Age | Commit message (Collapse) | Author | Files | Lines |
|
Type: improvement
Change-Id: I8c28c845c75657852f1e513e2832771fad6b90b7
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
In one's complement, there are two representations of zero: the all
zero and the all one bit values, often referred to as +0 and -0. See
RFC 1624 section 3 for more details.
This used to be taken care of in ip4_header_checksum(), but it is no
longer the case. The check ip->checksum == ip4_header_checksum (ip) is
no longer correct in the -0 case.
Always use ip4_header_checksum_is_valid() instead (which behaves
correctly since 9a79a1ab931c3b5a7ae07d6f0fcfef7c4368a2c4).
Type: fix
Fixes: e5f0050c7a5d411f96af6401797529d58825e2af
Change-Id: Iacc6b60645a834287b085aecb9e3fdb4554cf0cf
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
No change in default behavior. To use htlb pages for the ip4 mtrie,
use the "ip" command-line option "mtrie-hugetlb".
Type: improvement
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I5497e426a47200edff2c7e15563ed6a42af12e7f
|
|
VXLAN uses csum_offload for IPv6 packets.
But without gso node we have csum calculated only for inner
packet.
This patch adds support for outer header csum calculation.
Checksum for inner packet should be calculated before
interface-output node (for example in vxlan node).
Type: fix
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
Signed-off-by: Vladimir Isaev <visaev@netgate.com>
Change-Id: Ica68429ede4426293769207cd83c791ebe72fe56
|
|
Fix and optimize DMAC check in ethernet-input node to utilize NIC or
driver which support L3 DMAC-filtering mode so that DMAC check can be
bypassed safely for interfaces/sub-interfaces in L3 mode.
Checking of interface in L3-DMAC-filtering state to avoid DMAC check
require the following:
a) Fix interface driver init sequence for devices which supports L3
DMAC-filtering to indicate its capability and initialize interface
to L3 DMAC-filtering state.
b) Fix ethernet_set_flags() function and its associated callback
flags_change() functions registered by various drivers in interface
infra to provide proper L3 DMAC filtering status.
Maintain interface/sub-interface L3 config count so DMAC checks can be
bypassed if L3 forwarding is not setup on any main/sub-interfaces.
Type: fix
Ticket: VPP-1868
Signed-off-by: John Lo <loj@cisco.com>
Change-Id: I204d90459c13e9e486cfcba4e64e3d479bc9f2ae
|
|
Type: fix
Change-Id: Idf7c80b7d81f796bd0512bca4276bcfcf2af241a
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Type: fix
Signed-off-by: Ye donggang <yedg@wangsu.com>
Change-Id: Ia9f72ff2be455ecd4ff3d16e884c5a50f9df69fe
|
|
Type: improvement
- inline some common encap fixup functions into the midchain
rewrite node so we don't incur the cost of the virtual function call
- change the copy 'guess' from ethernet_header (which will never happen) to an ip4 header
- add adj-midchain-tx to multiarch sources
- don't run adj-midchain-tx as a feature, instead put this node as the
adj's next and at the end of the feature arc.
- cache the feature arc config index (to save the cache miss going to fetch it)
- don't check if features are enabled when taking the arc (since we know they are)
the last two changes will also benefit normal adjacencies taking the arc (i.e. for NAT, ACLs, etc)
for IPSec:
- don't run esp_encrypt as a feature, instead when required insert this
node into the adj's next and into the end of the feature arc. this
implies that encrypt is always 'the last feature' run, which is
symmetric with decrypt always being the first.
- esp_encrpyt for tunnels has adj-midchain-tx as next node
Change-Id: Ida0af56a704302cf2d7797ded5f118a781e8acb7
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Type: feature
- replace functions for prefixes attached to interfaces
- add ip_interface.[ch] to consoldate the functions
Signed-off-by: Neale Ranns <nranns@cisco.com>
Change-Id: I9c0c39c09dbf80ea1aadefee02c9bd16f094b6ad
|
|
Merge two mildly incompatible static inlines, and rename the results
vnet_calc_checksums_inline (...).
The resulting inline has three additional parameters: int is_ip4, int
is_ip6, and int with_gso. All calls manage to pass one or more as
compile-time constants, which causes a certain amount of code to
disappear in each instantiation.
Type: refactor
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I1a2a0e353b9a7bca20bc81318e8c915617261e1a
|
|
Type: feature
- move the IP4 code to plugin
- add ip6 support
- add suport for uRPF on TX
- add tests
Change-Id: I074c2debc486d3e79c12fad4b8dbd72c41e841a0
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Type: fix
- doing so in MTU dec, means mtu_inc is broken
- there's no need to. if a packet encounters ipX-rewrite a second time then it went through a tunnel the first time and is still locally originated.
