virtualization.lists.linux-foundation.org archive mirror
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Parav Pandit <parav@nvidia.com>
Cc: "virtio-dev@lists.oasis-open.org"
	<virtio-dev@lists.oasis-open.org>,
	"sridhar.samudrala@intel.com" <sridhar.samudrala@intel.com>,
	"jesse.brandeburg@intel.com" <jesse.brandeburg@intel.com>,
	Gavi Teitz <gavi@nvidia.com>,
	"virtualization@lists.linux-foundation.org"
	<virtualization@lists.linux-foundation.org>,
	"stephen@networkplumber.org" <stephen@networkplumber.org>,
	"loseweigh@gmail.com" <loseweigh@gmail.com>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	"kuba@kernel.org" <kuba@kernel.org>,
	"davem@davemloft.net" <davem@davemloft.net>,
	Gavin Li <gavinl@nvidia.com>
Subject: Re: [PATCH v5 2/2] virtio-net: use mtu size as buffer length for big packets
Date: Wed, 7 Sep 2022 15:23:57 -0400	[thread overview]
Message-ID: <20220907152156-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <PH0PR12MB54811F1234CB7822F47DD1B9DC419@PH0PR12MB5481.namprd12.prod.outlook.com>

On Wed, Sep 07, 2022 at 07:18:06PM +0000, Parav Pandit wrote:
> 
> > From: Michael S. Tsirkin <mst@redhat.com>
> > Sent: Wednesday, September 7, 2022 3:12 PM
> 
> > > Because of shallow queue of 16 entries deep.
> > 
> > but why is the queue just 16 entries?
> I explained the calculation in [1] about 16 entries.
> 
> [1] https://lore.kernel.org/netdev/PH0PR12MB54812EC7F4711C1EA4CAA119DC419@PH0PR12MB5481.namprd12.prod.outlook.com/
> 
> > does the device not support indirect?
> >
> Yes, indirect feature bit is disabled on the device.

OK that explains it.

> > because with indirect you get 256 entries, with 16 s/g each.
> > 
> Sure. I explained below that indirect comes with 7x memory cost that is not desired.
> (Ignored the table memory allocation cost and extra latency).

Oh sure, it's a waste. I wonder what effect does the patch have
on bandwidth with indirect enabled though.


> Hence don't want to enable indirect in this scenario.
> This optimization also works with indirect with smaller indirect table.
> 
> > 
> > > With driver turn around time to repost buffers, device is idle without any
> > RQ buffers.
> > > With this improvement, device has 85 buffers instead of 16 to receive
> > packets.
> > >
> > > Enabling indirect in device can help at cost of 7x higher memory per VQ in
> > the guest VM.

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

  reply	other threads:[~2022-09-07 19:24 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20220901021038.84751-1-gavinl@nvidia.com>
     [not found] ` <20220901021038.84751-3-gavinl@nvidia.com>
2022-09-07  2:17   ` [virtio-dev] [PATCH v5 2/2] virtio-net: use mtu size as buffer length for big packets Jason Wang
2022-09-07  5:31   ` Michael S. Tsirkin
     [not found]     ` <0355d1e4-a3cf-5b16-8c7f-b39b1ec14ade@nvidia.com>
2022-09-07  9:26       ` Michael S. Tsirkin
2022-09-07 14:08         ` Parav Pandit via Virtualization
2022-09-07 14:29           ` Michael S. Tsirkin
2022-09-07 14:33             ` Parav Pandit via Virtualization
2022-09-07 14:40               ` Michael S. Tsirkin
2022-09-07 16:12                 ` Parav Pandit via Virtualization
2022-09-07 18:15                   ` Michael S. Tsirkin
2022-09-07 19:06                     ` Parav Pandit via Virtualization
2022-09-07 19:11                       ` Michael S. Tsirkin
2022-09-07 19:18                         ` Parav Pandit via Virtualization
2022-09-07 19:23                           ` Michael S. Tsirkin [this message]
2022-09-07 19:27                             ` Parav Pandit via Virtualization
2022-09-07 19:36                               ` Michael S. Tsirkin
2022-09-07 19:37                                 ` Michael S. Tsirkin
2022-09-07 19:54                                   ` Parav Pandit via Virtualization
2022-09-07 19:51                                 ` Parav Pandit via Virtualization
2022-09-07 21:39                                   ` [virtio-dev] " Si-Wei Liu
2022-09-07 22:11                                     ` Parav Pandit via Virtualization
2022-09-07 22:57                                       ` Si-Wei Liu
2022-09-22  9:26                                   ` Michael S. Tsirkin
2022-09-22 10:07                                     ` Parav Pandit via Virtualization
2022-09-07 20:04                                 ` Parav Pandit via Virtualization
2022-09-22  9:35   ` Michael S. Tsirkin
2022-09-22 10:04     ` Parav Pandit via Virtualization
2022-09-22 10:14       ` Michael S. Tsirkin
2022-09-22 10:29         ` Parav Pandit via Virtualization
     [not found]         ` <20220922053458.66f31136@kernel.org>
2022-10-05 10:29           ` Parav Pandit via Virtualization

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=20220907152156-mutt-send-email-mst@kernel.org \
    --to=mst@redhat.com \
    --cc=davem@davemloft.net \
    --cc=gavi@nvidia.com \
    --cc=gavinl@nvidia.com \
    --cc=jesse.brandeburg@intel.com \
    --cc=kuba@kernel.org \
    --cc=loseweigh@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=parav@nvidia.com \
    --cc=sridhar.samudrala@intel.com \
    --cc=stephen@networkplumber.org \
    --cc=virtio-dev@lists.oasis-open.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
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).