2 Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
5 Redistribution and use in source and binary forms, with or without
6 modification, are permitted provided that the following conditions
9 * Redistributions of source code must retain the above copyright
10 notice, this list of conditions and the following disclaimer.
11 * Redistributions in binary form must reproduce the above copyright
12 notice, this list of conditions and the following disclaimer in
13 the documentation and/or other materials provided with the
15 * Neither the name of Intel Corporation nor the names of its
16 contributors may be used to endorse or promote products derived
17 from this software without specific prior written permission.
19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
31 .. _Environment_Abstraction_Layer:
33 Environment Abstraction Layer
34 =============================
36 The Environment Abstraction Layer (EAL) is responsible for gaining access to low-level resources such as hardware and memory space.
37 It provides a generic interface that hides the environment specifics from the applications and libraries.
38 It is the responsibility of the initialization routine to decide how to allocate these resources
39 (that is, memory space, PCI devices, timers, consoles, and so on).
41 Typical services expected from the EAL are:
43 * DPDK Loading and Launching:
44 The DPDK and its application are linked as a single application and must be loaded by some means.
46 * Core Affinity/Assignment Procedures:
47 The EAL provides mechanisms for assigning execution units to specific cores as well as creating execution instances.
49 * System Memory Reservation:
50 The EAL facilitates the reservation of different memory zones, for example, physical memory areas for device interactions.
52 * PCI Address Abstraction: The EAL provides an interface to access PCI address space.
54 * Trace and Debug Functions: Logs, dump_stack, panic and so on.
56 * Utility Functions: Spinlocks and atomic counters that are not provided in libc.
58 * CPU Feature Identification: Determine at runtime if a particular feature, for example, IntelĀ® AVX is supported.
59 Determine if the current CPU supports the feature set that the binary was compiled for.
61 * Interrupt Handling: Interfaces to register/unregister callbacks to specific interrupt sources.
63 * Alarm Functions: Interfaces to set/remove callbacks to be run at a specific time.
65 EAL in a Linux-userland Execution Environment
66 ---------------------------------------------
68 In a Linux user space environment, the DPDK application runs as a user-space application using the pthread library.
69 PCI information about devices and address space is discovered through the /sys kernel interface and through kernel modules such as uio_pci_generic, or igb_uio.
70 Refer to the UIO: User-space drivers documentation in the Linux kernel. This memory is mmap'd in the application.
72 The EAL performs physical memory allocation using mmap() in hugetlbfs (using huge page sizes to increase performance).
73 This memory is exposed to DPDK service layers such as the :ref:`Mempool Library <Mempool_Library>`.
75 At this point, the DPDK services layer will be initialized, then through pthread setaffinity calls,
76 each execution unit will be assigned to a specific logical core to run as a user-level thread.
78 The time reference is provided by the CPU Time-Stamp Counter (TSC) or by the HPET kernel API through a mmap() call.
80 Initialization and Core Launching
81 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
83 Part of the initialization is done by the start function of glibc.
84 A check is also performed at initialization time to ensure that the micro architecture type chosen in the config file is supported by the CPU.
85 Then, the main() function is called. The core initialization and launch is done in rte_eal_init() (see the API documentation).
86 It consist of calls to the pthread library (more specifically, pthread_self(), pthread_create(), and pthread_setaffinity_np()).
90 **Figure 2. EAL Initialization in a Linux Application Environment**
92 .. image3_png has been replaced
98 Initialization of objects, such as memory zones, rings, memory pools, lpm tables and hash tables,
99 should be done as part of the overall application initialization on the master lcore.
100 The creation and initialization functions for these objects are not multi-thread safe.
101 However, once initialized, the objects themselves can safely be used in multiple threads simultaneously.
103 Multi-process Support
104 ~~~~~~~~~~~~~~~~~~~~~
106 The Linuxapp EAL allows a multi-process as well as a multi-threaded (pthread) deployment model.
108 :ref:`Multi-process Support <Multi-process_Support>` for more details.
