public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Leon Romanovsky <leon@kernel.org>
To: "Longpeng (Mike,
	Cloud Infrastructure Service Product Dept.)" 
	<longpeng2@huawei.com>
Cc: bhelgaas@google.com, linux-pci@vger.kernel.org,
	linux-kernel@vger.kernel.org, jianjay.zhou@huawei.com,
	zhuangshengen@huawei.com, arei.gonglei@huawei.com,
	yechuan@huawei.com, huangzhichao@huawei.com, xiehong@huawei.com
Subject: Re: [RFC 0/4] pci/sriov: support VFs dynamic addition
Date: Mon, 14 Nov 2022 09:04:43 +0200	[thread overview]
Message-ID: <Y3Hoi4zGFY4Fz1l4@unreal> (raw)
In-Reply-To: <0b2202bf-18d3-b288-9605-279208165080@huawei.com>

On Sun, Nov 13, 2022 at 09:47:12PM +0800, Longpeng (Mike, Cloud Infrastructure Service Product Dept.) wrote:
> Hi leon,
> 
> 在 2022/11/12 0:39, Leon Romanovsky 写道:
> > On Fri, Nov 11, 2022 at 10:27:18PM +0800, Longpeng(Mike) wrote:
> > > From: Longpeng <longpeng2@huawei.com>
> > > 
> > > We can enable SRIOV and add VFs by /sys/bus/pci/devices/..../sriov_numvfs, but
> > > this operation needs to spend lots of time if there has a large amount of VFs.
> > > For example, if the machine has 10 PFs and 250 VFs per-PF, enable all the VFs
> > > concurrently would cost about 200-250ms. However most of them are not need to be
> > > used at the moment, so we can enable SRIOV first but add VFs on demand.
> > 
> > It is unclear what took 200-250ms, is it physical VF creation or bind of
> > the driver to these VFs?
> > 
> It is neither. In our test, we already created physical VFs before, so we
> skipped the 100ms waiting when writing PCI_SRIOV_CTRL. And our driver only
> probes PF, it just returns an error if the function is VF.

It means that you didn't try sriov_drivers_autoprobe. Once it is set to
true, It won't even try to probe VFs.

> 
> The hotspot is the sriov_add_vfs (but no driver probe in fact) which is a
> long procedure. Each step costs only a little, but the total cost is not
> acceptable in some time-sensitive cases.

This is also cryptic to me. In standard SR-IOV deployment, all VFs are
created and configured while operator booted the machine with sriov_drivers_autoprobe
set to false. Once this machine is ready, VFs are assigned to relevant VMs/users
through orchestration SW (IMHO, it is supported by all orchestration SW). 

And only last part (assigning to users) is time-sensitive operation.

> 
> What’s more, the sriov_add_vfs adds the VFs of a PF one by one. So we can
> mostly support 10 concurrent calls if there has 10 PFs.

I wondered, are you using real HW? or QEMU SR-IOV? What is your server
that supports such large number of VFs?

BTW, Your change will probably break all SR-IOV devices in the market as
they rely on PCI subsystem to have VFs ready and configured.

Thanks

  reply	other threads:[~2022-11-14  7:04 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-11 14:27 [RFC 0/4] pci/sriov: support VFs dynamic addition Longpeng(Mike)
2022-11-11 14:27 ` [RFC 1/4] pci/sriov: extract sriov_numvfs common helper Longpeng(Mike)
2022-11-11 14:27 ` [RFC 2/4] pci/sriov: add vf_bitmap to mark the vf id allocation Longpeng(Mike)
2022-11-11 14:27 ` [RFC 3/4] pci/sriov: add sriov_numfs_no_scan interface Longpeng(Mike)
2022-11-11 14:27 ` [RFC 4/4] pci/sriov: add sriov_scan_vf_id interface Longpeng(Mike)
2022-11-11 16:39 ` [RFC 0/4] pci/sriov: support VFs dynamic addition Leon Romanovsky
2022-11-13 13:47   ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)
2022-11-14  7:04     ` Leon Romanovsky [this message]
2022-11-14 12:38       ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)
2022-11-14 13:09         ` Leon Romanovsky
2022-11-14 14:06           ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)
2022-11-14 14:20             ` Leon Romanovsky
2022-11-15  1:38               ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)
2022-11-15  1:50               ` Oliver O'Halloran
2022-11-15  8:32                 ` Leon Romanovsky
2022-11-15  9:36                   ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)
2022-11-15 10:02                     ` Leon Romanovsky
2022-11-15 10:27                       ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)
2022-11-15 12:49                   ` Oliver O'Halloran
2022-11-15  2:06             ` Oliver O'Halloran
2022-11-16  0:52               ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)
2022-11-11 23:07 ` Bjorn Helgaas
2022-11-13 13:49   ` Longpeng (Mike, Cloud Infrastructure Service Product Dept.)

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=Y3Hoi4zGFY4Fz1l4@unreal \
    --to=leon@kernel.org \
    --cc=arei.gonglei@huawei.com \
    --cc=bhelgaas@google.com \
    --cc=huangzhichao@huawei.com \
    --cc=jianjay.zhou@huawei.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=longpeng2@huawei.com \
    --cc=xiehong@huawei.com \
    --cc=yechuan@huawei.com \
    --cc=zhuangshengen@huawei.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