the bihash key compare was doing the extra step of loading both keys into
vector regs and comparing those - instead of using the temporary values
created.
Change-Id: Ic557bfa56cd6aa60c71e7f28c880f85e1eb1e6ec
Signed-off-by: Eyal Bari <ebari@cisco.com>
This adds byte_swap (variant of shuffle) and shuffle vector intrinsic
for ARM based on Neon, concuring with same signature as SSE vector
intrinsic.
Change-Id: I386fd2b1dcc83654e4ad9f90a6065d7736e4ce5c
Signed-off-by: Sirshak Das <sirshak.das@arm.com>
When compiled w/ -DCLIB_DEBUG=1, paint 0x13 across freed memory
blocks. Should result in a characteristic SIGSEGV if someone
dereferences a pointer in a freed vector element, etc.
Change-Id: I5f78970f4342310bfbe7adeddb56feff21f0de2c
Signed-off-by: Dave Barach <dave@barachs.net>
Several test plugins report undefined symbol while being loaded
by load_one_vat_plugin. Fix this by adding VPP_API_TEST_BUILTIN
into CFLAGS for building these plugins.
Change-Id: I908720fd0d01da1ead30ba17027ba10358f6bdf1
Signed-off-by: Yi He <yi.he@arm.com>
create runtime dir when adding default socket filename.
Change-Id: Iad7e751fcab7ce773e0907302414a06ea44dd6cd
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
this receipe:
/* Get a line of input. */
if (!unformat_user (input, unformat_line_input, line_input))
return 0;
only works if there is more data following the registered command name.
So it is not so good for show commands...
Change-Id: I54249865a44526ade4b40e2d6207138a2d056e40
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
Avoids crashes on restarts if svm root region backing file was not
cleaned up.
Change-Id: I608cf5711aa8c3f9620900473bdf76bde8b918de
Signed-off-by: Florin Coras <fcoras@cisco.com>
Without inlining [TAG=vpp_debug], gcc7 (at least) refuses to produce
the indicated vector unit instruction.
Change-Id: I0f0400ad74b1e498dce7963a85c47d33afe0a768
Signed-off-by: Dave Barach <dave@barachs.net>
Have vcl poll and wait on the event message queues as opposed to
constantly polling the session fifos. This also adds event signaling to
cut through sessions.
On the downside, because we can't wait on multiple condvars, i.e., when
we have multiple message queues because of cut-through registrations, we
do timed waits.
Change-Id: I29ade95dba449659fe46008bb1af502276a7c5fd
Signed-off-by: Florin Coras <fcoras@cisco.com>
Also correct types.h for mips64 which could never be hit
as _mips was part of the previous ifdef.
Change-Id: Id0435c8fc960c5d25c43129b9d9f1606e39ba8e3
Signed-off-by: Carl Smith <carl.smith@alliedtelesis.co.nz>
Also: call os_panic() on a heap botch crash, attempt to generate a
post-mortem API dump, etc.
Add an "ugly" test case to vec_test.c, to cause a configurable block
overrun.
Change-Id: I7b29a7645277f9e485e06ff83335306fedc24b71
Signed-off-by: Dave Barach <dave@barachs.net>
Reduce default queue size to twice the vlib frame size. This throttles
the PP from transferring more packets than are necessary for the next
round of graph execution on a core. When combined with DMA'ing into L3
cache, Mpps for 64B packets increases 35%.
Change-Id: I0afe426d23560f262fdc56accb6302b099bd0076
Signed-off-by: Brian Brooks <brian.brooks@arm.com>
Needed a spinlock to protect the data vector. Cleaned up debug cli so
the output makes sense, and so that various parameters exist in one
place. Removed a nonsense memset-to-zero which led to ultra-confusing
results.
Change-Id: I91cd14ce7fe84fd2eceab86e016b5ee001993be4
Signed-off-by: Dave Barach <dbarach@cisco.com>
Three separate implementations, which vary by nearly a factor of two
in performance. Most of the performance difference is due to swapping
the src/dst mac addresses with an avx2 vector shuffle instruction.
Change-Id: Ieb36546d6074e4ac720d452a99d013c698135c57
Signed-off-by: Dave Barach <dave@barachs.net>
calling thread cancel on the timeout thread whilst it was
sleep on condwait and then send the cond signal did not reliably
wake up the thread.
instead don;t cancel the thread, use a loop variable to terminate it.
Change-Id: Ibc8ab6f21db7e4a98266bdf88b8b208b887820dd
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
Wrong vl_msg_id set in vl_api_stats_get_poller_delay_t_handler while reporting stats delay.
Change-Id: I7d6ad14359e41c717b976390c58e254e3602e0ff
Signed-off-by: Lukasz Majczak <lma@semihalf.com>
- assign rx thread on interface create, not on admin up
- keep interrupts unmasked all the time as they are needed for writeback
- forward interrupt to graph dispatcher only if queue is in the interrupt mode
Change-Id: I5cc5afe24960143bef08f12d59fbd80ae0c6ccf8
Signed-off-by: Damjan Marion <damarion@cisco.com>
in case of relative path, create subdirs inside runtime dir.
in case of absolute path, if folder does not exist return error.
Change-Id: I01fe018102c69deb105160cfa18b741541d76c82
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
avf_create_reply returns software index for the new interface
Change-Id: I8a6b1a1985b072efafa24eb258b1f2cb1bea1110
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
Move the tcp unit test to the plugin
Add a bihash unit test and a "make test" program to call it
Adjust framework.py to load the plugin, which is disabled by default
Change-Id: Ic229d386a56a9d28dbd54974f231149053ca8f93
Signed-off-by: Dave Barach <dave@barachs.net>
One handler is good enough when engine sends out a retry status
Thus this patch will just go one branch
Change-Id: Id81cb3fa67d2b322b0fe1b2f62cd866cf3491eb4
Signed-off-by: Ping Yu <ping.yu@intel.com>
It's not clear that fine-graind pipelining will ever produce optimal
performance. That having been written, it's an easier coding model for
folks interesting in decent results without having to manually
implement 2 or 4 way fine-grained parallelism.
Change-Id: Ida877292832eaf759ffbb6d2e16012f8fbd427e5
Signed-off-by: Dave Barach <dave@barachs.net>
Can happen if code bypasses the per-interface output node, and
dispatches packets directly to the tx node.
Switch to vlib_get_buffers(...) ... vlib_buffer_enqueue_to_next (...),
quad/single loop coding pattern.
Change-Id: Ic0e5d3b9748230f4e545a54186e6e64e7a782bb1
Signed-off-by: Dave Barach <dave@barachs.net>