Age | Commit message (Collapse) | Author | Files | Lines |
|
Type: feature
You can enable tso in starup.conf like this:
dev 0000:86:00.0{
tso on
}
TSO is disabled by default.
Change-Id: Ifdbaf5322f768c384aa54e532d7bf45e810ca01c
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
|
|
Type: feature
Change-Id: I21f8f4563f5545a684b2666f7410847e0f7bc403
Signed-off-by: Jim Thompson <jim@netgate.com>
|
|
Type: fix
Fixes: ce3e971
Change-Id: I30bbeced2f5ae7613e65546f2b9b41e2fb514208
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Explains a variety of hard-to-diagnose problems with certain Atom and
Denverton NIC types.
I finally tripped over a highly-repeatable failure: home gateway
use-case bitten by refusal to negotiate a DHCP lease for the trunk
port.
The dhcp client won't send pkts unless VNET_HW_INTERFACE_FLAG_LINK_UP
is set on the tx hw interface:
/* Interface(s) down? */
if ((hw->flags & VNET_HW_INTERFACE_FLAG_LINK_UP) == 0)
return;
Change-Id: I17ef2ba7b39078555fa27d2d874a60c67e1530ee
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
This patch enables bonding numa awareness on multi-socket
server working in active-backeup mode.
The VPP adds capability for automatically preferring slave
with local numa node in order to reduces the load on the
QPI-bus and improve system overall performance in multi-socket
use cases. Users doesn't need to add any extra operation as
usual.
Change-Id: Iec267375fc399a9a0c0a7dca649fadb994d36671
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
We have native implementation and we should not maintain both....
Change-Id: Ic09ebffda52cdc733b3cfeff06690e0d3cc08084
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
copy vlan strip config from default device
Change-Id: I4ad1c159bad964fd1900b5ae4960b7014dd9f9b1
Signed-off-by: Chenmin Sun <chenmin.sun@intel.com>
|
|
The vlib init function subsystem now supports a mix of procedural and
formally-specified ordering constraints. We should eliminate procedural
knowledge wherever possible.
The following schemes are *roughly* equivalent:
static clib_error_t *init_runs_first (vlib_main_t *vm)
{
clib_error_t *error;
... do some stuff...
if ((error = vlib_call_init_function (init_runs_next)))
return error;
...
}
VLIB_INIT_FUNCTION (init_runs_first);
and
static clib_error_t *init_runs_first (vlib_main_t *vm)
{
... do some stuff...
}
VLIB_INIT_FUNCTION (init_runs_first) =
{
.runs_before = VLIB_INITS("init_runs_next"),
};
The first form will [most likely] call "init_runs_next" on the
spot. The second form means that "init_runs_first" runs before
"init_runs_next," possibly much earlier in the sequence.
Please DO NOT construct sets of init functions where A before B
actually means A *right before* B. It's not necessary - simply combine
A and B - and it leads to hugely annoying debugging exercises when
trying to switch from ad-hoc procedural ordering constraints to formal
ordering constraints.
Change-Id: I5e4353503bf43b4acb11a45fb33c79a5ade8426c
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: Ia092a93a7ac0cbf9338f9d4a5db8b94b23549a13
Signed-off-by: Florin Coras <fcoras@cisco.com>
|
|
Change-Id: I65e7188c6893acca67455ff37f2dfbd0bedd5c09
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
As DVN has fewer DTLB entries supported for 2M page, default numbers of
RX/TX descriptors are changed to 512 if nums of RX/TX descriptors are not
specified by VPP users.
Change-Id: I076493b802b15d12750a5b49d1554da4d19ad460
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
|
|
Change-Id: I48cd8052f9509efdf13f64ab279edb66a2d4a0a9
Signed-off-by: Simon Zhang <yuwei1.zhang@intel.com>
|
|
With the following local patch, VIC adapters remove default vlan tags
from ingress packets. So, it is no longer necessary to enable VLAN
stripping by default. This change also allows VLAN sub interfaces to
work with VIC adapters.
patches/dpdk_19.02/0001-net-enic-untag-default-vlan-by-default.patch
Change-Id: I2e7d62c62120c351c27d827d90de4a8335efa044
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
|
|
Change-Id: Id5b30d7a394551844a79b3d222d2d26194d033df
Signed-off-by: ChenminSun <chenmin.sun@intel.com>
|
|
Change-Id: I572cbba817275d85c200a4b09a63f4650075f638
Signed-off-by: Jay Lubomirski <jlubomir@cisco.com>
|
|
The log-level dpdk config value should be transparently
forwarded to DPDK via EAL argument. Since DPDK now supports
naming log-levels, VPP no longer needs to parse and call
rte_set_loglevel().
