Age | Commit message (Collapse) | Author | Files | Lines |
|
Memory allocation changed in DPDK 18.05. The mlx4 and mlx5
PMDs did not support using externally allocated memory.
The patch for mlx5 was generated by Mellanox. That patch was
modified to apply to the mlx4 PMD and tested on Microsoft
Azure.
Change-Id: I92116b1d71a3896d5bf7b1f10c40c898d72540d6
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Change-Id: Ia1b188492138a0ca0e95daf6eb8e761e8db081ef
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I04cdd296bc3de0460308351d0bbb00d7cbbf023e
Signed-off-by: Marco Varlese <marco.varlese@suse.de>
|
|
GCC 7 plus Ubuntu-16.04 default ccache 3.2.4 reports warnings on switch case fall-through code,
which is commonly seen in DPDK code, which will cause image building failure finally.
It requires newer ccache version to support -Wimplicit-fallthrough.
To suppress the warning, if GCC is version 7 or higher, and it's old ccache version,
will disable the fall-through check.
dpdk-18.05/drivers/net/ark/ark_ethdev_rx.h:39:0,
from dpdk-18.05/drivers/net/ark/ark_ethdev_rx.c:36:
dpdk-18.05/arm64-armv8a-linuxapp-gcc/include/rte_mbuf.h: In function ‘rte_pktmbuf_alloc_bulk’:
dpdk-18.05/arm64-armv8a-linuxapp-gcc/include/rte_mbuf.h:1292:7: error: this statement may fall through [-Werror=implicit-fallthrough=]
idx++;
~~~^~
Change-Id: I4d12492471fadef9d349ba9e853a6799512f76f5
Signed-off-by: Lijian Zhang <Lijian.Zhang@arm.com>
|
|
Change-Id: If1b93341c222160b9a08f127620c024620e55c37
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Both vpp-devel and vpp-dpdk-devel have /usr/include/dpdk
in their files list. CentOS 7 won't allow them to both be
installed at the same time as a result.
Change vpp-dpdk-devel to /usr/include/dpdk/* so both can
be installed. It is useful to have both installed on a
development system.
Change-Id: I94b6cca299dfbd991638e43c41006db0d8f7ee8b
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Add patch for DPDK 18.05 that was previously applied
to DPDK 18.02.1. Issue with ixgbe on x550 SFP+ still
exists.
Bug report submitted to DPDK:
https://bugs.dpdk.org/show_bug.cgi?id=69
Change-Id: I9b005709ddf2a72192b1288ba8b4bac85bf12685
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Change-Id: I739d3e6c25efe8d32b2f4a60557c644edfe958e0
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I00b0e4d7f7b597760a898c895b1a80bfac3a47fb
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
An x550 with SFP+ interfaces attached to some switches can
have problems bringing the port up.
After configuring the link, there is a wait for 500 ms for
the link to come up. Some switches don't bring their ports up
that quickly. So the link is never observed to come up and is
reconfigured again the next time dpdk_update_link_state() is
called. Subsequent attempts time out also indefinitely.
Instead of waiting through 5 iterations of a 100 ms delay, wait
through 10 iterations. The i40e PMD does this when updating
link status.
This issue & patch will be reported to Intel so this or
some better solution can be applied upstream in the future.
Change-Id: I16d706a2790e51d695edc43c0ca17f1eff1dcf5e
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Change-Id: I205932bc727c990011bbbe1dc6c0cf5349d19806
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Added configure argument "--with-log2-cache-line-bytes=5|6|7|auto"
AKA 32, 64, or 128 bytes, or use the inferred value from the build host.
produces build-xxx/vpp/vppinfra/config.h, which .../src/vppinfra/cache.h
Kernels which implement the following pseudo-file (aka x86_64) are
easy: /sys/devices/system/cpu/cpu0/cache/index0/coherency_line_size
Otherwise, extract the cpuid from /proc/cpuinfo and map it to the
cache line size.
Change-Id: I7ff861e042faf82c3901fa1db98864fbdea95b74
Signed-off-by: Dave Barach <dave@barachs.net>
Signed-off-by: Nitin Saxena <nitin.saxena@cavium.com>
|
|
Fixes clang error: argument unused during compilation: '-pie'
by replacing it with -fPIE
Change-Id: Ic89a5e325e019d4d794d35556a07ebcf0b718dd3
Signed-off-by: Sirshak Das <sirshak.das@arm.com>
Reviewed-by: Brian Brooks <brian.brooks@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
|
|
This compile issue was first reported by Sirshak Das in following thread:
https://lists.fd.io/g/vpp-dev/message/8384
The issue was caused by auto-config shell script auto-config-h.sh regard
quotation mark "" as $CROSS prefix for $CC when CROSS is empty.
Change-Id: Ied535c6d18c4dffacbddabc3ad2087dffe19438d
Signed-off-by: Bin Huang <huangbin.mails@gmail.com>
|
|
Added build related keywords for TAP, FAILSAFE PMD
and also added some missing keywords for mlx4 PMD
This is part of initial effort to enable vpp running over
dpdk on failsafe PMD in Microsoft Azure (1/4).
