1.. SPDX-License-Identifier: BSD-3-Clause 2 Copyright(c) 2010-2014 Intel Corporation. 3 4Environment Abstraction Layer (EAL) Library 5=========================================== 6 7The Environment Abstraction Layer (EAL) is responsible for gaining access to low-level resources such as hardware and memory space. 8It provides a generic interface that hides the environment specifics from the applications and libraries. 9It is the responsibility of the initialization routine to decide how to allocate these resources 10(that is, memory space, devices, timers, consoles, and so on). 11 12Typical services expected from the EAL are: 13 14* DPDK Loading and Launching: 15 The DPDK and its application are linked as a single application and must be loaded by some means. 16 17* Core Affinity/Assignment Procedures: 18 The EAL provides mechanisms for assigning execution units to specific cores as well as creating execution instances. 19 20* System Memory Reservation: 21 The EAL facilitates the reservation of different memory zones, for example, physical memory areas for device interactions. 22 23* Trace and Debug Functions: Logs, dump_stack, panic and so on. 24 25* Utility Functions: Spinlocks and atomic counters that are not provided in libc. 26 27* CPU Feature Identification: Determine at runtime if a particular feature, for example, Intel® AVX is supported. 28 Determine if the current CPU supports the feature set that the binary was compiled for. 29 30* Interrupt Handling: Interfaces to register/unregister callbacks to specific interrupt sources. 31 32* Alarm Functions: Interfaces to set/remove callbacks to be run at a specific time. 33 34EAL in a Linux-userland Execution Environment 35--------------------------------------------- 36 37In a Linux user space environment, the DPDK application runs as a user-space application using the pthread library. 38 39The EAL performs physical memory allocation using mmap() in hugetlbfs (using huge page sizes to increase performance). 40This memory is exposed to DPDK service layers such as the :doc:`mempool_lib`. 41 42At this point, the DPDK services layer will be initialized, then through pthread setaffinity calls, 43each execution unit will be assigned to a specific logical core to run as a user-level thread. 44 45The time reference is provided by the CPU Time-Stamp Counter (TSC) or by the HPET kernel API through a mmap() call. 46 47Initialization and Core Launching 48~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 49 50Part of the initialization is done by the start function of glibc. 51A check is also performed at initialization time to ensure that the micro architecture type chosen in the config file is supported by the CPU. 52Then, the main() function is called. The core initialization and launch is done in rte_eal_init() (see the API documentation). 53It consist of calls to the pthread library (more specifically, pthread_self(), pthread_create(), and pthread_setaffinity_np()). 54 55.. _figure_linux_launch: 56 57.. figure:: img/linuxapp_launch.* 58 59 EAL Initialization in a Linux Application Environment 60 61 62.. note:: 63 64 Initialization of objects, such as memory zones, rings, memory pools, lpm tables and hash tables, 65 should be done as part of the overall application initialization on the main lcore. 66 The creation and initialization functions for these objects are not multi-thread safe. 67 However, once initialized, the objects themselves can safely be used in multiple threads simultaneously. 68 69Shutdown and Cleanup 70~~~~~~~~~~~~~~~~~~~~ 71 72During the initialization of EAL resources such as hugepage backed memory can be 73allocated by core components. The memory allocated during ``rte_eal_init()`` 74can be released by calling the ``rte_eal_cleanup()`` function. Refer to the 75API documentation for details. 76 77Multi-process Support 78~~~~~~~~~~~~~~~~~~~~~ 79 80The Linux EAL allows a multi-process as well as a multi-threaded (pthread) deployment model. 81See chapter :doc:`multi_proc_support` for more details. 82 83Memory Mapping Discovery and Memory Reservation 84~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 85 86The allocation of large contiguous physical memory is done using hugepages. 87The EAL provides an API to reserve named memory zones in this contiguous memory. 88The physical address of the reserved memory for that memory zone is also returned to the user by the memory zone reservation API. 89 90There are two modes in which DPDK memory subsystem can operate: dynamic mode, 91and legacy mode. Both modes are explained below. 92 93.. note:: 94 95 Memory reservations done using the APIs provided by rte_malloc 96 are also backed by hugepages unless ``--no-huge`` option is given. 97 98Dynamic Memory Mode 99^^^^^^^^^^^^^^^^^^^ 100 101Currently, this mode is only supported on Linux and Windows. 102 103In this mode, usage of hugepages by DPDK application will grow and shrink based 104on application's requests. Any memory allocation through ``rte_malloc()``, 105``rte_memzone_reserve()`` or other methods, can potentially result in more 106hugepages being reserved from the system. Similarly, any memory deallocation can 107potentially result in hugepages being released back to the system. 108 109Memory allocated in this mode is not guaranteed to be IOVA-contiguous. If large 110chunks of IOVA-contiguous are required (with "large" defined as "more than one 111page"), it is recommended to either use VFIO driver for all physical devices (so 112that IOVA and VA addresses can be the same, thereby bypassing physical addresses 113entirely), or use legacy memory mode. 114 115For chunks of memory which must be IOVA-contiguous, it is recommended to use 116``rte_memzone_reserve()`` function with ``RTE_MEMZONE_IOVA_CONTIG`` flag 117specified. This way, memory allocator will ensure that, whatever memory mode is 118in use, either reserved memory will satisfy the requirements, or the allocation 119will fail. 120 121There is no need to preallocate any memory at startup using ``-m`` or 122``--socket-mem`` command-line parameters, however it is still possible to do so, 123in which case preallocate memory will be "pinned" (i.e. will never be released 124by the application back to the system). It will be possible to allocate more 125hugepages, and deallocate those, but any preallocated pages will not be freed. 126If neither ``-m`` nor ``--socket-mem`` were specified, no memory will be 127preallocated, and all memory will be allocated at runtime, as needed. 128 129Another available option to use in dynamic memory mode is 130``--single-file-segments`` command-line option. This option will put pages in 131single files (per memseg list), as opposed to creating a file per page. This is 132normally not needed, but can be useful for use cases like userspace vhost, where 133there is limited number of page file descriptors that can be passed to VirtIO. 134 135If the application (or DPDK-internal code, such as device drivers) wishes to 136receive notifications about newly allocated memory, it is possible to register 137for memory event callbacks via ``rte_mem_event_callback_register()`` function. 