This was the other part of the DPDK log-level change.
It must have got missed during my initial checkin.
Without it passing dynamic log-level values like are silently
ignored.
Fixes: 6ca6ac6c887e ("dpdk: support passing log-level")
Change-Id: I732cec5f638c9924e3ffb04c4753f957e3633d64
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
|
|
Recent VIC models can support 25, 50, and 100Gbps links. Use the
helper (port_type_from_link_speed) to set the port type as it supports
all possible link speeds.
Change-Id: I748d8ac716a6393d116a9db8a599151c70a9000a
Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
|
|
Change-Id: Ieab56e0a20696b8cc97f783f08f10a94a83644eb
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
HQoS requires fixes to work with dpdk 19.02 so code is disabled and
pending deprecation unless active maintainer is found.
Change-Id: I3569c4287b6dfdd2c29e02375eb53bf01fa6ae84
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This patch introduces following changes:
- deprecated free lists which are not used and not compatible
with external buffer managers (i.e. DPDK)
- introduces native support for per-numa buffer pools
- significantly improves performance of buffer alloc and free
Change-Id: I4a8e723ae47056717afd6cac0efe87cb731b5be7
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I998658ad7860b23425444e218ce2e1ec655b885a
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ia083050389853c25b069f0f8286d50d3f4aef527
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I3896bf9d71dc300520c53dbe3c2fd8fcd1470881
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This patch adds support for VMBus to the VPP infrastructure.
Since the only device that matters is the netvsc Poll Mode Driver
in DPDK, the infrastructure is much simpler than PCI.
Change-Id: Ie96c897ad9c426716c2398e4528688ce2217419b
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
|
|
Change-Id: I89695c1ad47131ed830f35c677937ce12025a40d
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Change-Id: Ib6458e56f546bb5b11c23aa5e1afe0f4b5011c08
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Example:
dpdk {
dev 0000:01:00.0 { name eth0 }
dev 0000:02:00.0 { name eth1 }
}
Change-Id: I11e60e969a7e3548c99ac0c0c3531767819cb157
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I4ec7750ef58363bd8966a16a2baeec6db18b7e9e
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I6a6ab6d8d9aeca9df3ba8a3ac519be25aaa796e8
Signed-off-by: Federico Claramonte <fede.claramonte@caviumnetworks.com>
|
|
Change-Id: I3daf8d473aa37b4597d130d19913b782cf7b8511
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Idd4471a3adf7023e48e85717f00c786b1dde0cca
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
If no PCI address is specified in dpdk config, the default to automatically
put all PCIs in the whitelist.
For vmxnet3 PCIs, we want to change its default to exclude the vmxnet3 PCIs.
That is to put them in the blacklist instead of whitelist.
Change-Id: I2b7061d6437910eb0e1b16df19a770cab968c602
Signed-off-by: Steven <sluong@cisco.com>
|
|
Change-Id: I88577615bfd3cddca834a7b881979ab32bfd5574
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I5ff713ad0b254c74c5622e3b9425cca365b5ee97
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
In DPDK 18.11 the device flags for keeping/stripping CRC flags has
changed. The old strip flag is gone, and replaced by a CRC keep
flag.
Change-Id: Iaa162854862a2a0855b418ee8029383fc116d3a2
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
|
|
Change-Id: Ied34720ca5a6e6e717eea4e86003e854031b6eab
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
- logging
- pass vlib_main_t to all APIs
- open vfio container only when needed
Change-Id: I897e53e0af3f91c3a99f0c827401d1c0ec2e478a
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ibec32c6df32f4cd9889d378e244f170c93ad295b
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Teach DPDK plugin about the netvsc Poll Mode Driver.
The speed of the Netvsc device matches the speed of the external
port on the underlying vswitch. Therefore 1G, 10G, 25G, 56G and
even 100G are possible.
