From 5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Feb 2023 18:24:12 -0800 Subject: Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next Pull networking updates from Jakub Kicinski: "Core: - Add dedicated kmem_cache for typical/small skb->head, avoid having to access struct page at kfree time, and improve memory use. - Introduce sysctl to set default RPS configuration for new netdevs. - Define Netlink protocol specification format which can be used to describe messages used by each family and auto-generate parsers. Add tools for generating kernel data structures and uAPI headers. - Expose all net/core sysctls inside netns. - Remove 4s sleep in netpoll if carrier is instantly detected on boot. - Add configurable limit of MDB entries per port, and port-vlan. - Continue populating drop reasons throughout the stack. - Retire a handful of legacy Qdiscs and classifiers. Protocols: - Support IPv4 big TCP (TSO frames larger than 64kB). - Add IP_LOCAL_PORT_RANGE socket option, to control local port range on socket by socket basis. - Track and report in procfs number of MPTCP sockets used. - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path manager. - IPv6: don't check net.ipv6.route.max_size and rely on garbage collection to free memory (similarly to IPv4). - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986). - ICMP: add per-rate limit counters. - Add support for user scanning requests in ieee802154. - Remove static WEP support. - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate reporting. - WiFi 7 EHT channel puncturing support (client & AP). BPF: - Add a rbtree data structure following the "next-gen data structure" precedent set by recently added linked list, that is, by using kfunc + kptr instead of adding a new BPF map type. - Expose XDP hints via kfuncs with initial support for RX hash and timestamp metadata. - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to better support decap on GRE tunnel devices not operating in collect metadata. - Improve x86 JIT's codegen for PROBE_MEM runtime error checks. - Remove the need for trace_printk_lock for bpf_trace_printk and bpf_trace_vprintk helpers. - Extend libbpf's bpf_tracing.h support for tracing arguments of kprobes/uprobes and syscall as a special case. - Significantly reduce the search time for module symbols by livepatch and BPF. - Enable cpumasks to be used as kptrs, which is useful for tracing programs tracking which tasks end up running on which CPUs in different time intervals. - Add support for BPF trampoline on s390x and riscv64. - Add capability to export the XDP features supported by the NIC. - Add __bpf_kfunc tag for marking kernel functions as kfuncs. - Add cgroup.memory=nobpf kernel parameter option to disable BPF memory accounting for container environments. Netfilter: - Remove the CLUSTERIP target. It has been marked as obsolete for years, and we still have WARN splats wrt races of the out-of-band /proc interface installed by this target. - Add 'destroy' commands to nf_tables. They are identical to the existing 'delete' commands, but do not return an error if the referenced object (set, chain, rule...) did not exist. Driver API: - Improve cpumask_local_spread() locality to help NICs set the right IRQ affinity on AMD platforms. - Separate C22 and C45 MDIO bus transactions more clearly. - Introduce new DCB table to control DSCP rewrite on egress. - Support configuration of Physical Layer Collision Avoidance (PLCA) Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of shared medium Ethernet. - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing preemption of low priority frames by high priority frames. - Add support for controlling MACSec offload using netlink SET. - Rework devlink instance refcounts to allow registration and de-registration under the instance lock. Split the code into multiple files, drop some of the unnecessarily granular locks and factor out common parts of netlink operation handling. - Add TX frame aggregation parameters (for USB drivers). - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning messages with notifications for debug. - Allow offloading of UDP NEW connections via act_ct. - Add support for per action HW stats in TC. - Support hardware miss to TC action (continue processing in SW from a specific point in the action chain). - Warn if old Wireless Extension user space interface is used with modern cfg80211/mac80211 