Age | Commit message (Collapse) | Author | Files | Lines |
|
Add shared memory APIs for MPLS routes and MPLS to IP prefix bindings.
Change-Id: I85b074a4dadc8249c410fdabd8ea019d20479cf8
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
|
|
When an interface is removed, and another one is created,
the nodes from the previous interface are reused. The current
code fails when the new interface type differs from the original.
It is necessary to update the node runtime, in particular the
node function, and the node format function.
Jira: VPP-460
Change-Id: I2fc3c609e0173a7f62597230768669039853ab03
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
|
|
Change-Id: I7b70d52d11c228ec5e790632a91b042b34b38d28
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I9d1d02ffbb3cdbe8c4a3f335ece40f3dcbd848a7
Signed-off-by: Billy McFall <bmcfall@redhat.com>
|
|
Note that the new way to provision MPLSoGRE is using the "normal"
GRE tunnel such as in the following example:
create gre tunnel src 10.0.3.1 dst 10.0.3.3
set in state gre0 up
set int ip addr gre0 10.0.4.1/30
set int mpls gre0 enable
ip route table 4 6.0.0.0/24 via 10.0.4.2 gre0 out-label 30
mpls local-label add 30 eos ip4-lookup-in-table 4
The previous CLIs/APIs used to configure MPLSoGRE tunnel such as
"create mpls gre tunnel...", "mpls encap|decap add label .." and
"show mpls tunnel", etc. can not be used. They will be deprecated
in a later update to the VPP code base.
Change-Id: I244916841924dc2b87d2143691cd8476716c06b1
Signed-off-by: John Lo <loj@cisco.com>
|
|
Change-Id: I9c23ddbbd576e8c900ba82425572ed7a3c4f56e3
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Fixed Coverity complaining about the clib_memcpy used in
/vnet/vnet/interface.c
Change-Id: I401f05aea1066d829abac1555021e1180b4a161e
Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
|
|
Change-Id: Id17fcb9154c5337908f29b7ce3a282b6a4b72d64
Signed-off-by: Billy McFall <bmcfall@redhat.com>
|
|
Change-Id: I003f95db85ce5085045c607f0b1daaefa792c910
Signed-off-by: Billy McFall <bmcfall@redhat.com>
|
|
Change-Id: If4c47b91e949483675cf569cd474b6c85fecef05
Signed-off-by: Billy McFall <bmcfall@redhat.com>
|
|
Change the adjacency completion model to pull not push.
A complete adjacency has a rewirte string, an incomplete one does not. the re-write string for a peer comes either from a discovery protocol (i.e. ARP/ND) or can be directly derived from the link type (i.e. GRE tunnels). Which method it is, is interface type specific.
For each packet type sent on a link to a peer there is a corresponding adjacency. For example, if there is a peer 10.0.0.1 on Eth0 and we need to send to it IPv4 and MPLS packets, there will be two adjacencies; one for the IPv4 and one for the MPLS packets. The adjacencies are thus distinguished by the packets the carry, this is known as the adjacency's 'link-type'. It is not an L3 packet type, since the adjacency can have a link type of Ethernet (for L2 over GRE).
The discovery protocols are not aware of all the link types required - only the FIB is. the FIB will create adjacencies as and when they are required, and it is thus then desirable to 'pull' from the discovery protocol the re-write required. The alternative (that we have now) is that the discovery protocol pushes (i.e. creates) adjacencies for each link type - this creates more adjacencies than we need.
To pull, FIB now requests from the interface-type to 'complete' the adjacency. The interface can then delegate to the discovery protocol (on ethernet links) or directly build the re-write (i.e on GRE).
Change-Id: I61451789ae03f26b1012d8d6524007b769b6c6ee
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: Idf68266f705b0455e5ab0ac73d23c7e0f4120d5b
Signed-off-by: Dave Barach <dave@barachs.net>
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ie1be9dc057d07c250852952ea159ed3c44c25f5f
Signed-off-by: Billy McFall <bmcfall@redhat.com>
|
|
QEMU 2.7 introduces reconnect support. However it doesn't send
last state of last_avail_idx in the SET_VRING_BASE.
With this patch we are overriding SET_VRING_BASE with value taken
from used->idx.
It is a bit hackish as it basically overrides SET_VRING_BASE
provided value but it should work in most of the cases.
This is what author of reconnect code is also suggesting in his post:
https://lists.gnu.org/archive/html/qemu-devel/2016-06/msg01495.html
Change-Id: I10e5f401c409d2989bf6c43eaea377800a9c76df
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I570f57218293a4824f83b38575eacdbe3da540d8
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Proper cpu pinning in vpp_lite platform, like in normal vpp image.
Extended “show threads” command to show propper information.
Changed handling of coreID and socketID for threads in "show threads"
CLI, pthread_getaffinity is used instead of info stored in DPDK.
