Fix a day-1 bug, possibly dating back as far as 2002. The zap64() game
involves fetching 8 byte chunks, and clearing octets not to be
included in the key.
That's fine *unless* the 8-byte fetch happens to cross a page boundary
into unmapped or no-access space.
Type: fix
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I4607e9840032257c96ba7387f86c931c0921749d
(cherry picked from commit 7e2cea3d26)
vnet_feature_enable_disable takes sw_if_index, not hw_if_index. If there
is a subinterface created prior to the slave interface is created,
sw_if_index and hw_if_index start to diverge and the problem will happen.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I11e1f099378832f83b748526c6cbeb56960fad3c
(cherry picked from commit 1a41a35b27)
The non-extern declaration confuses clang linker in debug mode.
The function is defined as inline above anyway.
Type: fix
Fixes: c6215d902f
Change-Id: Ic7e4477631cf0bcfb31ab3f81effe3642dd4223e
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 5b1379be3e)
otherwise they get installed twice and the reference counting means they are not removed.
This is the same behaviour as IPv4.
Type: fix
Change-Id: I9266e04ccff6ff06a577e85973a2ddbeb9dfc52b
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit 1ff3c15b3c)
they can use the 'auto' adj for all traffic
Type: fix
Change-Id: Id2b9557683252a94badc8f9dfab5f7b2ae26f1ee
Signed-off-by: Neale Ranns <nranns@cisco.com>
(cherry picked from commit da0e7497ca)
If VPP_EXTRA_CMAKE_ARGS is set, its content will be
appended to the vpp cmake command cli
Type: feature
Change-Id: I825d4239e62b0a2fb70a652f0671f6c559630aad
Signed-off-by: Nathan Skrzypczak <nathan.skrzypczak@gmail.com>
(cherry picked from commit 2973654033)
The old code modified the node next array prior to obtaining the thread
barrier. Then it updated the runtime node data, and upon barrier release
caused reforking of each worker thread. The reforking clones the main
thread nodes and reconstructs the runtime node structure. This cloning
is not 100% "deep" in the sense that the node next array is
shared (i.e., only the pointer is copied). So prior to the barrier being
obtained the node's next array is being changed while workers are
actively using it (bad). Treating the node next array as read-only in
the workers and sharing it is a decent optimization so instead of trying
to fix that just move the barrier a little earlier in the process to
protect the node next array as well.
This was tripping an assert in next frame ownership change by way of the
ip4-arp node. The assert verifies that the node's next array length is
equal to the runtime next node count. The race above was lost and the
node next array data was updated in the main thread while the arp code
was still executing in a worker.
This was being hit when many arp requests were being sent from both ends
of a tunnel during which the add next node function was called, which
often led to an assert b/c the next node array was out of sync with the
runtime next node count.
- PS#2 update - move barrier sync to just above code that modifies state.
Ticket: VPP-1783
Type: fix
Signed-off-by: Christian E. Hopps <chopps@chopps.org>
Change-Id: I868784e28f994ee0922aaaae11c4894a3f4f1fe7
Signed-off-by: Christian E. Hopps <chopps@chopps.org>
(cherry picked from commit d3122ef4ec)
Prints the interior node vector rate, rx / tx / drop rates
Type: feature
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I57130db0f99e852a8498aa90d01e52f7ac33dcc9
(cherry picked from commit ac78f8a902)
In bond RX quad loop, when all packets within the frame have the same incoming
interface, we cannot skip calling bond_update_next because that function calls
vnet_feature_next() to update the b->current_config_index. The next node needs
the correct b->current_config_index to work with.
Type: fix
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I3d8b3d4e0f95490f406fae7638f0c43c301ce664
(cherry picked from commit 71e5b47102)
Allow for setting the maximum number of generated packets to be included
in the frame passed to next nodes. This is very important for testing
code which may be susceptible to multi-frame vs single-frame bugs (e.g.,
code that is doing re-ordering where packets may be buffered between
frames).
Update:
- remove redundant packet "rate" option.
- reduce n_max_frame to u32 as that's what pulled from the CLI.
Type: feature
Signed-off-by: Christian E. Hopps <chopps@chopps.org>
Change-Id: Ie362bbb110b2cf01d9f65c559bbe9101e17b7fdc
Signed-off-by: Christian Hopps <chopps@labn.net>
(cherry picked from commit 87d7bac5cf)
If no Linux PCI driver module is loaded, then the driver_name in the PCI
info struct is NULL. This can triggers crash when checking driver name
eg. in vlib_pci_device_open().
Default to "<NONE>" as driver name, which should never match.
