1.. BSD LICENSE 2 Copyright(c) 2010-2014 Intel Corporation. All rights reserved. 3 All rights reserved. 4 5 Redistribution and use in source and binary forms, with or without 6 modification, are permitted provided that the following conditions 7 are met: 8 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 14 distribution. 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. 18 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. 30 31.. _Environment_Abstraction_Layer: 32 33Environment Abstraction Layer 34============================= 35 36The Environment Abstraction Layer (EAL) is responsible for gaining access to low-level resources such as hardware and memory space. 37It provides a generic interface that hides the environment specifics from the applications and libraries. 38It 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). 40 41Typical services expected from the EAL are: 42 43* DPDK Loading and Launching: 44 The DPDK and its application are linked as a single application and must be loaded by some means. 45 46* Core Affinity/Assignment Procedures: 47 The EAL provides mechanisms for assigning execution units to specific cores as well as creating execution instances. 48 49* System Memory Reservation: 50 The EAL facilitates the reservation of different memory zones, for example, physical memory areas for device interactions. 51 52* PCI Address Abstraction: The EAL provides an interface to access PCI address space. 53 54* Trace and Debug Functions: Logs, dump_stack, panic and so on. 55 56* Utility Functions: Spinlocks and atomic counters that are not provided in libc. 57 58* CPU Feature Identification: Determine at runtime if a particular feature, for example, Intel® AVX is supported. 59 Determine if the current CPU supports the feature set that the binary was compiled for. 60 61* Interrupt Handling: Interfaces to register/unregister callbacks to specific interrupt sources. 62 63* Alarm Functions: Interfaces to set/remove callbacks to be run at a specific time. 64 65EAL in a Linux-userland Execution Environment 66--------------------------------------------- 67 68In a Linux user space environment, the DPDK application runs as a user-space application using the pthread library. 69PCI 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. 70Refer to the UIO: User-space drivers documentation in the Linux kernel. This memory is mmap'd in the application. 71 72The EAL performs physical memory allocation using mmap() in hugetlbfs (using huge page sizes to increase performance). 73This memory is exposed to DPDK service layers such as the :ref:`Mempool Library <Mempool_Library>`. 74 75At this point, the DPDK services layer will be initialized, then through pthread setaffinity calls, 76each execution unit will be assigned to a specific logical core to run as a user-level thread. 77 78The time reference is provided by the CPU Time-Stamp Counter (TSC) or by the HPET kernel API through a mmap() call. 79 80Initialization and Core Launching 81~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 82 83Part of the initialization is done by the start function of glibc. 84A check is also performed at initialization time to ensure that the micro architecture type chosen in the config file is supported by the CPU. 85Then, the main() function is called. The core initialization and launch is done in rte_eal_init() (see the API documentation). 86It consist of calls to the pthread library (more specifically, pthread_self(), pthread_create(), and pthread_setaffinity_np()). 87 88.. _figure_linuxapp_launch: 89 90.. figure:: img/linuxapp_launch.* 91 92 EAL Initialization in a Linux Application Environment 93 94 95.. note:: 96 97 Initialization of objects, such as memory zones, rings, memory pools, lpm tables and hash tables, 98 should be done as part of the overall application initialization on the master lcore. 99 The creation and initialization functions for these objects are not multi-thread safe. 100 However, once initialized, the objects themselves can safely be used in multiple threads simultaneously. 101 102Multi-process Support 103~~~~~~~~~~~~~~~~~~~~~ 104 105The Linuxapp EAL allows a multi-process as well as a multi-threaded (pthread) deployment model. 106See chapter 2.20 107:ref:`Multi-process Support <Multi-process_Support>` for more details. 108 109Memory Mapping Discovery and Memory Reservation 110~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 111 112The allocation of large contiguous physical memory is done using the hugetlbfs kernel filesystem. 113The EAL provides an API to reserve named memory zones in this contiguous memory. 114The physical address of the reserved memory for that memory zone is also returned to the user by the memory zone reservation API. 115 116.. note:: 117 118 Memory reservations done using the APIs provided by the rte_malloc library are also backed by pages from the hugetlbfs filesystem. 119 However, physical address information is not available for the blocks of memory allocated in this way. 120 121Xen Dom0 support without hugetbls 122~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 123 124The existing memory management implementation is based on the Linux kernel hugepage mechanism. 