d/numam-dpdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
3d38e3dcf1
numam-dpdk/doc/guides/nics/enic.rst
Hyong Youb Kim 8a6ff33d6d net/enic: add AVX2 based vectorized Rx handler 
Add the vectorized version of the no-scatter Rx handler. It aims to
process 8 descriptors per loop using AVX2 SIMD instructions. This
handler is in its own file enic_rxtx_vec_avx2.c, and makefile and
meson.build are modified to compile it when the compiler supports
AVX2. Under ideal conditions, the vectorized handler reduces
cycles/packet by more than 30%, when compared against the no-scatter
Rx handler. Most implementation ideas come from i40e's AVX2 based
handler, so credit goes to its authors.

At this point, the new handler is meant for field trials, and is not
selected by default. So add a new devarg enable-avx2-rx to allow the
user to request the use of the new handler. When enable-avx2-rx=1, the
driver will consider using the new handler.

Also update the guide doc and introduce the vectorized handler.

Signed-off-by: Hyong Youb Kim <hyonkim@cisco.com>
Reviewed-by: John Daley <johndale@cisco.com>
2018-10-11 18:53:49 +02:00

593 lines
23 KiB
ReStructuredText
Raw Blame History

.. SPDX-License-Identifier: BSD-3-Clause
Copyright (c) 2017, Cisco Systems, Inc.
All rights reserved.
ENIC Poll Mode Driver
=====================
ENIC PMD is the DPDK poll-mode driver for the Cisco System Inc. VIC Ethernet
NICs. These adapters are also referred to as vNICs below. If you are running
or would like to run DPDK software applications on Cisco UCS servers using
Cisco VIC adapters the following documentation is relevant.
How to obtain ENIC PMD integrated DPDK
--------------------------------------
ENIC PMD support is integrated into the DPDK suite. dpdk-<version>.tar.gz
should be downloaded from http://dpdk.org
Configuration information
-------------------------
- **DPDK Configuration Parameters**
The following configuration options are available for the ENIC PMD:
- **CONFIG_RTE_LIBRTE_ENIC_PMD** (default y): Enables or disables inclusion
of the ENIC PMD driver in the DPDK compilation.
- **vNIC Configuration Parameters**
- **Number of Queues**
The maximum number of receive queues (RQs), work queues (WQs) and
completion queues (CQs) are configurable on a per vNIC basis
through the Cisco UCS Manager (CIMC or UCSM).
These values should be configured as follows:
- The number of WQs should be greater or equal to the value of the
expected nb_tx_q parameter in the call to
rte_eth_dev_configure()
- The number of RQs configured in the vNIC should be greater or
equal to *twice* the value of the expected nb_rx_q parameter in
the call to rte_eth_dev_configure(). With the addition of Rx
scatter, a pair of RQs on the vnic is needed for each receive
queue used by DPDK, even if Rx scatter is not being used.
Having a vNIC with only 1 RQ is not a valid configuration, and
will fail with an error message.
- The number of CQs should set so that there is one CQ for each
WQ, and one CQ for each pair of RQs.
For example: If the application requires 3 Rx queues, and 3 Tx
queues, the vNIC should be configured to have at least 3 WQs, 6
RQs (3 pairs), and 6 CQs (3 for use by WQs + 3 for use by the 3
pairs of RQs).
- **Size of Queues**
Likewise, the number of receive and transmit descriptors are configurable on
a per-vNIC basis via the UCS Manager and should be greater than or equal to
the nb_rx_desc and nb_tx_desc parameters expected to be used in the calls
to rte_eth_rx_queue_setup() and rte_eth_tx_queue_setup() respectively.
An application requesting more than the set size will be limited to that
size.
Unless there is a lack of resources due to creating many vNICs, it
is recommended that the WQ and RQ sizes be set to the maximum. This
gives the application the greatest amount of flexibility in its
queue configuration.
