Target-devel archive on lore.kernel.org
 help / color / Atom feed
From: Stefan Hajnoczi <stefanha@redhat.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Mike Christie <michael.christie@oracle.com>,
	qemu-devel@nongnu.org, fam@euphon.net,
	linux-scsi@vger.kernel.org, target-devel@vger.kernel.org,
	jasowang@redhat.com, pbonzini@redhat.com,
	virtualization@lists.linux-foundation.org
Subject: Re: [PATCH 00/10] vhost/qemu: thread per IO SCSI vq
Date: Thu, 19 Nov 2020 14:00:10 +0000
Message-ID: <20201119140010.GD838210@stefanha-x1.localdomain> (raw)
In-Reply-To: <20201118044620-mutt-send-email-mst@kernel.org>


[-- Attachment #1: Type: text/plain, Size: 2747 bytes --]

On Wed, Nov 18, 2020 at 04:54:07AM -0500, Michael S. Tsirkin wrote:
> On Tue, Nov 17, 2020 at 01:13:14PM -0600, Mike Christie wrote:
> > On 11/17/20 10:40 AM, Stefan Hajnoczi wrote:
> > > On Thu, Nov 12, 2020 at 05:18:59PM -0600, Mike Christie wrote:
> > >> The following kernel patches were made over Michael's vhost branch:
> > >>
> > >> https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git/log/?h=vhost
> > >>
> > >> and the vhost-scsi bug fix patchset:
> > >>
> > >> https://lore.kernel.org/linux-scsi/20201112170008.GB1555653@stefanha-x1.localdomain/T/#t
> > >>
> > >> And the qemu patch was made over the qemu master branch.
> > >>
> > >> vhost-scsi currently supports multiple queues with the num_queues
> > >> setting, but we end up with a setup where the guest's scsi/block
> > >> layer can do a queue per vCPU and the layers below vhost can do
> > >> a queue per CPU. vhost-scsi will then do a num_queue virtqueues,
> > >> but all IO gets set on and completed on a single vhost-scsi thread.
> > >> After 2 - 4 vqs this becomes a bottleneck.
> > >>
> > >> This patchset allows us to create a worker thread per IO vq, so we
> > >> can better utilize multiple CPUs with the multiple queues. It
> > >> implments Jason's suggestion to create the initial worker like
> > >> normal, then create the extra workers for IO vqs with the
> > >> VHOST_SET_VRING_ENABLE ioctl command added in this patchset.
> > > 
> > > How does userspace find out the tids and set their CPU affinity?
> > > 
> > 
> > When we create the worker thread we add it to the device owner's cgroup,
> > so we end up inheriting those settings like affinity.
> > 
> > However, are you more asking about finer control like if the guest is
> > doing mq, and the mq hw queue is bound to cpu0, it would perform
> > better if we could bind vhost vq's worker thread to cpu0? I think the
> > problem might is if you are in the cgroup then we can't set a specific
> > threads CPU affinity to just one specific CPU. So you can either do
> > cgroups or not.
> 
> Something we wanted to try for a while is to allow userspace
> to create threads for us, then specify which vqs it processes.

Do you mean an interface like a blocking ioctl(vhost_fd,
VHOST_WORKER_RUN) where the vhost processing is done in the context of
the caller's userspace thread?

What is neat about this is that it removes thread configuration from the
kernel vhost code. On the other hand, userspace still needs an interface
indicating which vqs should be processed. Maybe it would even require an
int worker_fd = ioctl(vhost_fd, VHOST_WORKER_CREATE) and then
ioctl(worker_fd, VHOST_WORKER_BIND_VQ, vq_idx)? So then it becomes
complex again...

Stefan

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

  reply index

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-12 23:18 Mike Christie
2020-11-12 23:19 ` [PATCH 1/1] qemu vhost scsi: add VHOST_SET_VRING_ENABLE support Mike Christie
2020-11-17 11:53   ` Stefan Hajnoczi
2020-12-02  9:59   ` Michael S. Tsirkin
2020-12-02 16:05     ` Michael Christie
2020-11-12 23:19 ` [PATCH 01/10] vhost: remove work arg from vhost_work_flush Mike Christie
2020-11-17 13:04   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 02/10] vhost scsi: remove extra flushes Mike Christie
2020-11-17 13:07   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 03/10] vhost poll: fix coding style Mike Christie
2020-11-17 13:07   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 04/10] vhost: support multiple worker threads Mike Christie
2020-11-12 23:19 ` [PATCH 05/10] vhost: poll support support multiple workers Mike Christie
2020-11-17 15:32   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 06/10] vhost scsi: make SCSI cmd completion per vq Mike Christie
2020-11-17 16:04   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 07/10] vhost, vhost-scsi: flush IO vqs then send TMF rsp Mike Christie
2020-11-17 16:05   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 08/10] vhost: move msg_handler to new ops struct Mike Christie
2020-11-17 16:08   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 09/10] vhost: add VHOST_SET_VRING_ENABLE support Mike Christie
2020-11-17 16:14   ` Stefan Hajnoczi
2020-11-12 23:19 ` [PATCH 10/10] vhost-scsi: create a woker per IO vq Mike Christie
2020-11-17 16:40 ` [PATCH 00/10] vhost/qemu: thread per IO SCSI vq Stefan Hajnoczi
2020-11-17 19:13   ` Mike Christie
2020-11-18  9:54     ` Michael S. Tsirkin
2020-11-19 14:00       ` Stefan Hajnoczi [this message]
2020-11-18 11:31     ` Stefan Hajnoczi
2020-11-19 14:46       ` Michael S. Tsirkin
2020-11-19 16:11         ` Mike Christie
2020-11-19 16:24           ` Stefan Hajnoczi
2020-11-19 16:43             ` Mike Christie
2020-11-19 17:08               ` Stefan Hajnoczi
2020-11-20  8:45                 ` Stefan Hajnoczi
2020-11-20 12:31                   ` Michael S. Tsirkin
2020-12-01 12:59                     ` Stefan Hajnoczi
2020-12-01 13:45                       ` Stefano Garzarella
2020-12-01 17:43                         ` Stefan Hajnoczi
2020-12-02 10:35                           ` Stefano Garzarella
2020-11-23 15:17                   ` Stefano Garzarella
2020-11-18  5:17   ` Jason Wang
2020-11-18  6:57     ` Mike Christie
2020-11-18  7:19       ` Mike Christie
2020-11-18  7:54       ` Jason Wang
2020-11-18 20:06         ` Mike Christie
2020-11-19  4:35           ` Jason Wang
2020-11-19 15:49             ` Mike Christie

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=20201119140010.GD838210@stefanha-x1.localdomain \
    --to=stefanha@redhat.com \
    --cc=fam@euphon.net \
    --cc=jasowang@redhat.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=michael.christie@oracle.com \
    --cc=mst@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=target-devel@vger.kernel.org \
    --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

Target-devel archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/target-devel/0 target-devel/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 target-devel target-devel/ https://lore.kernel.org/target-devel \
		target-devel@vger.kernel.org
	public-inbox-index target-devel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.target-devel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git