qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Anthony Liguori <anthony@codemonkey.ws>
To: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 2/2 v2] Direct IDE I/O
Date: Mon, 03 Dec 2007 12:06:00 -0600	[thread overview]
Message-ID: <47544588.10700@codemonkey.ws> (raw)
In-Reply-To: <20071203170800.GC3797@implementation>

Samuel Thibault wrote:
> Anthony Liguori, le Mon 03 Dec 2007 09:54:47 -0600, a écrit :
>   
>> Have you done any performance testing?  Buffered IO should absolutely 
>> beat direct IO simply because buffered IO allows writes to complete 
>> before they actually hit disk.
>>     
>
> Since qemu can use the aio interface, that shouldn't matter.
>   

Well, let's separate a few things.  QEMU uses posix-aio which uses 
threads and normal read/write operations.  It also limits the number of 
threads that aio uses to 1 which effectively makes everything 
synchronous anyway.

But it still doesn't matter.  When you issue a write() on an O_DIRECT 
fd, the write does not complete until the data has made it's way to 
disk.  The guest can still run if you're using O_NONBLOCK but the IDE 
device will not submit another IO request until you complete the DMA 
operation.

The SCSI device supports multiple outstanding operations but it's 
limited to 16 but you'll never see more than one request at a time in 
QEMU currently because of the limitation to a single thread.

Regards,

Anthony Liguori

> Samuel
>
>
>
>   

  parent reply	other threads:[~2007-12-03 18:06 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-12-03 10:09 [Qemu-devel] [PATCH 0/2 v2] Open disk images with O_DIRECT Laurent Vivier
2007-12-03 10:09 ` [Qemu-devel] [PATCH 1/2 v2] Add "cache" parameter to "-drive" Laurent Vivier
2007-12-03 10:09   ` [Qemu-devel] [PATCH 2/2 v2] Direct IDE I/O Laurent Vivier
2007-12-03 10:23     ` Fabrice Bellard
2007-12-03 10:30       ` Laurent Vivier
2007-12-03 11:40         ` Markus Hitter
2007-12-03 15:39           ` Paul Brook
2007-12-03 19:26             ` Samuel Thibault
2007-12-03 15:54         ` Anthony Liguori
2007-12-03 17:08           ` Samuel Thibault
2007-12-03 17:17             ` Paul Brook
2007-12-03 17:49               ` Jamie Lokier
2007-12-03 18:08                 ` Anthony Liguori
2007-12-03 18:40                   ` Jamie Lokier
2007-12-03 18:06             ` Anthony Liguori [this message]
2007-12-03 19:10               ` Laurent Vivier
2007-12-03 19:16                 ` Paul Brook
2007-12-03 21:36                   ` Anthony Liguori
2007-12-04 12:49                     ` Gerd Hoffmann
2007-12-04 13:02                       ` Laurent Vivier
2007-12-04  8:13                   ` Laurent Vivier
2007-12-03 21:13                 ` Gerd Hoffmann
2007-12-03 21:23                   ` Samuel Thibault
2007-12-03 21:38                   ` Anthony Liguori
2007-12-04 13:21                     ` Gerd Hoffmann
2007-12-04 15:03                       ` Anthony Liguori
2007-12-04 16:18                         ` Gerd Hoffmann
2007-12-05 14:47                           ` Anthony Liguori
2007-12-03 19:14               ` Paul Brook
2007-12-03 19:00           ` Laurent Vivier
2007-12-03 11:14       ` Johannes Schindelin

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=47544588.10700@codemonkey.ws \
    --to=anthony@codemonkey.ws \
    --cc=qemu-devel@nongnu.org \
    /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).