xref: /dpdk/doc/guides/linux_gsg/sys_reqs.rst (revision 089e5ed727a15da2729cfee9b63533dd120bd04c)
1..  SPDX-License-Identifier: BSD-3-Clause
2    Copyright(c) 2010-2014 Intel Corporation.
3
4System Requirements
5===================
6
7This chapter describes the packages required to compile the DPDK.
8
9.. note::
10
11    If the DPDK is being used on an Intel® Communications Chipset 89xx Series platform,
12    please consult the *Intel® Communications Chipset 89xx Series Software for Linux Getting Started Guide*.
13
14BIOS Setting Prerequisite on x86
15--------------------------------
16
17For the majority of platforms, no special BIOS settings are needed to use basic DPDK functionality.
18However, for additional HPET timer and power management functionality,
19and high performance of small packets, BIOS setting changes may be needed.
20Consult the section on :ref:`Enabling Additional Functionality <Enabling_Additional_Functionality>`
21for more information on the required changes.
22
23.. note::
24
25   If UEFI secure boot is enabled, the Linux kernel may disallow the use of
26   UIO on the system. Therefore, devices for use by DPDK should be bound to the
27   ``vfio-pci`` kernel module rather than ``igb_uio`` or ``uio_pci_generic``.
28   For more details see :ref:`linux_gsg_binding_kernel`.
29
30Compilation of the DPDK
31-----------------------
32
33**Required Tools and Libraries:**
34
35.. note::
36
37    The setup commands and installed packages needed on various systems may be different.
38    For details on Linux distributions and the versions tested, please consult the DPDK Release Notes.
39
40*   GNU ``make``.
41
42*   coreutils: ``cmp``, ``sed``, ``grep``, ``arch``, etc.
43
44*   gcc: versions 4.9 or later is recommended for all platforms.
45    On some distributions, some specific compiler flags and linker flags are enabled by
46    default and affect performance (``-fstack-protector``, for example). Please refer to the documentation
47    of your distribution and to ``gcc -dumpspecs``.
48
49*   libc headers, often packaged as ``gcc-multilib`` (``glibc-devel.i686`` / ``libc6-dev-i386``;
50    ``glibc-devel.x86_64`` / ``libc6-dev`` for 64-bit compilation on Intel architecture;
51    ``glibc-devel.ppc64`` for 64 bit IBM Power architecture;)
52
53*   Linux kernel headers or sources required to build kernel modules. (kernel - devel.x86_64;
54    kernel - devel.ppc64)
55
56*   Additional packages required for 32-bit compilation on 64-bit systems are:
57
58    * glibc.i686, libgcc.i686, libstdc++.i686 and glibc-devel.i686 for Intel i686/x86_64;
59
60    * glibc.ppc64, libgcc.ppc64, libstdc++.ppc64 and glibc-devel.ppc64 for IBM ppc_64;
61
62    .. note::
63
64       x86_x32 ABI is currently supported with distribution packages only on Ubuntu
65       higher than 13.10 or recent Debian distribution. The only supported  compiler is gcc 4.9+.
66
67*   Library for handling NUMA (Non Uniform Memory Access).
68
69    * numactl-devel in Red Hat/Fedora;
70
71    * libnuma-dev in Debian/Ubuntu;
72
73    .. note::
74
75        On systems with NUMA support, `libnuma-dev` (aka `numactl-devel`)
76        is a recommended dependency when `--legacy-mem` switch is used,
77        and a *required* dependency if default memory mode is used.
78        While DPDK will compile and run without `libnuma`
79        even on NUMA-enabled systems,
80        both usability and performance will be degraded.
81
82*   Python, version 2.7+ or 3.2+, to use various helper scripts included in the DPDK package.
83
84.. note::
85
86   Please ensure that the latest patches are applied to third party libraries
87   and software to avoid any known vulnerabilities.
88
89
90**Optional Tools:**
91
92*   Intel® C++ Compiler (icc). For installation, additional libraries may be required.
93    See the icc Installation Guide found in the Documentation directory under the compiler installation.
94
95*   IBM® Advance ToolChain for Powerlinux. This is a set of open source development tools and runtime libraries
96    which allows users to take leading edge advantage of IBM's latest POWER hardware features on Linux. To install
97    it, see the IBM official installation document.
98
99*   libpcap headers and libraries (libpcap-devel) to compile and use the libpcap-based poll-mode driver.
100    This driver is disabled by default and can be enabled by setting ``CONFIG_RTE_LIBRTE_PMD_PCAP=y`` in the build time config file.
101
102*   libarchive headers and library are needed for some unit tests using tar to get their resources.
