linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff Moyer <jmoyer@redhat.com>
To: Zach Brown <zach.brown@oracle.com>
Cc: linux-fsdevel@vger.kernel.org, linux-aio@kvack.org,
	linux-kernel@vger.kernel.org
Subject: Re: [RFC 0/5] dio: clean up completion phase of direct_io_worker()
Date: Wed, 06 Sep 2006 10:57:09 -0400	[thread overview]
Message-ID: <x49hczl11ru.fsf@segfault.boston.devel.redhat.com> (raw)
In-Reply-To: <20060905235732.29630.3950.sendpatchset@tetsuo.zabbo.net> (Zach Brown's message of "Tue,  5 Sep 2006 16:57:32 -0700 (PDT)")

==> Regarding [RFC 0/5] dio: clean up completion phase of direct_io_worker(); Zach Brown <zach.brown@oracle.com> adds:

zach.brown> There have been a lot of bugs recently due to the way
zach.brown> direct_io_worker() tries to decide how to finish direct IO
zach.brown> operations.  In the worst examples it has failed to call
zach.brown> aio_complete() at all (hang) or called it too many times
zach.brown> (oops).

zach.brown> This set of patches cleans up the completion phase with the
zach.brown> goal of removing the complexity that lead to these bugs.  We
zach.brown> end up with one path that calculates the result of the
zach.brown> operation after all off the bios have completed.  We decide
zach.brown> when to generate a result of the operation using that path
zach.brown> based on the final release of a refcount on the dio structure.

[...]

zach.brown> I hoped to get some feedback (and maybe volunteers for
zach.brown> testing!) by sending the patches out before waiting for the
zach.brown> stress tests.

This all looks good, the code is much easier to follow.  What do you think
about making dio->result an unsigned quantity?  It should never be negative
now that there is an io_error field.

ACK.

Jeff

  parent reply	other threads:[~2006-09-06 14:57 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-09-05 23:57 [RFC 0/5] dio: clean up completion phase of direct_io_worker() Zach Brown
2006-09-05 23:57 ` [RFC 1/5] dio: centralize completion in dio_complete() Zach Brown
2006-09-05 23:57 ` [RFC 2/5] dio: call blk_run_address_space() once per op Zach Brown
2006-09-05 23:57 ` [RFC 3/5] dio: formalize bio counters as a dio reference count Zach Brown
2006-09-05 23:57 ` [RFC 4/5] dio: remove duplicate bio wait code Zach Brown
2006-09-05 23:57 ` [RFC 5/5] dio: only call aio_complete() after returning -EIOCBQUEUED Zach Brown
2006-09-06  4:35 ` bogofilter ate 3/5 Zach Brown
2006-09-06  5:00   ` Willy Tarreau
2006-09-06  7:22   ` Mike Galbraith
2006-09-08 22:16   ` Matthias Andree
2006-09-06  7:36 ` [RFC 0/5] dio: clean up completion phase of direct_io_worker() Suparna Bhattacharya
2006-09-06 16:36   ` Zach Brown
2006-09-06 14:57 ` Jeff Moyer [this message]
2006-09-06 16:46   ` Zach Brown
2006-09-06 18:13     ` Jeff Moyer
  -- strict thread matches above, loose matches on Subject: below --
2006-09-21 12:24 Veerendra Chandrappa
     [not found] <OFBE544A3C.7C1B2C64-ON652571F0.003C21B6-652571F0.003C2DF3@in.ibm.com>
2006-09-21 18:38 ` Zach Brown

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=x49hczl11ru.fsf@segfault.boston.devel.redhat.com \
    --to=jmoyer@redhat.com \
    --cc=linux-aio@kvack.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=zach.brown@oracle.com \
    /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;
as well as URLs for NNTP newsgroup(s).