Change-Id: Ic8299ec5e284472bb10a37a95fadeed57b6edae8
Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
|
|
Change-Id: I7a14d9a28514cdb49f65fab1b120b53239fed7e3
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
The classifier was written with the assumption that next-indicies
of IP4 and IP6 IP_LOOKUP_NEXT nodes are equal. That's not true,
and this patch splits the classifier session for IP4 and IP6.
Change-Id: Id0368f17bb1d3f145b771d2dc283b56871264e99
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: Ie221e246dd8c6a1b3c76e7aa85924762d4397782
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: Ieea56f3bf9e749878d9f2b35d39d9f7a9cdabde4
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
|
|
Eliminate an unused parameter in vnet_feature_arc_init(...) which
eliminates a spurious relationship between feature arcs and ip.
Add doxygen tags for 2x debug CLI commands, and for
vnet_feature_arc_init(...)
Change-Id: If83589539bcbbbd4d85e84458dd7eb7eb08837a0
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: Iff32c488af9b71acbc4e572c6741afae0a67333c
Signed-off-by: Juraj Sloboda <jsloboda@cisco.com>
|
|
In order to have meaningfull IPFIX implementation we should be able
to classify all packets flowing through vpp. But existing IPv4 and IPv6
classifier nodes are called only if destination IP address is local
to vpp. This commit adds new IPv4 and IPv6 classifier nodes that should
be used for collecting flow statistics.
Change-Id: I60e60105663ba15b5200862a23bb817047fe4d1a
Signed-off-by: Juraj Sloboda <jsloboda@cisco.com>
|
|
some other fixes and enhancemets to address failures in the IPv6 Scale testing:
- The rate at which VPP sends ARP/ND requests, 1 per-millisecond, is too high. This is reduced to 1 every 10 ms. probably still too high.
- If an ARP/ND response is received that does not contain a change to the known MAC address, no further processing is done.
- Added stats to get info about the async FIB walks.
- When walking FIB entries and performing a load-balance update, it is not necessary to re-insert the LB in the forwarding table.
Change-Id: Ifd47e4bdbce94495f44aaf1b78e7d1395ed870d1
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
The api was allocating a vector for the name, passing it, then freeing
it, on create. The cli allocated, passed then forgot about it.
af_packet_create_if was storing a reference to the name, which in the
case of the api, meant it was referencing dead memory. On
af_packet_delete_if this reference was freed, so in the api case, there
was a double free.
Also, the cli for delete leaked the name.
Change-Id: I4d572bd2936eaf8ea7a0a8ff282e83ac2bf1b062
Signed-off-by: Ivan Kelly <ivan@midokura.com>
|
|
In a heirarchical FIB performing a unicast RPF check would require the traversal of the data-plane graph to seek out all the adjacency objects and then read those to find their interface. This is not efficient. Instead, for each path-list we construct a list of unique input interfaces and link this uRPF-list against the entry in the prefix table. In the data-plane the uRPF list can be retrieved from the load-balance lookup result and the RPF check is a simple and efficient walk across the minimal interface list. The uRPF-list is maintained as the routing heirarchy changes, in a similar way to the data-plane object graph.
We also provide a knob to allow an arbitrary prefix to pass the loose check.
Change-Id: Ie7c0ae3c4483ef467cfd5b136ee0315ff98ec15b
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Uses existing UDP local API in order to register
requested UDP port punt to the host.
CLI: set punt udp [del] <port>
API: punt protocol <l4-protocol> [ip <ver>] [port <l4-port>] [del]
* Only UDP (l4-protocol = 17) is supported at this time
Change-Id: I9232af1c891d1ed174d77f3e0dfe60c4b9d85e40
Signed-off-by: Alex Popovsky <apopovsk@cisco.com>
|
|
Change-Id: Ic29592d39f1d672cb6e7e32b4af6123683e022a7
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Move check for link-local source IP address for MAC/IP notification
only. Allow generation of response to IP6 ND solicitation request
if a match is found for target IP irrespective of source IP type.
Change-Id: Ib79d4b75fb4fe8aece625fd8cd26c8b9fc75ea47
Signed-off-by: John Lo <loj@cisco.com>
|
|
Change-Id: Ice0a6939b399eed150b0c58139b79dd4bed91113
Signed-off-by: Vengada Govindan <venggovi@cisco.com>
|
|
Change-Id: I2cc57079741279669241b88d0317853e1312e8c5
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: Ibbcce6f54bc76b8922b1c649278643c6294d13f9
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Commit of the patch from Neale Ranns
Change-Id: Ia0480ef56036dd0aec990c3892a93b99eb9fb893
Signed-off-by: Igor Mikhailov (imichail) <imichail@cisco.com>
|
|
- correct formatting errors
- add to user_doc tree
- unix line endings, not windows(!)