Change-Id: I0f279c2837b608c1677485fe93f63398ab2737b3
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Type: improvement
Signed-off-by: Neale Ranns <nranns@cisco.com>
Change-Id: Ia8dff15855a81cf29729bdaa3ff28fbe3254fa97
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ie8bc5c9a03e858487cf565b4e9b520e6b496337a
|
|
Type: refactor
currently vnet_classify.h is included in ip.h where it's not required.
Change-Id: Id55682637601655aa2edda681536a979c8e323bd
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Type: fix
Fixes: be2286b0
This patch does following:
- If terminating frame has consecutive packets with same source IP, this patch
avoids fib lookup for those packets in ip4-local node. This drops cycle count
for ip4-local node on both ARM and x86. It being done by enabling dead code in
else {} case of ip4_local_check_src_x2() and ip4_local_check_src() functions.
- In case all packets in terminating frame have unique source IP (e.g:
incrementing), ip4-local is costlier by 2 cycles (broadwell)
Change-Id: I472ddc324716cec8bfe601568b8aeb7565f97ab3
Signed-off-by: Nitin Saxena <nsaxena@marvell.com>
|
|
Type: feature
- ip-neighbour: generic neighbour handling; APIs, DBs, event handling,
aging
- arp: ARP protocol implementation
- ip6-nd; IPv6 neighbor discovery implementation; separate ND,
MLD, RA
- ip6-link; manage link-local addresses
- l2-arp-term; events separated from IP neighbours, since they are not
the same.
vnet retains just enough education to perform ND/ARP packet
construction.
arp and ip6-nd to be moved to plugins soon.
Change-Id: I88dedd0006b299344f4c7024a0aa5baa6b9a8bbe
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Type: refactor
Change-Id: I51405b9d09fb6fb03d08569369fdd4e11c647908
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Type: refactor
Signed-off-by: Neale Ranns <nranns@cisco.com>
Change-Id: I3aad20b35d89fc541fdf185096d71ca12b09a6e2
|
|
This provides a functional interface to IP fragmentation.
Allowing external features to fragment. Supports
arbitrary encap size, for e.g. MPLS or inner fragmentation
of tunnels.
This also removed dual loop in MAP that was fundamentally broken.
Type: fix
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: Ia89ecec8ee3cbe2416edbe87630fdb714898c2a8
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Type: fix
if the packet is about to be fragmented, then don't call any of the
actions that expect the rewrite to have been written.
1) don't double count packets thru the adjacency (original & fragments)
2) don't double decrement the TTL for fragments
3) return to ip4-midchain post ip-frag if that's where we started.
4) only run midchain/mcast fixups if not fragmenting (if no errors)
Change-Id: Ib2866787a42713ee5871b87b597d8f74b901044b
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Type: feature
Based on the configuration, we can disable checksum offload capability
and calculate checksum while pushing the TCP & IP header.
This saves some cycles when VPP stack is used in legacy hardware devices.
Signed-off-by: Srikanth A <srakula@cisco.com>
Change-Id: Ic1b3fcf3040917e47ee65263694ebf7437ac5668
(cherry picked from commit 3642782a2748503f5b5ccf89d1575c1d489948ef)
|
|
this is a preparation step for introducing other reassembly types
Type: refactor
Change-Id: I197e299dbd729b00eead31667913b8ceff915d63
Signed-off-by: Klement Sekera <ksekera@cisco.com>
|
|
In addition to the external vppapitrace tool, VPP itself supports dumping of API trace files.
In two formats, "custom-dump" and "dump". "dump" gives a human friendly list,
and "custom-dump" is meant to give a list of commands that can be fed to VAT.
This patch only deals with "dump".
Prior to this fix, auto-generation was only done for the basic types.
This fix adds support for any type, including lists, and supports pretty-printing
of enums, strings, IP addresses, MAC addresses and so on.
