qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Marc-André Lureau" <marcandre.lureau@redhat.com>,
	"Sebastien Boeuf" <sebastien.boeuf@intel.com>,
	"Cathy Zhang" <cathy.zhang@intel.com>,
	"Stefan Hajnoczi" <stefanha@redhat.com>,
	"Michael S. Tsirkin" <mst@redhat.com>
Subject: [Qemu-devel] [PATCH] docs: clarify multiqueue vs multiple virtqueues
Date: Mon, 24 Jun 2019 10:13:04 +0100	[thread overview]
Message-ID: <20190624091304.666-1-stefanha@redhat.com> (raw)

The vhost-user specification does not explain when
VHOST_USER_PROTOCOL_F_MQ must be implemented.  This may lead
implementors of vhost-user masters to believe that this protocol feature
is required for any device that has multiple virtqueues.  That would be
a mistake since existing vhost-user slaves offer multiple virtqueues but
do not advertise VHOST_USER_PROTOCOL_F_MQ.

For example, a vhost-net device with one rx/tx queue pair is not
multiqueue.  The slave does not need to advertise
VHOST_USER_PROTOCOL_F_MQ.  Therefore the master must assume it has these
virtqueues and cannot rely on askingt the slave how many virtqueues
exist.

Extend the specification to explain the different between true
multiqueue and regular devices with a fixed virtqueue layout.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
Based-on: <20190621094005.4134-1-stefanha@redhat.com>
---
 docs/interop/vhost-user.rst | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst
index 5750668aba..7827b710aa 100644
--- a/docs/interop/vhost-user.rst
+++ b/docs/interop/vhost-user.rst
@@ -324,6 +324,15 @@ must support changing some configuration aspects on the fly.
 Multiple queue support
 ----------------------
 
+Many devices have a fixed number of virtqueues.  In this case the master
+already knows the number of available virtqueues without communicating with the
+slave.
+
+Some devices do not have a fixed number of virtqueues.  Instead the maximum
+number of virtqueues is chosen by the slave.  The number can depend on host
+resource availability or slave implementation details.  Such devices are called
+multiple queue devices.
+
 Multiple queue support allows the slave to advertise the maximum number of
 queues.  This is treated as a protocol extension, hence the slave has to
 implement protocol features first. The multiple queues feature is supported
@@ -339,6 +348,14 @@ queue in the sent message to identify a specified queue.
 The master enables queues by sending message ``VHOST_USER_SET_VRING_ENABLE``.
 vhost-user-net has historically automatically enabled the first queue pair.
 
+Slaves should always implement the ``VHOST_USER_PROTOCOL_F_MQ`` protocol
+feature, even for devices with a fixed number of virtqueues, since it is simple
+to implement and offers a degree of introspection.
+
+Masters must not rely on the ``VHOST_USER_PROTOCOL_F_MQ`` protocol feature for
+devices with a fixed number of virtqueues.  Only true multiqueue devices
+require this protocol feature.
+
 Migration
 ---------
 
-- 
2.21.0



WARNING: multiple messages have this Message-ID (diff)
From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
	"Stefan Hajnoczi" <stefanha@redhat.com>,
	"Marc-André Lureau" <marcandre.lureau@redhat.com>
Subject: [Qemu-devel] [PULL 01/12] docs: clarify multiqueue vs multiple virtqueues
Date: Thu, 25 Jul 2019 11:31:28 -0400	[thread overview]
Message-ID: <20190624091304.666-1-stefanha@redhat.com> (raw)
Message-ID: <20190725153128.ZS35uro8mbDsHoaQvKgrW5vv1sFCsId6jyapjXIW6AQ@z> (raw)
In-Reply-To: <20190725153059.7313-1-mst@redhat.com>

From: Stefan Hajnoczi <stefanha@redhat.com>

The vhost-user specification does not explain when
VHOST_USER_PROTOCOL_F_MQ must be implemented.  This may lead
implementors of vhost-user masters to believe that this protocol feature
is required for any device that has multiple virtqueues.  That would be
a mistake since existing vhost-user slaves offer multiple virtqueues but
do not advertise VHOST_USER_PROTOCOL_F_MQ.