- *Note*: Since the introduction of Rx scatter, for performance
reasons, this PMD uses two RQs on the vNIC per receive queue in
DPDK. One RQ holds descriptors for the start of a packet, and the
second RQ holds the descriptors for the rest of the fragments of
a packet. This means that the nb_rx_desc parameter to
rte_eth_rx_queue_setup() can be a greater than 4096. The exact
amount will depend on the size of the mbufs being used for
receives, and the MTU size.
For example: If the mbuf size is 2048, and the MTU is 9000, then
receiving a full size packet will take 5 descriptors, 1 from the
start-of-packet queue, and 4 from the second queue. Assuming
that the RQ size was set to the maximum of 4096, then the
application can specify up to 1024 + 4096 as the nb_rx_desc
parameter to rte_eth_rx_queue_setup().
- **Interrupts**
At least one interrupt per vNIC interface should be configured in the UCS
manager regardless of the number receive/transmit queues. The ENIC PMD
uses this interrupt to get information about link status and errors
in the fast path.
In addition to the interrupt for link status and errors, when using Rx queue
interrupts, increase the number of configured interrupts so that there is at
least one interrupt for each Rx queue. For example, if the app uses 3 Rx
queues and wants to use per-queue interrupts, configure 4 (3 + 1) interrupts.
- **Receive Side Scaling**
In order to fully utilize RSS in DPDK, enable all RSS related settings in
CIMC or UCSM. These include the following items listed under
Receive Side Scaling:
TCP, IPv4, TCP-IPv4, IPv6, TCP-IPv6, IPv6 Extension, TCP-IPv6 Extension.
.. _enic-flow-director:
Flow director support
---------------------
Advanced filtering support was added to 1300 series VIC firmware starting
with version 2.0.13 for C-series UCS servers and version 3.1.2 for UCSM
managed blade servers. In order to enable advanced filtering the 'Advanced
filter' radio button should be enabled via CIMC or UCSM followed by a reboot
of the server.
With advanced filters, perfect matching of all fields of IPv4, IPv6 headers
as well as TCP, UDP and SCTP L4 headers is available through flow director.
Masking of these fields for partial match is also supported.
Without advanced filter support, the flow director is limited to IPv4
perfect filtering of the 5-tuple with no masking of fields supported.
SR-IOV mode utilization
-----------------------
UCS blade servers configured with dynamic vNIC connection policies in UCSM
are capable of supporting SR-IOV. SR-IOV virtual functions (VFs) are
specialized vNICs, distinct from regular Ethernet vNICs. These VFs can be
directly assigned to virtual machines (VMs) as 'passthrough' devices.
In UCS, SR-IOV VFs require the use of the Cisco Virtual Machine Fabric Extender
(VM-FEX), which gives the VM a dedicated
interface on the Fabric Interconnect (FI). Layer 2 switching is done at
the FI. This may eliminate the requirement for software switching on the
host to route intra-host VM traffic.
Please refer to `Creating a Dynamic vNIC Connection Policy
<http://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/sw/vm_fex/vmware/gui/config_guide/b_GUI_VMware_VM-FEX_UCSM_Configuration_Guide/b_GUI_VMware_VM-FEX_UCSM_Configuration_Guide_chapter_010.html#task_433E01651F69464783A68E66DA8A47A5>`_
for information on configuring SR-IOV adapter policies and port profiles
using UCSM.
Once the policies are in place and the host OS is rebooted, VFs should be
visible on the host, E.g.:
.. code-block:: console
# lspci | grep Cisco | grep Ethernet
0d:00.0 Ethernet controller: Cisco Systems Inc VIC Ethernet NIC (rev a2)
0d:00.1 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
0d:00.2 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
0d:00.3 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
0d:00.4 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
0d:00.5 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
0d:00.6 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
0d:00.7 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
Enable Intel IOMMU on the host and install KVM and libvirt, and reboot again as
required. Then, using libvirt, create a VM instance with an assigned device.
Below is an example ``interface`` block (part of the domain configuration XML)
that adds the host VF 0d:00:01 to the VM. ``profileid='pp-vlan-25'`` indicates
the port profile that has been configured in UCSM.
