Age | Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: I798e4fb6470ae9e763f8de1c290ff0fc3c0b7f9e
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I363370b9d4a27b992bad55c48fc930a2fbea2165
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I03bb47a2baa4375b7bf9347d95c4cc8de37fe510
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
Currently, there are three variants available on aarch64, qdf24xx, thunderx2t99, and cortex-a72.
-DCLIB_N_PREFETCHES is passed to source code to select dual/quad implementation.
Besides, different compiler options are applied on these critical functions.
gcc-7.3.0 reports ICE(internal compiler error) with -mtune=thunderx2t99,
so -mtune=thunderx2t99 is enabled only when gcc version is greater than 7.3.0
Cavium ThunderX2, Impermenter 0x43, Part 0x0af
-march=armv8-a+crc+crypto -mtune=thunderx2t99
Qualcomm Centriq 2400, Impermenter 0x51, Part 0xc00
-march=armv8.1-a+crc+crypto -mtune=qdf24xx
Cortex-A72, Impermenter 0x41, Part 0xd08
-march=armv8-a+crc+crypto -mtune=cortex-a72
Change-Id: Id5649c6325c1e642d0fd42535e3908793b13e02a
Signed-off-by: Lijian Zhang <Lijian.Zhang@arm.com>
Reviewed-by: Sirshak Das <sirshak.das@arm.com>
Reviewed-by: Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com>
|
|
Change-Id: I2d3b8d5a7192ff68bee443a99346ecb807b2d833
Signed-off-by: Damjan Marion <damarion@cisco.com>
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: I8aba11142daeacce892ec529c2229ec54ee427e2
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Change-Id: I3e62c787882d93c3bfb398ed0d04ef56b3a60b2b
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: I229ce9c306a97956c87ccb8f7a1c5c9dac94ab31
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
- Reduce per packet cost by buffering the output packet buffer indexes in the queue and
process the queue outside the packet processing loop.
- Move unnecessary variable initialization outside of the while loop.
- There is no need to save the old interface if tracing is not enabled.
Test result for 256 bytes packet comparison. Other packet size shows similar improvement.
With the patch
--------------
BondEthernet0-output active 52836 13526016 0 1.71e1 256.00
BondEthernet0-tx active 52836 13526016 0 2.68e1 256.00
TenGigabitEthernet6/0/0-output active 52836 6762896 0 9.17e0 127.99
TenGigabitEthernet6/0/0-tx active 52836 6762896 0 6.97e1 127.99
TenGigabitEthernet6/0/1-output active 52836 6763120 0 9.40e0 128.00
TenGigabitEthernet6/0/1-tx active 52836 6763120 0 7.00e1 128.00
bond-input active 52836 13526016 0 1.76e1 256.00
Without the patch
-----------------
BondEthernet0-output active 60858 15579648 0 1.73e1 256.00
BondEthernet0-tx active 60858 15579648 0 2.94e1 256.00
TenGigabitEthernet6/0/0-output active 60858 7789626 0 9.29e0 127.99
TenGigabitEthernet6/0/0-tx active 60858 7789626 0 7.01e1 127.99
TenGigabitEthernet6/0/1-output active 60858 7790022 0 9.31e0 128.00
TenGigabitEthernet6/0/1-tx active 60858 7790022 0 7.10e1 128.00
bond-input active 60858 15579648 0 1.77e1 256.00
Change-Id: Ib6d73a63ceeaa2f1397ceaf4c5391c57fd865b04
Signed-off-by: Steven <sluong@cisco.com>
|
|
And more unit-tests.
Change-Id: I4667d82d928b7ba8d96b5a5648d464115b3ed216
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: Idc17b2f8794d37cd3242a97395ab56bd633ca575
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Split the stat_segment.c code from stats.c.
Rename stats.[ch] to prepare for removing (19.01?)
In addition stats.api can be removed.
Since the stats aggregation for the stat segment does not use the API, that
part is now done on the main thread. (Old stats aggregator is also left in place).
Change-Id: I9867429f4fc547b1a7ab7f88bc4f3625428d681b
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: I17b19875547cd6056aaf8f8fde439caa8e42d274
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Change-Id: I261e11f1006039e4a4019fec2fd3271763ae476a
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
This is a new commit for code under a different maintainer.
