linux-kernel - Re: [PATCH v1 2/6] fs: use on-stack-bio if backing device has BDI_CAP

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]

Date:   Mon, 14 Aug 2017 17:48:37 +0900
From:   Minchan Kim <minchan@...nel.org>
To:     Christoph Hellwig <hch@....de>
Cc:     Dan Williams <dan.j.williams@...el.com>,
        Matthew Wilcox <willy@...radead.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linux-mm <linux-mm@...ck.org>,
        Ross Zwisler <ross.zwisler@...ux.intel.com>,
        "karam . lee" <karam.lee@....com>, seungho1.park@....com,
        Dave Chinner <david@...morbit.com>, Jan Kara <jack@...e.cz>,
        Jens Axboe <axboe@...nel.dk>,
        Vishal Verma <vishal.l.verma@...el.com>,
        "linux-nvdimm@...ts.01.org" <linux-nvdimm@...ts.01.org>,
        kernel-team <kernel-team@....com>
Subject: Re: [PATCH v1 2/6] fs: use on-stack-bio if backing device has
 BDI_CAP_SYNC capability

Hi Christoph,

On Fri, Aug 11, 2017 at 12:46:15PM +0200, Christoph Hellwig wrote:
> On Wed, Aug 09, 2017 at 08:06:24PM -0700, Dan Williams wrote:
> > I like it, but do you think we should switch to sbvec[<constant>] to
> > preclude pathological cases where nr_pages is large?
> 
> Yes, please.

Still, I don't understand how sbvec[nr_pages] with on-stack bio in
do_mpage_readpage can help the performance.

IIUC, do_mpage_readpage works with page-base. IOW, it passes just one
page, not multiple pages so if we use on-stack bio, we just add *a page*
via bio_add_page and submit the bio before the function returning.

So, rather than sbvec[1], why de we need sbvec[nr_pages]?

Please, let me open my eyes. :)

Thanks.