Change-Id: I14ab6907b7d8d350b63a083409d45fb9c348a364
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
|
|
The rte_device is use as a base type of all DPDK devices.
It is not valid to use container_of to find PCI information
unless the bus of the rte_device is pci. Otherwise, the
pointer is looking at some other data, which may or may not
be zero.
This change introduces a helper function to get rte_pci_device
pointer. If device is on PCI bus it returns pointer to
rte_pci_device info, otherwise it returns NULL.
Change-Id: Ia7446006bb93a7a54844969f3b3dd3b918890dfd
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
|
|
Change-Id: I085615fde1f966490f30ed5d32017b8b088cfd59
Signed-off-by: Paul Vinciguerra <pvinci@vinciconsulting.com>
|
|
Warning messsage is displayed in system log:
vpp# show log
1970/ 1/ 1 01:00:01:198 warn dpdk unsupported rx offloads requested on port 0: scatter
Change-Id: I40021066daf2d37ca5233e3adce55e412f0d3932
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
The scatter/gather rxmode flag was set for ENA when building
against DPDK >= 18.08. ENA does not support this, so disable
it. It looks like enabling it was a copy/paste error.
Also, after offloads are adjusted based on whether "no-multi-seg"
is set, those configurations are overwritten by copying
port_conf_template over the port config. That should only happen
for versions of DPDK older than 18.08 because 18.08 and newer
make changes directly on the port config instead of making changes
to the template. Make the clib_memcpy() conditional on the DPDK
version being less than 18.08. After doing so, compiler
errors complain about port_conf_template being declared but not
used, so make it's declaration conditional.
Change-Id: If81980d71c379a565b51dd700b953f8c811a8703
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
supporting TCP/UDP
The DPDK plugin sets all of the offload flags, which may cause an initialization failure
on the NICs that do not support SCTP offload. The VPP code does not deal with the SCTP
offload at the moment at all, so after discussing with Damjan, we agreed
the best approach to fix the issue is to not request the SCTP offload.
The output of "show hardware" for the NIC in question before this patch:
Name Idx Link Hardware
GigabitEthernet1/0/0 1 down GigabitEthernet1/0/0
Ethernet address 00:e0:67:09:90:4b
Intel 82540EM (e1000)
carrier down
flags: pmd pmd-init-fail maybe-multiseg tx-offload intel-phdr-cksum
rx queues 1, rx desc 1024, tx queues 1, tx desc 1024
cpu socket 0
Errors:
rte_eth_dev_configure[port:0, errno:-22]: Unknown error -22
And the excerpt from "show log":
1970/ 1/ 1 00:00:00:739 notice dpdk Ethdev port_id=0 requested Tx offloads 0x1c doesn't match Tx offloads capabilities 0xf in rte_eth_dev_configure()
Change-Id: I159d65c02fc3f044441972205f1f0ac08e52050c
Signed-off-by: Andrew Yourtchenko <ayourtch@gmail.com>
|
|
Change-Id: I1f54b994425c58776e1445c8d9fe142e7a644d3d
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This adds a `num-mem-channels` option for DPDK enabling support for
choosing the number of memory channels used.
Change-Id: I1663dd866ac60592b6dd02261af66d87c64acdb1
Signed-off-by: Jessica Tallon <tsyesika@igalia.com>
|
|
DPDK 18.08 verifies if all set bits are supported and fails if not....
Change-Id: Ia87242fcda11147dff3bebe2e2bef32f0a8891fb
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: If1b93341c222160b9a08f127620c024620e55c37
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Fix a typo from previous patch. Change 0x104 to 0x1004.
Change-Id: I82230a8a0ec01567eb1d4bc12ac02062c2a98347
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|
|
If a PMD writes too many log messages using rte_vlog(), the
pipe for logging can fill and then rte_vlog() will block
on fflush() while it waits for something to read from the other
side of the pipe. That will never happen since the process node
that would read the other side of the pipe runs in the
same thread.
Set the write fd to non-blocking before the call to
rte_openlog_stream(). If the pipe is full, calls to write() or
fflush() will fail but execution will continue.
Change-Id: I0e5d710629633acda5617ff29897d6582c255d57
Signed-off-by: Matthew Smith <mgsmith@netgate.com>
|