Change-Id: I2aebf209fbc6db030185f41971b51a593a003a3a
Signed-off-by: Rui Cai <rucai@microsoft.com>
|
|
Change-Id: I9585603e9fa1904658d84d226bd50370f1dbc889
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
For further details, please see
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2018-1059
Change-Id: Icd207f129e5fdcc3d9d8ad56ba5a368926f2804d
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
DPDK rte_cryptodev_socket_id() is returning zero for QAT devices.
Apply DPDK patch where correct NUMA node can be obtained with pci_dev->device.numa_node.
Change-Id: I1c7a77bb13e2db8615189e97b67d68d043127787
Signed-off-by: Lee Roberts <lee.roberts@hpe.com>
|
|
dlopen linkage allow more transparent use for Mellanox nics.
Mellanox shared library librte_pmd_mlx5/4_glue.so* placed in LD_LIBRARY_PATH
At run time Mellanox code will be loaded only when Mellanox nics explicty used.
i.e if VPP is used with other vendor Mellanox code is not loaded.
Change-Id: Ib05bdbfc4cbb6e447c67186c98361f9c5b447140
Signed-off-by: Amir Zeidner <amirzei@mellanox.com>
|
|
Change-Id: Ib3361eded05babfc17ead28af7d252e7503ce141
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I1e0cea8e7ea6d8a777ca38abb61f4c093f29c722
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
Looks like bug in ipsec-mb library when DEBUG=yes is passed
so simply we stop doing that.
Change-Id: Ifedd6d8a2aecf5af902ab4fa80ef197aebd5f829
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I3764f57a4b8df96d6bd20753b86fc0119d833bd9
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
17.11 is still default.
Change-Id: I524d232579db8a59c717c5d760398b6b7f811d03
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
On rx, the ENA PMD sets tx offload flags based on the
received packet's l4 protocol. This means you need
to turn off those offloads for every packet if you
encapsulate packets arriving on an interface using
that PMD. Disable this behavior.
Change-Id: Icae9f32e3d292d767da440ae5c1280902bdaa083
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
Change-Id: Icb931de82cb5969fa4976611629e2f882c720a99
Signed-off-by: Ed Warnicke <eaw@cisco.com>
|
|
- fix ESP transport mode
- safely free crypto sessions
- use rte_mempool_virt2phy/rte_mempool_virt2iova
- align DPDK QAT capabilities for IPsec usage (DPDK 17.08)
- reserve 16B for aad (reference cryptodev doc)
Change-Id: I3822a7456fb5a255c767f5a44a429f91a140fe64
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
Change-Id: I84fafa369c6f16295e1c24d9712de2229bf78a91
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
DPDK tarball was changed by mistake and now it is reverted back.
This reverts commit 3d786efcb087533320e89f80077127fc507cfd99.
Change-Id: I1a07b96fbc3f4fe13bb4a5c401036cae4ac5d346
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
The dpdk package posted on static.dpdk.org for 17.08
was updated 11/27. This updates the checksum thats
statically included in makefile. Looks like they
also changed the dir structure to add -stable.
fast.dpdk.org has issues with its mirrors being in
sync...changing to static.dpdk.org for now
Change-Id: Id81e328b07873700ae3f76e1ca819f94f26f38c8
Signed-off-by: Ed Kern <ejk@cisco.com>
|
|
Also remove DPDK 17.05 support.
Change-Id: I4f96cb3f002cd90b12d800d6904f2364d7c4e270
Signed-off-by: Damjan Marion <damarion@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>
|
|
As not all distros have the minimum required nasm version (2.12.01)
available, build nasm from sources when building Intel IPsec MB library.
Change-Id: Iaa9da87f612c0f84da5704162c3bf430b3351076
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
At present, creating bonding devices using --vdev is broken for PMD like
mlx5 as it is neither UIO nor VFIO based and hence PMD driver is unknown
to find_port_id_by_pci_addr().
This DPDK patch fixes parsing PCI ID from bonding device params by verifying
it in RTE PCI bus, rather than checking dev->kdrv.
Change-Id: If575f63ef31733102566610d769ddd212d74736a
Signed-off-by: Steve Shin <jonshin@cisco.com>
|
|
This patch reworks the DPDK ipsec implementation including the cryptodev
management as well as replacing new cli commands for better usability.
For the data path:
- The dpdk-esp-encrypt-post node is not necessary anymore.
- IPv4 packets in the decrypt path are sent to ip4-input-no-checksum instead
of ip4-input.
The DPDK cryptodev cli commands are replaced by the following new commands:
- show dpdk crypto devices
- show dpdk crypto placement [verbose]
- set dpdk crypto placement (<device> <thread> | auto)
- clear dpdk crypto placement <device> [<thread>]
- show dpdk crypto pools
Change-Id: I47324517ede82d3e6e0e9f9c71c1a3433714b27b
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
JIRA VPP-498
This patch also allows RPMs to be built without multi-
buffer crypto for some RPM based downstream distros that
don't have sufficiently new nasm or don't have an USA
export license for multi-buffer crypto.
