Erlang -- erts

erts_alloc

An Erlang runtime system internal memory allocator library.

erts_alloc is an Erlang runtime system internal memory allocator library. erts_alloc provides the Erlang runtime system with a number of memory allocators.

The following allocators are present:

temp_alloc: Allocator used for temporary allocations.
eheap_alloc: Allocator used for Erlang heap data, such as Erlang process heaps.
binary_alloc: Allocator used for Erlang binary data.
ets_alloc: Allocator used for ets data.
driver_alloc: Allocator used for driver data.
literal_alloc: Allocator used for constant terms in Erlang code.
sl_alloc: Allocator used for memory blocks that are expected to be short-lived.
ll_alloc: Allocator used for memory blocks that are expected to be long-lived, for example, Erlang code.
fix_alloc: A fast allocator used for some frequently used fixed size data types.
exec_alloc: Allocator used by the HiPE application for native executable code on specific architectures (x86_64).
std_alloc: Allocator used for most memory blocks not allocated through any of the other allocators described above.
sys_alloc: This is normally the default malloc implementation used on the specific OS.
mseg_alloc: A memory segment allocator. It is used by other allocators for allocating memory segments and is only available on systems that have the mmap system call. Memory segments that are deallocated are kept for a while in a segment cache before they are destroyed. When segments are allocated, cached segments are used if possible instead of creating new segments. This to reduce the number of system calls made.

sys_alloc and literal_alloc are always enabled and cannot be disabled. exec_alloc is only available if it is needed and cannot be disabled. mseg_alloc is always enabled if it is available and an allocator that uses it is enabled. All other allocators can be enabled or disabled. By default all allocators are enabled. When an allocator is disabled, sys_alloc is used instead of the disabled allocator.

The main idea with the erts_alloc library is to separate memory blocks that are used differently into different memory areas, to achieve less memory fragmentation. By putting less effort in finding a good fit for memory blocks that are frequently allocated than for those less frequently allocated, a performance gain can be achieved.

Internally a framework called alloc_util is used for implementing allocators. sys_alloc and mseg_alloc do not use this framework, so the following does not apply to them.

An allocator manages multiple areas, called carriers, in which memory blocks are placed. A carrier is either placed in a separate memory segment (allocated through mseg_alloc), or in the heap segment (allocated through sys_alloc).

Multiblock carriers are used for storage of several blocks.
Singleblock carriers are used for storage of one block.
Blocks that are larger than the value of the singleblock carrier threshold (sbct) parameter are placed in singleblock carriers.
Blocks that are smaller than the value of parameter sbct are placed in multiblock carriers.

Normally an allocator creates a "main multiblock carrier". Main multiblock carriers are never deallocated. The size of the main multiblock carrier is determined by the value of parameter mmbcs.

Sizes of multiblock carriers allocated through mseg_alloc are decided based on the following parameters:

The values of the largest multiblock carrier size (lmbcs)
The smallest multiblock carrier size (smbcs)
The multiblock carrier growth stages (mbcgs)

If nc is the current number of multiblock carriers (the main multiblock carrier excluded) managed by an allocator, the size of the next mseg_alloc multiblock carrier allocated by this allocator is roughly smbcs+nc*(lmbcs-smbcs)/mbcgs when nc <= mbcgs, and lmbcs when nc > mbcgs. If the value of parameter sbct is larger than the value of parameter lmbcs, the allocator may have to create multiblock carriers that are larger than the value of parameter lmbcs, though. Singleblock carriers allocated through mseg_alloc are sized to whole pages.

Sizes of carriers allocated through sys_alloc are decided based on the value of the sys_alloc carrier size (ycs) parameter. The size of a carrier is the least number of multiples of the value of parameter ycs satisfying the request.

Coalescing of free blocks are always performed immediately. Boundary tags (headers and footers) in free blocks are used, which makes the time complexity for coalescing constant.

The memory allocation strategy used for multiblock carriers by an allocator can be configured using parameter as. The following strategies are available:

Best fit

Strategy: Find the smallest block satisfying the requested block size.

Implementation: A balanced binary search tree is used. The time complexity is proportional to log N, where N is the number of sizes of free blocks.

