[dpdk-dev] [PATCH 3/3] doc: update mlx guides

Shahaf Shuler shahafs at mellanox.com
Mon Jul 24 14:36:37 CEST 2017
Previous message: [dpdk-dev] [PATCH 2/3] doc: update release notes for mlx driver
Next message: [dpdk-dev] [PATCH 3/3] doc: update mlx guides
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Update the guides with:
   * New supported features.
   * Supported OFED and FW versions.
   * Quick start guide.
   * Performance tunning guide.

Signed-off-by: Shahaf Shuler <shahafs at mellanox.com>
Acked-by: Nelio Laranjeiro <nelio.laranjeiro at 6wind.com>
---
 doc/guides/nics/mlx4.rst | 161 +++++++++++++++++++++++++++++++---
 doc/guides/nics/mlx5.rst | 220 +++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 343 insertions(+), 38 deletions(-)

diff --git a/doc/guides/nics/mlx4.rst b/doc/guides/nics/mlx4.rst
index f1f26d4f9..23e14e52a 100644
--- a/doc/guides/nics/mlx4.rst
+++ b/doc/guides/nics/mlx4.rst
@@ -1,5 +1,6 @@
 ..  BSD LICENSE
     Copyright 2012-2015 6WIND S.A.
+    Copyright 2015 Mellanox.
 
     Redistribution and use in source and binary forms, with or without
     modification, are permitted provided that the following conditions
@@ -76,6 +77,7 @@ Compiling librte_pmd_mlx4 causes DPDK to be linked against libibverbs.
 Features
 --------
 
+- Multi arch support: x86 and Power8.
 - RSS, also known as RCA, is supported. In this mode the number of
   configured RX queues must be a power of two.
 - VLAN filtering is supported.
@@ -87,16 +89,7 @@ Features
 - Inner L3/L4 (IP, TCP and UDP) TX/RX checksum offloading and validation.
 - Outer L3 (IP) TX/RX checksum offloading and validation for VXLAN frames.
 - Secondary process TX is supported.
-
-Limitations
------------
-
-- RSS hash key cannot be modified.
-- RSS RETA cannot be configured
-- RSS always includes L3 (IPv4/IPv6) and L4 (UDP/TCP). They cannot be
-  dissociated.
-- Hardware counters are not implemented (they are software counters).
-- Secondary process RX is not supported.
+- Rx interrupts.
 
 Configuration
 -------------
@@ -244,8 +237,8 @@ DPDK and must be installed separately:
 
 Currently supported by DPDK:
 
-- Mellanox OFED **4.0-2.0.0.0**.
-- Firmware version **2.40.7000**.
+- Mellanox OFED **4.1**.
+- Firmware version **2.36.5000** and above.
 - Supported architectures:  **x86_64** and **POWER8**.
 
 Getting Mellanox OFED
@@ -273,6 +266,150 @@ Supported NICs
 
 * Mellanox(R) ConnectX(R)-3 Pro 40G MCX354A-FCC_Ax (2*40G)
 