drivers. Do not support Wireless Extensions for Wi-Fi 7 devices at all. Everyone should switch to using nl80211 interface instead. - Improve the CAN bit timing configuration. Use extack to return error messages directly to user space, update the SJW handling, including the definition of a new default value that will benefit CAN-FD controllers, by increasing their oscillator tolerance. New hardware / drivers: - Ethernet: - nVidia BlueField-3 support (control traffic driver) - Ethernet support for imx93 SoCs - Motorcomm yt8531 gigabit Ethernet PHY - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA) - Microchip LAN8841 PHY (incl. cable diagnostics and PTP) - Amlogic gxl MDIO mux - WiFi: - RealTek RTL8188EU (rtl8xxxu) - Qualcomm Wi-Fi 7 devices (ath12k) - CAN: - Renesas R-Car V4H Drivers: - Bluetooth: - Set Per Platform Antenna Gain (PPAG) for Intel controllers. - Ethernet NICs: - Intel (1G, igc): - support TSN / Qbv / packet scheduling features of i226 model - Intel (100G, ice): - use GNSS subsystem instead of TTY - multi-buffer XDP support - extend support for GPIO pins to E823 devices - nVidia/Mellanox: - update the shared buffer configuration on PFC commands - implement PTP adjphase function for HW offset control - TC support for Geneve and GRE with VF tunnel offload - more efficient crypto key management method - multi-port eswitch support - Netronome/Corigine: - add DCB IEEE support - support IPsec offloading for NFP3800 - Freescale/NXP (enetc): - support XDP_REDIRECT for XDP non-linear buffers - improve reconfig, avoid link flap and waiting for idle - support MAC Merge layer - Other NICs: - sfc/ef100: add basic devlink support for ef100 - ionic: rx_push mode operation (writing descriptors via MMIO) - bnxt: use the auxiliary bus abstraction for RDMA - r8169: disable ASPM and reset bus in case of tx timeout - cpsw: support QSGMII mode for J721e CPSW9G - cpts: support pulse-per-second output - ngbe: add an mdio bus driver - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing - r8152: handle devices with FW with NCM support - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation - virtio-net: support multi buffer XDP - virtio/vsock: replace virtio_vsock_pkt with sk_buff - tsnep: XDP support - Ethernet high-speed switches: - nVidia/Mellanox (mlxsw): - add support for latency TLV (in FW control messages) - Microchip (sparx5): - separate explicit and implicit traffic forwarding rules, make the implicit rules always active - add support for egress DSCP rewrite - IS0 VCAP support (Ingress Classification) - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS etc.) - ES2 VCAP support (Egress Access Control) - support for Per-Stream Filtering and Policing (802.1Q, 8.6.5.1) - Ethernet embedded switches: - Marvell (mv88e6xxx): - add MAB (port auth) offload support - enable PTP receive for mv88e6390 - NXP (ocelot): - support MAC Merge layer - support for the the vsc7512 internal copper phys - Microchip: - lan9303: convert to PHYLINK - lan966x: support TC flower filter statistics - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x - lan937x: support Credit Based Shaper configuration - ksz9477: support Energy Efficient Ethernet - other: - qca8k: convert to regmap read/write API, use bulk operations - rswitch: Improve TX timestamp accuracy - Intel WiFi (iwlwifi): - EHT (Wi-Fi 7) rate reporting - STEP equalizer support: transfer some STEP (connection to radio on platforms with integrated wifi) related parameters from the BIOS to the firmware. - Qualcomm 802.11ax WiFi (ath11k): - IPQ5018 support - Fine Timing Measurement (FTM) responder role support - channel 177 support - MediaTek WiFi (mt76): - per-PHY LED support - mt7996: EHT (Wi-Fi 7) support - Wireless Ethernet Dispatch (WED) reset support - switch to using page pool allocator - RealTek WiFi (rtw89): - support new version of Bluetooth co-existance - Mobile: - rmnet: support TX aggregation" * tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits) page_pool: add a comment explaining the fragment counter usage net: ethtool: fix __ethtool_dev_mm_supported() implementation ethtool: pse-pd: Fix double word in comments xsk: add linux/vmalloc.h to xsk.c sefltests: netdevsim: wait for devlink instance after netns removal selftest: fib_tests: Always cleanup before exit net/mlx5e: Align IPsec ASO result memory to be as required by hardware net/mlx5e: TC, Set CT miss to the specific