From: "Michael S. Tsirkin" <mst@redhat.com>
To: Hannes Reinecke <hare@suse.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
Linux Virtualization <virtualization@lists.linux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
qemu-devel <qemu-devel@nongnu.org>,
Rusty Russell <rusty@rustcorp.com.au>,
Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>,
Christoph Hellwig <chellwig@redhat.com>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Subject: Re: virtio scsi host draft specification, v3
Date: Sun, 12 Jun 2011 10:51:41 +0300 [thread overview]
Message-ID: <20110612075140.GB11941@redhat.com> (raw)
In-Reply-To: <4DF21447.6090005@suse.de>
On Fri, Jun 10, 2011 at 02:55:35PM +0200, Hannes Reinecke wrote:
> >Device operation: request queues
> >--------------------------------
> >
> >The driver queues requests to an arbitrary request queue, and they are
> >used by the device on that same queue.
> >
> What about request ordering?
> If requests are placed on arbitrary queues you'll inevitably run on
> locking issues to ensure strict request ordering.
> I would add here:
>
> If a device uses more than one queue it is the responsibility of the
> device to ensure strict request ordering.
Maybe I misunderstand - how can this be the responsibility of
the device if the device does not get the information about
the original ordering of the requests?
For example, if the driver is crazy enough to put
all write requests on one queue and all barriers
on another one, how is the device supposed to ensure
ordering?
--
MST
WARNING: multiple messages have this Message-ID (diff)
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <chellwig@redhat.com>,
Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
Rusty Russell <rusty@rustcorp.com.au>,
qemu-devel <qemu-devel@nongnu.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Paolo Bonzini <pbonzini@redhat.com>,
Linux Virtualization <virtualization@lists.linux-foundation.org>
Subject: Re: [Qemu-devel] virtio scsi host draft specification, v3
Date: Sun, 12 Jun 2011 10:51:41 +0300 [thread overview]
Message-ID: <20110612075140.GB11941@redhat.com> (raw)
In-Reply-To: <4DF21447.6090005@suse.de>
On Fri, Jun 10, 2011 at 02:55:35PM +0200, Hannes Reinecke wrote:
> >Device operation: request queues
> >--------------------------------
> >
> >The driver queues requests to an arbitrary request queue, and they are
> >used by the device on that same queue.
> >
> What about request ordering?
> If requests are placed on arbitrary queues you'll inevitably run on
> locking issues to ensure strict request ordering.
> I would add here:
>
> If a device uses more than one queue it is the responsibility of the
> device to ensure strict request ordering.
Maybe I misunderstand - how can this be the responsibility of
the device if the device does not get the information about
the original ordering of the requests?
For example, if the driver is crazy enough to put
all write requests on one queue and all barriers
on another one, how is the device supposed to ensure
ordering?
--
MST
next prev parent reply other threads:[~2011-06-12 7:51 UTC|newest]
Thread overview: 91+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-07 13:43 virtio scsi host draft specification, v3 Paolo Bonzini
2011-06-07 13:43 ` [Qemu-devel] " Paolo Bonzini
2011-06-08 23:28 ` Rusty Russell
2011-06-08 23:28 ` [Qemu-devel] " Rusty Russell
2011-06-08 23:28 ` Rusty Russell
2011-06-09 6:59 ` Paolo Bonzini
2011-06-09 6:59 ` Paolo Bonzini
2011-06-09 6:59 ` [Qemu-devel] " Paolo Bonzini
2011-06-10 11:33 ` Rusty Russell
2011-06-10 11:33 ` Rusty Russell
2011-06-10 11:33 ` [Qemu-devel] " Rusty Russell
