|
Flow-IPC 3.0.0
Flow-IPC project: Full implementation reference.
|
#include "ipc/shm/arena_lend/detail/arena_lend_fwd.hpp"#include "ipc/shm/arena_lend/shm_pool.hpp"#include "ipc/shm/arena_lend/arena_lend_stats.hpp"#include <flow/util/thread_lcl.hpp>#include <flow/util/util.hpp>#include <flow/util/stat/stat_set_list.hpp>#include <flow/util/stat/stat_set.hpp>#include <boost/container/flat_map.hpp>#include <boost/unordered/unordered_flat_map.hpp>#include <type_traits>#include <cstdint>Go to the source code of this file.
Classes | |
| struct | ipc::shm::arena_lend::detail::Pool_by_base |
Value type in the reverse-lookup map: identifies the pool and, when LOOKUP_CAN_FAIL is true for the containing lookup structure, also carries the pool size so that a range-check can determine whether a given vaddr actually falls inside the pool (as opposed to merely being above its base). More... | |
| class | ipc::shm::arena_lend::detail::Pool_lookup_global_stats< Shm_arena_t, OWNER_ELSE_BORROWER > |
Global stats singleton for either Owner_pool_lookup_global_stats or Borrower_pool_lookup_global_stats, per SHM-provider (as identified by Shm_arena type). More... | |
| class | ipc::shm::arena_lend::detail::Thread_local_pool_lookup_rev< Repository_t, LOOKUP_CAN_FAIL, Shm_arena_t > |
Thread_local_pool_lookup_rev<R>.s_registry, where R is a distinct pool-repository type, enables high-perf reverse-lookup from a vaddr to the containing SHM-pool's ID and offset in bytes from the pool-base vaddr. More... | |
| class | ipc::shm::arena_lend::detail::Thread_local_pool_lookup_fwd< Repository_t, Shm_arena_t > |
| Analogous to Thread_local_pool_lookup_rev but for forward lookup: pool ID -> base vaddr. More... | |
Namespaces | |
| namespace | ipc |
| Catch-all namespace for the Flow-IPC project: A library/API in modern C++17 providing high-performance communication between processes. | |
| namespace | ipc::shm |
| Modules for SHared Memory (SHM) support. | |
| namespace | ipc::shm::arena_lend |
| ipc::shm sub-module providing the SHM-provider family based on the arena-lending paradigm: each process allocates in its own arena(s), lending the resulting objects to other processes, which can read them but not allocate (nor deallocate) in that arena. | |
| namespace | ipc::shm::arena_lend::detail |
| Segregated private stuff for ipc::shm::arena_lend. | |