.. code-block:: console
<interface type='hostdev' managed='yes'>
<mac address='52:54:00:ac:ff:b6'/>
<driver name='vfio'/>
<source>
<address type='pci' domain='0x0000' bus='0x0d' slot='0x00' function='0x1'/>
</source>
<virtualport type='802.1Qbh'>
<parameters profileid='pp-vlan-25'/>
</virtualport>
</interface>
Alternatively, the configuration can be done in a separate file using the
``network`` keyword. These methods are described in the libvirt documentation for
`Network XML format <https://libvirt.org/formatnetwork.html>`_.
When the VM instance is started, libvirt will bind the host VF to
vfio, complete provisioning on the FI and bring up the link.
.. note::
It is not possible to use a VF directly from the host because it is not
fully provisioned until libvirt brings up the VM that it is assigned
to.
In the VM instance, the VF will now be visible. E.g., here the VF 00:04.0 is
seen on the VM instance and should be available for binding to a DPDK.
.. code-block:: console
# lspci | grep Ether
00:04.0 Ethernet controller: Cisco Systems Inc VIC SR-IOV VF (rev a2)
Follow the normal DPDK install procedure, binding the VF to either ``igb_uio``
or ``vfio`` in non-IOMMU mode.
In the VM, the kernel enic driver may be automatically bound to the VF during
boot. Unbinding it currently hangs due to a known issue with the driver. To
work around the issue, blacklist the enic module as follows.
Please see :ref:`Limitations <enic_limitations>` for limitations in
the use of SR-IOV.
.. code-block:: console
# cat /etc/modprobe.d/enic.conf
blacklist enic
# dracut --force
.. note::
Passthrough does not require SR-IOV. If VM-FEX is not desired, the user
may create as many regular vNICs as necessary and assign them to VMs as
passthrough devices. Since these vNICs are not SR-IOV VFs, using them as
passthrough devices do not require libvirt, port profiles, and VM-FEX.
.. _enic-genic-flow-api:
Generic Flow API support
------------------------
Generic Flow API is supported. The baseline support is:
- **1200 series VICs**
5-tuple exact flow support for 1200 series adapters. This allows:
- Attributes: ingress
- Items: ipv4, ipv6, udp, tcp (must exactly match src/dst IP
addresses and ports and all must be specified)
- Actions: queue and void
- Selectors: 'is'
- **1300 and later series VICS with advanced filters disabled**
With advanced filters disabled, an IPv4 or IPv6 item must be specified
in the pattern.
- Attributes: ingress
- Items: eth, ipv4, ipv6, udp, tcp, vxlan, inner eth, ipv4, ipv6, udp, tcp
- Actions: queue and void
- Selectors: 'is', 'spec' and 'mask'. 'last' is not supported
- In total, up to 64 bytes of mask is allowed across all headers
- **1300 and later series VICS with advanced filters enabled**
- Attributes: ingress
- Items: eth, ipv4, ipv6, udp, tcp, vxlan, inner eth, ipv4, ipv6, udp, tcp
- Actions: queue, mark, drop, flag and void
- Selectors: 'is', 'spec' and 'mask'. 'last' is not supported
- In total, up to 64 bytes of mask is allowed across all headers
- **1400 and later series VICS with advanced filters enabled**
All the above plus:
- Action: count
More features may be added in future firmware and new versions of the VIC.
Please refer to the release notes.
.. _overlay_offload:
Overlay Offload
---------------
Recent hardware models support overlay offload. When enabled, the NIC performs
the following operations for VXLAN, NVGRE, and GENEVE packets. In all cases,
inner and outer packets can be IPv4 or IPv6.
- TSO for VXLAN and GENEVE packets.
Hardware supports NVGRE TSO, but DPDK currently has no NVGRE offload flags.
- Tx checksum offloads.
The NIC fills in IPv4/UDP/TCP checksums for both inner and outer packets.