+Quick Start guide
+------------------
+
+1. Download latest Mellanox OFED. For more info check the  `prerequisites`_.
+
+2. Install the required libraries and kernel modules either by installing
+   only the required set, or by installing the entire Mellanox OFED:
+
+   For Bare metal use:
+
+   .. code-block:: console
+
+        ./mlnxofedinstall
+
+   For SR-IOV Hypervisors use:
+
+   .. code-block:: console
+
+        ./mlnxofedinstall --enable-sriov -hypervisor
+
+   For SR-IOV Virtual machine use:
+
+   .. code-block:: console
+
+        ./mlnxofedinstall --guest
+
+3. Verify the firmware is the correct one:
+
+   .. code-block:: console
+
+        ibv_devinfo
+
+4. Set all ports links to ethernet, follow instruction on the screen:
+
+   .. code-block:: console
+
+        connectx_port_config
+
+5. In case of bare metal or Hypervisor, config the optimized steering mode
+   by adding the following line to ``/etc/modprobe.d/mlx4_core.conf``:
+
+   .. code-block:: console
+
+        options mlx4_core log_num_mgm_entry_size=-7
+
+   .. note::
+
+        If VLAN filtering is used, set log_num_mgm_entry_size=-1.
+        Performance degradation can occur on this case
+
+6. Restart the driver:
+
+   .. code-block:: console
+
+        /etc/init.d/openibd restart
+   or:
+
+   .. code-block:: console
+
+        service openibd restart
+
+7. Enable MLX4 PMD on the ``.config`` file:
+
+    .. code-block:: console
+
+        CONFIG_RTE_LIBRTE_MLX4_PMD=y
+
+8. Compile DPDK and you are ready to go:
+
+    .. code-block:: console
+
+        make config T=<cpu arch, compiler, ..>
+        make
+
+
+Limitations and known issues
+----------------------------
+
+- RSS hash key cannot be modified.
+- RSS RETA cannot be configured
+- RSS always includes L3 (IPv4/IPv6) and L4 (UDP/TCP). They cannot be
+  dissociated.
+- Hardware counters are not implemented (they are software counters).
+- Secondary process RX is not supported.
+
+Performance tunning
+-------------------
+
+1. Verify the optimized steering mode is configured
+
+  .. code-block:: console
+
+        cat /sys/module/mlx4_core/parameters/log_num_mgm_entry_size
+
+2. Use environment variable MLX4_INLINE_RECV_SIZE=64 to get maximum
+   performance for 64B messages.
+
+3. Use the CPU near local NUMA node to which the PCIe adapter is connected,
+   for better performance. For Virtual Machines (VM), verify that the right CPU
+   and NUMA node are pinned for the VM according to the above. Run
+
+   .. code-block:: console
+
+        lstopo-no-graphics
+
+   to identify the NUMA node to which the PCIe adapter is connected.
+
+4. If more than one adapter is used, and root complex capabilities enables
+   to put both adapters on the same NUMA node without PCI bandwidth degredation,
+   it is recommended to locate both adapters on the same NUMA node.
+   This in order to forward packets from one to the other without
+   NUMA performance penalty.
+
+5. Disable pause frames
+
+   .. code-block:: console
+
+        ethtool -A <netdev> rx off tx off
+
+6. Verify IO non-posted prefetch is disabled by default. This can be checked
+   via the BIOS configuration. Please contact you server provider for more
+   information about the settings.
+
+.. hint::
+
+        On Some machines, depends on the machine intergrator, it is beneficial
+        to set the PCI max read request parameter to 1K. This can be
+        done in the following way:
+
+        To query the read request size use:
+
+        .. code-block:: console
+
+                setpci -s <NIC PCI address> 68.w
+
+        If the output is different than 3XXX, set it by:
+
+        .. code-block:: console
+
+                setpci -s <NIC PCI address> 68.w=3XXX
+
+        The XXX can be different on different systems. Make sure to configure
+        according to the setpci output.
+
 Usage example
 -------------
 
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index a68b7adc0..8accd754b 100644
--- a/doc/guides/nics/mlx5.rst
+++ b/doc/guides/nics/mlx5.rst
@@ -1,5 +1,6 @@
 ..  BSD LICENSE
     Copyright 2015 6WIND S.A.
+    Copyright 2015 Mellanox.
 
     Redistribution and use in source and binary forms, with or without
     modification, are permitted provided that the following conditions
@@ -64,6 +65,9 @@ physical memory (or memory that does not belong to the current process).
 This capability allows the PMD to coexist with kernel network interfaces
 which remain functional, although they stop receiving unicast packets as
 long as they share the same MAC address.
+This means legacy linux control tools (for example:  ethtool, ifconfig and
+more) can operate on the same network interfaces that owned by the DPDK
+application.
 
 Enabling librte_pmd_mlx5 causes DPDK applications to be linked against
 libibverbs.
@@ -71,6 +75,7 @@ libibverbs.
 Features
 --------
 
+- Multi arch support: x86, Power8, ARMv8.
 - Multiple TX and RX queues.
 - Support for scattered TX and RX frames.
 - IPv4, IPv6, TCPv4, TCPv6, UDPv4 and UDPv6 RSS on any number of queues.
@@ -92,14 +97,8 @@ Features
 - RSS hash result is supported.
 - Hardware TSO.
 - Hardware checksum TX offload for VXLAN and GRE.
-
-Limitations
------------
-
-- Inner RSS for VXLAN frames is not supported yet.
-- Port statistics through software counters only.
-- Hardware checksum RX offloads for VXLAN inner header are not supported yet.
-- Secondary process RX is not supported.
+- RX interrupts
+- Statistics query including Basic, Extended and per queue.
 
