Message ID | 1499792313-27606-5-git-send-email-billy.o.mahony@intel.com |
---|---|
State | Superseded |
Headers | show |
On 7/11/17, 9:58 AM, "ovs-dev-bounces@openvswitch.org on behalf of Billy O'Mahony" <ovs-dev-bounces@openvswitch.org on behalf of billy.o.mahony@intel.com> wrote:
Signed-off-by: Billy O'Mahony <billy.o.mahony@intel.com>
---
Documentation/howto/dpdk.rst | 31 +++++++++++++++++++++++++++++++
vswitchd/vswitch.xml | 31 +++++++++++++++++++++++++++++++
2 files changed, 62 insertions(+)
diff --git a/Documentation/howto/dpdk.rst b/Documentation/howto/dpdk.rst
index 93248b4..07fb97d 100644
--- a/Documentation/howto/dpdk.rst
+++ b/Documentation/howto/dpdk.rst
@@ -188,6 +188,37 @@ respective parameter. To disable the flow control at tx side, run::
$ ovs-vsctl set Interface dpdk-p0 options:tx-flow-ctrl=false
+Ingress Scheduling
+------------------
+
+The ingress scheduling feature is described in general in
+``ovs-vswitchd.conf.db (5)``.
+
+Interfaces of type ``dpdk`` support ingress scheduling only for
+either ether_type or else a fully specificed combination of src and
+dst ip address and port numbers for TCP or UDP packets.
+
+To prioritize packets for Precision Time Protocol:
+
+ $ ovs-vsctl set Interface dpdk-p0 \
+ other_config:ingress_sched=eth_type=0x88F7
+
+To prioritize UDP packets between specific IP source and destination:
+
+ $ ovs-vsctl set Interface dpdk-p0 \
+ other_config:ingress_sched=udp,ip_src=1.1.1.1,ip_dst=2.2.2.2,\
+ udp_src=11,udp_dst=22
+
+If unsupported ingress scheduling configuration is specified or it cannot be
+applied for any reason a warning message is logged and the Interface operates
+as if no ingress scheduling was configured.
+
+Interfaces of type ``dpdkvhostuserclient``, ``dpdkr`` and ``dpdkvhostuser`` do
+not support ingress scheduling.
+
+Currently only the match fields listed above are supported. No wildcarding of
+fields is supported.
+
I had a previous comment about this in Patch 2.
No wildcarding ?; meaning we need to specify either the 5 tuple exact match and ethertype
or just ethertype ?
pdump
-----
diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
index 0bf986d..299d725 100644
--- a/vswitchd/vswitch.xml
+++ b/vswitchd/vswitch.xml
@@ -2842,6 +2842,37 @@
</column>
</group>
+ <group title="Ingress Scheduling">
+ <p>
+ Packets matching the ingress_sched value are prioritized. This means
+ some combination of:
+ </p>
+ <ul>
+ <li>
+ prioritized packets are forwarded to their destination port before
+ non-prioritized
+ </li>
+ <li>
+ prioritized packets are less likely to be dropped in an overloaded
+ situation than prioritized packets
+ </li>
+ </ul>
+ <p>
+ Ingress scheduling is supported with the best effort of the Interface.
+ It may be dependant on the interface type and it's supporting
+ implementation devices. Different interface types may have different
+ levels of support for the feature and the same interface type attached
+ to different devices (physical NICs or vhost ports, device driver,
+ NIC model) may also offer different levels of support.
+ </p>
+ <column name="other_config" key="ingress_sched">
+ <p>
+ The format of the ingress_sched field is specified in ovs-fields(7) in
+ the ``Matching'' and ``FIELD REFERENCE'' sections.