- Rx checksum offloads.
The NIC validates IPv4/UDP/TCP checksums of both inner and outer packets.
Good checksum flags (e.g. ``PKT_RX_L4_CKSUM_GOOD``) indicate that the inner
packet has the correct checksum, and if applicable, the outer packet also
has the correct checksum. Bad checksum flags (e.g. ``PKT_RX_L4_CKSUM_BAD``)
indicate that the inner and/or outer packets have invalid checksum values.
- Inner Rx packet type classification
PMD sets inner L3/L4 packet types (e.g. ``RTE_PTYPE_INNER_L4_TCP``), and
``RTE_PTYPE_TUNNEL_GRENAT`` to indicate that the packet is tunneled.
PMD does not set L3/L4 packet types for outer packets.
- Inner RSS
RSS hash calculation, therefore queue selection, is done on inner packets.
In order to enable overlay offload, the 'Enable VXLAN' box should be checked
via CIMC or UCSM followed by a reboot of the server. When PMD successfully
enables overlay offload, it prints the following message on the console.
.. code-block:: console
Overlay offload is enabled
By default, PMD enables overlay offload if hardware supports it. To disable
it, set ``devargs`` parameter ``disable-overlay=1``. For example::
-w 12:00.0,disable-overlay=1
By default, the NIC uses 4789 as the VXLAN port. The user may change
it through ``rte_eth_dev_udp_tunnel_port_{add,delete}``. However, as
the current NIC has a single VXLAN port number, the user cannot
configure multiple port numbers.
Ingress VLAN Rewrite
--------------------
VIC adapters can tag, untag, or modify the VLAN headers of ingress
packets. The ingress VLAN rewrite mode controls this behavior. By
default, it is set to pass-through, where the NIC does not modify the
VLAN header in any way so that the application can see the original
header. This mode is sufficient for many applications, but may not be
suitable for others. Such applications may change the mode by setting
``devargs`` parameter ``ig-vlan-rewrite`` to one of the following.
- ``pass``: Pass-through mode. The NIC does not modify the VLAN
header. This is the default mode.
- ``priority``: Priority-tag default VLAN mode. If the ingress packet
is tagged with the default VLAN, the NIC replaces its VLAN header
with the priority tag (VLAN ID 0).
- ``trunk``: Default trunk mode. The NIC tags untagged ingress packets
with the default VLAN. Tagged ingress packets are not modified. To
the application, every packet appears as tagged.
- ``untag``: Untag default VLAN mode. If the ingress packet is tagged
with the default VLAN, the NIC removes or untags its VLAN header so
that the application sees an untagged packet. As a result, the
default VLAN becomes `untagged`. This mode can be useful for
applications such as OVS-DPDK performance benchmarks that utilize
only the default VLAN and want to see only untagged packets.
Vectorized Rx Handler
---------------------
ENIC PMD includes a version of the receive handler that is vectorized using
AVX2 SIMD instructions. It is meant for bulk, throughput oriented workloads
where reducing cycles/packet in PMD is a priority. In order to use the
vectorized handler, take the following steps.
- Use a recent version of gcc, icc, or clang and build 64-bit DPDK. If
the compiler is known to support AVX2, DPDK build system
automatically compiles the vectorized handler. Otherwise, the
handler is not available.
- Set ``devargs`` parameter ``enable-avx2-rx=1`` to explicitly request that
PMD consider the vectorized handler when selecting the receive handler.
For example::
-w 12:00.0,enable-avx2-rx=1
As the current implementation is intended for field trials, by default, the
vectorized handler is not considerd (``enable-avx2-rx=0``).
- Run on a UCS M4 or later server with CPUs that support AVX2.
PMD selects the vectorized handler when the handler is compiled into
the driver, the user requests its use via ``enable-avx2-rx=1``, CPU
supports AVX2, and scatter Rx is not used. To verify that the
vectorized handler is selected, enable debug logging
(``--log-level=pmd,debug``) and check the following message.
