kiper220/mimalloc
1
0
Fork 0
mirror of https://github.com/microsoft/mimalloc.git synced 2025-07-06 11:34:38 +03:00
Code Issues Projects Releases Packages Wiki Activity Actions

merge from dev

Browse source
This commit is contained in:
Daan Leijen 2022-12-19 17:08:45 -08:00
commit 92ffc25d79
65 changed files with 793 additions and 741 deletions
Download patch file Download diff file Expand all files Collapse all files

12
doc/mimalloc-doc.h
View file

@ -40,7 +40,7 @@ Notable aspects of the design include:
per mimalloc page, but for each page we have multiple free lists. In particular, there
is one list for thread-local `free` operations, and another one for concurrent `free`
operations. Free-ing from another thread can now be a single CAS without needing
sophisticated coordination between threads. Since there will be
sophisticated coordination between threads. Since there will be
thousands of separate free lists, contention is naturally distributed over the heap,
and the chance of contending on a single location will be low -- this is quite
similar to randomized algorithms like skip lists where adding
@ -414,7 +414,7 @@ void mi_register_error(mi_error_fun* errfun, void* arg);
bool mi_is_in_heap_region(const void* p);
/// Reserve OS memory for use by mimalloc. Reserved areas are used
/// before allocating from the OS again. By reserving a large area upfront,
/// before allocating from the OS again. By reserving a large area upfront,
/// allocation can be more efficient, and can be better managed on systems
/// without `mmap`/`VirtualAlloc` (like WASM for example).
/// @param size The size to reserve.
@ -423,7 +423,7 @@ bool mi_is_in_heap_region(const void* p);
/// @return \a 0 if successful, and an error code otherwise (e.g. `ENOMEM`).
int mi_reserve_os_memory(size_t size, bool commit, bool allow_large);
/// Manage a particular memory area for use by mimalloc.
/// Manage a particular memory area for use by mimalloc.
/// This is just like `mi_reserve_os_memory` except that the area should already be
/// allocated in some manner and available for use my mimalloc.
/// @param start Start of the memory area
@ -499,7 +499,7 @@ void mi_process_info(size_t* elapsed_msecs, size_t* user_msecs, size_t* system_m
/// \{
/// The maximum supported alignment size (currently 1MiB).
#define MI_ALIGNMENT_MAX (1024*1024UL)
#define MI_ALIGNMENT_MAX (1024*1024UL)
/// Allocate \a size bytes aligned by \a alignment.
/// @param size number of bytes to allocate.
@ -813,7 +813,7 @@ typedef enum mi_option_e {
mi_option_page_reset, ///< Reset page memory after \a mi_option_reset_delay milliseconds when it becomes free.
mi_option_abandoned_page_reset, //< Reset free page memory when a thread terminates.
mi_option_use_numa_nodes, ///< Pretend there are at most N NUMA nodes; Use 0 to use the actual detected NUMA nodes at runtime.
mi_option_eager_commit_delay, ///< the first N segments per thread are not eagerly committed (=1).
mi_option_eager_commit_delay, ///< the first N segments per thread are not eagerly committed (=1).
mi_option_os_tag, ///< OS tag to assign to mimalloc'd memory
mi_option_limit_os_alloc, ///< If set to 1, do not use OS memory for allocation (but only pre-reserved arenas)
@ -1097,7 +1097,7 @@ or via environment variables.
`MIMALLOC_EAGER_COMMIT_DELAY=N` (`N` is 1 by default) to delay the initial `N` segments (of 4MiB)
of a thread to not allocate in the huge OS pages; this prevents threads that are short lived
and allocate just a little to take up space in the huge OS page area (which cannot be reset).
- `MIMALLOC_RESERVE_HUGE_OS_PAGES_AT=N`: where N is the numa node. This reserves the huge pages at a specific numa node.
- `MIMALLOC_RESERVE_HUGE_OS_PAGES_AT=N`: where N is the numa node. This reserves the huge pages at a specific numa node.
(`N` is -1 by default to reserve huge pages evenly among the given number of numa nodes (or use the available ones as detected))
Use caution when using `fork` in combination with either large or huge OS pages: on a fork, the OS uses copy-on-write