Type: fix
Change-Id: I9e69889a7566467bd8220b92bbbaa72ada957257
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 0eae2bb1f1)
In a rare event, after the vhost protocol message exchange has finished and
the interface had been brought up successfully, the driver MAY still change
its mind about the memory regions by sending new memory maps via
SET_MEM_TABLE. Upon processing SET_MEM_TABLE, VPP invalidates the old memory
regions and the descriptor tables. But it does not re-compute the new
descriptor tables based on the new memory maps. Since VPP does not have the
descriptor tables, it does not read the packets from the vring.
In the normal working case, after SET_MEM_TABLE, the driver follows up with
SET_VRING_ADDRESS which VPP computes the descriptor tables.
The fix is to stash away the descriptor table addresses from
SET_VRING_ADDRESS. Re-compute the new descriptor tables when processing
SET_MEM_TABLE if descriptor table addresses are known.
Type: fix
Ticket: VPP-1784
Signed-off-by: Steven Luong <sluong@cisco.com>
Change-Id: I3361f14c3a0372b8d07943eb6aa4b3a3f10708f9
(cherry picked from commit 61b8ba69f7)
gid_ip4_table_t's and gid_ip6_table_t's are allocated from pools. They
MUST NOT be listed on the clib_all_bihash list to avoid dangling
references.
Switch to the clib_bihash_init2 API, which has the required knob.
Type: fix
Ticket: VPP-1788
Signed-off-by: Dave Barach <dave@barachs.net>
Change-Id: I49a17e937922c3af2e1c46b24e20883af51584a8
There is no bt->samples for this test, do not use it.
Type: fix
Change-Id: I2090290887bc5c0b5cdb0561cf2bf72a87781089
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit b0a7c484ee)
ACL tests use random port number in the tests.
A port number 6081 causes the decode in scapy
to consume some of the Raw payload into GENEVE
encoding, which breaks the test.
Solution: bring up the lower range of random
port to 16384, so that it does not touch any
of the well known ports.
Type: test
Change-Id: I022660d8ec147857924b436f1871b0b5ddcf4c47
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
(cherry picked from commit ec574ff912)
- set msgid to 0 not random.
- allow for no DH in ESP child SA
Ticket: VPP-1781
Type: fix
Signed-off-by: Christian E. Hopps <chopps@chopps.org>
Change-Id: Ibe26009d38f444eeaec5b042097f145d161c7672
(cherry picked from commit 0e182c5b1d)
Make sure to reinitialize data before free-ing it.
Type: fix
Change-Id: I45727c456d0345204d4825ecdd9690c5ebeb5e94
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit d4aeb84c3f)
Type: fix
The scenario is touch of a dpdk source file in places like
<TOP>/build-root/build-vpp-native/external/dpdk-19.08/. The subsequent
build will trigger an rm -rf $(PACKAGE_INSTALL_DIR) by
build-root/Makefile at around line 709. Because the package in
question, "external", consists of various "sub" packages, all of which
are controlled with files like .quicly.install.ok located in
build-root/build-vpp[_debug]-native/external, we end up not handling the
.*.install.ok files to reflect the removal of that PACKAGE_INSTALL_DIR.
This fix will correct that, at least for the case of touching/changing
dpdk files. A better long term solution would be to change the "sub"
packages to be full fledged build system packages, as Dave Barach points
out.
Signed-off-by: Burt Silverman <burtms@gmail.com>
Change-Id: Ia94ccaa2795cbf354b5921c3001a7141313b66be
(cherry picked from commit 839a7bce38)
Define CLIB_PAUSE () to generate the "yield" instruction. No significant
performance changes were observed for clib_spinlock_t and clib_rwlock_t.
Type: feature
Change-Id: I59eb996e61c7a16007517e57e6996567302c1657
Signed-off-by: Jason Zhang <jason.zhang2@arm.com>
Reviewed-by: Lijian Zhang <Lijian.Zhang@arm.com>
(cherry picked from commit 18512b002d)
Type: feature
Based on the configuration, we can disable checksum offload capability
and calculate checksum while pushing the TCP & IP header.
This saves some cycles when VPP stack is used in legacy hardware devices.
Signed-off-by: Srikanth A <srakula@cisco.com>
Change-Id: Ic1b3fcf3040917e47ee65263694ebf7437ac5668
TX queues must be created before RX queues on Mellanox cards in order to
not receive our own broadcast packets.
Type: fix
Change-Id: I32ae25a47d819f715feda621a5ecddcf4efd71ba
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit df213385d3)
The rdma driver use the pci subsystem, make sure the dependency is
recorded.
Type: fix
Change-Id: Ibd613f623d355612881acc31b9423f2de13793a3
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 4a98388dae)
When deleting interface from the API, 'hi' gets removed before
'h->sw_if_index' is copied.
Type: fix
Change-Id: I8e10108e9bdf95ab2fe002790d98262d583ca58c
Signed-off-by: Benoît Ganne <bganne@cisco.com>
(cherry picked from commit 8133c780a2)