.. code-block:: console
enic_use_vector_rx_handler use the non-scatter avx2 Rx handler
.. _enic_limitations:
Limitations
-----------
- **VLAN 0 Priority Tagging**
If a vNIC is configured in TRUNK mode by the UCS manager, the adapter will
priority tag egress packets according to 802.1Q if they were not already
VLAN tagged by software. If the adapter is connected to a properly configured
switch, there will be no unexpected behavior.
In test setups where an Ethernet port of a Cisco adapter in TRUNK mode is
connected point-to-point to another adapter port or connected though a router
instead of a switch, all ingress packets will be VLAN tagged. Programs such
as l3fwd may not account for VLAN tags in packets and may misbehave. One
solution is to enable VLAN stripping on ingress so the VLAN tag is removed
from the packet and put into the mbuf->vlan_tci field. Here is an example
of how to accomplish this:
.. code-block:: console
vlan_offload = rte_eth_dev_get_vlan_offload(port);
vlan_offload |= ETH_VLAN_STRIP_OFFLOAD;
rte_eth_dev_set_vlan_offload(port, vlan_offload);
Another alternative is modify the adapter's ingress VLAN rewrite mode so that
packets with the default VLAN tag are stripped by the adapter and presented to
DPDK as untagged packets. In this case mbuf->vlan_tci and the PKT_RX_VLAN and
PKT_RX_VLAN_STRIPPED mbuf flags would not be set. This mode is enabled with the
``devargs`` parameter ``ig-vlan-rewrite=untag``. For example::
-w 12:00.0,ig-vlan-rewrite=untag
- Limited flow director support on 1200 series and 1300 series Cisco VIC
adapters with old firmware. Please see :ref:`enic-flow-director`.
- Flow director features are not supported on generation 1 Cisco VIC adapters
(M81KR and P81E)
- **SR-IOV**
- KVM hypervisor support only. VMware has not been tested.
- Requires VM-FEX, and so is only available on UCS managed servers connected
to Fabric Interconnects. It is not on standalone C-Series servers.
- VF devices are not usable directly from the host. They can only be used
as assigned devices on VM instances.
- Currently, unbind of the ENIC kernel mode driver 'enic.ko' on the VM
instance may hang. As a workaround, enic.ko should be blacklisted or removed
from the boot process.
- pci_generic cannot be used as the uio module in the VM. igb_uio or
vfio in non-IOMMU mode can be used.
- The number of RQs in UCSM dynamic vNIC configurations must be at least 2.
- The number of SR-IOV devices is limited to 256. Components on target system
might limit this number to fewer than 256.
- **Flow API**
- The number of filters that can be specified with the Generic Flow API is
dependent on how many header fields are being masked. Use 'flow create' in
a loop to determine how many filters your VIC will support (not more than
1000 for 1300 series VICs). Filters are checked for matching in the order they
were added. Since there currently is no grouping or priority support,
'catch-all' filters should be added last.
- **Statistics**
- ``rx_good_bytes`` (ibytes) always includes VLAN header (4B) and CRC bytes (4B).
This behavior applies to 1300 and older series VIC adapters.
1400 series VICs do not count CRC bytes, and count VLAN header only when VLAN
stripping is disabled.
- When the NIC drops a packet because the Rx queue has no free buffers,
``rx_good_bytes`` still increments by 4B if the packet is not VLAN tagged or
VLAN stripping is disabled, or by 8B if the packet is VLAN tagged and stripping
is enabled.
This behavior applies to 1300 and older series VIC adapters. 1400 series VICs
do not increment this byte counter when packets are dropped.
- **RSS Hashing**
- Hardware enables and disables UDP and TCP RSS hashing together. The driver
cannot control UDP and TCP hashing individually.
How to build the suite
----------------------
The build instructions for the DPDK suite should be followed. By default
the ENIC PMD library will be built into the DPDK library.
Refer to the document :ref:`compiling and testing a PMD for a NIC
<pmd_build_and_test>` for details.
