DPDK patches and discussions
 help / color / mirror / Atom feed
From: Matan Azrad <matan@mellanox.com>
To: Maxime Coquelin <maxime.coquelin@redhat.com>,
	"dev@dpdk.org" <dev@dpdk.org>
Cc: Slava Ovsiienko <viacheslavo@mellanox.com>,
	Shahaf Shuler <shahafs@mellanox.com>,
	Xiao Wang <xiao.w.wang@intel.com>,
	Tiwei Bie <tiwei.bie@intel.com>
Subject: Re: [dpdk-dev] [PATCH] [RFC] vhost: inroduce operation to get vDPA queue stats
Date: Fri, 10 Apr 2020 14:15:23 +0000	[thread overview]
Message-ID: <AM0PR0502MB40196DD266D12BCE0FC13E73D2DE0@AM0PR0502MB4019.eurprd05.prod.outlook.com> (raw)
In-Reply-To: <5e9ad602-e60c-1098-9171-19948be86c2a@redhat.com>



From: Maxime Coquelin
> Sent: Friday, April 10, 2020 5:04 PM
> To: Matan Azrad <matan@mellanox.com>; dev@dpdk.org
> Cc: Slava Ovsiienko <viacheslavo@mellanox.com>; Shahaf Shuler
> <shahafs@mellanox.com>; Xiao Wang <xiao.w.wang@intel.com>; Tiwei Bie
> <tiwei.bie@intel.com>
> Subject: Re: [PATCH] [RFC] vhost: inroduce operation to get vDPA queue
> stats
> 
> 
> 
> On 4/10/20 3:54 PM, Matan Azrad wrote:
> > Hi Maxime
> >
> > From: Maxime Coquelin
> >> Hi Matan,
> >>
> >> On 3/16/20 4:12 PM, Matan Azrad wrote:
> >>> The vDPA device offloads all the datapath of the vhost device to the
> >>> HW device.
> >>>
> >>> In order to expose to the user traffic information this patch
> >>> introduce new API to get traffic statistics per virtio queue.
> >>>
> >>> The statistics are taken directly from the vDPA driver managing the
> >>> HW device.
> >>
> >> Shouldn't the stats be collected from the PF like for "legacy" SR-IOV
> >> via the sysfs?
> >> Something like:
> >> /sys/class/net/<PF name>/device/sriov/<VF ID>/stats/
> >
> > Yes, probably, statistics can be taken from sysfs.
> >
> > But we are introducing a DPDK API to get the virtio statistics.
> > So, driver which can take statistics from there can take them from sysfs in
> the driver operation.
> >
> > In addition, the API includes per virtio queue statistics and in virtio data-
> path terms, it is very helpful for debuggability.
> 
> Ok, the per queue stats is a valid point for queue debug.
> 
> Even if we add this API, I still think we should ensure the stats will be
> available through the sysfs, so that it makes Kubernetes integration seemless
> wrt to legacy SR-IOV[0].

For Mellanox legacy SRIOV the sysfs stats you mentioned are available.

> Also, it would provide a single way to get statistics for both DPDK and Kernel
> vDPA.

Sure.

> Thanks,
> Maxime
> 
> >
> >
> >> Regards,
> >> Maxime
> >


  reply	other threads:[~2020-04-10 14:15 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-16 15:12 Matan Azrad
2020-04-09 14:34 ` Maxime Coquelin
2020-04-10 13:54   ` Matan Azrad
2020-04-10 14:04     ` Maxime Coquelin
2020-04-10 14:15       ` Matan Azrad [this message]
2020-04-10 14:20         ` Maxime Coquelin

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=AM0PR0502MB40196DD266D12BCE0FC13E73D2DE0@AM0PR0502MB4019.eurprd05.prod.outlook.com \
    --to=matan@mellanox.com \
    --cc=dev@dpdk.org \
    --cc=maxime.coquelin@redhat.com \
    --cc=shahafs@mellanox.com \
    --cc=tiwei.bie@intel.com \
    --cc=viacheslavo@mellanox.com \
    --cc=xiao.w.wang@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).