125However, Xen Dom0 does not support hugepages, so a new Linux kernel module rte_dom0_mm is added to workaround this limitation. 126 127The EAL uses IOCTL interface to notify the Linux kernel module rte_dom0_mm to allocate memory of specified size, 128and get all memory segments information from the module, 129and the EAL uses MMAP interface to map the allocated memory. 130For each memory segment, the physical addresses are contiguous within it but actual hardware addresses are contiguous within 2MB. 131 132PCI Access 133~~~~~~~~~~ 134 135The EAL uses the /sys/bus/pci utilities provided by the kernel to scan the content on the PCI bus. 136To access PCI memory, a kernel module called uio_pci_generic provides a /dev/uioX device file 137and resource files in /sys 138that can be mmap'd to obtain access to PCI address space from the application. 139The DPDK-specific igb_uio module can also be used for this. Both drivers use the uio kernel feature (userland driver). 140 141Per-lcore and Shared Variables 142~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 143 144.. note:: 145 146 lcore refers to a logical execution unit of the processor, sometimes called a hardware *thread*. 147 148Shared variables are the default behavior. 149Per-lcore variables are implemented using *Thread Local Storage* (TLS) to provide per-thread local storage. 150 151Logs 152~~~~ 153 154A logging API is provided by EAL. 155By default, in a Linux application, logs are sent to syslog and also to the console. 156However, the log function can be overridden by the user to use a different logging mechanism. 157 158Trace and Debug Functions 159^^^^^^^^^^^^^^^^^^^^^^^^^ 160 161There are some debug functions to dump the stack in glibc. 162The rte_panic() function can voluntarily provoke a SIG_ABORT, 163which can trigger the generation of a core file, readable by gdb. 164 165CPU Feature Identification 166~~~~~~~~~~~~~~~~~~~~~~~~~~ 167 168The EAL can query the CPU at runtime (using the rte_cpu_get_feature() function) to determine which CPU features are available. 169 170User Space Interrupt and Alarm Handling 171~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 172 173The EAL creates a host thread to poll the UIO device file descriptors to detect the interrupts. 174Callbacks can be registered or unregistered by the EAL functions for a specific interrupt event 175and are called in the host thread asynchronously. 176The EAL also allows timed callbacks to be used in the same way as for NIC interrupts. 177 178.. note:: 179 180 The only interrupts supported by the DPDK Poll-Mode Drivers are those for link status change, 181 i.e. link up and link down notification. 182 183Blacklisting 184~~~~~~~~~~~~ 185 186The EAL PCI device blacklist functionality can be used to mark certain NIC ports as blacklisted, 187so they are ignored by the DPDK. 188The ports to be blacklisted are identified using the PCIe* description (Domain:Bus:Device.Function). 189 190Misc Functions 191~~~~~~~~~~~~~~ 192 193Locks and atomic operations are per-architecture (i686 and x86_64). 194 195Memory Segments and Memory Zones (memzone) 196------------------------------------------ 197 198The mapping of physical memory is provided by this feature in the EAL. 199As physical memory can have gaps, the memory is described in a table of descriptors, 200and each descriptor (called rte_memseg ) describes a contiguous portion of memory. 201 202On top of this, the memzone allocator's role is to reserve contiguous portions of physical memory. 203These zones are identified by a unique name when the memory is reserved. 204 205The rte_memzone descriptors are also located in the configuration structure. 206This structure is accessed using rte_eal_get_configuration(). 207The lookup (by name) of a memory zone returns a descriptor containing the physical address of the memory zone. 208 209Memory zones can be reserved with specific start address alignment by supplying the align parameter 210(by default, they are aligned to cache line size). 211The alignment value should be a power of two and not less than the cache line size (64 bytes). 212Memory zones can also be reserved from either 2 MB or 1 GB hugepages, provided that both are available on the system. 213 214 215Multiple pthread 216---------------- 217 218DPDK usually pins one pthread per core to avoid the overhead of task switching. 219This allows for significant performance gains, but lacks flexibility and is not always efficient. 220 221Power management helps to improve the CPU efficiency by limiting the CPU runtime frequency. 222However, alternately it is possible to utilize the idle cycles available to take advantage of 223the full capability of the CPU. 224 225By taking advantage of cgroup, the CPU utilization quota can be simply assigned. 226This gives another way to improve the CPU efficiency, however, there is a prerequisite; 227DPDK must handle the context switching between multiple pthreads per core. 228 229For further flexibility, it is useful to set pthread affinity not only to a CPU but to a CPU set. 230 231EAL pthread and lcore Affinity 232~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 233 234The term "lcore" refers to an EAL thread, which is really a Linux/FreeBSD pthread. 235"EAL pthreads" are created and managed by EAL and execute the tasks issued by *remote_launch*. 