Usage: api trace dump <api-trace-file>
For example
Change-Id: I4e485680e6dcfce7489299ae6cf31d835071ac40
---------- trace 48 -----------
vl_api_sw_interface_set_flags_t:
_vl_msg_id: 75
client_index: 0
context: 10
sw_if_index: 1
flags: IF_STATUS_API_FLAG_ADMIN_UP
---------- trace 49 -----------
vl_api_sw_interface_add_del_address_t:
_vl_msg_id: 88
client_index: 0
context: 11
sw_if_index: 1
is_add: 1
del_all: 0
prefix: 172.16.1.1/24
---------- trace 51 -----------
vl_api_cli_inband_t:
_vl_msg_id: 819
client_index: 0
context: 13
cmd: packet-generator capture pg0 pcap /tmp/vpp-unittest-TestMAP-YhcmDX/pg0_out.pcap disable
---------- trace 58 -----------
vl_api_ip_neighbor_add_del_t:
_vl_msg_id: 199
client_index: 0
context: 20
is_add: 1
neighbor:
sw_if_index: 2
flags: IP_API_NEIGHBOR_FLAG_NONE
mac_address: 0202.0000.ff02
ip_address: fd01:2::2
Signed-off-by: Ole Troan <ot@cisco.com>
Change-Id: I5556d06008de2762e7c2d35a8b0963ae670b3db1
Type: fix
Signed-off-by: Ole Troan <ot@cisco.com>
Signed-off-by: Ole Troan <ot@cisco.com>
Signed-off-by: Ole Troan <ot@cisco.com>
Signed-off-by: Ole Troan <ot@cisco.com>
Signed-off-by: Ole Troan <ot@cisco.com>
Signed-off-by: Ole Troan <ot@cisco.com>
Signed-off-by: Ole Troan <ot@cisco.com>
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Too many prefetches within loop unrollings induce bottleneck and
performance degradation on some CPUs which have less cache line fill
buffers, e.g, Arm Cortex-A72.
Apply dual loop unrolling and tune prefetches manually to remove
hot-spot with prefetch instructions, to get throughput improvement.
It brings about 7% throughput improvement and saves 28% clocks with
ip4_rewrite nodes on Cortex-A72 CPUs.
Type: feature
Change-Id: I0d35ef19faccbd7a5a4647f50bc369bfcb01a20d
Signed-off-by: Lijian Zhang <Lijian.Zhang@arm.com>
|
|
When checksumming chained buffers with odd lengths: insert a
NULL byte, or the calculation fails.
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Signed-off-by: John Lo <loj@cisco.com>
Change-Id: I380f7c42897bdb28c8c29aa1c4cdaaa849cc9ecc
|
|
The function ip4_rewrite_inline_with_gso can leverage
the existing vlib_buffer_advance to simplify code.
Type: fix
Change-Id: I25d943dc78aba2f67654385cf3f693eb01e0210e
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
Type: feature
Adding a prefix to an interface was not permitted if it overlapped
with another prefix on an interface which used the same FIB.
Loosen the restriction. Allow 2 or more addresses from the same
prefix on a single interface. Reference count the prefix to figure
out when a glean/connected route for the prefix needs to be added
or removed.
Added unit tests to check that the route is only removed when all
addresses in the prefix are removed from the interface.
Change-Id: I1a962ecb5e1ee65fc6d41f98a4cc097a51a55321
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Fix ip6_tcp_udp_icmp_compute_checksum to work properly for packets
with multiple buffers.
Fix ip4_tcp_udp_compute_checksum to exit upon detecting error.
Type: fix
Signed-off-by: John Lo <loj@cisco.com>
Change-Id: I673547f4479d72cd60757383343fc562cff10265
|
|
Type: feature
When admin status is changed on an interface, add or delete the
routes for the IPv4 addresses configured on that interface.
This is already being done for IPv6 interface addresses.
Change-Id: Ib1e7dc49c499921dd287e075640243520ffa5589
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Type: feature
ip4_local_check_src() was overwriting vnet buffer opaque data
on the adjacency for packets with "local" (dpo-receive) destination
addresses.
Retain the dpo receive index in vnet_buffer()->adj_index[VLIB_TX].
This can allow a graph node to distinguish the interface where the
destination address is configured from the interface where the
packet was received. This can be useful in correctly handling
packets that have been sent to an address configured on a loopback
interface.
Change-Id: I52a942e85b5302b338a2d0404a37c5ea1a99e89f
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Type: refactor
Change-Id: I791cc63e989df049104420faf74fc28f965b6648
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I0c2d6fccd95146e52bb88ca4a6e84554d5d6b2ed
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
- arp-input, registered with the ethernet protocol dispatcher, performs
basic checks and starts the arc
- arp-reply; first feature on the arc replies to requests and learns
from responses (no functional change)
- arp-proxy; checks against the proxy DB
arp-reply and arp-proxy are enabled when the interface is appropriately
configured.
Change-Id: I7d1bbabdb8c8b8187cac75e663daa4a5a7ce382a
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
TBH, this looks like merge damage or some such. Perfectly fine NULL pointer
check, about three lines after it was needed.
Change-Id: I52831062e30533a59fb76b644ee5ae389676d2ae
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
before and after:
ip4-load-balance 1.54e1
ip4-load-balance 1.36e1
p.s. Quad loops were not beneficial
Change-Id: I7bc01fc26288f0490af74db2b1b7993526c3d982
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
vlib_get_buffers can save 1.2 clocks/pkt from 16.1 to 14.9 clocks/pkt
on Skylake.