The default is to build WITH multi-buffer crypto
for x86-64. This patch allows optional building without
multi-buffer crypto.
To build without multi-buffer crypto, set the AESNI
environment variable to n.
To build rpm packages without multi-buffer crypto,
build the rpms with the option turned off.
make build AESNI=n
or..
make pkg-rpm --without aesni
---How to test this patch on a Centos build.---
Build as above and verify that nasm isn't executed during
the build process.
vpp may be installed and the dpdk plugin may be inspected to
verify that the multi-buffer code isn't present.
Change-Id: I8c5cfd4cdd9eb2b96772a687eaa54560806e001b
Signed-off-by: Thomas F Herbert <therbert@redhat.com>
|
|
Use ARMv8 Main ID Register (exposed thru /proc/cpuinfo) to identify
the CPU implementor and part number. For further details, see the
ARMv8 ARM D7.2.66.
Change-Id: I2b0d0b165cda4ab9fc57c645af90e9e354b73f44
Signed-off-by: Brian Brooks <brian.brooks@arm.com>
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: Ola Liljedahl <ola.liljedahl@arm.com>
Reviewed-by: Song Zhu <song.zhu@arm.com>
|
|
With this change, the status of `make build':
Huawei D02, Linux 4.4.0, gcc 5.4.1 - success
AMD Seattle, Linux 4.4.6, gcc 5.3.1 - compiler ICEs
Cavium ThunderX, Linux 4.4.49, gcc 5.4.0 - success
Before:
Huawei D02, Linux 4.4.0, gcc 5.4.1 - fail
AMD Seattle, Linux 4.4.6, gcc 5.3.1 - fail
Cavium ThunderX, Linux 4.4.49, gcc 5.4.0 - success
Change-Id: I49db34a33f9ca0725c7511d4f796706892b5b2da
Signed-off-by: Brian Brooks <brian.brooks@arm.com>
|
|
Change-Id: I674fb1212e48693939045523df085326a4dd1809
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
DPDK 17.08 breaks ethdev and cryptodev APIs.
Address those changes while keeping backwards compatibility for
DPDK 17.02 and 17.05.
Change-Id: Idd6ac264d0d047fe586c41d4c4ca74e8fc778a54
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
This fixes build on non-x86 platforms like arm64.
Change-Id: I7ff5df92f89e34c27889d82f35924dc28cde8c39
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Beside the fact that we don't need it, it fails to build on ARM64.
Change-Id: Iefae8bf234b588d8005df5e053b9152b6611929c
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
libraries.
This patch addresses the misbehaviour.
Change-Id: I41f1ece3ca21c5a8f2c95533ed3d77a535233ea6
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
Depending on the OS, the default libdir might change.
RHEL/Ubuntu:
libdir={exec_prefix}/lib
OpenSUSE:
libdir={exec_prefix}/lib64
Change-Id: I5f1672e5815ad821e6ac5fff95de5232ab735b67
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
Change-Id: Ic73b857c4e3d5a3f695e93924de5a5bed0af5019
Signed-off-by: Marco Varlese <marco.varlese@suse.com>
|
|
Change-Id: If559747ad59c82c81d15734f27e15548eca0962b
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
Current optional DPDK PMDs are:
- AESNI MB PMD (SW crypto)
- AESNI GCM PMD (SW crypto)
- MLX4 PMD
- MLX5 PMD
This change will always build DPDK SW crypto PMDs and required SW crypto
libraries, while MLX PMDs are still optional and the user has to build
required libraries.
Now the configure script detects if any of the optional DPDK PMDs were
built and link against their required libraries/dependencies.
Change-Id: I1560bebd71035d6486483f22da90042ec2ce40a1
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|
|
With some Linux container platforms /proc/cpuinfo reads as an empty
file. (Aside: stat on /proc/cpuinfo always indicates a length of
zero bytes, regardless of its content).
This has the effect that the make '-j' parameter being passed the
unhelpful value of '0' both in build-root/Makefile and dpdk/Makefile.
Make complains with the error:
make: the '-j' option requires a positive integer argument
This patch checks for '0' and replaces it with '2' as a reasonable
number of jobs to run in parallel when the CPU count isn't known
(and assumed to be one). It also makes the value determination
consistent between VPP and DPDK (2*ncpu).
Change-Id: I78b89420114a825fab4d339e4f9291d486b7b9c8
Signed-off-by: Chris Luke <chrisy@flirble.org>
|
|
This reverts commit 0e2e10b77d63196bfb93ae5be1251bbc1a1b561a.
Change-Id: I3c1737f391b6ed127f92416f06449216e79859bb
Signed-off-by: Peter Mikus <pmikus@cisco.com>
|
|
Change-Id: Ica95b5d3d44563c93c89b2a3233171c3aa1f048d
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
|