138This will call a callback function any time DPDK's memory map has changed. 139 140If the application (or DPDK-internal code, such as device drivers) wishes to be 141notified about memory allocations above specified threshold (and have a chance 142to deny them), allocation validator callbacks are also available via 143``rte_mem_alloc_validator_callback_register()`` function. 144 145A default validator callback is provided by EAL, which can be enabled with a 146``--socket-limit`` command-line option, for a simple way to limit maximum amount 147of memory that can be used by DPDK application. 148 149.. warning:: 150 Memory subsystem uses DPDK IPC internally, so memory allocations/callbacks 151 and IPC must not be mixed: it is not safe to allocate/free memory inside 152 memory-related or IPC callbacks, and it is not safe to use IPC inside 153 memory-related callbacks. See chapter 154 :doc:`multi_proc_support` for more details about DPDK IPC. 155 156Legacy Memory Mode 157^^^^^^^^^^^^^^^^^^ 158 159This mode is enabled by specifying ``--legacy-mem`` command-line switch to the 160EAL. This switch will have no effect on FreeBSD as FreeBSD only supports 161legacy mode anyway. 162 163This mode mimics historical behavior of EAL. That is, EAL will reserve all 164memory at startup, sort all memory into large IOVA-contiguous chunks, and will 165not allow acquiring or releasing hugepages from the system at runtime. 166 167If neither ``-m`` nor ``--socket-mem`` were specified, the entire available 168hugepage memory will be preallocated. 169 170Hugepage Allocation Matching 171^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 172 173This behavior is enabled by specifying the ``--match-allocations`` command-line 174switch to the EAL. This switch is Linux-only and not supported with 175``--legacy-mem`` nor ``--no-huge``. 176 177Some applications using memory event callbacks may require that hugepages be 178freed exactly as they were allocated. These applications may also require 179that any allocation from the malloc heap not span across allocations 180associated with two different memory event callbacks. Hugepage allocation 181matching can be used by these types of applications to satisfy both of these 182requirements. This can result in some increased memory usage which is 183very dependent on the memory allocation patterns of the application. 184 18532-bit Support 186^^^^^^^^^^^^^^ 187 188Additional restrictions are present when running in 32-bit mode. In dynamic 189memory mode, by default maximum of 2 gigabytes of VA space will be preallocated, 190and all of it will be on main lcore NUMA node unless ``--socket-mem`` flag is 191used. 192 193In legacy mode, VA space will only be preallocated for segments that were 194requested (plus padding, to keep IOVA-contiguousness). 195 196Maximum Amount of Memory 197^^^^^^^^^^^^^^^^^^^^^^^^ 198 199All possible virtual memory space that can ever be used for hugepage mapping in 200a DPDK process is preallocated at startup, thereby placing an upper limit on how 201much memory a DPDK application can have. DPDK memory is stored in segment lists, 202each segment is strictly one physical page. It is possible to change the amount 203of virtual memory being preallocated at startup by editing the following config 204variables: 205 206* ``RTE_MAX_MEMSEG_LISTS`` controls how many segment lists can DPDK have 207* ``RTE_MAX_MEM_MB_PER_LIST`` controls how much megabytes of memory each 208 segment list can address 209* ``RTE_MAX_MEMSEG_PER_LIST`` controls how many segments each segment list 210 can have 211* ``RTE_MAX_MEMSEG_PER_TYPE`` controls how many segments each memory type 212 can have (where "type" is defined as "page size + NUMA node" combination) 213* ``RTE_MAX_MEM_MB_PER_TYPE`` controls how much megabytes of memory each 214 memory type can address 215* ``RTE_MAX_MEM_MB`` places a global maximum on the amount of memory 216 DPDK can reserve 217 218Normally, these options do not need to be changed. 219 220.. note:: 221 222 Preallocated virtual memory is not to be confused with preallocated hugepage 223 memory! All DPDK processes preallocate virtual memory at startup. Hugepages 224 can later be mapped into that preallocated VA space (if dynamic memory mode 225 is enabled), and can optionally be mapped into it at startup. 226 227.. _hugepage_mapping: 228 229Hugepage Mapping 230^^^^^^^^^^^^^^^^ 231 232Below is an overview of methods used for each OS to obtain hugepages, 233explaining why certain limitations and options exist in EAL. 234See the user guide for a specific OS for configuration details. 235 236FreeBSD uses ``contigmem`` kernel module 237to reserve a fixed number of hugepages at system start, 238which are mapped by EAL at initialization using a specific ``sysctl()``. 239 240Windows EAL allocates hugepages from the OS as needed using Win32 API, 241so available amount depends on the system load. 242It uses ``virt2phys`` kernel module to obtain physical addresses, 243unless running in IOVA-as-VA mode (e.g. forced with ``--iova-mode=va``). 244 245Linux allows to select any combination of the following: 246 247* use files in hugetlbfs (the default) 248 or anonymous mappings (``--in-memory``); 249* map each hugepage from its own file (the default) 250 or map multiple hugepages from one big file (``--single-file-segments``). 251 252Mapping hugepages from files in hugetlbfs is essential for multi-process, 253because secondary processes need to map the same hugepages. 254EAL creates files like ``rtemap_0`` 255in directories specified with ``--huge-dir`` option 256(or in the mount point for a specific hugepage size). 257The ``rte`` prefix can be changed using ``--file-prefix``. 258This may be needed for running multiple primary processes 259that share a hugetlbfs mount point. 260Each backing file by default corresponds to one hugepage, 261it is opened and locked for the entire time the hugepage is used. 262This may exhaust the number of open files limit (``NOFILE``). 263See :ref:`segment-file-descriptors` section 264on how the number of open backing file descriptors can be reduced. 265 266In dynamic memory mode, EAL removes a backing hugepage file 267when all pages mapped from it are freed back to the system. 268However, backing files may persist after the application terminates 269in case of a crash or a leak of DPDK memory (e.g. ``rte_free()`` is missing). 270This reduces the number of hugepages available to other processes 271as reported by ``/sys/kernel/mm/hugepages/hugepages-*/free_hugepages``. 272EAL can remove the backing files after opening them for mapping 273if ``--huge-unlink`` is given to avoid polluting hugetlbfs. 274However, since it disables multi-process anyway, 275using anonymous mapping (``--in-memory``) is recommended instead. 276 277:ref:`EAL memory allocator <malloc>` relies on hugepages being zero-filled. 278Hugepages are cleared by the kernel when a file in hugetlbfs or its part 279is mapped for the first time system-wide 280to prevent data leaks from previous users of the same hugepage. 