For example, a vhost-net device with one rx/tx queue pair is not
multiqueue.  The slave does not need to advertise
VHOST_USER_PROTOCOL_F_MQ.  Therefore the master must assume it has these
virtqueues and cannot rely on askingt the slave how many virtqueues
exist.

Extend the specification to explain the different between true
multiqueue and regular devices with a fixed virtqueue layout.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20190624091304.666-1-stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 docs/interop/vhost-user.rst | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/docs/interop/vhost-user.rst b/docs/interop/vhost-user.rst
index 5750668aba..7827b710aa 100644
--- a/docs/interop/vhost-user.rst
+++ b/docs/interop/vhost-user.rst
@@ -324,6 +324,15 @@ must support changing some configuration aspects on the fly.
 Multiple queue support
 ----------------------
 
+Many devices have a fixed number of virtqueues.  In this case the master
+already knows the number of available virtqueues without communicating with the
+slave.
+
+Some devices do not have a fixed number of virtqueues.  Instead the maximum
+number of virtqueues is chosen by the slave.  The number can depend on host
+resource availability or slave implementation details.  Such devices are called
+multiple queue devices.
+
 Multiple queue support allows the slave to advertise the maximum number of
 queues.  This is treated as a protocol extension, hence the slave has to
 implement protocol features first. The multiple queues feature is supported
@@ -339,6 +348,14 @@ queue in the sent message to identify a specified queue.
 The master enables queues by sending message ``VHOST_USER_SET_VRING_ENABLE``.
 vhost-user-net has historically automatically enabled the first queue pair.
 
+Slaves should always implement the ``VHOST_USER_PROTOCOL_F_MQ`` protocol
+feature, even for devices with a fixed number of virtqueues, since it is simple
+to implement and offers a degree of introspection.
+
+Masters must not rely on the ``VHOST_USER_PROTOCOL_F_MQ`` protocol feature for
+devices with a fixed number of virtqueues.  Only true multiqueue devices
+require this protocol feature.
+
 Migration
 ---------
 