ct action instance net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG net/mlx5: Refactor tc miss handling to a single function net/mlx5: Kconfig: Make tc offload depend on tc skb extension net/sched: flower: Support hardware miss to tc action net/sched: flower: Move filter handle initialization earlier net/sched: cls_api: Support hardware miss to tc action net/sched: Rename user cookie and act cookie sfc: fix builds without CONFIG_RTC_LIB sfc: clean up some inconsistent indentings net/mlx4_en: Introduce flexible array to silence overflow warning net: lan966x: Fix possible deadlock inside PTP net/ulp: Remove redundant ->clone() test in inet_clone_ulp(). ... --- .../device_drivers/ethernet/netronome/nfp.rst | 374 +++++++++++++++++++++ 1 file changed, 374 insertions(+) create mode 100644 Documentation/networking/device_drivers/ethernet/netronome/nfp.rst (limited to 'Documentation/networking/device_drivers/ethernet/netronome') diff --git a/Documentation/networking/device_drivers/ethernet/netronome/nfp.rst b/Documentation/networking/device_drivers/ethernet/netronome/nfp.rst new file mode 100644 index 000000000..650b57742 --- /dev/null +++ b/Documentation/networking/device_drivers/ethernet/netronome/nfp.rst @@ -0,0 +1,374 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +.. include:: + +=========================================== +Network Flow Processor (NFP) Kernel Drivers +=========================================== + +:Copyright: |copy| 2019, Netronome Systems, Inc. +:Copyright: |copy| 2022, Corigine, Inc. + +Contents +======== + +- `Overview`_ +- `Acquiring Firmware`_ +- `Devlink Info`_ +- `Configure Device`_ +- `Statistics`_ + +Overview +======== + +This driver supports Netronome and Corigine's line of Network Flow Processor +devices, including the NFP3800, NFP4000, NFP5000, and NFP6000 models, which +are also incorporated in the companies' family of Agilio SmartNICs. The SR-IOV +physical and virtual functions for these devices are supported by the driver. + +Acquiring Firmware +================== + +The NFP3800, NFP4000 and NFP6000 devices require application specific firmware +to function. Application firmware can be located either on the host file system +or in the device flash (if supported by management firmware). + +Firmware files on the host filesystem contain card type (`AMDA-*` string), media +config etc. They should be placed in `/lib/firmware/netronome` directory to +load firmware from the host file system. + +Firmware for basic NIC operation is available in the upstream +`linux-firmware.git` repository. + +A more comprehensive list of firmware can be downloaded from the +`Corigine Support site `_. + +Firmware in NVRAM +----------------- + +Recent versions of management firmware supports loading application +firmware from flash when the host driver gets probed. The firmware loading +policy configuration may be used to configure this feature appropriately. + +Devlink or ethtool can be used to update the application firmware on the device +flash by providing the appropriate `nic_AMDA*.nffw` file to the respective +command. Users need to take care to write the correct firmware image for the +card and media configuration to flash. + +Available storage space in flash depends on the card being used. + +Dealing with multiple projects +------------------------------ + +NFP hardware is fully programmable therefore there can be different +firmware images targeting different applications. + +When using application firmware from host, we recommend placing +actual firmware files in application-named subdirectories in +`/lib/firmware/netronome` and linking the desired files, e.g.:: + + $ tree /lib/firmware/netronome/ + /lib/firmware/netronome/ + ├── bpf + │   ├── nic_AMDA0081-0001_1x40.nffw + │   └── nic_AMDA0081-0001_4x10.nffw + ├── flower + │   ├── nic_AMDA0081-0001_1x40.nffw + │   └── nic_AMDA0081-0001_4x10.nffw + ├── nic + │   ├── nic_AMDA0081-0001_1x40.nffw + │   └── nic_AMDA0081-0001_4x10.nffw + ├── nic_AMDA0081-0001_1x40.nffw -> bpf/nic_AMDA0081-0001_1x40.nffw + └── nic_AMDA0081-0001_4x10.nffw -> bpf/nic_AMDA0081-0001_4x10.nffw + + 3 directories, 8 files + +You may need to use hard instead of symbolic links on distributions +which use old `mkinitrd` command instead of `dracut` (e.g. Ubuntu). + +After changing firmware files you may need to regenerate the initramfs +image. Initramfs contains drivers and firmware files your