public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Dave Chinner <david@fromorbit.com>
Cc: Eric Sandeen <sandeen@sandeen.net>,
	Christoph Hellwig <hch@lst.de>,
	linux-xfs@vger.kernel.org, Ian Kent <raven@themaw.net>
Subject: Re: [PATCH 3/7] xfs: remove the m_readio_log field from struct xfs_mount
Date: Sat, 26 Oct 2019 07:47:22 +0200	[thread overview]
Message-ID: <20191026054722.GA14648@lst.de> (raw)
In-Reply-To: <20191025204329.GF4614@dread.disaster.area>

On Sat, Oct 26, 2019 at 07:43:29AM +1100, Dave Chinner wrote:
> NFSv2 had a maximum client IO size of 8kB and writes were
> synchronous. The Irix NFS server had some magic in it (enabled by
> the filesystem wsync mount option) that allowed clients to have two
> sequential 8k writes in flight at once, allowing XFS to optimise for
> 16KB write IOs instead of the normal default of 64kB. This
> optimisation was the reason that, at the time (early-mid 90s), SGI
> machines had double the NFS write throughput of any other Unix
> systems.
> 
> I'm surprised we still support NFSv2 at all in this day and age - I
> suspect we should just kill NFSv2 altogether. We need to keep the
> wsync option around for HA systems serving files to NFS and CIFS
> clients, but the 8kB IO size optimisations can certainly die....

Last time I talked to the NFS folks there still were some very obscure
v2 use cases left (embedded devices that can't be upgraded).

But yeah, I'll add a patch to stop overriding rsize/wsize with the
sync option.

  reply	other threads:[~2019-10-26  5:47 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-25 17:40 decruft misc mount related code Christoph Hellwig
2019-10-25 17:40 ` [PATCH 1/7] xfs: cleanup stat blksize reporting Christoph Hellwig
2019-10-25 17:56   ` Eric Sandeen
2019-10-25 17:40 ` [PATCH 2/7] xfs: remove the unused m_readio_blocks field from struct xfs_mount Christoph Hellwig
2019-10-25 18:02   ` Eric Sandeen
2019-10-25 17:40 ` [PATCH 3/7] xfs: remove the m_readio_log " Christoph Hellwig
2019-10-25 18:26   ` Eric Sandeen
2019-10-25 20:43     ` Dave Chinner
2019-10-26  5:47       ` Christoph Hellwig [this message]
2019-10-25 17:40 ` [PATCH 4/7] xfs: remove the dsunit and dswidth variables in xfs_parseargs Christoph Hellwig
2019-10-25 18:29   ` Eric Sandeen
2019-10-25 17:40 ` [PATCH 5/7] xfs: remove the iosizelog variable " Christoph Hellwig
2019-10-25 18:35   ` Eric Sandeen
2019-10-25 19:03     ` Eric Sandeen
2019-10-26  5:47       ` Christoph Hellwig
2019-10-25 17:40 ` [PATCH 6/7] xfs: clean up setting m_readio_* / m_writeio_* Christoph Hellwig
2019-10-25 19:18   ` Eric Sandeen
2019-10-25 20:53     ` Dave Chinner
2019-10-27  2:23       ` Eric Sandeen
2019-10-25 17:40 ` [PATCH 7/7] xfs: reverse the polarity of XFS_MOUNT_COMPAT_IOSIZE Christoph Hellwig
2019-10-25 19:24   ` Eric Sandeen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20191026054722.GA14648@lst.de \
    --to=hch@lst.de \
    --cc=david@fromorbit.com \
    --cc=linux-xfs@vger.kernel.org \
    --cc=raven@themaw.net \
    --cc=sandeen@sandeen.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox