Age | Commit message (Collapse) | Author | Files | Lines |
|
DPDK recently added a check in the virtio driver to make sure that
rxmode->mq_mode == ETH_MQ_RX_NONE. We were passing ETH_MQ_RX_RSS
and the device initialization was not accepted.
The reason for the change in DPDK was that there is no controls
(algorithm, redirection table, hash function). So they thought ETH_MQ_RX_NONE
was the best choice for the value of mq_mode.
Type: fix
Ticket: VPP-1853
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: Ifa0fc4206cedc56a851f94f6434a2a7500bbd419
(cherry picked from commit b32436aab9626cb8b7bc4099142135c5c9e19beb)
|
|
One actual bugfix.
Type: fix
Ticket: VPP-1837
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Icef25167f97a70cc795c0a481174de319ed79ad5
(cherry picked from commit c35f3e835b4078fedabc1ff5013bc4727f533e16)
|
|
This prevents a crash with quic listeners, and enables the display
of udp fifo status.
Change-Id: Ib9f48818ee3e51a3fa43ad8ab175e8aa7750df8f
Type: fix
Signed-off-by: Aloys Augustin <aloaugus@cisco.com>
(cherry picked from commit 6eef40bce3f5ae2b06ba75d5b4cf32f168a801e4)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I3f34011ca61ded310d0411e7b50548982bd164ac
(cherry picked from commit 936197467aac08b7620c9cb0614817b90466968e)
|
|
Type: fix
The data populated into an ipsec_tunnel_protect_details message includes
an outbound SA and a list of inbound SAs for a tunnel interface. These
are populated with SA indices. The values used by an API client
to refer to an SA in other messages is the SA id rather than the index.
Use the SA id instead of the index.
Change-Id: Ifaad32801092a7f87bd0dcf19de418d36613f8dd
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
(cherry picked from commit 5cee0bca5d0f01d3f26e90dee79780382e843d04)
|
|
Type: fix
Ticket: VPP-1837
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Ib5da600b7a40f775de7dc5b9aaa8d967e49e8632
(cherry picked from commit 4897d77c6d4d5d04eb7e02bda57dc6c7005a609f)
|
|
Type: fix
it was marked MP safe in the CLI (which it shouldn't be) but
it it not marked MP safe on the API.
Change-Id: I4bdea498a510a8b406d13d62a899b6d03656f7e8
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit 31d6c738c20d3baa619dbbb08af6125fab32bc08)
|
|
If buffer alloc fails, it may happend that rx queue will be stuck
as old code only refills if at least one packet is received.
Type: fix
Change-Id: I388c4f8a9fb2c208bdc222e31b443cbe6b94af82
Signed-off-by: Damjan Marion <damarion@cisco.com>
(cherry picked from commit e7f7190381c66cca6c1c3c77b3d43148818a5a4e)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I0875705d3a0c95f2781b0595ef27a30486438aae
(cherry picked from commit 2a6642e95dc18d7c56f87b74a70658b8312b8c08)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Iec5db8dcf3a019b731e15fd79d0208d6eb10943b
(cherry picked from commit eda1b8c32f9857bb749835b0189af002b224927d)
|
|
Type: fix
Change-Id: Id20a0fe77372602fd211156ccee01c18d829d8df
Signed-off-by: Damjan Marion <damarion@cisco.com>
(cherry picked from commit 1839fe165c7ffb834775b8582fe0ee2321ff2ab6)
|
|
Turn on gso, turn off both indirect and mrg_rxbuf caused traffic received
and sent with checksum error. The problem is we are not mapping the hdr
correctly in the shared memory address.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I7ef3bc2755544167b0e624365988111b17399e89
(cherry picked from commit b232d192f2f77114e26cf7f99b57bcd6140872d0)
|
|
Right now following configuration leads to crash:
cpu {
corelist-workers 2
workers 2
}
because threads count will be set to 2, but we have only
one core in coremask.
Type: fix
Signed-off-by: Vladimir Isaev <visaev@netgate.com>
Change-Id: Ia93b892733971e7c8ddfceaaec5f4eb8bf9063ac
(cherry picked from commit 18a4a371646bccfd299e6a509e801a524aeb4c92)
|
|
On SMP architecture, '/sys/bus/pci/devices/<devices id>/numa_node' file
will return -1 as a valid value if it does not have any NUMA node information.
Using -1 as a valid node id to access data structures will cause memory issue.
Fix the error by setting the value of numa_node to 0 if '/sys/bus/pci/devices/
<devices id>/numa_node' returns -1 and it is a SMP system.
Type: fix
Change-Id: Ib60e79c3656fe5b17e08fd9011122683e8b08b6f
Signed-off-by: Jieqiang Wang <jieqiang.wang@arm.com>
(cherry picked from commit 76c6159d83c2dfe29f84dc4b05d399cdbbdf2878)
|
|
Type: fix
Signed-off-by: Jawahar Santosh Gundapaneni <jgundapa@cisco.com>
Change-Id: Ife0f7749a72cc834a3d811f8cf6b5a0840157014
(cherry picked from commit 62ad2aaff3810f4ec0f010700091c2f59fbfe466)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ia2e29b4776c0f8df666da39a14c5ef2fa1693c21
(cherry picked from commit 7fd59cc79c9fb0cccd0cb5c0b4579d0f0a004f6b)
|
|
Rather than leaving 2 bytes of junk in the upper word
of a mac address represented as a u64, zero them out.
That way later compairsons stand a chance of matching
when deleting a bridge's arp termination entries.
The volatile qualifier shouldn't be needed here, but
without it the compiler removes the clib_memcpy() at -O2.
Bad compiler. No biscuit.
Type: fix
commit: faf22cb303b65e2a6bf8dad959d7f5ee6d031c4f
Change-Id: Iebcf35fdd421293dccbcaefadef767f7e139438e
Signed-off-by: Jon Loeliger <jdl@netgate.com>
(cherry picked from commit 6a32ce326495bfe48ebef74dfbb8a9c1cf37a530)
|
|
update packet generator test scripts to reflect latest code changes
Type: fix
Signed-off-by: ShivaShankarK <shivaashankar1204@gmail.com>
Change-Id: I048a5e8b9b0506dcb0b664e5549d13f51b2e2ea6
(cherry picked from commit b0f956c255de7aa61fead96b7b94bcf937d92c05)
|
|
Type: fix
Ticket: VPP-1837
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I357eb72d478d8175ab9c7cf502d312ab3112213e
(cherry picked from commit a53068905eb3f005a5a6d324c37f51056fb49b3e)
|
|
The port registry uses host byte order for while the session API
uses network order. In a single place the conversion was missing.
Type: fix
Signed-off-by: Andreas Schultz <andreas.schultz@travelping.com>
Change-Id: Ic8cfe2cb4e0711b3e0614060ff6b4f2fe4ed4391
(cherry picked from commit 49a11cff8e7892858d946d3f93ae19502236c358)
|
|
This path flag FIB_API_PATH_TYPE_INTERFACE_RX is not copied to the client bin_api.
Type: fix
Change-Id:
Signed-off-by: IJsbrand Wijnands <ice@cisco.com>
Change-Id: I612044d2f564c852f83fceb63ce750a6330e1365
(cherry picked from commit 79437c8dbc707e6f60e7a2425fac15c4153f71b3)
|
|
Type: fix
Signed-off-by: Mark Nelson <manelso2@cisco.com>
Change-Id: I4f121e49d3c05c21eed3fed2469bd88fc84e2271
(cherry picked from commit ea2abbaeaf34a4652e970fd1e2f60c0d377ebde4)
|
|
Type: fix
Change-Id: If50477f2487d8c13f91c119a1492370e6e937904
Signed-off-by: Ondrej Fabry <ofabry@cisco.com>
(cherry picked from commit eec5d48040695e45c4880adb59bb18ebacd667c1)
|
|
VIP prefix index becomes always 0 when adding a VIP which is already registered different port, causing LB config crash.
This change assigns the same VIP prefix index to the same VIP.
Ticket: https://jira.fd.io/browse/VPP-1834
Type: fix
Signed-off-by: Yasuhiro Nakamura <yanakamu@yahoo-corp.jp>
Change-Id: Ib63b3e58db9bd85714d68cd1292aadd0c8580da8
(cherry picked from commit 551775eaaa9c162c73e15690e4d7580935e9a70a)
|
|
build vat with src/vnet/ip/ip_types_api.c
Type: fix
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
Change-Id: Iab0f18bf7a89cf9512beab0629bc3a349edec383
Signed-off-by: Jakub Grajciar <jgrajcia@cisco.com>
(cherry picked from commit 23a386b71b287bc95000a8980856c5dbe7cdaf5a)
|
|
Without this patch offset for TCP/UDP headers was not calculated
correctly if there is one or more IPv6 extension headers.
Type: fix
Signed-off-by: Vladimir Isaev <visaev@netgate.com>
Change-Id: I04d6f5e42f8f072987192d6236085afbd74a4420
(cherry picked from commit 7d4cd0cf6f1a94953ef97ab885752424dea6948c)
|
|
coverity complains that the subject test may cause dst buffer overrun
problem and it is right. The problem is when __builtin_constant_p (n)
returns true, memcpy_s_inline skips all the errors checking and does the
copy blindly. Please see the code in memcpy_s_inline.
The fix is to skip the subject test when the aformentioned builtin function
returns true.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I50de91cc0c853a134b3bcf3b0cd8d45d7668b092
(cherry picked from commit 2da39718f560478678caacccd198ee4c0c9673c3)
|
|
Type: fix
Change-Id: Ia42ff39a0a33f89901b8333a9e6ca82ca9805cc6
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
(cherry picked from commit 41e831f5588fe5ebfd879f4e570e85e12770b360)
|
|
Type: fix
Signed-off-by: Simon Zhang <yuwei1.zhang@intel.com>
Change-Id: Ie2dc2653baec92347eb2cbcd197a2e5ec6a80c79
(cherry picked from commit 487507f40f4e443ff1e683641206db80875f3477)
|
|
Type: fix
Change-Id: If74ad528e68f45b00719295388e0e1399452ef93
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit e9a630a5248ee6e234c1bf1fdb2c29fc6239f60d)
|
|
Type: fix
Change-Id: Ia61d6fbf6e80977f83f1f6672e5e83b52ddeb0e5
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit ea96e92361b483962fd2a6b027cedc02f3bb6f93)
|
|
Type: fix
Change-Id: I008f23b5f0c7269ddd35cc747a867240fbe9c49b
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit d057625d499525625d60d2207665eaeb755e380e)
|
|
Type: fix
Change-Id: I048c9ed423ca2993d2179cdce364ac98980311bb
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit 39a233a0aa21c644c78ddd4ffa0ab3cdb1c10318)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I4153a9a377b8b1da1366a94ff791ad99617b7a6d
(cherry picked from commit 62a7fe28933b6310b9a4e3a0fab99949587576ac)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ie060b25b5e8c85d9b2037e300723a2b6023c65d1
(cherry picked from commit 6ce954f8e4446472689146024126ba42da182aa9)
|
|
Type: fix
Change-Id: I57f8bfbce4feed9d2775875cb8b1b729a47900a4
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit 24064d02aa9810ebc64c16dc778a179bb0ef5483)
(cherry picked from commit 3ebebc3a2fe47f1222ba035e04ccd8caed0cf58f)
|
|
With this commit, forward the translated packet directly to the
specified next-hop if pre-resolve param is enabled in MAP-T.
Type: fix
Change-Id: Ie26080c7820318c7982599577a4af6e4d01a0574
Signed-off-by: Alexander Chernavin <achernavin@netgate.com>
(cherry picked from commit f145c15631ba62e798395499f83a2f8a91ae83c7)
Signed-off-by: Dave Wallace <dwallacelf@gmail.com>
|
|
Add an ALWAYS_ASSERT (...) macro, to (a) shut up coverity, and (b)
check the indicated condition in production images.