system may +need to boot. Refer to the documentation of your distribution to find +out how to update initramfs. Good indication of stale initramfs +is system loading wrong driver or firmware on boot, but when driver is +later reloaded manually everything works correctly. + +Selecting firmware per device +----------------------------- + +Most commonly all cards on the system use the same type of firmware. +If you want to load a specific firmware image for a specific card, you +can use either the PCI bus address or serial number. The driver will +print which files it's looking for when it recognizes a NFP device:: + + nfp: Looking for firmware file in order of priority: + nfp: netronome/serial-00-12-34-aa-bb-cc-10-ff.nffw: not found + nfp: netronome/pci-0000:02:00.0.nffw: not found + nfp: netronome/nic_AMDA0081-0001_1x40.nffw: found, loading... + +In this case if file (or link) called *serial-00-12-34-aa-bb-5d-10-ff.nffw* +or *pci-0000:02:00.0.nffw* is present in `/lib/firmware/netronome` this +firmware file will take precedence over `nic_AMDA*` files. + +Note that `serial-*` and `pci-*` files are **not** automatically included +in initramfs, you will have to refer to documentation of appropriate tools +to find out how to include them. + +Running firmware version +------------------------ + +The version of the loaded firmware for a particular interface, +(e.g. enp4s0), or an interface's port (e.g. enp4s0np0) can +be displayed with the ethtool command:: + + $ ethtool -i + +Firmware loading policy +----------------------- + +Firmware loading policy is controlled via three HWinfo parameters +stored as key value pairs in the device flash: + +app_fw_from_flash + Defines which firmware should take precedence, 'Disk' (0), 'Flash' (1) or + the 'Preferred' (2) firmware. When 'Preferred' is selected, the management + firmware makes the decision over which firmware will be loaded by comparing + versions of the flash firmware and the host supplied firmware. + This variable is configurable using the 'fw_load_policy' + devlink parameter. + +abi_drv_reset + Defines if the driver should reset the firmware when + the driver is probed, either 'Disk' (0) if firmware was found on disk, + 'Always' (1) reset or 'Never' (2) reset. Note that the device is always + reset on driver unload if firmware was loaded when the driver was probed. + This variable is configurable using the 'reset_dev_on_drv_probe' + devlink parameter. + +abi_drv_load_ifc + Defines a list of PF devices allowed to load FW on the device. + This variable is not currently user configurable. + +Devlink Info +============ + +The devlink info command displays the running and stored firmware versions +on the device, serial number and board information. + +Devlink info command example (replace PCI address):: + + $ devlink dev info pci/0000:03:00.0 + pci/0000:03:00.0: + driver nfp + serial_number CSAAMDA2001-1003000111 + versions: + fixed: + board.id AMDA2001-1003 + board.rev 01 + board.manufacture CSA + board.model mozart + running: + fw.mgmt 22.10.0-rc3 + fw.cpld 0x1000003 + fw.app nic-22.09.0 + chip.init AMDA-2001-1003 1003000111 + stored: + fw.bundle_id bspbundle_1003000111 + fw.mgmt 22.10.0-rc3 + fw.cpld 0x0 + chip.init AMDA-2001-1003 1003000111 + +Configure Device +================ + +This section explains how to use Agilio SmartNICs running basic NIC firmware. + +Configure interface link-speed +------------------------------ +The following steps explains how to change between 10G mode and 25G mode on +Agilio CX 2x25GbE cards. The changing of port speed must be done in order, +port 0 (p0) must be set to 10G before port 1 (p1) may be set to 10G. + +Down the respective interface(s):: + + $ ip link set dev down + $ ip link set dev down + +Set interface link-speed to 10G:: + + $ ethtool -s speed 10000 + $ ethtool -s speed 10000 + +Set interface link-speed to 25G:: + + $ ethtool -s speed 25000 + $ ethtool -s speed 25000 + +Reload driver for changes to take effect:: + + $ rmmod nfp; modprobe nfp + +Configure interface Maximum Transmission Unit (MTU) +--------------------------------------------------- + +The MTU of interfaces can temporarily be set using the iproute2, ip link or +ifconfig tools. Note that this change will not persist. Setting this via +Network Manager, or another appropriate OS configuration tool, is +recommended as changes to the MTU using Network Manager can be made to +persist. + +Set