Change-Id: I79fa403fec6a312238a9a4b18b35dbcafaa05439
Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
|
|
Change-Id: Iacdff0a5006a0617e3c9559ce7b258f26e256abb
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
There are issues with VPP finding and linking the mlx5 shared glue
library which was built by default if mlx5 was enabled.
Runtime Errors this patch fixes:
net_mlx5: cannot load glue library: librte_pmd_mlx5_glue.so.18.05.0:
cannot open shared object file: No such file or directory
net_mlx5: cannot initialize PMD due to missing run-time dependency on
rdma-core libraries (libibverbs, libmlx5)
This patch introduces additional config parameter to disable glue
library building and instead statically link ibverbs and mlx5
libraries to the PMD and dpdk_plugin.
Change-Id: I0b2f67652a57854c778e991780903fb15706ace8
Signed-off-by: Sirshak Das <sirshak.das@arm.com>
Reviewed-by: Lijian Zhang <Lijian.Zhang@arm.com>
|
|
Change-Id: I36f42c03f778955dd543da6c878be090d0443922
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: Idf83fce8ca176e57b323e3741034e3223f1d195a
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
This code is orignally in automake, but it is missing in cmake.
Thus add it to make openssl async work in cmake build system
Change-Id: Ie69ee9c2099273e51ce13ccab27bdd2619db4814
Signed-off-by: Ping Yu <ping.yu@intel.com>
|
|
In avf_rxq_refill, we invoke vlib_buffer_alloc_to_ring which may fill buffers from the
end of the ring and continue to the beginning of the ring. If we fill some in the end and
continue to fill some in the beginning, but does not have enough buffers to fill the
whole request, n_alloc returns a value which is not equal to n_refill to indicate partial
refill. We don't like partial refill and invoke vlib_buffer_free to get rid of
the buffers that just got refilled. However, vlib_buffer_free API is to free the buffers
from the slot continuously. It does not know how to free some from rxq->bufs[slot], and then
continue to free the rest when it reaches the end of the ring.
The fix is to use vlib_buffer_free_from_ring which is smart enough to figure that stuff
out.
Change-Id: I93c28e0b0d8d8f22c321d1a5912e00c27b4e2e8d
Signed-off-by: Steven <sluong@cisco.com>
|
|
Change-Id: Ied63ebaec8e19189f8b1ab2a7f6d7474c3f2ad6e
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
- Error where ICMPv6 error code doesn't reset VLIB_TX = -1
Leading to crash for ICMP generated on tunnelled packets
- Missed setting VNET_BUFFER_F_LOCALLY_ORIGINATED, so
IP in IPv6 packets never got fragmented.
- Add support for fragmentation of buffer chains.
- Remove support for inner fragmentation in frag code itself.
Change-Id: If9a97301b7e35ca97ffa5c0fada2b9e7e7dbfb27
Signed-off-by: Ole Troan <ot@cisco.com>
|
|
Before
------
DBGvpp# sh int rx-mode
sh int rx-mode
Thread 1 (vpp_wk_0^@):
node vmxnet3-input:
vmxnet3-0/b/0/0 queue 0 (polling)
Thread 2 (vpp_wk_1^@):
node vmxnet3-input:
vmxnet3-0/13/0/0 queue 0 (polling)
DBGvpp#
After
-----
DBGvpp# sh int rx-placement
sh int rx-placement
Thread 1 (vpp_wk_0):
node vmxnet3-input:
vmxnet3-0/b/0/0 queue 0 (polling)
Thread 2 (vpp_wk_1):
node vmxnet3-input:
vmxnet3-0/13/0/0 queue 0 (polling)
DBGvpp#
Change-Id: I5910d502757054c3942fac9d20c5104e95fc6b56
Signed-off-by: Steven <sluong@cisco.com>
|
|
Need to free the buffers starting from ring->consume+1, not ring->consume
Make use of the cool API vlib_buffer_free_from_ring
Fix a memory leak in vmxnet3_txq_init
Change-Id: I0a539ea75211408d84ce433d97a0a0aec5a9618d
Signed-off-by: Steven <sluong@cisco.com>
|
|
The stateful ICMP/ICMPv6 handling got broken.