Address order best fit

Strategy: Find the smallest block satisfying the requested block size. If multiple blocks are found, choose the one with the lowest address.

Implementation: A balanced binary search tree is used. The time complexity is proportional to log N, where N is the number of free blocks.

Address order first fit

Strategy: Find the block with the lowest address satisfying the requested block size.

Implementation: A balanced binary search tree is used. The time complexity is proportional to log N, where N is the number of free blocks.

Address order first fit carrier best fit

Strategy: Find the carrier with the lowest address that can satisfy the requested block size, then find a block within that carrier using the "best fit" strategy.

Implementation: Balanced binary search trees are used. The time complexity is proportional to log N, where N is the number of free blocks.

Address order first fit carrier address order best fit

Strategy: Find the carrier with the lowest address that can satisfy the requested block size, then find a block within that carrier using the "address order best fit" strategy.

Implementation: Balanced binary search trees are used. The time complexity is proportional to log N, where N is the number of free blocks.

Age order first fit carrier address order first fit

Strategy: Find the oldest carrier that can satisfy the requested block size, then find a block within that carrier using the "address order first fit" strategy.

Implementation: A balanced binary search tree is used. The time complexity is proportional to log N, where N is the number of free blocks.

Age order first fit carrier best fit

Strategy: Find the oldest carrier that can satisfy the requested block size, then find a block within that carrier using the "best fit" strategy.

Implementation: Balanced binary search trees are used. The time complexity is proportional to log N, where N is the number of free blocks.

Age order first fit carrier address order best fit

Strategy: Find the oldest carrier that can satisfy the requested block size, then find a block within that carrier using the "address order best fit" strategy.

Implementation: Balanced binary search trees are used. The time complexity is proportional to log N, where N is the number of free blocks.

Good fit

Strategy: Try to find the best fit, but settle for the best fit found during a limited search.

Implementation: The implementation uses segregated free lists with a maximum block search depth (in each list) to find a good fit fast. When the maximum block search depth is small (by default 3), this implementation has a time complexity that is constant. The maximum block search depth can be configured using parameter mbsd.

A fit

Strategy: Do not search for a fit, inspect only one free block to see if it satisfies the request. This strategy is only intended to be used for temporary allocations.

Implementation: Inspect the first block in a free-list. If it satisfies the request, it is used, otherwise a new carrier is created. The implementation has a time complexity that is constant.

As from ERTS 5.6.1 the emulator refuses to use this strategy on other allocators than temp_alloc. This because it only causes problems for other allocators.

Apart from the ordinary allocators described above, some pre-allocators are used for some specific data types. These pre-allocators pre-allocate a fixed amount of memory for certain data types when the runtime system starts. As long as pre-allocated memory is available, it is used. When no pre-allocated memory is available, memory is allocated in ordinary allocators. These pre-allocators are typically much faster than the ordinary allocators, but can only satisfy a limited number of requests.

Warning

Only use these flags if you are sure what you are doing. Unsuitable settings can cause serious performance degradation and even a system crash at any time during operation.

Memory allocator system flags have the following syntax: +M<S><P> <V>, where <S> is a letter identifying a subsystem, <P> is a parameter, and <V> is the value to use. The flags can be passed to the Erlang emulator (erl(1)) as command-line arguments.

System flags effecting specific allocators have an uppercase letter as <S>. The following letters are used for the allocators:

B: binary_alloc
D: std_alloc
E: ets_alloc
F: fix_alloc
H: eheap_alloc
I: literal_alloc
L: ll_alloc
M: mseg_alloc
R: driver_alloc
S: sl_alloc
T: temp_alloc
X: exec_alloc
Y: sys_alloc

Flags for Configuration of mseg_alloc

+MMamcbf <size>: Absolute maximum cache bad fit (in kilobytes). A segment in the memory segment cache is not reused if its size exceeds the requested size with more than the value of this parameter. Defaults to 4096.
+MMrmcbf <ratio>: Relative maximum cache bad fit (in percent). A segment in the memory segment cache is not reused if its size exceeds the requested size with more than relative maximum cache bad fit percent of the requested size. Defaults to 20.
+MMsco true|false: Sets super carrier only flag. Defaults to true. When a super carrier is used and this flag is true, mseg_alloc only creates carriers in the super carrier. Notice that the alloc_util framework can create sys_alloc carriers, so if you want all carriers to be created in the super carrier, you therefore want to disable use of sys_alloc carriers by also passing +Musac false. When the flag is false, mseg_alloc tries to create carriers outside of the super carrier when the super carrier is full.