 Configuration
 -------------
@@ -156,13 +155,12 @@ Run-time configuration
 - ``rxq_cqe_comp_en`` parameter [int]
 
   A nonzero value enables the compression of CQE on RX side. This feature
-  allows to save PCI bandwidth and improve performance at the cost of a
-  slightly higher CPU usage.  Enabled by default.
+  allows to save PCI bandwidth and improve performance. Enabled by default.
 
   Supported on:
 
-  - x86_64 with ConnectX4 and ConnectX4 LX
-  - Power8 with ConnectX4 LX
+  - x86_64 with ConnectX-4, ConnectX-4LX and ConnectX-5.
+  - Power8 and ARMv8 with ConnectX-4LX and ConnectX-5.
 
 - ``txq_inline`` parameter [int]
 
@@ -170,17 +168,26 @@ Run-time configuration
   Can improve PPS performance when PCI back pressure is detected and may be
   useful for scenarios involving heavy traffic on many queues.
 
-  It is not enabled by default (set to 0) since the additional software
-  logic necessary to handle this mode can lower performance when back
+  Since the additional software logic necessary to handle this mode this
+  option should be used with care, as it can lower performance when back
   pressure is not expected.
 
 - ``txqs_min_inline`` parameter [int]
 
   Enable inline send only when the number of TX queues is greater or equal
   to this value.
-
   This option should be used in combination with ``txq_inline`` above.
 
+  On ConnectX-4/ConnectX-4LX:
+
+        - disabled by default. in case ``txq_inline`` is set recommendation is 4.
+
+  On ConnectX-5:
+
+        - when Enhanced MPW is enabled, it is set to 8 by default.
+        - otherwise disabled by default. in case ``txq_inline`` is set
+          use same values as ConnectX-4/ConnectX-4LX.
+
 - ``txq_mpw_en`` parameter [int]
 
   A nonzero value enables multi-packet send (MPS) for ConnectX-4 Lx and
@@ -221,9 +228,7 @@ Run-time configuration
 
   A nonzero value enables hardware TSO.
   When hardware TSO is enabled, packets marked with TCP segmentation
-  offload will be divided into segments by the hardware.
-
-  Disabled by default.
+  offload will be divided into segments by the hardware. Disabled by default.
 
 Prerequisites
 -------------
@@ -279,13 +284,13 @@ DPDK and must be installed separately:
 
 Currently supported by DPDK:
 
-- Mellanox OFED version: **4.0-2.0.0.0**
+- Mellanox OFED version: **4.1**.
 - firmware version:
 
-  - ConnectX-4: **12.18.2000**
-  - ConnectX-4 Lx: **14.18.2000**
-  - ConnectX-5: **16.19.1200**
-  - ConnectX-5 Ex: **16.19.1200**
+  - ConnectX-4: **12.20.1010** and above.
+  - ConnectX-4 Lx: **14.20.1010** and above.
+  - ConnectX-5: **16.20.1010** and above.
+  - ConnectX-5 Ex: **16.20.1010** and above.
 
