs.

-- 
Ming