281EAL ensures this behavior by removing existing backing files at startup 282and by recreating them before opening for mapping (as a precaution). 283 284One exception is ``--huge-unlink=never`` mode. 285It is used to speed up EAL initialization, usually on application restart. 286Clearing memory constitutes more than 95% of hugepage mapping time. 287EAL can save it by remapping existing backing files 288with all the data left in the mapped hugepages ("dirty" memory). 289Such segments are marked with ``RTE_MEMSEG_FLAG_DIRTY``. 290Memory allocator detects dirty segments and handles them accordingly, 291in particular, it clears memory requested with ``rte_zmalloc*()``. 292In this mode EAL also does not remove a backing file 293when all pages mapped from it are freed, 294because they are intended to be reusable at restart. 295 296Anonymous mapping does not allow multi-process architecture. 297This mode does not use hugetlbfs 298and thus does not require root permissions for memory management 299(the limit of locked memory amount, ``MEMLOCK``, still applies). 300It is free of filename conflict and leftover file issues. 301If ``memfd_create(2)`` is supported both at build and run time, 302DPDK memory manager can provide file descriptors for memory segments, 303which are required for VirtIO with vhost-user backend. 304This can exhaust the number of open files limit (``NOFILE``) 305despite not creating any visible files. 306See :ref:`segment-file-descriptors` section 307on how the number of open file descriptors used by EAL can be reduced. 308 309.. _segment-file-descriptors: 310 311Segment File Descriptors 312^^^^^^^^^^^^^^^^^^^^^^^^ 313 314On Linux, in most cases, EAL will store segment file descriptors in EAL. This 315can become a problem when using smaller page sizes due to underlying limitations 316of ``glibc`` library. For example, Linux API calls such as ``select()`` may not 317work correctly because ``glibc`` does not support more than certain number of 318file descriptors. 319 320There are two possible solutions for this problem. The recommended solution is 321to use ``--single-file-segments`` mode, as that mode will not use a file 322descriptor per each page, and it will keep compatibility with Virtio with 323vhost-user backend. This option is not available when using ``--legacy-mem`` 324mode. 325 326Another option is to use bigger page sizes. Since fewer pages are required to 327cover the same memory area, fewer file descriptors will be stored internally 328by EAL. 329 330Hugepage Worker Stacks 331^^^^^^^^^^^^^^^^^^^^^^ 332 333When the ``--huge-worker-stack[=size]`` EAL option is specified, worker 334thread stacks are allocated from hugepage memory local to the NUMA node 335of the thread. Worker stack size defaults to system pthread stack size 336if the optional size parameter is not specified. 337 338.. warning:: 339 Stacks allocated from hugepage memory are not protected by guard 340 pages. Worker stacks must be sufficiently sized to prevent stack 341 overflow when this option is used. 342 343 As with normal thread stacks, hugepage worker thread stack size is 344 fixed and is not dynamically resized. Therefore, an application that 345 is free of stack page faults under a given load should be safe with 346 hugepage worker thread stacks given the same thread stack size and 347 loading conditions. 348 349Support for Externally Allocated Memory 350~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 351 352It is possible to use externally allocated memory in DPDK. There are two ways in 353which using externally allocated memory can work: the malloc heap API's, and 354manual memory management. 355 356+ Using heap API's for externally allocated memory 357 358Using a set of malloc heap API's is the recommended way to use externally 359allocated memory in DPDK. In this way, support for externally allocated memory 360is implemented through overloading the socket ID - externally allocated heaps 361will have socket ID's that would be considered invalid under normal 362circumstances. Requesting an allocation to take place from a specified 363externally allocated memory is a matter of supplying the correct socket ID to 364DPDK allocator, either directly (e.g. through a call to ``rte_malloc``) or 365indirectly (through data structure-specific allocation API's such as 366``rte_ring_create``). Using these API's also ensures that mapping of externally 367allocated memory for DMA is also performed on any memory segment that is added 368to a DPDK malloc heap. 369 370Since there is no way DPDK can verify whether memory is available or valid, this 371responsibility falls on the shoulders of the user. All multiprocess 372synchronization is also user's responsibility, as well as ensuring that all 373calls to add/attach/detach/remove memory are done in the correct order. It is 374not required to attach to a memory area in all processes - only attach to memory 375areas as needed. 376 377The expected workflow is as follows: 378 379* Get a pointer to memory area 380* Create a named heap 381* Add memory area(s) to the heap 382 - If IOVA table is not specified, IOVA addresses will be assumed to be 383 unavailable, and DMA mappings will not be performed 384 - Other processes must attach to the memory area before they can use it 385* Get socket ID used for the heap 386* Use normal DPDK allocation procedures, using supplied socket ID 387* If memory area is no longer needed, it can be removed from the heap 388 - Other processes must detach from this memory area before it can be removed 389* If heap is no longer needed, remove it 390 - Socket ID will become invalid and will not be reused 391 392For more information, please refer to ``rte_malloc`` API documentation, 393specifically the ``rte_malloc_heap_*`` family of function calls. 394 395+ Using externally allocated memory without DPDK API's 396 397While using heap API's is the recommended method of using externally allocated 398memory in DPDK, there are certain use cases where the overhead of DPDK heap API 399is undesirable - for example, when manual memory management is performed on an 400externally allocated area. To support use cases where externally allocated 401memory will not be used as part of normal DPDK workflow, there is also another 402set of API's under the ``rte_extmem_*`` namespace. 403 404These API's are (as their name implies) intended to allow registering or 405unregistering externally allocated memory to/from DPDK's internal page table, to 406allow API's like ``rte_mem_virt2memseg`` etc. to work with externally allocated 407memory. Memory added this way will not be available for any regular DPDK 408allocators; DPDK will leave this memory for the user application to manage. 