This code was actually never working. It was just used as workaround for
Cisco ENIC issue with tagged default vlan frames.
Today Cisco ENIC provides solution to this problem with devargs flags...
Change-Id: Ia8284274117cb200bf6c7f7911d945d5a093d878
Type: refactor
Signed-off-by: Damjan Marion <damarion@cisco.com>
Type: improvement
Previously multiple sw crypto scheduler queues per core design
caused unaverage frame processing rate for each async op ID –
the lower the op ID is the highly likely they are processed first.
For example, when a RX core is feeding both encryption and
decryption jobs of the same crypto algorithm to the queues at a
high rate, in the mean time the crypto cores have no enough
cycles to process all: the jobs in the decryption queue are less
likely being processed, causing packet drop.
To improve the situation this patch makes every core only owning
a two queues, one for encrypt operations and one for decrypt.
The queue is changed either after checking each core
or after founding a frame to process.
All crypto jobs with different algorithm are pushed to
thoses queues and are treated evenly.
In addition, the crypto async infra now uses unified dequeue handler,
one per engine. Only the active engine will be registered its
dequeue handler in crypto main.
Signed-off-by: DariuszX Kazimierski <dariuszx.kazimierski@intel.com>
Signed-off-by: PiotrX Kleski <piotrx.kleski@intel.com>
Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Signed-off-by: Jakub Wysocki <jakubx.wysocki@intel.com>
Change-Id: I517ee8e31633980de5e0dd4b05e1d5db5dea760e
It is actually longer and slower...
Type: refactor
Change-Id: I0f126d4cdb13ecc60a2d370409f23820d7f7eb72
Signed-off-by: Damjan Marion <damarion@cisco.com>
- change 'port' to 'dst-port' to make the input self-explanatory
- add empty spaces to short help
Type: fix
Signed-off-by: Miroslav Miklus <miroslav.miklus@pantheon.tech>
Change-Id: I74c783e6e8629e61b1c100e9355d09cf5e35a750
We add the possibility to bind the destination UDP port of a Scapy SA
to the ESP layer in the IPsec tunnel protection tests, even if it is not
the default port for ESP (4500).
This allows to test IPSec tunnel protection with ports other than 4500
in the UDP header, without hardcoding them in the Scapy patch (ex: 4545)
Type: improvement
Change-Id: I1eea3d4660ed1b59d827250a419af6b7b41c4a72
Signed-off-by: Arthur de Kerhor <arthurdekerhor@gmail.com>
Change [1] put static mappings in flow hash. This change also broke
relationship between nat pool addresses and static mappings. Port and
address are no longer reserved in nat pool address records for a new
static mapping.
Because of this change both nat objects and their configuration can
function independently. This change also removed already broken logic of
having static-mapping-only configuration.
In this patch i have cleaned up and removed unnecessary logic for static
mapping configuration functions, address configuration functions,
interface configuraiton functions and all callback functions used for
resolving interface address bound records.
No more viable configuration option static-mapping-only is also removed
because there is no more option to run traffic through vpp for static
mappings without having flow hash table allocated. Instead user is now
able to create static mapping records without using nat pool addresses.
Fixed and improved management of required fib entries (ensuring VPP will
reply to ARP for all external addresses on outside interfaces) through
holding a refcount for their creation and removal.
[1] https://gerrit.fd.io/r/c/vpp/+/34077
Type: improvement
Change-Id: Ic16deefbede833d574c2a5972155c9afa5bc62ce
Signed-off-by: Filip Varga <fivarga@cisco.com>
This patch improves algorithm responsible for choosing nat pool address
during dynamic session creation.
Patch synchronizes nat pool address attributes with vpp interfaces doing
so gives nat option to determine correct nat pool address when creating
dynamic session.
Improvements:
* match dst address subnet and src fib with nat pool address
* for output-feature also fallback match of tx_sw_if_index with nat
pool address
Type: improvement
Change-Id: I594638bc76fc1153f582add376838b4b72ff573a
Signed-off-by: Filip Varga <fivarga@cisco.com>
Type: fix
Fixes: 4b1b13315a3c
When adding or deleting a VR, multicast routes can be added or deleted.
When the first VR is added, a local (dpo-receive) route is added. The
route is deleted when the last VR is deleted.
Perform the check on whether to add or delete the route on a per-FIB
basis. Otherwise, if the route is only added after the first VR is added
without regards to the FIB being used and a second VR is added later on
an interface attached to a different FIB, the necessary route will not be
added to the FIB used by the second interface.
Change-Id: Ib30925ecf45c714cfe3ac6a223754bea918f10e3
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
when bitmap vec len is 1 and all the bits has been
occupied, clib_bitmap_next_clear(ai, 63) will
return result:65 which should be 64. It will lead to
missing iteration for pool_foreach.
Type: fix
Signed-off-by: jiangxiaoming <jiangxiaoming@outlook.com>
Change-Id: Iadac7e6f6b4da357943c4c9d50bf22353c4a8408
Now DPDK have API to register external threads so we can remove this
mess...
Type: improvement
Change-Id: I71a21f0cd94bd668aa406710c75a0bcc63fdc840
Signed-off-by: Damjan Marion <damarion@cisco.com>
When trying to read files that are larger than the fifo
a timeout was reached on client side leading to an abort of the request
and a retry (sending another GET command).
The svm fifo notification request was set to notify when the fifo is no
longer full, this lead to an inefficient loop of sending small amounts
of data each time with a large overhead of context switch and waiting
for the next notification, eventually leading to a timeout.
Modifying the trigger on the svm-fifo to be notified on a preset threshold
value enabled sending larger amounts of data between context switches and
sending large files more efficiently. This solved the timeout issue.
In addition, cap the max write from application to 4MB to avoid running
into a case of trying to allocate chunks that are too large.
Reproduce:
Server:
http static server www-root /var/www/data uri tcp://0.0.0.0/80 cache-size 5m fifo-size 300
Client:
wget http://11.0.0.2/file_of_size_32M
Type: fix
Change-Id: Idfceedffd935da9486cde820e9dca5dad69d9ca5
Signed-off-by: Yuval Caduri <cyuval@marvell.com>
Signed-off-by: Michal Kalderon <mkalderon@marvell.com>
Add the ability to set a threshold on the fifo that will
be used to notify caller when de-queue from the fifo reached
a value that is less than or equal to the threshold.
Type: feature
Change-Id: I70ba1a05f783ce5247409e9beebe4e336b7c1eb5
Signed-off-by: Michal Kalderon <mkalderon@marvell.com>
Signed-off-by: Yuval Caduri <cyuval@marvell.com>
the receive dpo added by tcp src-address cli do not have a valid sw_if_index ,
ip4_local_check_src() and tcp_input_lookup_buffer() will set ~0 to vnet_buffer(b)->sw_if_index[VLIB_RX],
which will cause crash in tcp46_reset_inline,
Type: fix
Signed-off-by: Mercury <mercury124185@gmail.com>
Change-Id: Ie01c31f3575e14187c6380ebcfff96fcb6098cde
fix ipv4 key uninitialized in local_endpoints_table,
which will cause transport_endpoint_cleanup() failed
to lookup the endpoint and can not delete it,
as for ipv6, clib_memcpy_fast() will change all bytes of lcl_addr
and there maybe no need to initalize,
Type: fix
Signed-off-by: Mercury <mercury124185@gmail.com>
Change-Id: I56676493a393b1d64eaa438224e256094ca75d2f
Simplifies allocation of fifos as fifo segment and segment manager
indices can be set at alloc time.
Type: improvement
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ibd357b3ff0279d8deefcdcb17010b4068007ccb7
Avoid scenarios where a worker allocates a segment but while it drops
the segment manager writer lock and acquires the reader lock another
worker uses the segment and frees it.
Type: fix
Thanks to wanghanlin@corp.netease.com for the report.
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I0a88d738c51b33fd07c34916f125c98806861a06
The release of the map_lock is a bit too early, it needs to wrap
mprotect(), otherwise the access rights of hdr may be deprived of
other worker threads.
Type: fix
Signed-off-by: liuyacan <liuyacan@corp.netease.com>
Change-Id: I9cb89d49a097ec6d5ae9fd2af2bddf967ec946e9
Type: fix
Presently a local label associated with an attached or connected prefix will link to the glean. This is a problem since it will never use the adj-fibs that are installed for that attached prefix. Instead link the local label to a lookup in the table in which the attached link is bound.
Signed-off-by: Neale Ranns <neale@graphiant.com>
Change-Id: Iad49fb6168b9ba47216a9a52bd262363b49c3c43
There could be a race condition where two fragments of one chain end up
at the same time on different workers, one overwriting others hash
entry. Add a check for that and restart processing on the unlucky worker
who ends up being second from hash table POV. This will then result in a
proper handover to worker now owning this reassembly.
Type: fix
Fixes: de34c35fc73226943538149fae9dbc5cfbdc6e75
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Change-Id: I9eb29c5cb1ffe3b5eb1d5a638e17ab7ba2628d28
ip6_get_link_local_address() may return NULL if the local interface
is not (or not yet) configured
Type: fix
Signed-off-by: Frédéric Perrin <fred@fperrin.net>
Change-Id: I42bf2081582c4a36fa4e32145ca2f0ff73488110
Type: fix
hash_resize is available in hash.h file, but missing __clib_export in hash.c
Signed-off-by: Leung Lai Yung <benkerbuild@gmail.com>
Change-Id: Ibb741b532cd1080ec5d8314aae8dbbca87f42502
Detect mq congestion and handle it by queueing messages in a fifo and
postponing handling via rpcs. App workers with congested mqs cannot
accept nor connect additional sessions.
Type: feature
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I401d971a1a53896758b88fc60f158cbc31e0c7cb
Coverity complained that there is boundary checking in
add_del_punt_redirect_v2 handler. This test proves that such boundary
checking is not necessary as it is handled in the common path.
Type: test
Change-Id: Ibec054c01d4eb057accdc9d5732aba6fe6de51cc
Signed-off-by: Klement Sekera <ksekera@cisco.com>
Type: fix
The pub_sub_handler() stores registrations key'd with the message's client_index (which is in network-order).
The socket-client invokes the reaper function (to cleanup these registrations) using the pool index of the registration. hence the pub-sub registration is not removed.
change the socket-client to pass the network-order client_index.
This approach was chosen in preference to chaning the way the registrations are key'd since the memory-client also uses this registration, and correclty passes the client_index to the reaper.
Signed-off-by: Neale Ranns <neale@graphiant.com>
Change-Id: I6118555c8601149d05801f558e08bcc7aed4fc98
This adds an interface CLI 'monitor
interface <interface>' that periodically
reports rx/tx pps & bandwidth for a
given interface.
Type: feature
Change-Id: Ia9d59b3443913520a52b38d7bda012190be6f167
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