interface MTU to 9000 bytes:: + + $ ip link set dev mtu 9000 + +It is the responsibility of the user or the orchestration layer to set +appropriate MTU values when handling jumbo frames or utilizing tunnels. For +example, if packets sent from a VM are to be encapsulated on the card and +egress a physical port, then the MTU of the VF should be set to lower than +that of the physical port to account for the extra bytes added by the +additional header. If a setup is expected to see fallback traffic between +the SmartNIC and the kernel then the user should also ensure that the PF MTU +is appropriately set to avoid unexpected drops on this path. + +Configure Forward Error Correction (FEC) modes +---------------------------------------------- + +Agilio SmartNICs support FEC mode configuration, e.g. Auto, Firecode Base-R, +ReedSolomon and Off modes. Each physical port's FEC mode can be set +independently using ethtool. The supported FEC modes for an interface can +be viewed using:: + + $ ethtool + +The currently configured FEC mode can be viewed using:: + + $ ethtool --show-fec + +To force the FEC mode for a particular port, auto-negotiation must be disabled +(see the `Auto-negotiation`_ section). An example of how to set the FEC mode +to Reed-Solomon is:: + + $ ethtool --set-fec encoding rs + +Auto-negotiation +---------------- + +To change auto-negotiation settings, the link must first be put down. After the +link is down, auto-negotiation can be enabled or disabled using:: + + ethtool -s autoneg + +Statistics +========== + +Following device statistics are available through the ``ethtool -S`` interface: + +.. flat-table:: NFP device statistics + :header-rows: 1 + :widths: 3 1 11 + + * - Name + - ID + - Meaning + + * - dev_rx_discards + - 1 + - Packet can be discarded on the RX path for one of the following reasons: + + * The NIC is not in promisc mode, and the destination MAC address + doesn't match the interfaces' MAC address. + * The received packet is larger than the max buffer size on the host. + I.e. it exceeds the Layer 3 MRU. + * There is no freelist descriptor available on the host for the packet. + It is likely that the NIC couldn't cache one in time. + * A BPF program discarded the packet. + * The datapath drop action was executed. + * The MAC discarded the packet due to lack of ingress buffer space + on the NIC. + + * - dev_rx_errors + - 2 + - A packet can be counted (and dropped) as RX error for the following + reasons: + + * A problem with the VEB lookup (only when SR-IOV is used). + * A physical layer problem that causes Ethernet errors, like FCS or + alignment errors. The cause is usually faulty cables or SFPs. + + * - dev_rx_bytes + - 3 + - Total number of bytes received. + + * - dev_rx_uc_bytes + - 4 + - Unicast bytes received. + + * - dev_rx_mc_bytes + - 5 + - Multicast bytes received. + + * - dev_rx_bc_bytes + - 6 + - Broadcast bytes received. + + * - dev_rx_pkts + - 7 + - Total number of packets received. + + * - dev_rx_mc_pkts + - 8 + - Multicast packets received. + + * - dev_rx_bc_pkts + - 9 + - Broadcast packets received. + + * - dev_tx_discards + - 10 + - A packet can be discarded in the TX direction if the MAC is + being flow controlled and the NIC runs out of TX queue space. + + * - dev_tx_errors + - 11 + - A packet can be counted as TX error (and dropped) for one for the + following reasons: + + * The packet is an LSO segment, but the Layer 3 or Layer 4 offset + could not be determined. Therefore LSO could not continue. + * An invalid packet descriptor was received over PCIe. + * The packet Layer 3 length exceeds the device MTU. + * An error on the MAC/physical layer. Usually due to faulty cables or + SFPs. + * A CTM buffer could not be allocated. + * The packet offset was incorrect and could not be fixed by the NIC. + + * - dev_tx_bytes + - 12 + - Total number of bytes transmitted. + + * - dev_tx_uc_bytes + - 13 + - Unicast bytes transmitted. + + * - dev_tx_mc_bytes + - 14 + - Multicast bytes transmitted. + + * - dev_tx_bc_bytes + - 15 + - Broadcast bytes transmitted. + + * - dev_tx_pkts + - 16 + - Total number of packets transmitted. + + * - dev_tx_mc_pkts + - 17 + - Multicast packets transmitted. + + * - dev_tx_bc_pkts + - 18 + - Broadcast packets transmitted. + +Note that statistics unknown to the driver will be displayed as +``dev_unknown_stat$ID``, where ``$ID`` refers to the second column +above. -- cgit v1.2.3