409 410The expected workflow is as follows: 411 412* Get a pointer to memory area 413* Register memory within DPDK 414 - If IOVA table is not specified, IOVA addresses will be assumed to be 415 unavailable 416 - Other processes must attach to the memory area before they can use it 417* Perform DMA mapping with ``rte_dev_dma_map`` if needed 418* Use the memory area in your application 419* If memory area is no longer needed, it can be unregistered 420 - If the area was mapped for DMA, unmapping must be performed before 421 unregistering memory 422 - Other processes must detach from the memory area before it can be 423 unregistered 424 425Since these externally allocated memory areas will not be managed by DPDK, it is 426therefore up to the user application to decide how to use them and what to do 427with them once they're registered. 428 429Per-lcore and Shared Variables 430~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 431 432.. note:: 433 434 lcore refers to a logical execution unit of the processor, sometimes called a hardware *thread*. 435 436Shared variables are the default behavior. 437Per-lcore variables are implemented using *Thread Local Storage* (TLS) to provide per-thread local storage. 438 439Logs 440~~~~ 441 442While originally part of EAL, DPDK logging functionality is now provided by the :doc:`log_lib`. 443 444Trace and Debug Functions 445^^^^^^^^^^^^^^^^^^^^^^^^^ 446 447There are some debug functions to dump the stack in glibc. 448The rte_panic() function can voluntarily provoke a SIG_ABORT, 449which can trigger the generation of a core file, readable by gdb. 450 451CPU Feature Identification 452~~~~~~~~~~~~~~~~~~~~~~~~~~ 453 454The EAL can query the CPU at runtime (using the rte_cpu_get_features() function) to determine which CPU features are available. 455 456User Space Interrupt Event 457~~~~~~~~~~~~~~~~~~~~~~~~~~ 458 459+ User Space Interrupt and Alarm Handling in Host Thread 460 461The EAL creates a host thread to poll the UIO device file descriptors to detect the interrupts. 462Callbacks can be registered or unregistered by the EAL functions for a specific interrupt event 463and are called in the host thread asynchronously. 464The EAL also allows timed callbacks to be used in the same way as for NIC interrupts. 465 466.. note:: 467 468 In DPDK PMD, the only interrupts handled by the dedicated host thread are those for link status change 469 (link up and link down notification) and for sudden device removal. 470 471 472+ RX Interrupt Event 473 474The receive and transmit routines provided by each PMD don't limit themselves to execute in polling thread mode. 475To ease the idle polling with tiny throughput, it's useful to pause the polling and wait until the wake-up event happens. 476The RX interrupt is the first choice to be such kind of wake-up event, but probably won't be the only one. 477 478EAL provides the event APIs for this event-driven thread mode. 479Taking Linux as an example, the implementation relies on epoll. Each thread can monitor an epoll instance 480in which all the wake-up events' file descriptors are added. The event file descriptors are created and mapped to 481the interrupt vectors according to the UIO/VFIO spec. 482From FreeBSD's perspective, kqueue is the alternative way, but not implemented yet. 483 484EAL initializes the mapping between event file descriptors and interrupt vectors, while each device initializes the mapping 485between interrupt vectors and queues. In this way, EAL actually is unaware of the interrupt cause on the specific vector. 486The eth_dev driver takes responsibility to program the latter mapping. 487 488.. note:: 489 490 Per queue RX interrupt event is only allowed in VFIO which supports multiple MSI-X vector. In UIO, the RX interrupt 491 together with other interrupt causes shares the same vector. In this case, when RX interrupt and LSC(link status change) 492 interrupt are both enabled(intr_conf.lsc == 1 && intr_conf.rxq == 1), only the former is capable. 493 494The RX interrupt are controlled/enabled/disabled by ethdev APIs - 'rte_eth_dev_rx_intr_*'. They return failure if the PMD 495hasn't support them yet. The intr_conf.rxq flag is used to turn on the capability of RX interrupt per device. 496 497+ Device Removal Event 498 499This event is triggered by a device being removed at a bus level. Its 500underlying resources may have been made unavailable (i.e. PCI mappings 501unmapped). The PMD must make sure that on such occurrence, the application can 502still safely use its callbacks. 503 504This event can be subscribed to in the same way one would subscribe to a link 505status change event. The execution context is thus the same, i.e. it is the 506dedicated interrupt host thread. 507 508Considering this, it is likely that an application would want to close a 509device having emitted a Device Removal Event. In such case, calling 510``rte_eth_dev_close()`` can trigger it to unregister its own Device Removal Event 511callback. Care must be taken not to close the device from the interrupt handler 512context. It is necessary to reschedule such closing operation. 513 514Block list 515~~~~~~~~~~ 516 517The EAL PCI device block list functionality can be used to mark certain NIC ports as unavailable, 518so they are ignored by the DPDK. 519The ports to be blocked are identified using the PCIe* description (Domain:Bus:Device.Function). 520 521Misc Functions 522~~~~~~~~~~~~~~ 523 524Locks and atomic operations are per-architecture (i686 and x86_64). 525 526Lock annotations 527~~~~~~~~~~~~~~~~ 528 529R/W locks, seq locks and spinlocks have been instrumented to help developers in 530catching issues in DPDK. 531 532This instrumentation relies on 533`clang Thread Safety checks <https://clang.llvm.org/docs/ThreadSafetyAnalysis.html>`_. 534All attributes are prefixed with __rte and are fully described in the clang 535documentation. 536 537Some general comments: 538 539- it is important that lock requirements are expressed at the function 540 declaration level in headers so that other code units can be inspected, 541- when some global lock is necessary to some user-exposed API, it is preferred 542 to expose it via an internal helper rather than expose the global variable, 543- there are a list of known limitations with clang instrumentation, but before 544 waiving checks with ``__rte_no_thread_safety_analysis`` in your code, please 545 discuss it on the mailing list, 546 547The checks are enabled by default for libraries and drivers. 548They can be disabled by setting ``annotate_locks`` to ``false`` in 549the concerned library/driver ``meson.build``. 550 551IOVA Mode Detection 552~~~~~~~~~~~~~~~~~~~ 553 554IOVA Mode is selected by considering what the current usable Devices on the 555system require and/or support. 556 557On FreeBSD, RTE_IOVA_PA is always the default. On Linux, the IOVA mode is 558detected based on a 2-step heuristic detailed below. 