2011-06-10 12:14 ` Stefan Hajnoczi
2011-06-10 12:14 ` [Qemu-devel] " Stefan Hajnoczi
2011-06-10 12:22 ` Paolo Bonzini
2011-06-10 12:22 ` Paolo Bonzini
2011-06-10 12:22 ` [Qemu-devel] " Paolo Bonzini
2011-06-10 12:14 ` Stefan Hajnoczi
2011-06-08 23:28 ` Rusty Russell
2011-06-10 12:55 ` Hannes Reinecke
2011-06-10 12:55 ` Hannes Reinecke
2011-06-10 12:55 ` [Qemu-devel] " Hannes Reinecke
2011-06-10 14:35 ` Paolo Bonzini
2011-06-10 14:35 ` [Qemu-devel] " Paolo Bonzini
2011-06-14 8:39 ` Hannes Reinecke
2011-06-14 8:39 ` [Qemu-devel] " Hannes Reinecke
2011-06-14 15:53 ` Stefan Hajnoczi
2011-06-14 15:53 ` Stefan Hajnoczi
2011-06-14 15:53 ` [Qemu-devel] " Stefan Hajnoczi
2011-06-29 8:33 ` Paolo Bonzini
2011-06-29 8:33 ` [Qemu-devel] " Paolo Bonzini
2011-06-29 9:39 ` Stefan Hajnoczi
2011-06-29 9:39 ` Stefan Hajnoczi
2011-06-29 9:39 ` [Qemu-devel] " Stefan Hajnoczi
2011-06-29 10:07 ` Christoph Hellwig
2011-06-29 10:07 ` [Qemu-devel] " Christoph Hellwig
2011-06-29 10:23 ` Hannes Reinecke
2011-06-29 10:23 ` Hannes Reinecke
2011-06-29 10:23 ` [Qemu-devel] " Hannes Reinecke
2011-06-29 10:27 ` Christoph Hellwig
2011-06-29 10:27 ` [Qemu-devel] " Christoph Hellwig
2011-06-29 10:27 ` Christoph Hellwig
2011-06-29 10:07 ` Christoph Hellwig
2011-07-01 6:41 ` Paolo Bonzini
2011-07-01 6:41 ` Paolo Bonzini
2011-07-01 6:41 ` [Qemu-devel] " Paolo Bonzini
2011-07-01 7:14 ` Hannes Reinecke
2011-07-01 7:14 ` [Qemu-devel] " Hannes Reinecke
2011-07-01 8:35 ` Paolo Bonzini
2011-07-01 8:35 ` [Qemu-devel] " Paolo Bonzini
2011-07-04 13:38 ` Hai Dong,Li
2011-07-04 13:38 ` [Qemu-devel] " Hai Dong,Li
2011-07-04 14:22 ` Stefan Hajnoczi
2011-07-04 14:22 ` [Qemu-devel] " Stefan Hajnoczi
2011-07-04 14:22 ` Stefan Hajnoczi
2011-07-04 13:38 ` Hai Dong,Li
2011-07-01 8:35 ` Paolo Bonzini
2011-07-01 7:14 ` Hannes Reinecke
2011-06-29 8:33 ` Paolo Bonzini
2011-06-14 8:39 ` Hannes Reinecke
2011-06-10 14:35 ` Paolo Bonzini
2011-06-12 7:51 ` Michael S. Tsirkin [this message]
2011-06-12 7:51 ` [Qemu-devel] " Michael S. Tsirkin
2011-06-14 15:30 ` Hannes Reinecke
2011-06-14 15:30 ` [Qemu-devel] " Hannes Reinecke
2011-06-29 10:00 ` Christoph Hellwig
2011-06-29 10:00 ` Christoph Hellwig
2011-06-29 10:00 ` [Qemu-devel] " Christoph Hellwig
2011-06-14 15:30 ` Hannes Reinecke
2011-06-29 8:23 ` Paolo Bonzini
2011-06-29 8:23 ` [Qemu-devel] " Paolo Bonzini
2011-06-29 8:46 ` Michael S. Tsirkin
2011-06-29 8:46 ` [Qemu-devel] " Michael S. Tsirkin
2011-06-29 8:46 ` Michael S. Tsirkin
2011-06-29 10:03 ` Christoph Hellwig
2011-06-29 10:03 ` [Qemu-devel] " Christoph Hellwig
2011-06-29 10:06 ` Paolo Bonzini
2011-06-29 10:06 ` Paolo Bonzini
2011-06-29 10:06 ` [Qemu-devel] " Paolo Bonzini
2011-06-29 10:31 ` Michael S. Tsirkin
2011-06-29 10:31 ` Michael S. Tsirkin
2011-06-29 10:31 ` [Qemu-devel] " Michael S. Tsirkin
2011-06-29 10:35 ` Paolo Bonzini
2011-06-29 10:35 ` Paolo Bonzini
2011-06-29 10:35 ` [Qemu-devel] " Paolo Bonzini
2011-06-29 10:03 ` Christoph Hellwig
2011-06-29 8:23 ` Paolo Bonzini
2011-06-29 10:01 ` Christoph Hellwig
2011-06-29 10:01 ` Christoph Hellwig
2011-06-29 10:01 ` [Qemu-devel] " Christoph Hellwig
2011-06-12 7:51 ` Michael S. Tsirkin
-- strict thread matches above, loose matches on Subject: below --
2011-06-07 13:43 Paolo Bonzini
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=20110612075140.GB11941@redhat.com \
--to=mst@redhat.com \
--cc=chellwig@redhat.com \
--cc=hare@suse.de \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rusty@rustcorp.com.au \
--cc=stefanha@linux.vnet.ibm.com \
--cc=virtualization@lists.linux-foundation.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 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.