For configuring and using UIO and VFIO frameworks, please refer to the
documentation that comes with DPDK suite.
Supported Cisco VIC adapters
----------------------------
ENIC PMD supports all recent generations of Cisco VIC adapters including:
- VIC 1200 series
- VIC 1300 series
- VIC 1400 series
Supported Operating Systems
---------------------------
Any Linux distribution fulfilling the conditions described in Dependencies
section of DPDK documentation.
Supported features
------------------
- Unicast, multicast and broadcast transmission and reception
- Receive queue polling
- Port Hardware Statistics
- Hardware VLAN acceleration
- IP checksum offload
- Receive side VLAN stripping
- Multiple receive and transmit queues
- Flow Director ADD, UPDATE, DELETE, STATS operation support IPv4 and IPv6
- Promiscuous mode
- Setting RX VLAN (supported via UCSM/CIMC only)
- VLAN filtering (supported via UCSM/CIMC only)
- Execution of application by unprivileged system users
- IPV4, IPV6 and TCP RSS hashing
- UDP RSS hashing (1400 series and later adapters)
- Scattered Rx
- MTU update
- SR-IOV on UCS managed servers connected to Fabric Interconnects
- Flow API
- Overlay offload
- Rx/Tx checksum offloads for VXLAN, NVGRE, GENEVE
- TSO for VXLAN and GENEVE packets
- Inner RSS
Known bugs and unsupported features in this release
---------------------------------------------------
- Signature or flex byte based flow direction
- Drop feature of flow direction
- VLAN based flow direction
- Non-IPV4 flow direction
- Setting of extended VLAN
- MTU update only works if Scattered Rx mode is disabled
- Maximum receive packet length is ignored if Scattered Rx mode is used
Prerequisites
-------------
- Prepare the system as recommended by DPDK suite. This includes environment
variables, hugepages configuration, tool-chains and configuration.
- Insert vfio-pci kernel module using the command 'modprobe vfio-pci' if the
user wants to use VFIO framework.
- Insert uio kernel module using the command 'modprobe uio' if the user wants
to use UIO framework.
- DPDK suite should be configured based on the user's decision to use VFIO or
UIO framework.
- If the vNIC device(s) to be used is bound to the kernel mode Ethernet driver
use 'ip' to bring the interface down. The dpdk-devbind.py tool can
then be used to unbind the device's bus id from the ENIC kernel mode driver.
- Bind the intended vNIC to vfio-pci in case the user wants ENIC PMD to use
VFIO framework using dpdk-devbind.py.
- Bind the intended vNIC to igb_uio in case the user wants ENIC PMD to use
UIO framework using dpdk-devbind.py.
At this point the system should be ready to run DPDK applications. Once the
application runs to completion, the vNIC can be detached from vfio-pci or
igb_uio if necessary.
Root privilege is required to bind and unbind vNICs to/from VFIO/UIO.
VFIO framework helps an unprivileged user to run the applications.
For an unprivileged user to run the applications on DPDK and ENIC PMD,
it may be necessary to increase the maximum locked memory of the user.
The following command could be used to do this.
.. code-block:: console
sudo sh -c "ulimit -l <value in Kilo Bytes>"
The value depends on the memory configuration of the application, DPDK and
PMD. Typically, the limit has to be raised to higher than 2GB.
e.g., 2621440
The compilation of any unused drivers can be disabled using the
configuration file in config/ directory (e.g., config/common_linuxapp).
This would help in bringing down the time taken for building the
libraries and the initialization time of the application.
Additional Reference
--------------------
- https://www.cisco.com/c/en/us/products/servers-unified-computing/index.html
- https://www.cisco.com/c/en/us/products/interfaces-modules/unified-computing-system-adapters/index.html
Contact Information
-------------------
Any questions or bugs should be reported to DPDK community and to the ENIC PMD
maintainers:
- John Daley <johndale@cisco.com>
- Hyong Youb Kim <hyonkim@cisco.com>
Reference in New Issue View Git Blame Copy Permalink