559 560For the first step, EAL asks each bus its requirement in terms of IOVA mode 561and decides on a preferred IOVA mode. 562 563- if all buses report RTE_IOVA_PA, then the preferred IOVA mode is RTE_IOVA_PA, 564- if all buses report RTE_IOVA_VA, then the preferred IOVA mode is RTE_IOVA_VA, 565- if all buses report RTE_IOVA_DC, no bus expressed a preference, then the 566 preferred mode is RTE_IOVA_DC, 567- if the buses disagree (at least one wants RTE_IOVA_PA and at least one wants 568 RTE_IOVA_VA), then the preferred IOVA mode is RTE_IOVA_DC (see below with the 569 check on Physical Addresses availability), 570 571If the buses have expressed no preference on which IOVA mode to pick, then a 572default is selected using the following logic: 573 574- if physical addresses are not available, RTE_IOVA_VA mode is used 575- if /sys/kernel/iommu_groups is not empty, RTE_IOVA_VA mode is used 576- otherwise, RTE_IOVA_PA mode is used 577 578In the case when the buses had disagreed on their preferred IOVA mode, part of 579the buses won't work because of this decision. 580 581The second step checks if the preferred mode complies with the Physical 582Addresses availability since those are only available to root user in recent 583kernels. Namely, if the preferred mode is RTE_IOVA_PA but there is no access to 584Physical Addresses, then EAL init fails early, since later probing of the 585devices would fail anyway. 586 587.. note:: 588 589 The RTE_IOVA_VA mode is preferred as the default in most cases for the 590 following reasons: 591 592 - All drivers are expected to work in RTE_IOVA_VA mode, irrespective of 593 physical address availability. 594 - By default, the mempool, first asks for IOVA-contiguous memory using 595 ``RTE_MEMZONE_IOVA_CONTIG``. This is slow in RTE_IOVA_PA mode and it may 596 affect the application boot time. 597 - It is easy to enable large amount of IOVA-contiguous memory use cases 598 with IOVA in VA mode. 599 600 It is expected that all PCI drivers work in both RTE_IOVA_PA and 601 RTE_IOVA_VA modes. 602 603 If a PCI driver does not support RTE_IOVA_PA mode, the 604 ``RTE_PCI_DRV_NEED_IOVA_AS_VA`` flag is used to dictate that this PCI 605 driver can only work in RTE_IOVA_VA mode. 606 607 608IOVA Mode Configuration 609~~~~~~~~~~~~~~~~~~~~~~~ 610 611Auto detection of the IOVA mode, based on probing the bus and IOMMU configuration, may not report 612the desired addressing mode when virtual devices that are not directly attached to the bus are present. 613To facilitate forcing the IOVA mode to a specific value the EAL command line option ``--iova-mode`` can 614be used to select either physical addressing('pa') or virtual addressing('va'). 615 616.. _max_simd_bitwidth: 617 618 619Max SIMD bitwidth 620~~~~~~~~~~~~~~~~~ 621 622The EAL provides a single setting to limit the max SIMD bitwidth used by DPDK, 623which is used in determining the vector path, if any, chosen by a component. 624The value can be set at runtime by an application using the 625'rte_vect_set_max_simd_bitwidth(uint16_t bitwidth)' function, 626which should only be called once at initialization, before EAL init. 627The value can be overridden by the user using the EAL command-line option '--force-max-simd-bitwidth'. 628 629When choosing a vector path, along with checking the CPU feature support, 630the value of the max SIMD bitwidth must also be checked, and can be retrieved using the 631'rte_vect_get_max_simd_bitwidth()' function. 632The value should be compared against the enum values for accepted max SIMD bitwidths: 633 634.. code-block:: c 635 636 enum rte_vect_max_simd { 637 RTE_VECT_SIMD_DISABLED = 64, 638 RTE_VECT_SIMD_128 = 128, 639 RTE_VECT_SIMD_256 = 256, 640 RTE_VECT_SIMD_512 = 512, 641 RTE_VECT_SIMD_MAX = INT16_MAX + 1, 642 }; 643 644 if (rte_vect_get_max_simd_bitwidth() >= RTE_VECT_SIMD_512) 645 /* Take AVX-512 vector path */ 646 else if (rte_vect_get_max_simd_bitwidth() >= RTE_VECT_SIMD_256) 647 /* Take AVX2 vector path */ 648 649 650Memory Segments and Memory Zones (memzone) 651------------------------------------------ 652 653The mapping of physical memory is provided by this feature in the EAL. 654As physical memory can have gaps, the memory is described in a table of descriptors, 655and each descriptor (called rte_memseg ) describes a physical page. 656 657On top of this, the memzone allocator's role is to reserve contiguous portions of physical memory. 658These zones are identified by a unique name when the memory is reserved. 659 660The rte_memzone descriptors are also located in the configuration structure. 661This structure is accessed using rte_eal_get_configuration(). 662The lookup (by name) of a memory zone returns a descriptor containing the physical address of the memory zone. 663 664Memory zones can be reserved with specific start address alignment by supplying the align parameter 665(by default, they are aligned to cache line size). 666The alignment value should be a power of two and not less than the cache line size (64 bytes). 667Memory zones can also be reserved from either 2 MB or 1 GB hugepages, provided that both are available on the system. 668 669Both memsegs and memzones are stored using ``rte_fbarray`` structures. Please 670refer to *DPDK API Reference* for more information. 671 672 673Multiple pthread 674---------------- 675 676DPDK usually pins one pthread per core to avoid the overhead of task switching. 677This allows for significant performance gains, but lacks flexibility and is not always efficient. 678 679Power management helps to improve the CPU efficiency by limiting the CPU runtime frequency. 680However, alternately it is possible to utilize the idle cycles available to take advantage of 681the full capability of the CPU. 682 683By taking advantage of cgroup, the CPU utilization quota can be simply assigned. 684This gives another way to improve the CPU efficiency, however, there is a prerequisite; 685DPDK must handle the context switching between multiple pthreads per core. 686 687For further flexibility, it is useful to set pthread affinity not only to a CPU but to a CPU set. 688 689EAL pthread and lcore Affinity 690~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 691 692The term "lcore" refers to an EAL thread, which is really a Linux/FreeBSD pthread. 693"EAL pthreads" are created and managed by EAL and execute the tasks issued by *remote_launch*. 694In each EAL pthread, there is a TLS (Thread Local Storage) called *_lcore_id* for unique identification. 695As EAL pthreads usually bind 1:1 to the physical CPU, the *_lcore_id* is typically equal to the CPU ID. 696 697When using multiple pthreads, however, the binding is no longer always 1:1 between an EAL pthread and a specified physical CPU. 698The EAL pthread may have affinity to a CPU set, and as such the *_lcore_id* will not be the same as the CPU ID. 699For this reason, there is an EAL long option '--lcores' defined to assign the CPU affinity of lcores. 700For a specified lcore ID or ID group, the option allows setting the CPU set for that EAL pthread. 701 702The format pattern: 703 --lcores='<lcore_set>[@cpu_set][,<lcore_set>[@cpu_set],...]' 704 705'lcore_set' and 'cpu_set' can be a single number, range or a group. 706 707A number is a "digit([0-9]+)"; a range is "<number>-<number>"; a group is "(<number|range>[,<number|range>,...])". 708 709If a '\@cpu_set' value is not supplied, the value of 'cpu_set' will default to the value of 'lcore_set'. 