public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Chuck Lever <cel@citi.umich.edu>
To: Andrew Morton <akpm@osdl.org>
Cc: cel@netapp.com, linux-kernel@vger.kernel.org, trond.myklebust@fys.uio.no
Subject: Re: [PATCH 1/6] nfs: "open code" the NFS direct write rescheduler
Date: Fri, 19 May 2006 14:37:03 -0400	[thread overview]
Message-ID: <446E104F.3060900@citi.umich.edu> (raw)
In-Reply-To: <20060519111054.1842ccfb.akpm@osdl.org>

Andrew Morton wrote:
> Chuck Lever <cel@netapp.com> wrote:
>> +	 * Prevent I/O completion while we're still rescheduling
>> +	 */
>> +	dreq->outstanding++;
>> +
> 
> No locking.
> 
>>  	dreq->count = 0;
>> +	list_for_each(pos, &dreq->rewrite_list) {
>> +		struct nfs_write_data *data =
>> +			list_entry(dreq->rewrite_list.next, struct nfs_write_data, pages);
>> +
>> +		spin_lock(&dreq->lock);
>> +		dreq->outstanding++;
>> +		spin_unlock(&dreq->lock);
> 
> Locking.
> 
> Deliberate?

Yes.  At the top of the loop, there is no outstanding I/O, so no locking 
is needed while updating "outstanding."  Inside the loop, we've 
dispatched some I/O against "dreq" so locking is needed to ensure 
outstanding is updated properly.

-- 
corporate:	cel at netapp dot com
personal:	chucklever at bigfoot dot com

  reply	other threads:[~2006-05-19 18:37 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-05-19 17:56 [PATCH 0/6] Support scatter/gather I/O in NFS direct I/O path Chuck Lever
2006-05-19 18:00 ` [PATCH 1/6] nfs: "open code" the NFS direct write rescheduler Chuck Lever
2006-05-19 18:10   ` Andrew Morton
2006-05-19 18:37     ` Chuck Lever [this message]
2006-05-19 18:46       ` Andrew Morton
2006-05-19 18:56         ` Chuck Lever
2006-05-19 18:00 ` [PATCH 2/6] nfs: remove user_addr and user_count from nfs_direct_req Chuck Lever
2006-05-19 18:00 ` [PATCH 3/6] nfs: Eliminate nfs_get_user_pages() Chuck Lever
2006-05-19 18:17   ` Andrew Morton
2006-05-19 19:18     ` Chuck Lever
2006-05-19 18:00 ` [PATCH 4/6] nfs: alloc nfs_read/write_data as direct I/O is scheduled Chuck Lever
2006-05-19 18:00 ` [PATCH 5/6] nfs: check all iov segments for correct memory access rights Chuck Lever
2006-05-19 18:22   ` Andrew Morton
2006-05-19 18:46     ` Chuck Lever
2006-05-19 19:36     ` Chuck Lever
2006-05-19 20:07       ` Andrew Morton
2006-05-19 18:25   ` Badari Pulavarty
2006-05-22 11:27   ` Andi Kleen
2006-05-19 18:00 ` [PATCH 6/6] nfs: Support vector I/O throughout the NFS direct I/O path Chuck Lever

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=446E104F.3060900@citi.umich.edu \
    --to=cel@citi.umich.edu \
    --cc=akpm@osdl.org \
    --cc=cel@netapp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=trond.myklebust@fys.uio.no \
    /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