As in:
p = hash_get(...);
ALWAYS_ASSERT(p) /* was ASSERT(p) */
elt = pool_elt_at_index(pool, p[0]);
This may not be the best way to handle a specific case, but failure to
check return values at all followed by e.g. a pointer dereference
isn't ok.
Type: fix
Ticket: VPP-1837
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: Ia97c641cefcfb7ea7d77ea5a55ed4afea0345acb
(cherry picked from commit 47d41ad62c5d6008e72d2e9c137cf8f49ca86353)
|
|
Type: fix
Change-Id: I50329bda365d98f9f9d56a58187fb4fb2a4eb461
Signed-off-by: Yu Ping <ping.yu@intel.com>
(cherry picked from commit 985d9293a08dc3da016fbeeaa3f8fff10e1b504e)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I60be191866d20721951ad22f571a2a3275511e12
(cherry picked from commit 067f9544d52c95c0b60b0e8425fce1e295120180)
|
|
Type: fix
this patch fixes mark flow deletion crash issue, see below
test flow add src-ip any proto udp src-port 111 dst-port 222 mark 100
test flow enable index 0 1/1
test flow disable index 0 1/1
test flow enable index 0 1/1
test flow disable index 0 1/1 -> [crash]
This is because the code resets a wrong vector in flow lookup entry
recycle logic. See function dpdk_flow_ops_fn().
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
Change-Id: I2b0a1e531931ab25541d672d88da18dc2289f1ce
(cherry picked from commit cd120f9bbb2101dfd7eca11d1a28e06ac5ace479)
|
|
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I756170bd799d1f482186cbb4b5dff9373ae6e08f
(cherry picked from commit 3be33f17ecd14a12738a44f9c0e09cb3778b1345)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Idb513232e7b091c8f767726bfa1deb10a7e3b751
(cherry picked from commit e88845e4fea2c6b6743ff7790cc2247631d65189)
|
|
Type: fix
Change-Id: I5c81d2f55057f5fba780cb12154a3fb1aef79f20
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit aba4983ad48374a50cd93ba91f66be241f210279)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Idaa7cc26ad941be86daec4ed5920727237712f4a
(cherry picked from commit 16d974ec59776f0103ad62d0d04dc57989eef7ed)
|
|
Type: fix
Change-Id: I77b03efcac04cc46550d03657464ab8de5d7da78
Signed-off-by: Klement Sekera <ksekera@cisco.com>
(cherry picked from commit 90d28846f963a86d760b4a6b83aed62b862f1c61)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: Ie8bc5c9a03e858487cf565b4e9b520e6b496337a
(cherry picked from commit fa2a316663e622a2feeecf8ad7d32b242370a70a)
|
|
According to the description in the startup.conf, the assignment
of worker threads starts with the lcore following the main_lcore.
A non-zero skip_cores will correctly achieve this assignment.
However, prior to this patch when workers are assigned, the code
picks up and assigns core 0 even thought it shouldn't.
This patch determins if a non-zero number of workers are desired
and if so, marks CPU unavailable for a worker assignment.
Type: fix
Change-Id: I1fdf73a6f218dcbf146fda2efc90c553f7cd6d20
Signed-off-by: Jon Loeliger <jdl@netgate.com>
(cherry picked from commit 4a06846dd668d7f687e6770215c38e8feb5f1740)
|
|
Type: fix
Signed-off-by: Florin Coras <fcoras@cisco.com>
Change-Id: I87940d02813b66616f1944e41b16c8cc16f5dac0
(cherry picked from commit deb6f784edbaddd2e60e0d6ac9927f17c4214ad2)
|
|
Type: fix
Change-Id: I0c9640dab2c0eaba369bc8f3ff7ae56d8e97e170
Signed-off-by: Damjan Marion <damarion@cisco.com>
(cherry picked from commit f1ecb6555326f8a7d5bba6f56aa676f064675dc0)
|