Note

Setting this flag to false is not supported on all systems. The flag is then ignored.
+MMscrfsd <amount>: Sets super carrier reserved free segment descriptors. Defaults to 65536. This parameter determines the amount of memory to reserve for free segment descriptors used by the super carrier. If the system runs out of reserved memory for free segment descriptors, other memory is used. This can however cause fragmentation issues, so you want to ensure that this never happens. The maximum amount of free segment descriptors used can be retrieved from the erts_mmap tuple part of the result from calling erlang:system_info({allocator, mseg_alloc}).
+MMscrpm true|false: Sets super carrier reserve physical memory flag. Defaults to true. When this flag is true, physical memory is reserved for the whole super carrier at once when it is created. The reservation is after that left unchanged. When this flag is set to false, only virtual address space is reserved for the super carrier upon creation. The system attempts to reserve physical memory upon carrier creations in the super carrier, and attempt to unreserve physical memory upon carrier destructions in the super carrier.

Note

What reservation of physical memory means, highly depends on the operating system, and how it is configured. For example, different memory overcommit settings on Linux drastically change the behavior.

Setting this flag to false is possibly not supported on all systems. The flag is then ignored.
+MMscs <size in MB>: Sets super carrier size (in MB). Defaults to 0, that is, the super carrier is by default disabled. The super carrier is a large continuous area in the virtual address space. mseg_alloc always tries to create new carriers in the super carrier if it exists. Notice that the alloc_util framework can create sys_alloc carriers. For more information, see +MMsco.
+MMmcs <amount>: Maximum cached segments. The maximum number of memory segments stored in the memory segment cache. Valid range is [0, 30]. Defaults to 10.

Flags for Configuration of sys_alloc

+MYe true: Enables sys_alloc.

Note

sys_alloc cannot be disabled.
+MYm libc: malloc library to use. Only libc is available. libc enables the standard libc malloc implementation. By default libc is used.
+MYtt <size>: Trim threshold size (in kilobytes). This is the maximum amount of free memory at the top of the heap (allocated by sbrk) that is kept by malloc (not released to the operating system). When the amount of free memory at the top of the heap exceeds the trim threshold, malloc releases it (by calling sbrk). Trim threshold is specified in kilobytes. Defaults to 128.

Note

This flag has effect only when the emulator is linked with the GNU C library, and uses its malloc implementation.
+MYtp <size>: Top pad size (in kilobytes). This is the amount of extra memory that is allocated by malloc when sbrk is called to get more memory from the operating system. Defaults to 0.

Note

This flag has effect only when the emulator is linked with the GNU C library, and uses its malloc implementation.

Flags for Configuration of Allocators Based on alloc_util

If u is used as subsystem identifier (that is, <S> = u), all allocators based on alloc_util are effected. If B, D, E, F, H, L, R, S, or T is used as subsystem identifier, only the specific allocator identifier is effected.

+M<S>acul <utilization>|de

Abandon carrier utilization limit. A valid <utilization> is an integer in the range [0, 100] representing utilization in percent. When a utilization value > 0 is used, allocator instances are allowed to abandon multiblock carriers. If de (default enabled) is passed instead of a <utilization>, a recommended non-zero utilization value is used. The value chosen depends on the allocator type and can be changed between ERTS versions. Defaults to de, but this can be changed in the future.

Carriers are abandoned when memory utilization in the allocator instance falls below the utilization value used. Once a carrier is abandoned, no new allocations are made in it. When an allocator instance gets an increased multiblock carrier need, it first tries to fetch an abandoned carrier from an allocator instance of the same allocator type. If no abandoned carrier can be fetched, it creates a new empty carrier. When an abandoned carrier has been fetched, it will function as an ordinary carrier. This feature has special requirements on the allocation strategy used. Only the strategies aoff, aoffcbf, aoffcaobf, ageffcaoffm, ageffcbf and ageffcaobf support abandoned carriers.