710 711 :: 712 713 For example, "--lcores='1,2@(5-7),(3-5)@(0,2),(0,6),7-8'" which means start 9 EAL thread; 714 lcore 0 runs on cpuset 0x41 (cpu 0,6); 715 lcore 1 runs on cpuset 0x2 (cpu 1); 716 lcore 2 runs on cpuset 0xe0 (cpu 5,6,7); 717 lcore 3,4,5 runs on cpuset 0x5 (cpu 0,2); 718 lcore 6 runs on cpuset 0x41 (cpu 0,6); 719 lcore 7 runs on cpuset 0x80 (cpu 7); 720 lcore 8 runs on cpuset 0x100 (cpu 8). 721 722Using this option, for each given lcore ID, the associated CPUs can be assigned. 723It's also compatible with the pattern of corelist('-l') option. 724 725non-EAL pthread support 726~~~~~~~~~~~~~~~~~~~~~~~ 727 728It is possible to use the DPDK execution context with any user pthread (aka. non-EAL pthreads). 729There are two kinds of non-EAL pthreads: 730 731- a registered non-EAL pthread with a valid *_lcore_id* that was successfully assigned by calling ``rte_thread_register()``, 732- a non registered non-EAL pthread with a LCORE_ID_ANY, 733 734For non registered non-EAL pthread (with a LCORE_ID_ANY *_lcore_id*), 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). 735 736All these impacts are mentioned in :ref:`known_issue_label` section. 737 738Public Thread API 739~~~~~~~~~~~~~~~~~ 740 741There are two public APIs ``rte_thread_set_affinity()`` and ``rte_thread_get_affinity()`` introduced for threads. 742When they're used in any pthread context, the Thread Local Storage(TLS) will be set/get. 743 744Those TLS include *_cpuset* and *_socket_id*: 745 746* *_cpuset* stores the CPUs bitmap to which the pthread is affinitized. 747 748* *_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. 749 750 751Control Thread API 752~~~~~~~~~~~~~~~~~~ 753 754It is possible to create Control Threads using the public API 755``rte_thread_create_control()``. 756Those threads can be used for management/infrastructure tasks and are used 757internally by DPDK for multi process support and interrupt handling. 758 759Those threads will be scheduled on CPUs part of the original process CPU 760affinity from which the dataplane and service lcores are excluded. 761 762For example, on a 8 CPUs system, starting a dpdk application with -l 2,3 763(dataplane cores), then depending on the affinity configuration which can be 764controlled with tools like taskset (Linux) or cpuset (FreeBSD), 765 766- with no affinity configuration, the Control Threads will end up on 767 0-1,4-7 CPUs. 768- with affinity restricted to 2-4, the Control Threads will end up on 769 CPU 4. 770- with affinity restricted to 2-3, the Control Threads will end up on 771 CPU 2 (main lcore, which is the default when no CPU is available). 772 773.. _known_issue_label: 774 775Known Issues 776~~~~~~~~~~~~ 777 778+ rte_mempool 779 780 The rte_mempool uses a per-lcore cache inside the mempool. 781 For unregistered non-EAL pthreads, ``rte_lcore_id()`` will not return a valid number. 782 So for now, when rte_mempool is used with unregistered non-EAL pthreads, the put/get operations will bypass the default mempool cache and there is a performance penalty because of this bypass. 783 Only user-owned external caches can be used in an unregistered non-EAL context in conjunction with ``rte_mempool_generic_put()`` and ``rte_mempool_generic_get()`` that accept an explicit cache parameter. 784 785+ rte_ring 786 787 rte_ring supports multi-producer enqueue and multi-consumer dequeue. 788 However, it is non-preemptive, this has a knock on effect of making rte_mempool non-preemptible. 789 790 .. note:: 791 792 The "non-preemptive" constraint means: 793 794 - a pthread doing multi-producers enqueues on a given ring must not 795 be preempted by another pthread doing a multi-producer enqueue on 796 the same ring. 797 - a pthread doing multi-consumers dequeues on a given ring must not 798 be preempted by another pthread doing a multi-consumer dequeue on 799 the same ring. 800 801 Bypassing this constraint may cause the 2nd pthread to spin until the 1st one is scheduled again. 802 Moreover, if the 1st pthread is preempted by a context that has an higher priority, it may even cause a dead lock. 803 804 This means, use cases involving preemptible pthreads should consider using rte_ring carefully. 805 806 #. It CAN be used for preemptible single-producer and single-consumer use case. 807 808 #. It CAN be used for non-preemptible multi-producer and preemptible single-consumer use case. 809 810 #. It CAN be used for preemptible single-producer and non-preemptible multi-consumer use case. 811 812 #. It MAY be used by preemptible multi-producer and/or preemptible multi-consumer pthreads whose scheduling policy are all SCHED_OTHER(cfs), SCHED_IDLE or SCHED_BATCH. User SHOULD be aware of the performance penalty before using it. 813 814 #. It MUST not be used by multi-producer/consumer pthreads, whose scheduling policies are SCHED_FIFO or SCHED_RR. 815 816 Alternatively, applications can use the lock-free stack mempool handler. When 817 considering this handler, note that: 818 819 - It is currently limited to the aarch64 and x86_64 platforms, because it uses 820 an instruction (16-byte compare-and-swap) that is not yet available on other 821 platforms. 822 - It has worse average-case performance than the non-preemptive rte_ring, but 823 software caching (e.g. the mempool cache) can mitigate this by reducing the 824 number of stack accesses. 825 826+ rte_timer 827 828 Running ``rte_timer_manage()`` on an unregistered non-EAL pthread is not allowed. However, resetting/stopping the timer from a non-EAL pthread is allowed. 829 830+ rte_log 831 832 In unregistered non-EAL pthreads, there is no per thread loglevel and logtype, global loglevels are used. 833 834+ misc 835 836 The debug statistics of rte_ring, rte_mempool and rte_timer are not supported in an unregistered non-EAL pthread. 837 838Signal Safety 839~~~~~~~~~~~~~ 840 841 The Posix API defines an async-signal-safe function as one that can be safely 842 called from with a signal handler. Many DPDK functions are non-reentrant and 843 therefore are unsafe to call from a signal handler. 844 845 The kinds of issues that make DPDK functions unsafe can be understood when 846 one considers that much of the code in DPDK uses locks and other shared 847 resources. For example, calling ``rte_mempool_lookup()`` from a signal 848 would deadlock if the signal happened during previous call ``rte_mempool`` 849 routines. 850 851 Other functions are not signal safe because they use one or more 852 library routines that are not themselves signal safe. 853 For example, calling ``rte_panic()`` is not safe in a signal handler 854 because it uses ``rte_log()`` and ``rte_log()`` calls the 855 ``syslog()`` library function which is in the list of 856 signal safe functions in 857 `Signal-Safety manual page <https://man7.org/linux/man-pages/man7/signal-safety.7.html>`_. 