Fix that and introduce testcases to catch in the future.
Change-Id: Ie602e72d6ac613d64ab0bf6693b6d75afb1a9552
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
|
|
Change-Id: I6c0d5aec6ee96a0d40358f0e09a0901b22265063
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
|
|
Change-Id: I0f39477bbf88d490409fdcd5f58df55cfe2ec531
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: I35dcb987edf11097f34a633ac36d87cecd12088f
Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
|
|
bi0 always points to the first buffer in chain but b0 could be any fragment in a ip packet.
It causes a vnet_feature_next function call to the wrong buffer
and crashes sometime when there is >1 feature nodes under ip-unicast.
The fix is simple, makes b0 and bi0 both pointed to the first buffer
Change-Id: I7de36e68fb42b050fa63201abd98aeb6ba2e0cd3
Signed-off-by: Kingwel Xie <kingwel.xie@ericsson.com>
|
|
Change-Id: I89c7df778e66a5d2147190dc99445405d81964e5
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ie6a6dde1fdc0e4fa8560682072a69876867a88d3
Signed-off-by: Alexander Chernavin <achernavin@netgate.com>
|
|
Change-Id: Ib62e503f4eb5d72431288de32f417a4553df4e0c
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Change-Id: I0c3f2add35ad9fc11308b7a2a2c69ffd8472dd2e
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I7dd48c8a2c77f82a2c1aa8311b062f5f0bc4e3fd
Signed-off-by: Mohsin Kazmi <sykazmi@cisco.com>
|
|
Change-Id: I958bf057be751dc7b3a0d93080021b3addc405b7
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
|
|
Change-Id: I21ad6b04c19c8735d057174b1f260a59f2812241
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I9d3d5243841d5b888f079e3ea5dc1e2e8befd1dc
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
|
|
module: id SFP/SFP+/SFP28, compatibility: 40g_active_cable
vendor: Amphenol, part NDCCGF-I202
revision: C, serial: APF1711202351C, date code: 170318
cable length: 2m
Change-Id: Ife35607b4f078f7b56737fe066ad4cbd247a7504
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
set and check a special RX interface value as the packet enters and
exits a BIER domain
Change-Id: I5ff2f0e2d1b3ce0f3598b935f518fc11eb0896ee
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
|
|
This commit adds explicit signaling from a non-owning thread to the owning thread
to restart the session timer as necessary.
Consequently, we now can sweep the session lists at their respective timeouts,
rather than sweeping all the lists at the pace of the shortest timeout value,
just taking care to wake up if the session requeue to a different list results
in needing to wake up earlier.
Change-Id: Ifc8c500f6988748f4cd3dc184dd7824321aaaaca
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
|
|
Change-Id: Idbe5f4d7f37dbe4dbf8d2679c3d816aeaed148d8
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I49da8be88dd033aae1b190e8e2163069ef480442
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: Icff3d688506e7658330db004c58bcfcac273fcec
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
After the slave interface is removed from bond, bond input node still receives traffic for
the slave interface.
We have to disable feature arc for the corresponding slave interface.
Change-Id: I44e7001e6685e290b032c48147d02911a55d547b
Signed-off-by: Steven <sluong@cisco.com>
|
|
Under extreme condition, it is possible for input node to have more than 256 packets available
to process. Add a check to the while loop to prevent overrun buffer_indices and nexts.
Change-Id: Id9830e5c254f04216eb79e864a5460e08cf2e4e5
Signed-off-by: Steven <sluong@cisco.com>
|
|
Change-Id: I47e1fc789ddf3dbfdf9768b99d5c3a8804d6b750
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Change-Id: I630f3da1ea4e6e50a50f1352c097becef1efe3c0
Signed-off-by: Matus Fabian <matfabia@cisco.com>
|
|
Change-Id: Iedd79eeba35fb7385c03a177f8ac7d3ddfeb6a84
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ifa4fceef7edbe43d444790a624957db0817064de
Signed-off-by: Florin Coras <fcoras@cisco.com>
|