[dpdk-dev] [PATCH v2 0/5] Introduce virtual PMD for Hyper-V/Azure platforms

Stephen Hemminger stephen at networkplumber.org
Sat Dec 23 03:06:58 CET 2017


Why does this need to be a PMD?
Maybe we need some platform infrastructure?
My definition of PMD is it can send and receive

On Dec 22, 2017 10:01, "Adrien Mazarguil" <adrien.mazarguil at 6wind.com>
wrote:

> Virtual machines hosted by Hyper-V/Azure platforms are fitted with
> simplified virtual network devices named NetVSC that are used for fast
> communication between VM to VM, VM to hypervisor, and the outside.
>
> They appear as standard system netdevices to user-land applications, the
> main difference being they are implemented on top of VMBUS [1] instead of
> emulated PCI devices.
>
> While this reads like a case for a standard DPDK PMD, there is more to it.
>
> To accelerate outside communication, NetVSC devices as they appear in a VM
> can be paired with physical SR-IOV virtual function (VF) devices owned by
> that same VM [2]. Both netdevices share the same MAC address in that case.
>
> When paired, egress and most of the ingress traffic flow through the VF
> device, while part of it (e.g. multicasts, hypervisor control data) still
> flows through NetVSC. Moreover VF devices are not retained and disappear
> during VM migration; from a VM standpoint, they can be hot-plugged anytime
> with NetVSC acting as a fallback.
>
> Running DPDK applications in such a context involves driving VF devices
> using their dedicated PMDs in a vendor-independent fashion (to benefit from
> maximum performance without writing dedicated code) while simultaneously
> listening to NetVSC and handling the related hot-plug events.
>
> This new virtual PMD (referred to as "vdev_netvsc" from this point on)
> automatically coordinates the Hyper-V/Azure-specific management part
> described above by relying on vendor-specific, failsafe and tap PMDs to
> expose a single consolidated Ethernet device usable directly by existing
> applications.
>
>          .------------------.
>          | DPDK application |
>          `--------+---------'
>                   |
>            .------+------.
>            | DPDK ethdev |
>            `------+------'       Control
>                   |                 |
>      .------------+------------.    v    .-----------------.
>      |       failsafe PMD      +---------+ vdev_netvsc PMD |
>      `--+-------------------+--'         `-----------------'
>         |                   |
>         |          .........|.........
>         |          :        |        :
>    .----+----.     :   .----+----.   :
>    | tap PMD |     :   | any PMD |   :
>    `----+----'     :   `----+----'   : <-- Hot-pluggable
>         |          :        |        :
>  .------+-------.  :  .-----+-----.  :
>  | NetVSC-based |  :  | SR-IOV VF |  :
>  |   netdevice  |  :  |   device  |  :
>  `--------------'  :  `-----------'  :
>                    :.................:
>
> Note this diagram differs from that of the original RFC [3], with
> vdev_netvsc no longer acting as a data plane layer.
>
> This initial version of the driver only works in whitelist mode. Users have
> to provide the --vdev net_vdev_netvsc EAL option at least once to trigger
> it.
>
> Subsequent work will add support for blacklist mode based on automatic
> detection of the host environment.
>
> [1] http://dpdk.org/ml/archives/dev/2017-January/054165.html
> [2] https://docs.microsoft.com/en-us/windows-hardware/drivers/
> network/overview-of-hyper-v
> [3] http://dpdk.org/ml/archives/dev/2017-November/082339.html
>
> v2 changes:
>
> - Renamed driver from "hyperv" to "vdev_netvsc". This change covers
>   documentation and symbols prefix.
> - Driver is now tagged EXPERIMENTAL.
> - Replaced ether_addr_from_str() with a basic sscanf() call.
> - Removed debugging code (memset() poisoning).
> - Fixed hyperv_iface_is_netvsc()'s buffer allocation according to comments.
> - Removed hyperv_basename().
> - Discarded unused variables through __rte_unused.
> - Added separate but necessary free() bugfix for failsafe PMD.
> - Added file descriptor input support to failsafe PMD.
> - Replaced temporary bash execution; failsafe now reads device definitions
>   directly through a pipe without an intermediate bash one-liner.
> - Expanded DEBUG/INFO/WARN/ERROR() macros as PMD_DRV_LOG().
> - Added dynamic log type (pmd.vdev_netvsc).
> - Modified initialization code to probe devices immediately during startup.
> - Fixed several snprintf() return value checks ("ret >= sizeof(foo)" is
> more
>   appropriate than "ret >= sizeof(foo) - 1").
>
> Adrien Mazarguil (5):
>   net/failsafe: fix invalid free
>   net/failsafe: add "fd" parameter
>   net/vdev_netvsc: introduce Hyper-V platform driver
>   net/vdev_netvsc: implement core functionality
>   net/vdev_netvsc: add "force" parameter
>
>  MAINTAINERS                                     |   6 +
>  config/common_base                              |   5 +
>  config/common_linuxapp                          |   1 +
>  doc/guides/nics/fail_safe.rst                   |   9 +
>  doc/guides/nics/features/vdev_netvsc.ini        |  12 +
>  doc/guides/nics/index.rst                       |   1 +
>  doc/guides/nics/vdev_netvsc.rst                 | 116 +++
>  drivers/net/Makefile                            |   1 +
>  drivers/net/failsafe/failsafe_args.c            |  88 ++-
>  drivers/net/failsafe/failsafe_private.h         |   3 +
>  drivers/net/vdev_netvsc/Makefile                |  58 ++
>  .../vdev_netvsc/rte_pmd_vdev_netvsc_version.map |   4 +
>  drivers/net/vdev_netvsc/vdev_netvsc.c           | 722 +++++++++++++++++++
>  mk/rte.app.mk                                   |   1 +
>  14 files changed, 1025 insertions(+), 2 deletions(-)
>  create mode 100644 doc/guides/nics/features/vdev_netvsc.ini
>  create mode 100644 doc/guides/nics/vdev_netvsc.rst
>  create mode 100644 drivers/net/vdev_netvsc/Makefile
>  create mode 100644 drivers/net/vdev_netvsc/rte_
> pmd_vdev_netvsc_version.map
>  create mode 100644 drivers/net/vdev_netvsc/vdev_netvsc.c
>
> --
> 2.11.0
>


More information about the dev mailing list