858 859 The set of functions that are expected to be async-signal-safe in DPDK 860 is shown in the following table. The functions not otherwise noted 861 are not async-signal-safe. 862 863.. csv-table:: **Signal Safe Functions** 864 :header: "Function" 865 :widths: 32 866 867 rte_dump_stack 868 rte_eal_get_lcore_state 869 rte_eal_get_runtime_dir 870 rte_eal_has_hugepages 871 rte_eal_has_pci 872 rte_eal_lcore_role 873 rte_eal_process_type 874 rte_eal_using_phys_addrs 875 rte_get_hpet_cycles 876 rte_get_hpet_hz 877 rte_get_main_lcore 878 rte_get_next_lcore 879 rte_get_tsc_hz 880 rte_hypervisor_get 881 rte_hypervisor_get_name 882 rte_lcore_count 883 rte_lcore_cpuset 884 rte_lcore_has_role 885 rte_lcore_index 886 rte_lcore_is_enabled 887 rte_lcore_to_cpu_id 888 rte_lcore_to_socket_id 889 rte_log_get_global_level 890 rte_log_get_level 891 rte_memory_get_nchannel 892 rte_memory_get_nrank 893 rte_reciprocal_value 894 rte_reciprocal_value_u64 895 rte_socket_count 896 rte_socket_id 897 rte_socket_id_by_idx 898 rte_strerror 899 rte_strscpy 900 rte_strsplit 901 rte_sys_gettid 902 rte_uuid_compare 903 rte_uuid_is_null 904 rte_uuid_parse 905 rte_uuid_unparse 906 907 908cgroup control 909~~~~~~~~~~~~~~ 910 911The 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). 912We expect only 50% of CPU spend on packet IO. 913 914 .. code-block:: console 915 916 mkdir /sys/fs/cgroup/cpu/pkt_io 917 mkdir /sys/fs/cgroup/cpuset/pkt_io 918 919 echo $cpu > /sys/fs/cgroup/cpuset/cpuset.cpus 920 921 echo $t0 > /sys/fs/cgroup/cpu/pkt_io/tasks 922 echo $t0 > /sys/fs/cgroup/cpuset/pkt_io/tasks 923 924 echo $t1 > /sys/fs/cgroup/cpu/pkt_io/tasks 925 echo $t1 > /sys/fs/cgroup/cpuset/pkt_io/tasks 926 927 cd /sys/fs/cgroup/cpu/pkt_io 928 echo 100000 > pkt_io/cpu.cfs_period_us 929 echo 50000 > pkt_io/cpu.cfs_quota_us 930 931.. _malloc: 932 933Malloc 934------ 935 936The EAL provides a malloc API to allocate any-sized memory. 937 938The objective of this API is to provide malloc-like functions to allow 939allocation from hugepage memory and to facilitate application porting. 940The *DPDK API Reference* manual describes the available functions. 941 942Typically, these kinds of allocations should not be done in data plane 943processing because they are slower than pool-based allocation and make 944use of locks within the allocation and free paths. 945However, they can be used in configuration code. 946 947Refer to the rte_malloc() function description in the *DPDK API Reference* 948manual for more information. 949 950 951Alignment and NUMA Constraints 952~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 953 954The rte_malloc() takes an align argument that can be used to request a memory 955area that is aligned on a multiple of this value (which must be a power of two). 956 957On systems with NUMA support, a call to the rte_malloc() function will return 958memory that has been allocated on the NUMA socket of the core which made the call. 959A set of APIs is also provided, to allow memory to be explicitly allocated on a 960NUMA socket directly, or by allocated on the NUMA socket where another core is 961located, in the case where the memory is to be used by a logical core other than 962on the one doing the memory allocation. 963 964Use Cases 965~~~~~~~~~ 966 967This API is meant to be used by an application that requires malloc-like 968functions at initialization time. 969 970For allocating/freeing data at runtime, in the fast-path of an application, 971the memory pool library should be used instead. 972 973Internal Implementation 974~~~~~~~~~~~~~~~~~~~~~~~ 975 976Data Structures 977^^^^^^^^^^^^^^^ 978 979There are two data structure types used internally in the malloc library: 980 981* struct malloc_heap - used to track free space on a per-socket basis 982 983* struct malloc_elem - the basic element of allocation and free-space 984 tracking inside the library. 985 986Structure: malloc_heap 987"""""""""""""""""""""" 988 989The malloc_heap structure is used to manage free space on a per-socket basis. 990Internally, there is one heap structure per NUMA node, which allows us to 991allocate memory to a thread based on the NUMA node on which this thread runs. 992While this does not guarantee that the memory will be used on that NUMA node, 993it is no worse than a scheme where the memory is always allocated on a fixed 994or random node. 995 996The key fields of the heap structure and their function are described below 997(see also diagram above): 998 999* lock - the lock field is needed to synchronize access to the heap. 1000 Given that the free space in the heap is tracked using a linked list, 1001 we need a lock to prevent two threads manipulating the list at the same time. 1002 1003* free_head - this points to the first element in the list of free nodes for 1004 this malloc heap. 1005 1006* first - this points to the first element in the heap. 1007 1008* last - this points to the last element in the heap. 1009 1010.. _figure_malloc_heap: 1011 1012.. figure:: img/malloc_heap.* 1013 1014 Example of a malloc heap and malloc elements within the malloc library 1015 1016 1017.. _malloc_elem: 1018 1019Structure: malloc_elem 1020"""""""""""""""""""""" 1021 1022The malloc_elem structure is used as a generic header structure for various 1023blocks of memory. 1024It is used in two different ways - all shown in the diagram above: 1025 1026#. As a header on a block of free or allocated memory - normal case 1027 1028#. As a padding header inside a block of memory 1029 1030The most important fields in the structure and how they are used are described below. 1031 1032Malloc heap is a doubly-linked list, where each element keeps track of its 1033previous and next elements. Due to the fact that hugepage memory can come and 1034go, neighboring malloc elements may not necessarily be adjacent in memory. 1035Also, since a malloc element may span multiple pages, its contents may not 1036necessarily be IOVA-contiguous either - each malloc element is only guaranteed 1037to be virtually contiguous. 1038 1039.. note:: 1040 1041 If the usage of a particular field in one of the above three usages is not 1042 described, the field can be assumed to have an undefined value in that 1043 situation, for example, for padding headers only the "state" and "pad" 1044 fields have valid values. 1045 1046* heap - this pointer is a reference back to the heap structure from which 1047 this block was allocated. 1048 It is used for normal memory blocks when they are being freed, to add the 1049 newly-freed block to the heap's free-list. 1050 1051* prev - this pointer points to previous header element/block in memory. When 1052 freeing a block, this pointer is used to reference the previous block to 1053 check if that block is also free. If so, and the two blocks are immediately 1054 adjacent to each other, then the two free blocks are merged to form a single 1055 larger block. 1056 1057* next - this pointer points to next header element/block in memory. When 1058 freeing a block, this pointer is used to reference the next block to check 1059 if that block is also free. If so, and the two blocks are immediately 1060 adjacent to each other, then the two free blocks are merged to form a single 1061 larger block. 