The exiting explanation didn't tell user the conntrack capability
and user may be unaware of the stateful feature of OVS.
Signed-off-by: Han Zhou
---
FAQ.md | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/FAQ.md b/FAQ.md
index 420e40e..776b8f6 100644
[snip]
> > +
> > +static int
> > netdev_dpdk_vhost_construct(struct netdev *netdev)
> > {
> > struct netdev_dpdk *dev = netdev_dpdk_cast(netdev);
> > @@ -904,7 +1051,7 @@ netdev_dpdk_vhost_construct(struct netdev
> *netdev)
> > /* 'name' is appended to 'vhost_sock_dir' and used to
On Tue, Oct 18, 2016 at 10:03 PM, Stephen Finucane
wrote:
> Signed-off-by: Stephen Finucane
> ---
> Makefile.am | 2 +-
> README-native-tunneling.md | 85 ---
> README-native-tunneling.rst | 176
The vHost PMD allows vHost User ports to be controlled by the
librte_ether API, like physical 'dpdk' ports and IVSHM 'dpdkr' ports.
This commit integrates this PMD into OVS and removes direct calls to the
librte_vhost DPDK library.
This commit requires DPDK v16.11 functionality that isn't
Previous: http://openvswitch.org/pipermail/dev/2016-October/080902.html
This RFC patchset adds support for DPDK 16.11 (rc2) and the vHost PMD.
Thanks to a lack of API changes that affect OVS, the DPDK 16.11 support
patch is mostly documentation updates, with no code changes necessary.
The vHost
This commit announces support for DPDK 16.11. Compaitibilty with DPDK
v16.07 is not broken yet thanks to no code changes being needed for the
upgrade. This may change soon however with the addition of subsequent
commits that integrate 16.11-only features.
Signed-off-by: Ciara Loftus
On 28.10.2016 11:36, Loftus, Ciara wrote:
>>
>> 'dev->requested_{rxq,txq}_size' and 'dev->{rxq,txq}_size' are
>> relevant only for DPDK_DEV_ETH devices and should be skipped
>> in 'netdev_dpdk_get_config()' for other ports.
>>
>> CC: Ciara Loftus
>> Fixes: b685696b8c81
>
>When building with DPDK, and using xmalloc() to get a new packet,
>field mbuf of the packet will not be initialized, but it's very important for
>DPDK port when copying the data to DPDK mbuf, because if ol_flags
>and other info are random values, DPDK driver may hang.
>
>Signed-off-by: Michael
>
This patch didn't apply cleanly either; 'apply -3' resolved the issue, but I
recommend rebasing to HEAD of master for next version.
Other than that, just some minor cosmetic comments inline.
>When a packet is from DPDK source, and it contains
>multipule segments, data_len is not equal to the
>
> 'dev->requested_{rxq,txq}_size' and 'dev->{rxq,txq}_size' are
> relevant only for DPDK_DEV_ETH devices and should be skipped
> in 'netdev_dpdk_get_config()' for other ports.
>
> CC: Ciara Loftus
> Fixes: b685696b8c81 ("netdev-dpdk: Allow configurable queue sizes for
>
>Currently, when doing packet copy, lots of DPDK mbuf's info
>will be missed, like packet type, ol_flags, etc.
>Those information is very important for DPDK to do
>packets processing.
>
>Signed-off-by: Michael Qiu
>Signed-off-by: Jijiang Liu
>---
>
Still not a full review.
Comments inline.
Bet regards, Ilya Maximets.
On 28.10.2016 16:03, Ciara Loftus wrote:
> The vHost PMD allows vHost User ports to be controlled by the
> librte_ether API, like physical 'dpdk' ports and IVSHM 'dpdkr' ports.
> This commit integrates this PMD into OVS and
>When doing packet clone, if packet source is from DPDK driver,
>multi-segment must be considered, and copy the segment's
>data one by one.
>
>Signed-off-by: Michael Qiu
>Signed-off-by: Jijiang Liu
>---
> lib/dp-packet.c | 25 ++---
>
The non-pmd thread static_tx_qid is assumed to be equal to the highest
core ID + 1. The function dp_netdev_del_pmds_on_numa() invalidates
this assumption by re-distributing the static_tx_qid:s on all pmd and
non-pmd threads of the "other" numa.
There might be a number of unwanted effects due to
>
>Currently, one packet is only copied to one segment
>in function dpdk_do_tx_copy(), this could be an issue
>when a jumboframe comes, especially for multipile segments.
Typos - 'jumboframe', 'multipile'
This patch doesn't apply cleanly - change 'mbufs' to 'pkts' as previously
described to
>
> > I'll post few comments to v4 here.
> >
> > > static int
> > > +dpdk_attach_vhost_pmd(struct netdev_dpdk *dev, int mode)
> > > +{
> > > +char *devargs;
> > > +int err = 0;
> > > +uint8_t port_no = 0;
> > > +uint32_t driver_id = -1;
> > > +
> > > +if
when boot vswitchd, we can pass log_file and log level as parameters, but it
seems these params only take effect for vswitchd itself, but not dpdk.
I can only see EAL logs in the console when launch vswitchd, but I can't see
any EAL logs in vswitchd.log and /var/log/syslog.
so is there any
'dpdkvdev' ports are for use with virtual DPDK Poll Mode Drivers eg.
null, pcap.
To add a DPDK vdev, a valid 'dpdk-devargs' must be set. The format
expected is 'eth_' where 'x' is a number between 0 and
RTE_MAX_ETHPORTS -1.
For example to add a 'dpdkvdev' port that uses the 'null' DPDK PMD
'dpdk' ports no longer have naming restrictions. Now, instead
of specifying the dpdk port ID as part of the name, the PCI
address of the device must be specified via the 'dpdk-devargs'
option. eg.
ovs-vsctl add-port br0 my-port
ovs-vsctl set Interface my-port type=dpdk
ovs-vsctl set Interface
This RFC series consists of 3 patches.
1. Port Hotplug (Mauricio Vasquez) (v8)
Previous: http://openvswitch.org/pipermail/dev/2016-July/075350.html
2. Arbitrary Port Naming (Ciara Loftus) (v4)
Previous: http://openvswitch.org/pipermail/dev/2016-July/075385.html
3. Experimental vDev PMD (Ciara
In order to use dpdk ports in ovs they have to be bound to a DPDK
compatible driver before ovs is started.
This patch adds the possibility to hotplug (or hot-unplug) a device
after ovs has been started. The implementation adds two appctl commands:
netdev-dpdk/port-attach and
Hi Cascardo,
Thank you for replying. Please excuse the disclaimer at the bottom of the email
as it is automatically added by the company mail server.
Thanks for clarifying the revalidator error that I saw.
Regarding the datapath ID, I've turned on debug logging for ovsdb-server and
On 28.10.2016 13:44, Loftus, Ciara wrote:
>>
>>> I'll post few comments to v4 here.
>>>
static int
+dpdk_attach_vhost_pmd(struct netdev_dpdk *dev, int mode)
+{
+char *devargs;
+int err = 0;
+uint8_t port_no = 0;
+uint32_t driver_id = -1;
+
23 matches
Mail list logo