+ </p>
+ </column>
+ </group>
+
<group title="Bidirectional Forwarding Detection (BFD)">
<p>
BFD, defined in RFC 5880 and RFC 5881, allows point-to-point
--
2.7.4
_______________________________________________
dev mailing list
dev@openvswitch.org
https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.openvswitch.org_mailman_listinfo_ovs-2Ddev&d=DwICAg&c=uilaK90D4TOVoH58JNXRgQ&r=BVhFA09CGX7JQ5Ih-uZnsw&m=cYvXqJfbNkGW_Hacwk5ZCIh4KMLFD_2r5ssimwwuVmA&s=5-_svanz4DDwQU7zBthfGajskziYym_GEuVpDJogHaA&e=
Hi Darrell, Thanks for reviewing. > -----Original Message----- > From: Darrell Ball [mailto:dball@vmware.com] > Sent: Tuesday, July 11, 2017 7:29 PM > To: O Mahony, Billy <billy.o.mahony@intel.com>; dev@openvswitch.org > Subject: Re: [ovs-dev] [RFC v2 4/4] docs: Document ingress scheduling > feature > > > > On 7/11/17, 9:58 AM, "ovs-dev-bounces@openvswitch.org on behalf of Billy > O'Mahony" <ovs-dev-bounces@openvswitch.org on behalf of > billy.o.mahony@intel.com> wrote: > > Signed-off-by: Billy O'Mahony <billy.o.mahony@intel.com> > --- > Documentation/howto/dpdk.rst | 31 > +++++++++++++++++++++++++++++++ > vswitchd/vswitch.xml | 31 +++++++++++++++++++++++++++++++ > 2 files changed, 62 insertions(+) > > diff --git a/Documentation/howto/dpdk.rst > b/Documentation/howto/dpdk.rst > index 93248b4..07fb97d 100644 > --- a/Documentation/howto/dpdk.rst > +++ b/Documentation/howto/dpdk.rst > @@ -188,6 +188,37 @@ respective parameter. To disable the flow control > at tx side, run:: > > $ ovs-vsctl set Interface dpdk-p0 options:tx-flow-ctrl=false > > +Ingress Scheduling > +------------------ > + > +The ingress scheduling feature is described in general in > +``ovs-vswitchd.conf.db (5)``. > + > +Interfaces of type ``dpdk`` support ingress scheduling only for > +either ether_type or else a fully specificed combination of src and > +dst ip address and port numbers for TCP or UDP packets. > + > +To prioritize packets for Precision Time Protocol: > + > + $ ovs-vsctl set Interface dpdk-p0 \ > + other_config:ingress_sched=eth_type=0x88F7 > + > +To prioritize UDP packets between specific IP source and destination: > + > + $ ovs-vsctl set Interface dpdk-p0 \ > + other_config:ingress_sched=udp,ip_src=1.1.1.1,ip_dst=2.2.2.2,\ > + udp_src=11,udp_dst=22 > + > +If unsupported ingress scheduling configuration is specified or it cannot > be > +applied for any reason a warning message is logged and the Interface > operates > +as if no ingress scheduling was configured. > + > +Interfaces of type ``dpdkvhostuserclient``, ``dpdkr`` and ``dpdkvhostuser`` > do > +not support ingress scheduling. > + > +Currently only the match fields listed above are supported. No > wildcarding of > +fields is supported. > + > > I had a previous comment about this in Patch 2. > No wildcarding ?; meaning we need to specify either the 5 tuple exact match > and ethertype or just ethertype ? > [[BO'M]] What I mean is that saying something like nw_src=10.1.0.0/255.255.0.0 is not supported by dpdk_netdev devices (currently). > > > pdump > ----- > > diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml > index 0bf986d..299d725 100644 > --- a/vswitchd/vswitch.xml > +++ b/vswitchd/vswitch.xml > @@ -2842,6 +2842,37 @@ > </column> > </group> > > + <group title="Ingress Scheduling"> > + <p> > + Packets matching the ingress_sched value are prioritized. This means > + some combination of: > + </p> > + <ul> > + <li> > + prioritized packets are forwarded to their destination port before > + non-prioritized > + </li> > + <li> > + prioritized packets are less likely to be dropped in an overloaded > + situation than prioritized packets > + </li> > + </ul> > + <p> > + Ingress scheduling is supported with the best effort of the Interface. > + It may be dependant on the interface type and it's supporting > + implementation devices. Different interface types may have different > + levels of support