110 Memory Mapping Discovery and Memory Reservation
111 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
113 The allocation of large contiguous physical memory is done using the hugetlbfs kernel filesystem.
114 The EAL provides an API to reserve named memory zones in this contiguous memory.
115 The physical address of the reserved memory for that memory zone is also returned to the user by the memory zone reservation API.
119 Memory reservations done using the APIs provided by the rte_malloc library are also backed by pages from the hugetlbfs filesystem.
120 However, physical address information is not available for the blocks of memory allocated in this way.
122 Xen Dom0 support without hugetbls
123 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
125 The existing memory management implementation is based on the Linux kernel hugepage mechanism.
126 However, Xen Dom0 does not support hugepages, so a new Linux kernel module rte_dom0_mm is added to workaround this limitation.
128 The EAL uses IOCTL interface to notify the Linux kernel module rte_dom0_mm to allocate memory of specified size,
129 and get all memory segments information from the module,
130 and the EAL uses MMAP interface to map the allocated memory.
131 For each memory segment, the physical addresses are contiguous within it but actual hardware addresses are contiguous within 2MB.
136 The EAL uses the /sys/bus/pci utilities provided by the kernel to scan the content on the PCI bus.
137 To access PCI memory, a kernel module called uio_pci_generic provides a /dev/uioX device file
138 and resource files in /sys
139 that can be mmap'd to obtain access to PCI address space from the application.
140 The DPDK-specific igb_uio module can also be used for this. Both drivers use the uio kernel feature (userland driver).
142 Per-lcore and Shared Variables
143 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
147 lcore refers to a logical execution unit of the processor, sometimes called a hardware *thread*.
149 Shared variables are the default behavior.
150 Per-lcore variables are implemented using *Thread Local Storage* (TLS) to provide per-thread local storage.
155 A logging API is provided by EAL.
156 By default, in a Linux application, logs are sent to syslog and also to the console.
157 However, the log function can be overridden by the user to use a different logging mechanism.
159 Trace and Debug Functions
160 ^^^^^^^^^^^^^^^^^^^^^^^^^
162 There are some debug functions to dump the stack in glibc.
163 The rte_panic() function can voluntarily provoke a SIG_ABORT,
164 which can trigger the generation of a core file, readable by gdb.
166 CPU Feature Identification
167 ~~~~~~~~~~~~~~~~~~~~~~~~~~
169 The EAL can query the CPU at runtime (using the rte_cpu_get_feature() function) to determine which CPU features are available.
171 User Space Interrupt and Alarm Handling
172 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
174 The EAL creates a host thread to poll the UIO device file descriptors to detect the interrupts.
175 Callbacks can be registered or unregistered by the EAL functions for a specific interrupt event
176 and are called in the host thread asynchronously.
177 The EAL also allows timed callbacks to be used in the same way as for NIC interrupts.
181 The only interrupts supported by the DPDK Poll-Mode Drivers are those for link status change,
182 i.e. link up and link down notification.
187 The EAL PCI device blacklist functionality can be used to mark certain NIC ports as blacklisted,
188 so they are ignored by the DPDK.
189 The ports to be blacklisted are identified using the PCIe* description (Domain:Bus:Device.Function).
194 Locks and atomic operations are per-architecture (i686 and x86_64).
196 Memory Segments and Memory Zones (memzone)
197 ------------------------------------------
199 The mapping of physical memory is provided by this feature in the EAL.
200 As physical memory can have gaps, the memory is described in a table of descriptors,
201 and each descriptor (called rte_memseg ) describes a contiguous portion of memory.
