Remove fedora dependency restriction on subunit and subunit-devel packages.
libsunit is only called in test/ext/Makefile on all platforms safe opensuse.
Change-Id: I483f42f20f6ba8bd112c18a9c51753a5d816d1c1
Signed-off-by: Gabriel Ganne <gabriel.ganne@enea.com>
Signed-off-by: Romain Ly <romain.ly@enea.com>
L3 egress path does not set the ethernet flags reflecting the count of VLANs,
but rather has the offset explicitly, so use that.
Change-Id: Id3f6562dcd52ca24137c305f1a1c88c1f125da78
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
The original version printed just a few u64s, which is useful for
directly working on the code, but not when figuring out what is
possibly a config or environment-related issue. So, add printing
the 5-tuple struct in a way that is usable by an operator.
Change-Id: I84cc3a239cdaff05ed31c3458cea198e38b58e03
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
Enhence support of DHCP VSS (Virtual Subnet Selection) to include
VSS type 0 where VSS info is a NVT (Network Virtual Terminal)
ASCII VPN ID where the ASCII string MUST NOT be terminated with a
zero byte. Existing code already support VSS type 1, where VSS
information is a RFC 2685 VPN-ID of 7 bytes with 3 bytes OUI
and 4 bytes VPN index, and VSS type 255 indicating global VPN.
Change-Id: I54edbc447c89a2aacd1cc9fc72bd5ba386037608
Signed-off-by: John Lo <loj@cisco.com>
Set the content of tmp.sock prior to calling memif_msg_send_disconnect.
Also fix the problem socket was not close in the same spot due to
error encountered.
Change-Id: I8f54ebad2250d1944afcc52e71d2a59da05362af
Signed-off-by: Steven <sluong@cisco.com>
Package is already in the RPM_DEPENDS list, and is required for test-* targets
Change-Id: I3846e68855402b50a51adf3daf218e0244562cd6
Signed-off-by: Gabriel Ganne <gabriel.ganne@enea.com>
Need to be NULL-terminated.
Fix declarations of:
- bier_disp_table_bier_nodes
- bier_table_mpls_nodes
- bier_fmask_mpls_nodes
This was crashing during make test on aarch64 platform:
During the API call to bier_table_add_del, the crash happens during
dpo_default_get_next_node().
Change-Id: I16207ba38fc9ab65bad787878c4608740c312257
Signed-off-by: Gabriel Ganne <gabriel.ganne@enea.com>
When BUM packets are flooded in the l2 domain, some data should be
kept and restored for recycling in the replication routine.
If l2 bridge domain has multiple interfaces mixed with normal and
vlan tagged, the vlan tag value of the vnet buffer can be changed
while flooding the replicated packets. The change is made to store
and restore the original vlan tag in the replication logic.
Change-Id: I399cf54cd2e74cb44a2be42241bdc4fba85032c5
Signed-off-by: Steve Shin <jonshin@cisco.com>
With recent introduction of C++ code required memory for each compiler
instance has significantly increased causing build issues.
Currently build system spins 2 compiler instances per logical CPU core.
As CPU can hardly execute more than one thread at a time, it should be
pretty safe to change that formula so it doesn't multiply number of cpu
cores by 2 and such change will signifucantly reduce amount of memory
needed.
Change-Id: Ic829fff6e45f4caf98a6d9c1c98c53ed003039ef
Signed-off-by: Damjan Marion <damarion@cisco.com>
- In the bug report, Docker was sometimes giving shells
a 0,0 terminal size. The minimum-term-size logic meant
that VPP assumed the terminal had 1 row. The pager
functioned, but of course overwrote the one line with its
own prompt.
- Instead of a minimum size, always use a default size when
the either terminal dimension is 0.
Change-Id: Iee5a465f0e4cbb618ef2222b40a52994aefa54bf
Signed-off-by: Chris Luke <chrisy@flirble.org>
Fixed out-of-bounds access in vcom_socket.c by limiting the copy to the
size of the address field that was passed. Truncation will occur if
the address field is not big enough.
Fixed inequality comparison in vppcom.c by using the predefined macro
MAP_FAILED.
Change-Id: I9517c29ae811d08058621bd548a352b4d4f05139
Signed-off-by: Steven <sluong@cisco.com>
This reverts commit 45a588fa3efaaf52360986360ab1f6827bae3164.
Change-Id: I7e541545791f7743ee827bdec8b6fc46cbb0938f
Signed-off-by: Damjan Marion <damarion@cisco.com>
Mapping shared virtual memory at 0x30000000, which appears to be derived
from x86-32, turns out to be too close to the heap on arm64 systems. The
symptoms of memory corruption were random and included crashes in the
Python runtime and what appeared to be corruption of malloc's internal
mutex. Thanks to Gabriel Ganne for pointing out that disabling ASLR seemed
to mitigate the situation.
This patch maps SVM regions at an offset from the arm64 kernel constant
TASK_UNMAPPED_BASE and also assumes a 48-bit VA (for Ubuntu).
Change-Id: I642e5fe83344ab9b5c66c93e0cf1575c17251f3b
Signed-off-by: Brian Brooks <brian.brooks@arm.com>
- makes the VAPI generated file more consumable.
- VOM build times improve.
Change-Id: I838488930bd23a0d3818adfdffdbca3eead382df
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
Hongjun Ni