Change-Id: I9aa5511c398fc33ecb25fb33ae64c8e666c56dd8
Signed-off-by: Chris Luke <chrisy@flirble.org>
|
|
Change-Id: I554041bf47aa82dd6f6e0563c68d8bbd3b18b255
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
- new API/CLI to define pbb tag rewrite on interface
- encapsulation/decapsulation of PBB tags
- tracing of PBB header
- PBB tag rewrite operations
Change-Id: I538b3025a8b2e41cdeed9f10fea94bbcd28b5f5f
Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
|
|
add two new CLI commands:
show fib memory
show dpo memory
to display the memory usage of the FIB and DPO object types respectively.
Change-Id: I759e149a0b6fbb58d59c139362221dc33531cffa
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I1f4421289b2f93b64a2470559d834717ef40e181
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
It is very likely that all packets are being stored in the same
memory area.
It is also likely that all indirect descriptors are being
allocated from the same area.
This patch takes this into account by predicting the value
returned by map_guest_mem will be the same as the previous one.
It falls back to slow mode otherwise.
Change-Id: Ic1044a172fc4abe77387ba2e19e11cd1f39ad1e3
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
|
|
Change-Id: Ie17b324492d61130690ba9d05821eaefe10072d4
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
|
|
Change-Id: I0d7f9c7f41a9f9e0acb0950adedb90d45df08c2a
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I046de0c00db75d25ed90e33e9910c9dd0ff95580
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Added ability to change interface address.
Added new CLI and API functions.
Change-Id: Ia336bc75ad8c5858c26f39af851485c4c6f19f58
Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
|
|
loop0 interfaces were transmitting the same packets multiple times,
eventually causing them to be freed multiple times
Change-Id: I32fcf5cfa10e58bb9e9bbbbc1b9312518d00ee9f
Signed-off-by: David Hotham <david.hotham@metaswitch.com>
|
|
GID dictionary IP prefix entries are rewritten with more narrow one
which is not desirable. This patch does exact matching instead of
longest prefix matching.
Change-Id: I0191e23229a69ffe86f82ea4d71e4a5534dbb5b0
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
|
|
Change-Id: If90c9c630629e727fffc5d450516045a04433661
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
|
|
This commit extends the vpp framework with new thread type "hqos-threads" that
runs the Hierarchical Quality of Service (HQoS) scheduler associted with output
interface. HQoS Scheduler prioritize the packets from different users and
ensures sufficient bandwidth to pass the more important traffic.
At high level, HQoS scheduler is a buffer that can temporarily store a
large number of packets. In otherwords, it is a collection of large number
of queues organized into hierarchy of 5 levels; the port (i.e. the physical
interface) is at the root of the hierarchy followed by the subport (a set
of users), the pipes (individual users), the traffic classes (each with a
strict priority) and at the leaves, the queues.
In each HQoS scheduler, three operations are performed; classification
(setting HQoS port, subport, pipe, traffic class and queue within traffic
class from packet fields), enqueue (selecting HQoS queue for the packet,
and to drop the packet if the queue is full) and dequeue (schedule the
packet based on its length and available credits, and handover the scheduled
packet to the output interface).
In vpp, the number of hqos threads will be equal to cpu cores specified in
corelist-hqos-threads parameter cpu section of the vpp configuration file.
One hqos thread can run HQoS for multiple output interfaces. A particular HQoS
instance is initialised with default parameters required to configure hqos port,
subport, pipe and queues. Some of them can be re-configured in run-time
through CLI commands as well binary APIs.
Following illustrates the sample startup configuration file with 4x worker
threads feeding 2x hqos threads that handle each HQoS for 1x output interface.
For more details on HQoS configuration please refer to DPDK Programmer's Guide.
dpdk {
socket-mem 16384,16384
dev 0000:02:00.0 {
num-rx-queues 2
hqos
}
dev 0000:06:00.0 {
num-rx-queues 2
hqos
}
num-mbufs 1000000
}
cpu {
main-core 0
corelist-workers 1, 2, 3, 4
corelist-hqos-threads 5, 6
}
Change-Id: I635c3395a7c4ddf0a239ef77b0b0a31a6dfc4767
Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
|
|
Change-Id: Ibf3d2e69dbac28fbef77d8ba20d7f5a60eea0198
Signed-off-by: John Lo <loj@cisco.com>
|
|
Output classify node should use sw_if_index[VLIB_TX]
Like in l2_output.c, we have to use sw_if_index[VLIB_TX]
to get the appropriate configuration.
Change-Id: I96b1a03ddf5c9e9c45fa3f5df3ea88895ed4345e
Signed-off-by: Christophe Fontaine <christophe.fontaine@qosmos.com>
|
|
Only rewind the buffer back to the start of the ethernet header,
rather than all the way. We don't want to undo earlier
decapsulation.
Change-Id: I5f4b183eb3e8df690695ce3a97e55f3b0579a266
Signed-off-by: David Hotham <david.hotham@metaswitch.com>
|