All of lore.kernel.org
 help / color / mirror / Atom feed
From: Trond Myklebust <trondmy@primarydata.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>,
	"xfs@oss.sgi.com" <xfs@oss.sgi.com>
Subject: Re: [PATCH 10/12] NFS: Do not serialise O_DIRECT reads and writes
Date: Wed, 15 Jun 2016 15:45:37 +0000	[thread overview]
Message-ID: <E6003989-FB5F-4FE8-A161-C6E52E031091@primarydata.com> (raw)
In-Reply-To: <20160615151422.GA28557@infradead.org>


[-- Attachment #1.1: Type: text/plain, Size: 1902 bytes --]



On 6/15/16, 11:14, "Christoph Hellwig" <hch@infradead.org> wrote:

>On Wed, Jun 15, 2016 at 03:09:23PM +0000, Trond Myklebust wrote:
>> As I said earlier, the problem with that is you end up artificially serialising buffered reads and buffered writes.
>
>If you actually want to be Posix compiant you need to serialize buffered
>reads against buffererd writes - it's just that most Linux file systems
>happen to get this wrong.
>
>> ??? The reads only need a shared lock in order to protect the I/O mode from flipping to O_DIRECT (and relying on page locks to protect against buffered writes).
>
>Which strictly speaking is not enough, although as said above most
>Linux filesystems get this wrong.  If you indeed want to keep that
>(incorrect) behavior you need another lock.  It's defintively not
>"simpler", though.
>

Serialisation is not mandatory in POSIX:

http://pubs.opengroup.org/onlinepubs/9699919799/functions/write.html

“Writes can be serialized with respect to other reads and writes. If a read() of file data can be proven (by any means) to occur after a write() of the data, it must reflect that write(), even if the calls are made by different processes. A similar requirement applies to multiple write operations to the same file position. This is needed to guarantee the propagation of data from write() calls to subsequent read() calls. This requirement is particularly significant for networked file systems, where some caching schemes violate these semantics.”

Disclaimer

The information contained in this communication from the sender is confidential. It is intended solely for use by the recipient and others authorized to receive it. If you are not the recipient, you are hereby notified that any disclosure, copying, distribution or taking action in relation of the contents of this information is strictly prohibited and may be unlawful.

[-- Attachment #1.2: Type: text/html, Size: 2351 bytes --]

[-- Attachment #2: Type: text/plain, Size: 121 bytes --]

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

WARNING: multiple messages have this Message-ID (diff)
From: Trond Myklebust <trondmy@primarydata.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>,
	"xfs@oss.sgi.com" <xfs@oss.sgi.com>
Subject: Re: [PATCH 10/12] NFS: Do not serialise O_DIRECT reads and writes
Date: Wed, 15 Jun 2016 15:45:37 +0000	[thread overview]
Message-ID: <E6003989-FB5F-4FE8-A161-C6E52E031091@primarydata.com> (raw)
In-Reply-To: <20160615151422.GA28557@infradead.org>
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  reply	other threads:[~2016-06-15 15:45 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-14 19:05 [PATCH 01/12] NFS: Don't flush caches for a getattr that races with writeback Trond Myklebust
2016-06-14 19:05 ` [PATCH 02/12] NFS: Cache access checks more aggressively Trond Myklebust
2016-06-14 19:05   ` [PATCH 03/12] NFS: Cache aggressively when file is open for writing Trond Myklebust
2016-06-14 19:05     ` [PATCH 04/12] NFS: Kill NFS_INO_NFS_INO_FLUSHING: it is a performance killer Trond Myklebust
2016-06-14 19:05       ` [PATCH 05/12] NFS: writepage of a single page should not be synchronous Trond Myklebust
2016-06-14 19:05         ` [PATCH 06/12] NFS: Don't hold the inode lock across fsync() Trond Myklebust
2016-06-14 19:05           ` [PATCH 07/12] NFS: Don't enable deep stack recursion when doing memory reclaim Trond Myklebust
2016-06-14 19:05             ` [PATCH 08/12] NFS: Fix O_DIRECT verifier problems Trond Myklebust
2016-06-14 19:05               ` [PATCH 09/12] NFS: Ensure we reset the write verifier 'committed' value on resend Trond Myklebust
2016-06-14 19:05                 ` [PATCH 10/12] NFS: Do not serialise O_DIRECT reads and writes Trond Myklebust
2016-06-14 19:05                   ` [PATCH 11/12] NFS: Don't count O_DIRECT reads in the inode->i_dio_count Trond Myklebust
2016-06-14 19:05                     ` [PATCH 12/12] NFS: Clean up nfs_direct_complete() Trond Myklebust
2016-06-15  7:16                     ` [PATCH 11/12] NFS: Don't count O_DIRECT reads in the inode->i_dio_count Christoph Hellwig
2016-06-15 14:36                       ` Trond Myklebust
2016-06-15 14:41                         ` Christoph Hellwig
2016-06-15 14:50                           ` Trond Myklebust
2016-06-15 14:53                             ` Christoph Hellwig
2016-06-15  7:13                   ` [PATCH 10/12] NFS: Do not serialise O_DIRECT reads and writes Christoph Hellwig
2016-06-15 14:29                     ` Trond Myklebust
2016-06-15 14:48                       ` Christoph Hellwig
2016-06-15 14:48                         ` Christoph Hellwig
2016-06-15 14:52                         ` Trond Myklebust
2016-06-15 14:52                           ` Trond Myklebust
2016-06-15 14:56                           ` Christoph Hellwig
2016-06-15 14:56                             ` Christoph Hellwig
2016-06-15 15:09                             ` Trond Myklebust
2016-06-15 15:09                               ` Trond Myklebust
2016-06-15 15:14                               ` Christoph Hellwig
2016-06-15 15:14                                 ` Christoph Hellwig
2016-06-15 15:45                                 ` Trond Myklebust [this message]
2016-06-15 15:45                                   ` Trond Myklebust
2016-06-16  9:12                                   ` Christoph Hellwig
2016-06-16  9:12                                     ` Christoph Hellwig
2016-06-15  7:09             ` [PATCH 07/12] NFS: Don't enable deep stack recursion when doing memory reclaim Christoph Hellwig
2016-06-15  7:08           ` [PATCH 06/12] NFS: Don't hold the inode lock across fsync() Christoph Hellwig
2016-06-15 14:47             ` Trond Myklebust
2016-06-15 14:54               ` Christoph Hellwig
2016-06-17  1:11     ` [PATCH 03/12] NFS: Cache aggressively when file is open for writing Oleg Drokin
2016-06-17 14:01       ` Trond Myklebust

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=E6003989-FB5F-4FE8-A161-C6E52E031091@primarydata.com \
    --to=trondmy@primarydata.com \
    --cc=hch@infradead.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=xfs@oss.sgi.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.