203 On top of this, the memzone allocator's role is to reserve contiguous portions of physical memory.
204 These zones are identified by a unique name when the memory is reserved.
206 The rte_memzone descriptors are also located in the configuration structure.
207 This structure is accessed using rte_eal_get_configuration().
208 The lookup (by name) of a memory zone returns a descriptor containing the physical address of the memory zone.
210 Memory zones can be reserved with specific start address alignment by supplying the align parameter
211 (by default, they are aligned to cache line size).
212 The alignment value should be a power of two and not less than the cache line size (64 bytes).
213 Memory zones can also be reserved from either 2 MB or 1 GB hugepages, provided that both are available on the system.
219 DPDK usually pins one pthread per core to avoid the overhead of task switching.
220 This allows for significant performance gains, but lacks flexibility and is not always efficient.
222 Power management helps to improve the CPU efficiency by limiting the CPU runtime frequency.
223 However, alternately it is possible to utilize the idle cycles available to take advantage of
224 the full capability of the CPU.
226 By taking advantage of cgroup, the CPU utilization quota can be simply assigned.
227 This gives another way to improve the CPU efficienct, however, there is a prerequisite;
228 DPDK must handle the context switching between multiple pthreads per core.
230 For further flexibility, it is useful to set pthread affinity not only to a CPU but to a CPU set.
232 EAL pthread and lcore Affinity
233 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
235 The term "lcore" refers to an EAL thread, which is really a Linux/FreeBSD pthread.
236 "EAL pthreads" are created and managed by EAL and execute the tasks issued by *remote_launch*.
237 In each EAL pthread, there is a TLS (Thread Local Storage) called *_lcore_id* for unique identification.
238 As EAL pthreads usually bind 1:1 to the physical CPU, the *_lcore_id* is typically equal to the CPU ID.
240 When using multiple pthreads, however, the binding is no longer always 1:1 between an EAL pthread and a specified physical CPU.
241 The EAL pthread may have affinity to a CPU set, and as such the *_lcore_id* will not be the same as the CPU ID.
242 For this reason, there is an EAL long option '--lcores' defined to assign the CPU affinity of lcores.
243 For a specified lcore ID or ID group, the option allows setting the CPU set for that EAL pthread.
246 --lcores='<lcore_set>[@cpu_set][,<lcore_set>[@cpu_set],...]'
248 'lcore_set' and 'cpu_set' can be a single number, range or a group.
250 A number is a "digit([0-9]+)"; a range is "<number>-<number>"; a group is "(<number|range>[,<number|range>,...])".
252 If a '\@cpu_set' value is not supplied, the value of 'cpu_set' will default to the value of 'lcore_set'.
256 For example, "--lcores='1,2@(5-7),(3-5)@(0,2),(0,6),7-8'" which means start 9 EAL thread;
257 lcore 0 runs on cpuset 0x41 (cpu 0,6);
258 lcore 1 runs on cpuset 0x2 (cpu 1);
259 lcore 2 runs on cpuset 0xe0 (cpu 5,6,7);
260 lcore 3,4,5 runs on cpuset 0x5 (cpu 0,2);
261 lcore 6 runs on cpuset 0x41 (cpu 0,6);
262 lcore 7 runs on cpuset 0x80 (cpu 7);
263 lcore 8 runs on cpuset 0x100 (cpu 8).
265 Using this option, for each given lcore ID, the associated CPUs can be assigned.
266 It's also compatible with the pattern of corelist('-l') option.
268 non-EAL pthread support
269 ~~~~~~~~~~~~~~~~~~~~~~~
271 It is possible to use the DPDK execution context with any user pthread (aka. Non-EAL pthreads).
272 In a non-EAL pthread, the *_lcore_id* is always LCORE_ID_ANY which identifies that it is not an EAL thread with a valid, unique, *_lcore_id*.
273 Some libraries will use an alternative unique ID (e.g. TID), some will not be impacted at all, and some will work but with limitations (e.g. timer and mempool libraries).
