[dpdk-dev] [PATCH v3 0/9] virtio/vhost: Add MTU feature support
Maxime Coquelin
maxime.coquelin at redhat.com
Thu Mar 30 13:34:10 CEST 2017
Hi Lei,
On 03/28/2017 07:39 AM, Yao, Lei A wrote:
>
>> -----Original Message-----
>> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Maxime Coquelin
>> Sent: Monday, March 13, 2017 12:34 AM
>> To: aconole at redhat.com; sodey at sonusnet.com;
>> yuanhan.liu at linux.intel.com; Tan, Jianfeng <jianfeng.tan at intel.com>;
>> thomas.monjalon at 6wind.com; dev at dpdk.org
>> Cc: Maxime Coquelin <maxime.coquelin at redhat.com>
>> Subject: [dpdk-dev] [PATCH v3 0/9] virtio/vhost: Add MTU feature support
>>
>> This series adds support to new Virtio's MTU feature[1]. The MTU
>> value is set via QEMU parameters.
>>
>> If the feature is negotiated (i.e supported by both host and guest,
>> and valid MTU value is set in QEMU via its host_mtu parameter), QEMU
>> shares the configured MTU value throught dedicated Vhost protocol
>> feature.
>>
>> On vhost side, the value is stored in the virtio_net structure, and
>> made available to the application thanks to new vhost lib's
>> rte_vhost_get_mtu() function.
>>
>> To be able to set eth_dev's MTU value at the right time, i.e. to call
>> rte_vhost_get_mtu() just after Virtio features have been negotiated
>> and before the device is really started, a new vhost flag has been
>> introduced (VIRTIO_DEV_READY), because the VIRTIO_DEV_RUNNING flag is
>> set too late (after .new_device() ops is called).
>>
>> Regarding valid MTU values, the maximum MTU value accepted on vhost
>> side is 65535 bytes, as defined in Virtio Spec and supported in
>> Virtio-net Kernel driver. But in Virtio PMD, current maximum frame
>> size is 9728 bytes (~9700 bytes MTU). So maximum MTU size accepted in
>> Virtio PMD is the minimum between ~9700 bytes and host's MTU.
>>
>> Finally, this series also adds MTU value printing in testpmd's
>> "show port info" command when non-zero.
>>
>> This series target v17.05 release.
>>
>> Cheers,
>> Maxime
>>
>> [1]: https://lists.oasis-open.org/archives/virtio-dev/201609/msg00128.html
>>
>> Changes since v1:
>> -----------------
>> * Rebased on top of v17.02
>> * Virtio PMD: ensure MTU value is valid before ack'ing the feature (Aaron)
>> * Vhost lib/PMD: Remove MTU setting API/op (Yuanhan)
>>
>> Changes since v2:
>> -----------------
>> * Update release notes (Thomas)
>> * s/rte_vhost_mtu_get/rte_vhost_get_mtu/ (Yuanhan)
>> * Use %"PRIu64" instead of %lu (Yuanhan)
>> * Add rte_vhost_get_mtu in rte_vhost_version.map
>>
>> Maxime Coquelin (9):
>> vhost: Enable VIRTIO_NET_F_MTU feature
>> vhost: vhost-user: Add MTU protocol feature support
>> vhost: Add new ready status flag
>> vhost: Add API to get MTU value
>> vhost: export MTU value
>> net/vhost: Fill rte_eth_dev's MTU property
>> net/virtio: Add MTU feature support
>> doc: announce Virtio and Vhost MTU support
>> app/testpmd: print MTU value in show port info
>>
>> app/test-pmd/config.c | 5 ++++
>> doc/guides/nics/features/virtio.ini | 1 +
>> doc/guides/rel_notes/release_17_05.rst | 8 ++++++
>> drivers/net/vhost/rte_eth_vhost.c | 2 ++
>> drivers/net/virtio/virtio_ethdev.c | 45
>> ++++++++++++++++++++++++++++++++--
>> drivers/net/virtio/virtio_ethdev.h | 3 ++-
>> drivers/net/virtio/virtio_pci.h | 3 +++
>> lib/librte_vhost/rte_vhost_version.map | 7 ++++++
>> lib/librte_vhost/rte_virtio_net.h | 15 ++++++++++++
>> lib/librte_vhost/vhost.c | 22 ++++++++++++++++-
>> lib/librte_vhost/vhost.h | 9 ++++++-
>> lib/librte_vhost/vhost_user.c | 44 +++++++++++++++++++++++++++---
>> ---
>> lib/librte_vhost/vhost_user.h | 5 +++-
>> 13 files changed, 156 insertions(+), 13 deletions(-)
>>
>> --
>> 2.9.3
> Hi, Maxime
>
> If I want have a try for this MTU function, is there any specific requirement for the settings?
> Such as the qemu version, kernel version or any others? Looks like this feature are very new
> in Qemu and linux side.
> Thanks a lot!
Sorry, for the delay.
You need QEMU v2.9.0-rc at least, as this is not in v2.8?
Regarding the Kernel, there are no version dependency for the host.
For the guest, this is only if you use the Kernel virtio-net driver.
In this case, your guest Kernel should be at least v4.10.
Regards,
Maxime
More information about the dev
mailing list