-- 
MST



         reply	other threads:[~2019-06-24  9:15 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-25 15:31 [Qemu-devel] [PULL 00/12] virtio, pc: fixes, cleanups Michael S. Tsirkin
2019-06-02 11:42 ` [Qemu-devel] [PATCH] ioapic: kvm: Skip route updates for masked pins Jan Kiszka
2019-06-02 12:10   ` Peter Xu
2019-06-03  6:30     ` Jan Kiszka
2019-06-03  0:36   ` Michael S. Tsirkin
2019-07-21  8:58     ` Jan Kiszka
2019-07-21 10:04       ` Michael S. Tsirkin
2019-07-21 16:55         ` Paolo Bonzini
2019-07-25 15:31   ` [Qemu-devel] [PULL 03/12] " Michael S. Tsirkin
2019-06-24  9:13 ` Stefan Hajnoczi [this message]
2019-06-24 10:19   ` [Qemu-devel] [PATCH] docs: clarify multiqueue vs multiple virtqueues Marc-André Lureau
2019-07-17 10:14   ` Stefan Hajnoczi
2019-07-17 10:35     ` Michael S. Tsirkin
2019-07-25 15:31   ` [Qemu-devel] [PULL 01/12] " Michael S. Tsirkin
2019-07-18 16:14 ` [Qemu-devel] [PATCH v2] i386/acpi: fix gint overflow in crs_range_compare Evgeny Yakovlev
2019-07-18 20:30   ` Michael S. Tsirkin
2019-07-25 15:31   ` [Qemu-devel] [PULL 02/12] " Michael S. Tsirkin
2019-07-19  8:54 ` [Qemu-devel] [PATCH] i386/acpi: show PCI Express bus on pxb-pcie expanders Evgeny Yakovlev
2019-07-19 12:14   ` Igor Mammedov
2019-07-25 15:31   ` [Qemu-devel] [PULL 04/12] " Michael S. Tsirkin
2019-07-25 15:32 ` [Qemu-devel] [PULL 12/12] virtio-balloon: free pbp more aggressively Michael S. Tsirkin
2019-07-26  9:53 ` [Qemu-devel] [PULL 00/12] virtio, pc: fixes, cleanups Peter Maydell
  -- strict thread matches above, loose matches on Subject: below --
2019-07-25 11:36 [Qemu-devel] [PATCH-for-4.1 v4 0/7] virtio-balloon: fixes David Hildenbrand
2019-07-25 11:36 ` [Qemu-devel] [PATCH-for-4.1 v4 1/7] virtio-balloon: Fix wrong sign extension of PFNs David Hildenbrand
2019-07-25 12:36   ` Pankaj Gupta
2019-07-25 11:36 ` [Qemu-devel] [PATCH-for-4.1 v4 2/7] virtio-balloon: Fix QEMU crashes on pagesize > BALLOON_PAGE_SIZE David Hildenbrand
2019-07-25 11:36 ` [Qemu-devel] [PATCH-for-4.1 v4 3/7] virtio-balloon: Simplify deflate with pbp David Hildenbrand
2019-07-25 11:36 ` [Qemu-devel] [PATCH-for-4.1 v4 4/7] virtio-balloon: Better names for offset variables in inflate/deflate code David Hildenbrand
2019-07-25 11:36 ` [Qemu-devel] [PATCH-for-4.1 v4 5/7] virtio-balloon: Rework pbp tracking data David Hildenbrand
2019-07-26  8:08   ` David Gibson
2019-07-25 11:36 ` [Qemu-devel] [PATCH-for-4.1 v4 6/7] virtio-balloon: Use temporary PBP only David Hildenbrand
2019-07-25 11:53   ` Michael S. Tsirkin
2019-07-25 11:56     ` David Hildenbrand
2019-07-25 11:36 ` [Qemu-devel] [PATCH-for-4.1 v4 7/7] virtio-balloon: No need to track subpages for the PBP anymore David Hildenbrand
2019-07-25 15:32   ` [Qemu-devel] [PULL 11/12] virtio-balloon: don't track subpages for the PBP Michael S. Tsirkin
2019-07-26  8:10   ` [Qemu-devel] [PATCH-for-4.1 v4 7/7] virtio-balloon: No need to track subpages for the PBP anymore David Gibson
2019-07-22 13:41 [Qemu-devel] [PATCH-for-4.1 v3 0/6] virtio-balloon: fixes David Hildenbrand
2019-07-22 13:41 ` [Qemu-devel] [PATCH-for-4.1 v3 1/6] virtio-balloon: Fix wrong sign extension of PFNs David Hildenbrand
2019-07-23  2:27   ` David Gibson
2019-07-25 15:31   ` [Qemu-devel] [PULL 05/12] " Michael S. Tsirkin
2019-07-22 13:41 ` [Qemu-devel] [PATCH-for-4.1 v3 2/6] virtio-balloon: Fix QEMU crashes on pagesize > BALLOON_PAGE_SIZE David Hildenbrand
2019-07-25 15:31   ` [Qemu-devel] [PULL 06/12] " Michael S. Tsirkin
2019-07-22 13:41 ` [Qemu-devel] [PATCH-for-4.1 v3 3/6] virtio-balloon: Simplify deflate with pbp David Hildenbrand
2019-07-25 15:32   ` [Qemu-devel] [PULL 07/12] " Michael S. Tsirkin
2019-07-22 13:41 ` [Qemu-devel] [PATCH-for-4.1 v3 4/6] virtio-balloon: Better names for offset variables in inflate/deflate code David Hildenbrand
2019-07-25 15:32   ` [Qemu-devel] [PULL 08/12] " Michael S. Tsirkin
2019-07-22 13:41 ` [Qemu-devel] [PATCH-for-4.1 v3 5/6] virtio-balloon: Rework pbp tracking data David Hildenbrand
2019-07-23  2:54   ` David Gibson
2019-07-23  7:38     ` David Hildenbrand
2019-07-25 15:32   ` [Qemu-devel] [PULL 09/12] " Michael S. Tsirkin
2019-07-22 13:41 ` [Qemu-devel] [PATCH-for-4.1 v3 6/6] virtio-balloon: Use temporary PBP only David Hildenbrand
2019-07-23  3:22   ` David Gibson
2019-07-25 15:32   ` [Qemu-devel] [PULL 10/12] " Michael S. Tsirkin

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=20190624091304.666-1-stefanha@redhat.com \
    --to=stefanha@redhat.com \
    --cc=cathy.zhang@intel.com \
    --cc=marcandre.lureau@redhat.com \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=sebastien.boeuf@intel.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).