275 All these impacts are mentioned in :ref:`known_issue_label` section.
280 There are two public APIs ``rte_thread_set_affinity()`` and ``rte_pthread_get_affinity()`` introduced for threads.
281 When they're used in any pthread context, the Thread Local Storage(TLS) will be set/get.
283 Those TLS include *_cpuset* and *_socket_id*:
285 * *_cpuset* stores the CPUs bitmap to which the pthread is affinitized.
287 * *_socket_id* stores the NUMA node of the CPU set. If the CPUs in CPU set belong to different NUMA node, the *_socket_id* will be set to SOCKTE_ID_ANY.
290 .. _known_issue_label:
297 The rte_mempool uses a per-lcore cache inside the mempool.
298 For non-EAL pthreads, ``rte_lcore_id()`` will not return a valid number.
299 So for now, when rte_mempool is used with non-EAL pthreads, the put/get operations will bypass the mempool cache and there is a performance penalty because of this bypass.
300 Support for non-EAL mempool cache is currently being enabled.
304 rte_ring supports multi-producer enqueue and multi-consumer dequeue.
305 However, it is non-preemptive, this has a knock on effect of making rte_mempool non-preemtable.
309 The "non-preemptive" constraint means:
311 - a pthread doing multi-producers enqueues on a given ring must not
312 be preempted by another pthread doing a multi-producer enqueue on
314 - a pthread doing multi-consumers dequeues on a given ring must not
315 be preempted by another pthread doing a multi-consumer dequeue on
318 Bypassing this constraint it may cause the 2nd pthread to spin until the 1st one is scheduled again.
319 Moreover, if the 1st pthread is preempted by a context that has an higher priority, it may even cause a dead lock.
321 This does not mean it cannot be used, simply, there is a need to narrow down the situation when it is used by multi-pthread on the same core.
323 1. It CAN be used for any single-producer or single-consumer situation.
325 2. It MAY be used by multi-producer/consumer pthread whose scheduling policy are all SCHED_OTHER(cfs). User SHOULD be aware of the performance penalty before using it.
327 3. It MUST not be used by multi-producer/consumer pthreads, whose scheduling policies are SCHED_FIFO or SCHED_RR.
329 ``RTE_RING_PAUSE_REP_COUNT`` is defined for rte_ring to reduce contention. It's mainly for case 2, a yield is issued after number of times pause repeat.
331 It adds a sched_yield() syscall if the thread spins for too long while waiting on the other thread to finish its operations on the ring.
332 This gives the pre-empted thread a chance to proceed and finish with the ring enqueue/dequeue operation.
336 Running ``rte_timer_manager()`` on a non-EAL pthread is not allowed. However, resetting/stopping the timer from a non-EAL pthread is allowed.
340 In non-EAL pthreads, there is no per thread loglevel and logtype, global loglevels are used.
344 The debug statistics of rte_ring, rte_mempool and rte_timer are not supported in a non-EAL pthread.
349 The following is a simple example of cgroup control usage, there are two pthreads(t0 and t1) doing packet I/O on the same core ($CPU).
350 We expect only 50% of CPU spend on packet IO.
352 .. code-block:: console
354 mkdir /sys/fs/cgroup/cpu/pkt_io
355 mkdir /sys/fs/cgroup/cpuset/pkt_io
357 echo $cpu > /sys/fs/cgroup/cpuset/cpuset.cpus
359 echo $t0 > /sys/fs/cgroup/cpu/pkt_io/tasks
360 echo $t0 > /sys/fs/cgroup/cpuset/pkt_io/tasks
362 echo $t1 > /sys/fs/cgroup/cpu/pkt_io/tasks
363 echo $t1 > /sys/fs/cgroup/cpuset/pkt_io/tasks
365 cd /sys/fs/cgroup/cpu/pkt_io
366 echo 100000 > pkt_io/cpu.cfs_period_us
367 echo 50000 > pkt_io/cpu.cfs_quota_us
370 .. |linuxapp_launch| image:: img/linuxapp_launch.*