From: "Bly, Mike" <mbly@ciena.com>
To: "dev@dpdk.org" <dev@dpdk.org>
Subject: [dpdk-dev] dpdk-devbind --status and PCI-PT
Date: Fri, 17 Apr 2020 17:57:55 +0000 [thread overview]
Message-ID: <DM6PR04MB4714BFE82E85B32868F402A3CFD90@DM6PR04MB4714.namprd04.prod.outlook.com> (raw)
Hello,
I am looking to know if folks were aware that running "dpdk-debind -status" on a host displays both NICs in host space as well as those "owned" by a VM via PCI-PT where that VM is internally running a DPDK enabled application. Per below there is no discernable difference indicated as to actual DPDK ownership for host vs guest designation.
In this example the first two NICs are being accessed via PCI-PT by the VM, which itself bound the NICs to its internal DPDK driver. The last two NICs listed are owned by the host DPDK application. The end result is there is no way via use of the dpdk-devind script to know which NICs are whose within the host space application. Is this known/intentional?
Example to show observation:
Prior to starting the VM, you can see we have two sets of NICs, some bound to host application DPDK drivers and some bound to host kernel drivers.
# dpdk-devbind --status-dev net
Network devices using DPDK-compatible driver
============================================
0000:05:00.0 'I350 Gigabit Network Connection 1521' drv=vfio-pci unused=uio_pci_generic
0000:05:00.1 'I350 Gigabit Network Connection 1521' drv=vfio-pci unused=uio_pci_generic
Network devices using kernel driver
===================================
0000:03:00.0 'Ethernet Connection X552/X557-AT 10GBASE-T 15ad' if=eth0 drv=ixgbe unused=vfio-pci,uio_pci_generic
0000:03:00.1 'Ethernet Connection X552/X557-AT 10GBASE-T 15ad' if=eth1 drv=ixgbe unused=vfio-pci,uio_pci_generic
After starting the VM up, its internal DPDK based application grabs and binds its two NICs to its own DPDK driver(s). After this occurs, the host view is now as shown below.
# dpdk-devbind --status-dev net
Network devices using DPDK-compatible driver
============================================
0000:03:00.0 'Ethernet Connection X552/X557-AT 10GBASE-T 15ad' drv=vfio-pci unused=uio_pci_generic
0000:03:00.1 'Ethernet Connection X552/X557-AT 10GBASE-T 15ad' drv=vfio-pci unused=uio_pci_generic
0000:05:00.0 'I350 Gigabit Network Connection 1521' drv=vfio-pci unused=uio_pci_generic
0000:05:00.1 'I350 Gigabit Network Connection 1521' drv=vfio-pci unused=uio_pci_generic
Regards,
Mike
reply other threads:[~2020-04-17 17:58 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=DM6PR04MB4714BFE82E85B32868F402A3CFD90@DM6PR04MB4714.namprd04.prod.outlook.com \
--to=mbly@ciena.com \
--cc=dev@dpdk.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).