X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=doc%2Fguides%2Fnics%2Fqede.rst;h=84becc988aed5cf7b67f99025dea736e50238f13;hb=db86fbe54d901f048e958c0f2ff86a39086886fa;hp=53d749c9b1c25d49179a3b78b084185c2c944fa7;hpb=a5d7a3f77ddc3c3ae18bce04d7555b458360cc65;p=dpdk.git diff --git a/doc/guides/nics/qede.rst b/doc/guides/nics/qede.rst index 53d749c9b1..84becc988a 100644 --- a/doc/guides/nics/qede.rst +++ b/doc/guides/nics/qede.rst @@ -1,5 +1,6 @@ .. BSD LICENSE Copyright (c) 2016 QLogic Corporation + Copyright (c) 2017 Cavium Inc. All rights reserved. Redistribution and use in source and binary forms, with or without @@ -32,8 +33,7 @@ QEDE Poll Mode Driver ====================== The QEDE poll mode driver library (**librte_pmd_qede**) implements support -for **QLogic FastLinQ QL4xxxx 25G/40G CNA** family of adapters as well -as their virtual functions (VF) in SR-IOV context. It is supported on +for **QLogic FastLinQ QL4xxxx 10G/25G/40G/50G/100G Intelligent Ethernet Adapters (IEA) and Converged Network Adapters (CNA)** family of adapters as well as SR-IOV virtual functions (VF). It is supported on several standard Linux distros like RHEL7.x, SLES12.x and Ubuntu. It is compile-tested under FreeBSD OS. @@ -47,44 +47,57 @@ Supported Features - Promiscuous mode - Allmulti mode - Port hardware statistics -- Jumbo frames (using single buffer) -- VLAN offload - Filtering and stripping -- Stateless checksum offloads (IPv4/TCP/UDP) -- Multiple Rx/Tx queues (queue-pairs) -- RSS (with user configurable table/key) -- TSS +- Jumbo frames - Multiple MAC address +- MTU change - Default pause flow control -- SR-IOV VF for 25G/40G modes +- Multiprocess aware +- Scatter-Gather +- Multiple Rx/Tx queues +- RSS (with RETA/hash table/key) +- TSS +- Stateless checksum offloads (IPv4/IPv6/TCP/UDP) +- LRO/TSO +- VLAN offload - Filtering and stripping +- N-tuple filter and flow director (limited support) +- NPAR (NIC Partitioning) +- SR-IOV VF +- VXLAN tunneling offload +- MPLSoUDP Tx tunnel offload Non-supported Features ---------------------- -- Scatter-Gather Rx/Tx frames -- Unequal number of Rx/Tx queues -- MTU change (dynamic) - SR-IOV PF -- Tunneling offloads -- Reload of the PMD after a non-graceful termination +- GENEVE and NVGRE Tunneling offloads Supported QLogic Adapters ------------------------- -- QLogic FastLinQ QL4xxxx 25G/40G/100G CNAs. +- QLogic FastLinQ QL4xxxx 10G/25G/40G/50G/100G Intelligent Ethernet Adapters (IEA) and Converged Network Adapters (CNA) Prerequisites ------------- -- Requires firmware version **8.7.x.** and management firmware - version **8.7.x or higher**. Firmware may be available +- Requires storm firmware version **8.30.12.0**. Firmware may be available inbox in certain newer Linux distros under the standard directory - ``E.g. /lib/firmware/qed/qed_init_values_zipped-8.7.7.0.bin`` + ``E.g. /lib/firmware/qed/qed_init_values-8.30.12.0.bin`` + If the required firmware files are not available then download it from + `QLogic Driver Download Center `_. + For downloading firmware file, select adapter category, model and DPDK Poll Mode Driver. + +- Requires management firmware (MFW) version **8.30.x.x** or higher to be + flashed on to the adapter. If the required management firmware is not + available then download from + `QLogic Driver Download Center `_. + For downloading firmware upgrade utility, select adapter category, model and Linux distro. + To flash the management firmware