| 1 |
.. SPDX-License-Identifier: GPL-2.0+ |
| 2 |
|
| 3 |
============================================================= |
| 4 |
Linux Base Driver for Intel(R) Ethernet Multi-host Controller |
| 5 |
============================================================= |
| 6 |
|
| 7 |
August 20, 2018 |
| 8 |
Copyright(c) 2015-2018 Intel Corporation. |
| 9 |
|
| 10 |
Contents |
| 11 |
======== |
| 12 |
- Identifying Your Adapter |
| 13 |
- Additional Configurations |
| 14 |
- Performance Tuning |
| 15 |
- Known Issues |
| 16 |
- Support |
| 17 |
|
| 18 |
Identifying Your Adapter |
| 19 |
======================== |
| 20 |
The driver in this release is compatible with devices based on the Intel(R) |
| 21 |
Ethernet Multi-host Controller. |
| 22 |
|
| 23 |
For information on how to identify your adapter, and for the latest Intel |
| 24 |
network drivers, refer to the Intel Support website: |
| 25 |
https://www.intel.com/support |
| 26 |
|
| 27 |
|
| 28 |
Flow Control |
| 29 |
------------ |
| 30 |
The Intel(R) Ethernet Switch Host Interface Driver does not support Flow |
| 31 |
Control. It will not send pause frames. This may result in dropped frames. |
| 32 |
|
| 33 |
|
| 34 |
Virtual Functions (VFs) |
| 35 |
----------------------- |
| 36 |
Use sysfs to enable VFs. |
| 37 |
Valid Range: 0-64 |
| 38 |
|
| 39 |
For example:: |
| 40 |
|
| 41 |
echo $num_vf_enabled > /sys/class/net/$dev/device/sriov_numvfs //enable VFs |
| 42 |
echo 0 > /sys/class/net/$dev/device/sriov_numvfs //disable VFs |
| 43 |
|
| 44 |
NOTE: Neither the device nor the driver control how VFs are mapped into config |
| 45 |
space. Bus layout will vary by operating system. On operating systems that |
| 46 |
support it, you can check sysfs to find the mapping. |
| 47 |
|
| 48 |
NOTE: When SR-IOV mode is enabled, hardware VLAN filtering and VLAN tag |
| 49 |
stripping/insertion will remain enabled. Please remove the old VLAN filter |
| 50 |
before the new VLAN filter is added. For example:: |
| 51 |
|
| 52 |
ip link set eth0 vf 0 vlan 100 // set vlan 100 for VF 0 |
| 53 |
ip link set eth0 vf 0 vlan 0 // Delete vlan 100 |
| 54 |
ip link set eth0 vf 0 vlan 200 // set a new vlan 200 for VF 0 |
| 55 |
|
| 56 |
|
| 57 |
Additional Features and Configurations |
| 58 |
====================================== |
| 59 |
|
| 60 |
Jumbo Frames |
| 61 |
------------ |
| 62 |
Jumbo Frames support is enabled by changing the Maximum Transmission Unit (MTU) |
| 63 |
to a value larger than the default value of 1500. |
| 64 |
|
| 65 |
Use the ifconfig command to increase the MTU size. For example, enter the |
| 66 |
following where <x> is the interface number:: |
| 67 |
|
| 68 |
ifconfig eth<x> mtu 9000 up |
| 69 |
|
| 70 |
Alternatively, you can use the ip command as follows:: |
| 71 |
|
| 72 |
ip link set mtu 9000 dev eth<x> |
| 73 |
ip link set up dev eth<x> |
| 74 |
|
| 75 |
This setting is not saved across reboots. The setting change can be made |
| 76 |
permanent by adding 'MTU=9000' to the file: |
| 77 |
|
| 78 |
- For RHEL: /etc/sysconfig/network-scripts/ifcfg-eth<x> |
| 79 |
- For SLES: /etc/sysconfig/network/<config_file> |
| 80 |
|
| 81 |
NOTE: The maximum MTU setting for Jumbo Frames is 15342. This value coincides |
| 82 |
with the maximum Jumbo Frames size of 15364 bytes. |
| 83 |
|
| 84 |
NOTE: This driver will attempt to use multiple page sized buffers to receive |
| 85 |
each jumbo packet. This should help to avoid buffer starvation issues when |
| 86 |
allocating receive packets. |
| 87 |
|
| 88 |
|
| 89 |
Generic Receive Offload, aka GRO |
| 90 |
-------------------------------- |
| 91 |
The driver supports the in-kernel software implementation of GRO. GRO has |
| 92 |
shown that by coalescing Rx traffic into larger chunks of data, CPU |
| 93 |
utilization can be significantly reduced when under large Rx load. GRO is an |
| 94 |
evolution of the previously-used LRO interface. GRO is able to coalesce |
| 95 |
other protocols besides TCP. It's also safe to use with configurations that |
| 96 |
are problematic for LRO, namely bridging and iSCSI. |
| 97 |
|
| 98 |
|
| 99 |
|
| 100 |
Supported ethtool Commands and Options for Filtering |
| 101 |
---------------------------------------------------- |
| 102 |
-n --show-nfc |
| 103 |
Retrieves the receive network flow classification configurations. |
| 104 |
|
| 105 |
rx-flow-hash tcp4|udp4|ah4|esp4|sctp4|tcp6|udp6|ah6|esp6|sctp6 |
| 106 |
Retrieves the hash options for the specified network traffic type. |
| 107 |
|
| 108 |
-N --config-nfc |
| 109 |
Configures the receive network flow classification. |
| 110 |
|
| 111 |
rx-flow-hash tcp4|udp4|ah4|esp4|sctp4|tcp6|udp6|ah6|esp6|sctp6 m|v|t|s|d|f|n|r |
| 112 |
Configures the hash options for the specified network traffic type. |
| 113 |
|
| 114 |
- udp4: UDP over IPv4 |
| 115 |
- udp6: UDP over IPv6 |
| 116 |
- f Hash on bytes 0 and 1 of the Layer 4 header of the rx packet. |
| 117 |
- n Hash on bytes 2 and 3 of the Layer 4 header of the rx packet. |
| 118 |
|
| 119 |
|
| 120 |
Known Issues/Troubleshooting |
| 121 |
============================ |
| 122 |
|
| 123 |
Enabling SR-IOV in a 64-bit Microsoft Windows Server 2012/R2 guest OS under Linux KVM |
| 124 |
------------------------------------------------------------------------------------- |
| 125 |
KVM Hypervisor/VMM supports direct assignment of a PCIe device to a VM. This |
| 126 |
includes traditional PCIe devices, as well as SR-IOV-capable devices based on |
| 127 |
the Intel Ethernet Controller XL710. |
| 128 |
|
| 129 |
|
| 130 |
Support |
| 131 |
======= |
| 132 |
For general information, go to the Intel support website at: |
| 133 |
https://www.intel.com/support/ |
| 134 |
|
| 135 |
If an issue is identified with the released source code on a supported kernel |
| 136 |
with a supported adapter, email the specific information related to the issue |
| 137 |
to [email protected]. |