for the feature and the same interface type attached > + to different devices (physical NICs or vhost ports, device driver, > + NIC model) may also offer different levels of support. > + </p> > + <column name="other_config" key="ingress_sched"> > + <p> > + The format of the ingress_sched field is specified in ovs-fields(7) in > + the ``Matching'' and ``FIELD REFERENCE'' sections. > + </p> > + </column> > + </group> > + > <group title="Bidirectional Forwarding Detection (BFD)"> > <p> > BFD, defined in RFC 5880 and RFC 5881, allows point-to-point > -- > 2.7.4 > > _______________________________________________ > dev mailing list > dev@openvswitch.org > https://urldefense.proofpoint.com/v2/url?u=https- > 3A__mail.openvswitch.org_mailman_listinfo_ovs- > 2Ddev&d=DwICAg&c=uilaK90D4TOVoH58JNXRgQ&r=BVhFA09CGX7JQ5Ih- > uZnsw&m=cYvXqJfbNkGW_Hacwk5ZCIh4KMLFD_2r5ssimwwuVmA&s=5- > _svanz4DDwQU7zBthfGajskziYym_GEuVpDJogHaA&e= >
diff --git a/Documentation/howto/dpdk.rst b/Documentation/howto/dpdk.rst index 93248b4..07fb97d 100644 --- a/Documentation/howto/dpdk.rst +++ b/Documentation/howto/dpdk.rst @@ -188,6 +188,37 @@ respective parameter. To disable the flow control at tx side, run:: $ ovs-vsctl set Interface dpdk-p0 options:tx-flow-ctrl=false +Ingress Scheduling +------------------ + +The ingress scheduling feature is described in general in +``ovs-vswitchd.conf.db (5)``. + +Interfaces of type ``dpdk`` support ingress scheduling only for +either ether_type or else a fully specificed combination of src and +dst ip address and port numbers for TCP or UDP packets. + +To prioritize packets for Precision Time Protocol: + + $ ovs-vsctl set Interface dpdk-p0 \ + other_config:ingress_sched=eth_type=0x88F7 + +To prioritize UDP packets between specific IP source and destination: + + $ ovs-vsctl set Interface dpdk-p0 \ + other_config:ingress_sched=udp,ip_src=1.1.1.1,ip_dst=2.2.2.2,\ + udp_src=11,udp_dst=22 + +If unsupported ingress scheduling configuration is specified or it cannot be +applied for any reason a warning message is logged and the Interface operates +as if no ingress scheduling was configured. + +Interfaces of type ``dpdkvhostuserclient``, ``dpdkr`` and ``dpdkvhostuser`` do +not support ingress scheduling. + +Currently only the match fields listed above are supported. No wildcarding of +fields is supported. + pdump ----- diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml index 0bf986d..299d725 100644 --- a/vswitchd/vswitch.xml +++ b/vswitchd/vswitch.xml @@ -2842,6 +2842,37 @@ </column> </group> + <group title="Ingress Scheduling"> + <p> + Packets matching the ingress_sched value are prioritized. This means + some combination of: + </p> + <ul> + <li> + prioritized packets are forwarded to their destination port before + non-prioritized + </li> + <li> + prioritized packets are less likely to be dropped in an overloaded + situation than prioritized packets + </li> + </ul> + <p> + Ingress scheduling is supported with the best effort of the Interface. + It may be dependant on the interface type and it's supporting + implementation devices. Different interface types may have different + levels of support for the feature and the same interface type attached + to different devices (physical NICs or vhost ports, device driver, + NIC model) may also offer different levels of support. + </p> + <column name="other_config" key="ingress_sched"> + <p> + The format of the ingress_sched field is specified in ovs-fields(7) in + the ``Matching'' and ``FIELD REFERENCE'' sections. + </p> + </column> + </group> + <group title="Bidirectional Forwarding Detection (BFD)"> <p> BFD, defined in RFC 5880 and RFC 5881, allows point-to-point
Signed-off-by: Billy O'Mahony <billy.o.mahony@intel.com> --- Documentation/howto/dpdk.rst | 31 +++++++++++++++++++++++++++++++ vswitchd/vswitch.xml | 31 +++++++++++++++++++++++++++++++ 2 files changed, 62 insertions(+)