1062 1063* free_list - this is a structure pointing to previous and next elements in 1064 this heap's free list. 1065 It is only used in normal memory blocks; on ``malloc()`` to find a suitable 1066 free block to allocate and on ``free()`` to add the newly freed element to 1067 the free-list. 1068 1069* state - This field can have one of three values: ``FREE``, ``BUSY`` or 1070 ``PAD``. 1071 The former two are to indicate the allocation state of a normal memory block 1072 and the latter is to indicate that the element structure is a dummy structure 1073 at the end of the start-of-block padding, i.e. where the start of the data 1074 within a block is not at the start of the block itself, due to alignment 1075 constraints. 1076 In that case, the pad header is used to locate the actual malloc element 1077 header for the block. 1078 1079* dirty - this flag is only meaningful when ``state`` is ``FREE``. 1080 It indicates that the content of the element is not fully zero-filled. 1081 Memory from such blocks must be cleared when requested via ``rte_zmalloc*()``. 1082 Dirty elements only appear with ``--huge-unlink=never``. 1083 1084* pad - this holds the length of the padding present at the start of the block. 1085 In the case of a normal block header, it is added to the address of the end 1086 of the header to give the address of the start of the data area, i.e. the 1087 value passed back to the application on a malloc. 1088 Within a dummy header inside the padding, this same value is stored, and is 1089 subtracted from the address of the dummy header to yield the address of the 1090 actual block header. 1091 1092* size - the size of the data block, including the header itself. 1093 1094Memory Allocation 1095^^^^^^^^^^^^^^^^^ 1096 1097On EAL initialization, all preallocated memory segments are setup as part of the 1098malloc heap. This setup involves placing an :ref:`element header<malloc_elem>` 1099with ``FREE`` at the start of each virtually contiguous segment of memory. 1100The ``FREE`` element is then added to the ``free_list`` for the malloc heap. 1101 1102This setup also happens whenever memory is allocated at runtime (if supported), 1103in which case newly allocated pages are also added to the heap, merging with any 1104adjacent free segments if there are any. 1105 1106When an application makes a call to a malloc-like function, the malloc function 1107will first index the ``lcore_config`` structure for the calling thread, and 1108determine the NUMA node of that thread. 1109The NUMA node is used to index the array of ``malloc_heap`` structures which is 1110passed as a parameter to the ``heap_alloc()`` function, along with the 1111requested size, type, alignment and boundary parameters. 1112 1113The ``heap_alloc()`` function will scan the free_list of the heap, and attempt 1114to find a free block suitable for storing data of the requested size, with the 1115requested alignment and boundary constraints. 1116 1117When a suitable free element has been identified, the pointer to be returned 1118to the user is calculated. 1119The cache-line of memory immediately preceding this pointer is filled with a 1120struct malloc_elem header. 1121Because of alignment and boundary constraints, there could be free space at 1122the start and/or end of the element, resulting in the following behavior: 1123 1124#. Check for trailing space. 1125 If the trailing space is big enough, i.e. > 128 bytes, then the free element 1126 is split. 1127 If it is not, then we just ignore it (wasted space). 1128 1129#. Check for space at the start of the element. 1130 If the space at the start is small, i.e. <=128 bytes, then a pad header is 1131 used, and the remaining space is wasted. 1132 If, however, the remaining space is greater, then the free element is split. 1133 1134The advantage of allocating the memory from the end of the existing element is 1135that no adjustment of the free list needs to take place - the existing element 1136on the free list just has its size value adjusted, and the next/previous elements 1137have their "prev"/"next" pointers redirected to the newly created element. 1138 1139In case when there is not enough memory in the heap to satisfy allocation 1140request, EAL will attempt to allocate more memory from the system (if supported) 1141and, following successful allocation, will retry reserving the memory again. In 1142a multiprocessing scenario, all primary and secondary processes will synchronize 1143their memory maps to ensure that any valid pointer to DPDK memory is guaranteed 1144to be valid at all times in all currently running processes. 1145 1146Failure to synchronize memory maps in one of the processes will cause allocation 1147to fail, even though some of the processes may have allocated the memory 1148successfully. The memory is not added to the malloc heap unless primary process 1149has ensured that all other processes have mapped this memory successfully. 1150 1151Any successful allocation event will trigger a callback, for which user 1152applications and other DPDK subsystems can register. Additionally, validation 1153callbacks will be triggered before allocation if the newly allocated memory will 1154exceed threshold set by the user, giving a chance to allow or deny allocation. 1155 1156.. note:: 1157 1158 Any allocation of new pages has to go through primary process. If the 1159 primary process is not active, no memory will be allocated even if it was 1160 theoretically possible to do so. This is because primary's process map acts 1161 as an authority on what should or should not be mapped, while each secondary 1162 process has its own, local memory map. Secondary processes do not update the 1163 shared memory map, they only copy its contents to their local memory map. 1164 1165Freeing Memory 1166^^^^^^^^^^^^^^ 1167 1168To free an area of memory, the pointer to the start of the data area is passed 1169to the free function. 1170The size of the ``malloc_elem`` structure is subtracted from this pointer to get 1171the element header for the block. 1172If this header is of type ``PAD`` then the pad length is further subtracted from 1173the pointer to get the proper element header for the entire block. 1174 1175From this element header, we get pointers to the heap from which the block was 1176allocated and to where it must be freed, as well as the pointer to the previous 1177and next elements. These next and previous elements are then checked to see if 1178they are also ``FREE`` and are immediately adjacent to the current one, and if 1179so, they are merged with the current element. This means that we can never have 1180two ``FREE`` memory blocks adjacent to one another, as they are always merged 1181into a single block. 1182 1183If deallocating pages at runtime is supported, and the free element encloses 1184one or more pages, those pages can be deallocated and be removed from the heap. 1185If DPDK was started with command-line parameters for preallocating memory 1186(``-m`` or ``--socket-mem``), then those pages that were allocated at startup 1187will not be deallocated. 1188 1189Any successful deallocation event will trigger a callback, for which user 1190applications and other DPDK subsystems can register. 1191