On 28/06/2019 15:39, Ian Stokes wrote: > On 6/27/2019 12:12 PM, Kevin Traynor wrote: >> vhost tx retries can provide some mitigation against >> dropped packets due to a temporarily slow guest/limited queue >> size for an interface, but on the other hand when a system >> is fully loaded those extra cycles retrying could mean >> packets are dropped elsewhere. >> >> Up to now max vhost tx retries have been hardcoded, which meant >> no tuning and no way to disable for debugging to see if extra >> cycles spent retrying resulted in rx drops on some other >> interface. >> >> Add an option to change the max retries, with a value of >> 0 effectively disabling vhost tx retries. >> >> Signed-off-by: Kevin Traynor <ktray...@redhat.com> >> --- >> Documentation/topics/dpdk/vhost-user.rst | 25 ++++++++++++++++ >> lib/netdev-dpdk.c | 36 +++++++++++++++++++++--- >> vswitchd/vswitch.xml | 10 +++++++ >> 3 files changed, 67 insertions(+), 4 deletions(-) >> >> diff --git a/Documentation/topics/dpdk/vhost-user.rst >> b/Documentation/topics/dpdk/vhost-user.rst >> index 3caa88231..d8508d8f8 100644 >> --- a/Documentation/topics/dpdk/vhost-user.rst >> +++ b/Documentation/topics/dpdk/vhost-user.rst >> @@ -392,4 +392,29 @@ The default value is ``false``. >> .. _dpdk-testpmd: > I'm still testing this but a minor suggestion below. > > in patch 2 of the series you introduce a section 'vhost tx retries' and > you specify the behavior and number of default retries. It would be nice > to link that section to this section below to flag to users that the > default can be changed. It should make navigation easy between the two > points as currently they are separated in the doc by other configuration > sections. > > Something like > > --- a/Documentation/topics/dpdk/vhost-user.rst > +++ b/Documentation/topics/dpdk/vhost-user.rst > @@ -82,8 +82,10 @@ When sending a batch of packets to a vhost-user or > vhost-user-client interface, > it may happen that some but not all of the packets in the batch are > able to be > sent to the guest. This is often because there is not enough free > descriptors > in the virtqueue for all the packets in the batch to be sent. In this case > -there will be a retry, with a default maximum of 8 occurring. If at any > time no > -packets can be sent, it may mean the guest is not accepting packets, so > there > +there will be a retry, with a default maximum of 8 occurring, for > information > +regarding how this can be configured please refer to > +`vhost-user-client tx retries config`_. If at any time no packets can > be sent, > +it may mean the guest is not accepting packets, so there > are no (more) retries. >
sure, I can add some sort of link between the sections > Ian > _______________________________________________ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev
