Age | Commit message (Collapse) | Author | Files | Lines |
|
Under rare scenario, epoll may still post an event to VPP although the file
descriptor is already deleted via epoll_ctl (EPOLL_CTL_DEL) and the file
descriptor is close. VPP tries to access the free file index entry and crash.
The fix is to throw away the events which the file descriptor is already deleted.
Change-Id: Ieca3a1873aecb28630c3abc42c40341f27c2faa7
Signed-off-by: Steven <sluong@cisco.com>
(cherry picked from commit aec7297ba012e1fe4bbf85cdaec8e810aa476cea)
|
|
also some moving of l2 headers to reduce dependencies
Change-Id: I7a700a411a91451ef13fd65f9c90de2432b793bb
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Otherwise the key destructor is not called on pthread_exit.
Change-Id: I11e6b9683a926eecd3f40a44aab41924ff9c3101
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: Ibc74e7f7587f8b17fc0dcec20cc4530b9dd4c3ca
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: I770607cb213d538a29d7bd5762682981b01e9d18
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: I72d400401a308012b43513179870823f6f921e44
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
clib_time_verify_frequency(...) rejects clock frequency changes
greater than 1%.
vlib_worker_thread_barrier_sync_int (...) continuously checks that the
barrier hold-down timer is not unreasonably far in the future.
Change-Id: I00ecb4c20e44de5d6a9c1499fa933e3dd834d11a
Signed-off-by: Dave Barach <dbarach@cisco.com>
|
|
Change-Id: I033a1f5cdb24f77462cb5ec90a62226d93f52925
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
According to Nitin Saxena ThunderX2 machine has 64B cache line whilst
all others are 128B. According to Damjan, Nitin's previous patch broke
compilation for all non-Cavium machines. This patch should make
everything happy again.
Change-Id: I8c5c2661f9f2f9c3e9b0965a277712f9a1eefa5f
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
This caused a spurious message to be sent on the API ring causing debug messages.
Change-Id: Ief4ca655ae048aad386a88fed11e8c1114378dc7
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: Ic31da442a0e0477569d53b4a72627bbb25e93365
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I9cec7486cb6e3c5261d74d2b15a4d19469285a30
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: Ic952ed5b837ac8409fd95e2b5cb92eb028ba0c40
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: Ifcc2717efd242ae2016563d6f3e5cd57fe161e00
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I7fcfddc2bc7d9a64f8aa0d57ba5d11d325a15ce1
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
This patch depends on https://gerrit.fd.io/r/#/c/13260
Change-Id: If94968342935ebd24cf4ebed238e6a27d94959b7
Signed-off-by: Hongjun Ni <hongjun.ni@intel.com>
|
|
It is not good enough to initialize sw_if_index0 = 0,
sw_if_index1 = 0, as it maybe causes the first two
incoming packets to miss necessary computation.
Change-Id: Ifcab408d9514820e0daa280f4c73956db13b59be
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
Change-Id: I7b7183b4603076e5afac096545d820091ee7c495
Signed-off-by: Nitin Saxena <nitin.saxena@cavium.com>
|
|
- We deliberately ignore the return value.
Change-Id: If467911b019e7336cf9dc6d4c95c2cd53a5af33f
Signed-off-by: Chris Luke <chrisy@flirble.org>
|
|
1. For vxlan, prefetching one cacheline is enough.
2. Reduce vlib_increment_combined_counter functtion calling
if possible.
Change-Id: If3a72ac40c8988caaf0d5915b695f86d799f15a9
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
Change-Id: I5a47d30d783dcf000e3ca2bcdc46e7d93654cc37
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
The patch can reduce 13 cycles per packet for the graph node
vxlan-gpe-encap and increases 5% or so vxlan_gpe encap throughput
on Haswell platform for the best case (All pkts have the same
sw_if_index).
Change-Id: I9c70fd3e0f2f0a9d922cf64970d0b0d51b772024
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
- A check for the length of the buffer should have used the provided
'len' variable, not 'vec_len' since the buffer pointer may be
within a vector, but not the start of one. 'vec_len' reports 0
in that case, causing premature exit from the options processing
loop and a wait for further input before it checks the next option.
- Also add TCP_NODELAY to CLI sockets to disable Nagle on TCP
connections for a possible improvement in interactive response.
Change-Id: Ie1f53c003ca7d66bed51f437d298183215bad38c
Signed-off-by: Chris Luke <chrisy@flirble.org>
|
|
bihash deletion operation may in turn do underlying memory operations,
so ensure it is using the correct (private) heap.
Change-Id: Ibef7ad7f9db6fa83da02316bf7509072ce579bc0
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit f74b4d2b559b1d5697fd625d9c8e0f76ba5a4463)
|
|
Change-Id: I27532b3ab244dc95955e836a42b229a6e4e32818
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Allows app workers to listen on the same session endpoint. Incoming
connects are spread across the workers in a round-robin fashion
Change-Id: Ib5f5817230d9abc6127a85cdbdcad70d980c0f7f
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: I877541ede6e26581c659821502f23b777903b82f
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
All fragments should be dropped when max_frag is 1 and 2 non-initial fragments are received before first fragment.
Change-Id: Id0c968f45629698e347e8226c5926f27b48b82d6
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Update temporary port allocation with fixed allocated from:
https://github.com/prometheus/prometheus/wiki/Default-port-allocations
Change-Id: I99a7e069fb95d00884458aeacaba06e4713fbb76
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: I4cb073f23a7f24f0fc16f2eda8b869dbe657f71b
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
applied ACLs
The partition_split() did not increment the refcount when using a mask type index,
thus subsequent modifications potentially resulted in double frees and in the best case
immediate crash, in the worst case delayed crash in another place.
Introduce the lock_mask_type_index() and call it, move the mask type index
related functions closer to the top of the file.
Make the assignment of the new mask type indices
for the tuplemerge case to use the assign_mask_type_index().
Keep some debugs in case we need to investigate this further at some point.
Change-Id: Iae370f5cd92e1fe1442480db34656a8a3442dbc0
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit 1edc406da3d4f6e63de2f278360b5753f55c00df)
|
|
Change-Id: I6882b6daa05db866fe6e78a62b380ec331507f74
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Change-Id: I622cdb969ea489d333888b90c15ab57c1820e2db
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
|
|
Change-Id: I7fcc082376b451c6179ec2ef58f98c931adfed27
Signed-off-by: Eyal Bari <ebari@cisco.com>
|
|
Change-Id: Ia931377004c2996826a5f504d16c3c934ca3b70d
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Some counters (bytes, pkts) are formatted as signed instead of unsigned
in "show hardware-interfaces" and "show lb".
These stats counters are declared as u64.
Change-Id: Id1b588188bff4e36402beb8d07f779e9a5193956
Signed-off-by: Naoyuki Mori <naoyuki.mori@intel.com>
|
|
Orignal code hard code TLS ciphers, and this patch allows user to
set ciphers via CLI, so that user can perform the TLS testing
without re-building the code.
Change-Id: I0d497f6d906af25bc7a33cee5747f9a1d63e0683
Signed-off-by: Ping Yu <ping.yu@intel.com>
|
|
Change-Id: I1f54b994425c58776e1445c8d9fe142e7a644d3d
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I634971f6376a7ea49de718ade9139e67eeed48e5
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit d039281e11cfc4580fe140e72390c1c48688c722)
|
|
Add missing calls to clib_mem_init to vppinfra test codes.
Change-Id: I53ffc6fc287d1a378065bb86c18b6e995ecdb775
Signed-off-by: Damjan Marion <damarion@cisco.com>
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Enable client-IP based session affinity per LB NAT rule with specific timeout.
Change-Id: I9aade152e330218d21dfda99cc5e984d769ab806
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Change-Id: Id45a3a37dd77771289622314406dccf50b64f04d
Signed-off-by: Klement Sekera <ksekera@cisco.com>
|
|
from vpp_papi.vpp_stats import VPPStats
s = VPPStats(socketname='/var/run/stats.sock')
c = s.ls('/if/rx')
counters = s.dump(c)
print(s.set_error_str())
Change-Id: I203ebe60b0c9ee5742aadc737c0f29051757959d
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: I982b69390c55b5ffbd744f355efc0aaf425b360c
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I9a0df8d15deefdf31cfead56c96433cd7220b802
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I8e2dd1902f57ee173dd266743b46f5602548a4c1
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I5d0777421debe6c3e4b93d62ad6790912063a9e6
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Previously, a service is specified by vip.
This patch extend that a service is specified
by both vip and per-port-vip cases.
Change-Id: Icbfd1f972c6bafde7d85c6abb498576bd9ba250d
Signed-off-by: Hongjun Ni <hongjun.ni@intel.com>
|
|
Change-Id: I45b0c48e89cb1ea556c795f6163a8346f2e2e0da
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: Iadf4ce27f56400b669b80f2e718b6d4330c949a5
Signed-off-by: Dave Barach <dbarach@cisco.com>
|