236In each EAL pthread, there is a TLS (Thread Local Storage) called *_lcore_id* for unique identification. 237As EAL pthreads usually bind 1:1 to the physical CPU, the *_lcore_id* is typically equal to the CPU ID. 238 239When using multiple pthreads, however, the binding is no longer always 1:1 between an EAL pthread and a specified physical CPU. 240The EAL pthread may have affinity to a CPU set, and as such the *_lcore_id* will not be the same as the CPU ID. 241For this reason, there is an EAL long option '--lcores' defined to assign the CPU affinity of lcores. 242For a specified lcore ID or ID group, the option allows setting the CPU set for that EAL pthread. 243 244The format pattern: 245 --lcores='<lcore_set>[@cpu_set][,<lcore_set>[@cpu_set],...]' 246 247'lcore_set' and 'cpu_set' can be a single number, range or a group. 248 249A number is a "digit([0-9]+)"; a range is "<number>-<number>"; a group is "(<number|range>[,<number|range>,...])". 250 251If a '\@cpu_set' value is not supplied, the value of 'cpu_set' will default to the value of 'lcore_set'. 252 253 :: 254 255 For example, "--lcores='1,2@(5-7),(3-5)@(0,2),(0,6),7-8'" which means start 9 EAL thread; 256 lcore 0 runs on cpuset 0x41 (cpu 0,6); 257 lcore 1 runs on cpuset 0x2 (cpu 1); 258 lcore 2 runs on cpuset 0xe0 (cpu 5,6,7); 259 lcore 3,4,5 runs on cpuset 0x5 (cpu 0,2); 260 lcore 6 runs on cpuset 0x41 (cpu 0,6); 261 lcore 7 runs on cpuset 0x80 (cpu 7); 262 lcore 8 runs on cpuset 0x100 (cpu 8). 263 264Using this option, for each given lcore ID, the associated CPUs can be assigned. 265It's also compatible with the pattern of corelist('-l') option. 266 267non-EAL pthread support 268~~~~~~~~~~~~~~~~~~~~~~~ 269 270It is possible to use the DPDK execution context with any user pthread (aka. Non-EAL pthreads). 271In 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*. 272Some 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). 273 274All these impacts are mentioned in :ref:`known_issue_label` section. 275 276Public Thread API 277~~~~~~~~~~~~~~~~~ 278 279There are two public APIs ``rte_thread_set_affinity()`` and ``rte_pthread_get_affinity()`` introduced for threads. 280When they're used in any pthread context, the Thread Local Storage(TLS) will be set/get. 281 282Those TLS include *_cpuset* and *_socket_id*: 283 284* *_cpuset* stores the CPUs bitmap to which the pthread is affinitized. 285 286* *_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 SOCKET_ID_ANY. 287 288 289.. _known_issue_label: 290 291Known Issues 292~~~~~~~~~~~~ 293 294+ rte_mempool 295 296 The rte_mempool uses a per-lcore cache inside the mempool. 297 For non-EAL pthreads, ``rte_lcore_id()`` will not return a valid number. 298 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. 299 Support for non-EAL mempool cache is currently being enabled. 300 301+ rte_ring 302 303 rte_ring supports multi-producer enqueue and multi-consumer dequeue. 304 However, it is non-preemptive, this has a knock on effect of making rte_mempool non-preemptable. 305 306 .. note:: 307 308 The "non-preemptive" constraint means: 309 310 - a pthread doing multi-producers enqueues on a given ring must not 311 be preempted by another pthread doing a multi-producer enqueue on 312 the same ring. 313 - a pthread doing multi-consumers dequeues on a given ring must not 314 be preempted by another pthread doing a multi-consumer dequeue on 315 the same ring. 316 317 Bypassing this constraint it may cause the 2nd pthread to spin until the 1st one is scheduled again. 318 Moreover, if the 1st pthread is preempted by a context that has an higher priority, it may even cause a dead lock. 319 320 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. 321 322 1. It CAN be used for any single-producer or single-consumer situation. 323 324 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. 325 326 3. It MUST not be used by multi-producer/consumer pthreads, whose scheduling policies are SCHED_FIFO or SCHED_RR. 327 328 ``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. 329 330 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. 331 This gives the preempted thread a chance to proceed and finish with the ring enqueue/dequeue operation. 332 333+ rte_timer 334 335 Running ``rte_timer_manager()`` on a non-EAL pthread is not allowed. However, resetting/stopping the timer from a non-EAL pthread is allowed. 336 337+ rte_log 338 339 In non-EAL pthreads, there is no per thread loglevel and logtype, global loglevels are used. 340 341+ misc 342 343 The debug statistics of rte_ring, rte_mempool and rte_timer are not supported in a non-EAL pthread. 344 345cgroup control 346~~~~~~~~~~~~~~ 347 348The 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). 349We expect only 50% of CPU spend on packet IO. 350 351 .. code-block:: console 352 353 mkdir /sys/fs/cgroup/cpu/pkt_io 354 mkdir /sys/fs/cgroup/cpuset/pkt_io 355 356 echo $cpu > /sys/fs/cgroup/cpuset/cpuset.cpus 357 358 echo $t0 > /sys/fs/cgroup/cpu/pkt_io/tasks 359 echo $t0 > /sys/fs/cgroup/cpuset/pkt_io/tasks 360 361 echo $t1 > /sys/fs/cgroup/cpu/pkt_io/tasks 362 echo $t1 > /sys/fs/cgroup/cpuset/pkt_io/tasks 363 364 cd /sys/fs/cgroup/cpu/pkt_io 365 echo 100000 > pkt_io/cpu.cfs_period_us 366 echo 50000 > pkt_io/cpu.cfs_quota_us 367 368 369