refer to the instructions in the QLogic Firmware Upgrade Utility Readme document. -- If the required firmware files are not available then visit - `QLogic Driver Download Center `_. +- SR-IOV requires Linux PF driver version **8.20.x.x** or higher. + If the required PF driver is not available then download it from + `QLogic Driver Download Center `_. + For downloading PF driver, select adapter category, model and Linux distro. -- This driver relies on external zlib library (-lz) for uncompressing - the firmware file. Performance note ~~~~~~~~~~~~~~~~ @@ -117,75 +130,113 @@ enabling debugging options may affect system performance. Toggle display of receive fast path run-time messages. +- ``CONFIG_RTE_LIBRTE_QEDE_VF_TX_SWITCH`` (default **"y"**) + + A knob to control per-VF Tx switching feature. + - ``CONFIG_RTE_LIBRTE_QEDE_FW`` (default **""**) Gives absolute path of firmware file. - ``Eg: "/lib/firmware/qed/qed_init_values_zipped-8.7.7.0.bin"`` + ``Eg: "/lib/firmware/qed/qed_init_values-8.30.12.0.bin"`` Empty string indicates driver will pick up the firmware file - from the default location. + from the default location /lib/firmware/qed. + CAUTION this option is more for custom firmware, it is not + recommended for use under normal condition. -Driver Compilation -~~~~~~~~~~~~~~~~~~ +Driver compilation and testing +------------------------------ -To compile QEDE PMD for Linux x86_64 gcc target, run the following ``make`` -command:: +Refer to the document :ref:`compiling and testing a PMD for a NIC ` +for details. - cd - make config T=x86_64-native-linuxapp-gcc install +SR-IOV: Prerequisites and Sample Application Notes +-------------------------------------------------- -To compile QEDE PMD for Linux x86_64 clang target, run the following ``make`` -command:: +This section provides instructions to configure SR-IOV with Linux OS. - cd - make config T=x86_64-native-linuxapp-clang install +**Note**: librte_pmd_qede will be used to bind to SR-IOV VF device and Linux native kernel driver (qede) will function as SR-IOV PF driver. Requires PF driver to be 8.10.x.x or higher. -To compile QEDE PMD for FreeBSD x86_64 clang target, run the following ``gmake`` -command:: +#. Verify SR-IOV and ARI capability is enabled on the adapter using ``lspci``: + + .. code-block:: console + + lspci -s -vvv - cd - gmake config T=x86_64-native-bsdapp-clang install + Example output: -To compile QEDE PMD for FreeBSD x86_64 gcc target, run the following ``gmake`` -command:: + .. code-block:: console - cd - gmake config T=x86_64-native-bsdapp-gcc install -Wl,-rpath=\ - /usr/local/lib/gcc48 CC=gcc48 + [...] + Capabilities: [1b8 v1] Alternative Routing-ID Interpretation (ARI) + [...] + Capabilities: [1c0 v1] Single Root I/O Virtualization (SR-IOV) + [...] + Kernel driver in use: igb_uio +#. Load the kernel module: -Sample Application Notes -~~~~~~~~~~~~~~~~~~~~~~~~ + .. code-block:: console -This section demonstrates how to launch ``testpmd`` with QLogic 4xxxx -devices managed by ``librte_pmd_qede`` in Linux operating system. + modprobe qede -#. Request huge pages: + Example output: .. code-block:: console - echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages/ \ - nr_hugepages + systemd-udevd[4848]: renamed network interface eth0 to ens5f0 + systemd-udevd[4848]: renamed network interface eth1 to ens5f1 -#. Load ``igb_uio`` driver: +#. Bring up the PF ports: .. code-block:: console - insmod ./x86_64-native-linuxapp-gcc/kmod/igb_uio.ko + ifconfig ens5f0 up + ifconfig ens5f1 up -#. Bind the QLogic 4xxxx adapters to ``igb_uio`` loaded in the - previous step: +#. Create VF device(s): + + Echo the number of VFs to be created into ``"sriov_numvfs"`` sysfs entry + of the parent PF. + + Example output: .. code-block:: console - ./tools/dpdk-devbind.py --bind igb_uio 0000:84:00.0 0000:84:00.1 \ - 0000:84:00.2 0000:84:00.3 + echo 2 > /sys/devices/pci0000:00/0000:00:03.0/0000:81:00.0/sriov_numvfs -#. Start ``testpmd`` with basic parameters: - (Enable QEDE_DEBUG_INFO=y to view informational messages) + +#. Assign VF MAC address: + + Assign MAC address to the VF using iproute2 utility. The syntax is:: + + ip link set vf mac + + Example output: .. code-block:: console - testpmd -c 0xff1 -n 4 -- -i --nb-cores=8 --portmask=0xf --rxd=4096 \ + ip link set ens5f0 vf 0 mac 52:54:00:2f:9d:e8 + + +#. PCI Passthrough: + + The VF devices may be passed through to the guest VM using ``virt-manager`` or + ``virsh``. QEDE PMD should be used to bind the VF devices in the guest VM + using the instructions from Driver compilation and testing section above. + + +#. Running testpmd + (Enable QEDE_DEBUG_INFO=y to view informational messages): + + Refer to the document + :ref:`compiling and testing a PMD for a NIC ` to run + ``testpmd`` application. + + Example output: + + .. code-block:: console + + testpmd -l 0,4-11 -n 4 -- -i --nb-cores=8 --portmask=0xf --rxd=4096 \ --txd=4096 --txfreet=4068 --enable-rx-cksum --rxq=4 --txq=4 \ --rss-ip --rss-udp @@ -236,79 +287,3 @@ devices managed by ``librte_pmd_qede`` in Linux operating system. Port 3 Link Up - speed 25000 Mbps - full-duplex Done testpmd> - - -SR-IOV: Prerequisites and Sample Application Notes -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - -This section provides instructions to configure SR-IOV with Linux OS. - -**Note**: librte_pmd_qede will be used to bind to SR-IOV VF device and Linux native kernel driver (QEDE) will function as SR-IOV PF driver. - -#. Verify SR-IOV and ARI capability is enabled on the adapter using ``lspci``: - - .. code-block:: console - - lspci -s -vvv - - Example output: - - .. code-block:: console - - [...] - Capabilities: [1b8 v1] Alternative Routing-ID Interpretation (ARI) - [...] - Capabilities: [1c0 v1] Single Root I/O Virtualization (SR-IOV) - [...] - Kernel driver in use: igb_uio - -#. Load the kernel module: - - .. code-block:: console - - modprobe qede - - Example output: - - .. code-block:: console - - systemd-udevd[4848]: renamed network interface eth0 to ens5f0 - systemd-udevd[4848]: renamed network interface eth1 to ens5f1 - -#. Bring up the PF ports: - - .. code-block:: console - - ifconfig ens5f0 up - ifconfig ens5f1 up - -#. Create VF device(s): - - Echo the number of VFs to be created into ``"sriov_numvfs"`` sysfs entry - of the parent PF. - - Example output: - - .. code-block:: console - - echo 2 > /sys/devices/pci0000:00/0000:00:03.0/0000:81:00.0/sriov_numvfs - - -#. Assign VF MAC address: - - Assign MAC address to the VF using iproute2 utility. The syntax is:: - - ip link set vf mac - - Example output: - - .. code-block:: console - - ip link set ens5f0 vf 0 mac 52:54:00:2f:9d:e8 - - -#. PCI Passthrough: - - The VF devices may be passed through to the guest VM using ``virt-manager`` or - ``virsh``. QEDE PMD should be used to bind the VF devices in the guest VM - using the instructions outlined in the Application notes above.