103
104
105Running DPDK Applications
106-------------------------
107
108To run an DPDK application, some customization may be required on the target machine.
109
110System Software
111~~~~~~~~~~~~~~~
112
113**Required:**
114
115*   Kernel version >= 3.16
116
117    The kernel version required is based on the oldest long term stable kernel available
118    at kernel.org when the DPDK version is in development.
119    Compatibility for recent distribution kernels will be kept, notably RHEL/CentOS 7.
120
121    The kernel version in use can be checked using the command::
122
123        uname -r
124
125*   glibc >= 2.7 (for features related to cpuset)
126
127    The version can be checked using the ``ldd --version`` command.
128
129*   Kernel configuration
130
131    In the Fedora OS and other common distributions, such as Ubuntu, or Red Hat Enterprise Linux,
132    the vendor supplied kernel configurations can be used to run most DPDK applications.
133
134    For other kernel builds, options which should be enabled for DPDK include:
135
136    *   HUGETLBFS
137
138    *   PROC_PAGE_MONITOR  support
139
140    *   HPET and HPET_MMAP configuration options should also be enabled if HPET  support is required.
141        See the section on :ref:`High Precision Event Timer (HPET) Functionality <High_Precision_Event_Timer>` for more details.
142
143.. _linux_gsg_hugepages:
144
145Use of Hugepages in the Linux Environment
146~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
147
148Hugepage support is required for the large memory pool allocation used for packet buffers
149(the HUGETLBFS option must be enabled in the running kernel as indicated the previous section).
150By using hugepage allocations, performance is increased since fewer pages are needed,
151and therefore less Translation Lookaside Buffers (TLBs, high speed translation caches),
152which reduce the time it takes to translate a virtual page address to a physical page address.
153Without hugepages, high TLB miss rates would occur with the standard 4k page size, slowing performance.
154
155Reserving Hugepages for DPDK Use
156^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
157
158The allocation of hugepages should be done at boot time or as soon as possible after system boot
159to prevent memory from being fragmented in physical memory.
160To reserve hugepages at boot time, a parameter is passed to the Linux kernel on the kernel command line.
161
162For 2 MB pages, just pass the hugepages option to the kernel. For example, to reserve 1024 pages of 2 MB, use::
163
164    hugepages=1024
165
166For other hugepage sizes, for example 1G pages, the size must be specified explicitly and
167can also be optionally set as the default hugepage size for the system.
168For example, to reserve 4G of hugepage memory in the form of four 1G pages, the following options should be passed to the kernel::
169
170    default_hugepagesz=1G hugepagesz=1G hugepages=4
171
172.. note::
173
174    The hugepage sizes that a CPU supports can be determined from the CPU flags on Intel architecture.
175    If pse exists, 2M hugepages are supported; if pdpe1gb exists, 1G hugepages are supported.
176    On IBM Power architecture, the supported hugepage sizes are 16MB and 16GB.
177
178.. note::
179
180    For 64-bit applications, it is recommended to use 1 GB hugepages if the platform supports them.
181
182In the case of a dual-socket NUMA system,
183the number of hugepages reserved at boot time is generally divided equally between the two sockets
184(on the assumption that sufficient memory is present on both sockets).
185
186See the Documentation/admin-guide/kernel-parameters.txt file in your Linux source tree for further details of these and other kernel options.
187
188**Alternative:**
189
190For 2 MB pages, there is also the option of allocating hugepages after the system has booted.
191This is done by echoing the number of hugepages required to a nr_hugepages file in the ``/sys/devices/`` directory.
192For a single-node system, the command to use is as follows (assuming that 1024 pages are required)::
193
194    echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
195
196On a NUMA machine, pages should be allocated explicitly on separate nodes::
197
198    echo 1024 > /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
199    echo 1024 > /sys/devices/system/node/node1/hugepages/hugepages-2048kB/nr_hugepages
200
201.. note::
202
203    For 1G pages, it is not possible to reserve the hugepage memory after the system has booted.
204
205Using Hugepages with the DPDK
206^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
207
208Once the hugepage memory is reserved, to make the memory available for DPDK use, perform the following steps::
209
210    mkdir /mnt/huge
211    mount -t hugetlbfs nodev /mnt/huge
212
213The mount point can be made permanent across reboots, by adding the following line to the ``/etc/fstab`` file::
214
215    nodev /mnt/huge hugetlbfs defaults 0 0
216
217For 1GB pages, the page size must be specified as a mount option::
218
219    nodev /mnt/huge_1GB hugetlbfs pagesize=1GB 0 0
220