 Getting Mellanox OFED
 ~~~~~~~~~~~~~~~~~~~~~
@@ -330,10 +335,103 @@ Supported NICs
 * Mellanox(R) ConnectX(R)-5 100G MCX556A-ECAT (2x100G)
 * Mellanox(R) ConnectX(R)-5 Ex EN 100G MCX516A-CDAT (2x100G)
 
-Known issues
-------------
+Quick Start guide
+------------------
+
+1. Download latest Mellanox OFED. For more info check the  `prerequisites`_.
+
+
+2. Install the required libraries and kernel modules either by installing
+   only the required set, or by installing the entire Mellanox OFED:
+
+   .. code-block:: console
+
+        ./mlnxofedinstall
+
+3. Verify the firmware is the correct one:
+
+   .. code-block:: console
+
+        ibv_devinfo
+
+4. Verify all ports links are set to Ethernet:
+
+   .. code-block:: console
+
+        mlxconfig -d <mst device> query | grep LINK_TYPE
+        LINK_TYPE_P1                        ETH(2)
+        LINK_TYPE_P2                        ETH(2)
+
+   If the Links are not in the current protocol move the to Ethernet:
+
+   .. code-block:: console
+
+        mlxconfig -d <mst device> set LINK_TYPE_P1/2=1/2/3
+
+        * LINK_TYPE_P1=<1|2|3> , 1=Infiniband 2=Ethernet 3=VPI(auto-sense)
+
+   For Hypervisors verify SR-IOV is enabled on the NIC:
+
+   .. code-block:: console
+
+        mlxconfig -d <mst device> query | grep SRIOV_EN
+        SRIOV_EN                            True(1)
+
+   If Needed, set enable the set the relevant fields:
 
-* **Flow pattern without any specific vlan will match for vlan packets as well.**
+   .. code-block:: console
+
+        mlxconfig -d <mst device> set SRIOV_EN=1 NUM_OF_VFS=16
+        mlxfwreset -d <mst device> reset
+
+5. Restart the driver:
+
+   .. code-block:: console
+
+        /etc/init.d/openibd restart
+   or:
+
+   .. code-block:: console
+
+        service openibd restart
+
+   If port link protocol was changed need to reset the fw as well:
+
+   .. code-block:: console
+
+        mlxfwreset -d <mst device> reset
+
+   For Hypervisors, after reset write the sysfs number of Virtual Functions
+   needed for the PF.
+   The following is an example of a standard Linux kernel generated file that
+   is available in the new kernels:
+
+   .. code-block:: console
+
+        echo [num_vfs] > /sys/class/infiniband/mlx5_0/device/sriov_numvfs
+
+
+6. Enable MLX5 PMD in the ``.config`` file :
+
+    .. code-block:: console
+
+        CONFIG_RTE_LIBRTE_MLX5_PMD=y
+
+7. Compile DPDK and you are ready to go:
+
+    .. code-block:: console
+
+        make config T=<cpu arch, compiler, ..>
+        make
+
+Limitations and Known issues
+----------------------------
+
+- Inner RSS for VXLAN frames is not supported yet.
+- Port statistics through software counters only.
+- Hardware checksum RX offloads for VXLAN inner header are not supported yet.
+- Secondary process RX is not supported.
+- Flow pattern without any specific vlan will match for vlan packets as well:
 
   When VLAN spec is not specified in the pattern, the matching rule will be created with VLAN as a wild card.
   Meaning, the flow rule::
@@ -350,6 +448,76 @@ Known issues
 
   Will match any ipv4 packet (VLAN included).
 
+Performance tunning
+-------------------
+
+1. Configure aggressive CQE Zipping for maximum performance
+
+  .. code-block:: console
+
+        mlxconfig -d <mst device> s CQE_COMPRESSION=1
+
+  To set it back to the default CQE Zipping mode use
+
+  .. code-block:: console
+
+        mlxconfig -d <mst device> s CQE_COMPRESSION=0
+
+2. In case of Virtualization:
+
+   - Make sure that Hypervisor kernel is 3.16 or newer.
+   - Configure boot with "iommu=pt".
+   - Use 1G huge pages.
+   - Make sure to allocate a VM on huge pages.
+   - Make sure to set CPU pinning.
+
+3. Use the CPU near local NUMA node to which the PCIe adapter is connected,
+   for better performance. For Virtual Machines (VM), verify that the right CPU
+   and NUMA node are pinned for the VM according to the above. Run
+
+   .. code-block:: console
+
+        lstopo-no-graphics
+
+   to identify the NUMA node to which the PCIe adapter is connected.
+
+4. If more than one adapter is used, and root complex capabilities enables
+   to put both adapters on the same NUMA node without PCI bandwidth degredation,
+   it is recommended to locate both adapters on the same NUMA node.
+   This in order to forward packets from one to the other without
+   NUMA performance penalty.
+
+5. Disable pause frames
+
+   .. code-block:: console
+
+        ethtool -A <netdev> rx off tx off
+
+6. Verify IO non-posted prefetch is disabled by default. This can be checked
+   via the BIOS configuration. Please contact you server provider for more
+   information about the settings.
+
+.. hint::
+
+        On Some machines, depends on the machine intergrator, it is beneficial
+        to set the PCI max read request parameter to 1K. This can be
+        done in the following way:
+
+        To query the read request size use:
+
+        .. code-block:: console
+
+                setpci -s <NIC PCI address> 68.w
+
+        If the output is different than 3XXX, set it by:
+
+        .. code-block:: console
+
+                setpci -s <NIC PCI address> 68.w=3XXX
+
+        The XXX can be different on different systems. Make sure to configure
+        according to the setpci output.
+
 Notes for testpmd
 -----------------
 
-- 
2.12.0
Previous message: [dpdk-dev] [PATCH 2/3] doc: update release notes for mlx driver
Next message: [dpdk-dev] [PATCH 3/3] doc: update mlx guides
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the dev mailing list