Change-Id: I79d8b58b192280af5e5a5f73562b6301e1821cec
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
Change-Id: I046e481a67fbeffdaa8504c8d77d232b986a61ee
Signed-off-by: John Lo <loj@cisco.com>
|
|
Change-Id: I215e1e0208a073db80ec6f87695d734cf40fabe3
Signed-off-by: Jim Thompson <jim@netgate.com>
|
|
Change-Id: I6527e3fd8bbbca2d5f728621fc66b3856b39d505
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
error should be recorded in buffer so that
process-error-punt can handle them correctly
Per Damjan's comments, move counter to under
else clause of last error0==NONE check. Both
v4 and v6 are changed.
Change-Id: I707c7877ccb12589337155173fc4a5200b42ee93
Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
|
|
This commit adds a "gso" parameter to existing "create tap..." CLI,
and a "no-gso" parameter for the compatibility with the future,
when/if defaults change.
It makes use of the lowest bit of the "tap_flags" field in the API call
in order to allow creation of GSO interfaces via API as well.
It does the necessary syscalls to enable the GSO
and checksum offload support on the kernel side and sets two flags
on the interface: virtio-specific virtio_if_t.gso_enabled,
and vnet_hw_interface_t.flags & VNET_HW_INTERFACE_FLAG_SUPPORTS_GSO.
The first one, if enabled, triggers the marking of the GSO-encapsulated
packets on ingress with VNET_BUFFER_F_GSO flag, and
setting vnet_buffer2(b)->gso_size to the desired L4 payload size.
VNET_HW_INTERFACE_FLAG_SUPPORTS_GSO determines the egress packet
processing in interface-output for such packets:
When the flag is set, they are sent out almost as usual (just taking
care to set the vnet header for virtio).
When the flag is not enabled (the case for most interfaces),
the egress path performs the re-segmentation such that
the L4 payload of the transmitted packets equals gso_size.
The operations in the datapath are enabled only when there is at least
one GSO-compatible interface in the system - this is done by tracking
the count in interface_main.gso_interface_count. This way the impact
of conditional checks for the setups that do not use GSO is minimized.
"show tap" CLI shows the state of the GSO flag on the interface, and
the total count of GSO-enabled interfaces (which is used to enable
the GSO-related processing in the packet path).
This commit lacks IPv6 extension header traversal support of any kind -
the L4 payload is assumed to follow the IPv6 header. Also it performs
the offloads only for TCP (TSO - TCP segmentation offload).
The UDP fragmentation offload (UFO) is not part of it.
For debug purposes it also adds the debug CLI:
"set tap gso {<interface> | sw_if_index <sw_idx>} <enable|disable>"
Change-Id: Ifd562db89adcc2208094b3d1032cee8c307aaef9
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
|
|
-fno-common makes sure we do not have multiple declarations of the same
global symbol across compilation units. It helps debug nasty linkage
bugs by guaranteeing that all reference to a global symbol use the same
underlying object.
It also helps avoiding benign mistakes such as declaring enum as global
objects instead of types in headers (hence the minor fixes scattered
across the source).
Change-Id: I55c16406dc54ff8a6860238b90ca990fa6b179f1
Signed-off-by: Benoît Ganne <bganne@cisco.com>
|
|
use address_t and mac_address_t for IPv6 and ARP entries
and all other API calls in ip.api aprat from the route ones,
that will follow in a separate commit
Change-Id: I67161737c2184d3f8fc1e79ebd2b55121c5b0191
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: Ic65f686aaccaf8450732d88d7471b587faccaa9d
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I0470b4b13095583fe018f565f100342fab45715e
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
ipsec4-output and ipsec6-output were conflicting with ipsec
interface names ("ipsec<id>") and vnet/interface.c autogenerated
output node ("<ifname>-output").
Changing feature names seems to be the less invasive option.
This patch also changes "input" feature names for consistency.
Change-Id: I4ba10d07e9ba09df20aa2500104252b06b55f8f7
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
|
|
Change-Id: Ib0d9b533d72c899b77c9a7bd1daa9b4a55b7221c
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Prefetching first 2 packets' header is useless cause of the prefetching
action is not done before using the packets.
There's no performance drop in Xeon platform and slightly performance
gain in Atom platform after rmoving the prefetch.
Change-Id: Ib4b074af20d7cd5053aecc7147b162141aec31f5
Signed-off-by: Simon Zhang <yuwei1.zhang@intel.com>
|
|
Verify that last node in the computed feature order matches
reality. This check doesn't make sense in all cases, so we skip it if
the newly-added vnet_feature_arc_registration_t ".last_in_arc" datum
is a NULL pointer.
Change-Id: Ia99c3e2b2da2e4780a7d5bc71670c5742a66fef2
Signed-off-by: Dave Barach <dave@barachs.net>
|