This feature also requires multiple thread specific instances to be enabled. When enabling this feature, multiple thread-specific instances are enabled if not already enabled, and the aoffcbf strategy is enabled if the current strategy does not support abandoned carriers. This feature can be enabled on all allocators based on the alloc_util framework, except temp_alloc (which would be pointless).

+M<S>acfml <bytes>

Abandon carrier free block min limit. A valid <bytes> is a positive integer representing a block size limit. The largest free block in a carrier must be at least bytes large, for the carrier to be abandoned. The default is zero but can be changed in the future.

Flags for Configuration of alloc_util

All allocators based on alloc_util are effected.

+Muycs <size>: sys_alloc carrier size. Carriers allocated through sys_alloc are allocated in sizes that are multiples of the sys_alloc carrier size. This is not true for main multiblock carriers and carriers allocated during a memory shortage, though.
+Mummc <amount>: Maximum mseg_alloc carriers. Maximum number of carriers placed in separate memory segments. When this limit is reached, new carriers are placed in memory retrieved from sys_alloc.
+Musac <bool>: Allow sys_alloc carriers. Defaults to true. If set to false, sys_alloc carriers are never created by allocators using the alloc_util framework.

Special Flag for literal_alloc

+MIscs <size in MB>: literal_alloc super carrier size (in MB). The amount of virtual address space reserved for literal terms in Erlang code on 64-bit architectures. Defaults to 1024 (that is, 1 GB), which is usually sufficient. The flag is ignored on 32-bit architectures.

Special Flag for exec_alloc

+MXscs <size in MB>: exec_alloc super carrier size (in MB). The amount of virtual address space reserved for native executable code used by the HiPE application on specific architectures (x86_64). Defaults to 512.

Instrumentation Flags

+Mim true|false: A map over current allocations is kept by the emulator. The allocation map can be retrieved through module instrument(3). +Mim true implies +Mis true. +Mim true is the same as flag -instr in erl(1).
+Mis true|false: Status over allocated memory is kept by the emulator. The allocation status can be retrieved through module instrument(3).
+Mit X: Reserved for future use. Do not use this flag.

Note

When instrumentation of the emulator is enabled, the emulator uses more memory and runs slower.

Other Flags

+Mea min|max|r9c|r10b|r11b|config

Options:

min: Disables all allocators that can be disabled.
max: Enables all allocators (default).
r9c|r10b|r11b: Configures all allocators as they were configured in respective Erlang/OTP release. These will eventually be removed.
config: Disables features that cannot be enabled while creating an allocator configuration with erts_alloc_config(3).

Note

This option is to be used only while running erts_alloc_config(3), not when using the created configuration.

+Mlpm all|no

Lock physical memory. Defaults to no, that is, no physical memory is locked. If set to all, all memory mappings made by the runtime system are locked into physical memory. If set to all, the runtime system fails to start if this feature is not supported, the user has not got enough privileges, or the user is not allowed to lock enough physical memory. The runtime system also fails with an out of memory condition if the user limit on the amount of locked memory is reached.

Only some default values have been presented here. For information about the currently used settings and the current status of the allocators, see erlang:system_info(allocator) and erlang:system_info({allocator, Alloc}).

Note

Most of these flags are highly implementation-dependent and can be changed or removed without prior notice.

erts_alloc is not obliged to strictly use the settings that have been passed to it (it can even ignore them).

The erts_alloc_config(3) tool can be used to aid creation of an erts_alloc configuration that is suitable for a limited number of runtime scenarios.

erl(1), erlang(3), erts_alloc_config(3), instrument(3)

Table of Contents

erts_alloc

C Library

Library Summary

Description

Allocators

The alloc_util Framework

System Flags Effecting erts_alloc

Flags for Configuration of mseg_alloc

Flags for Configuration of sys_alloc

Flags for Configuration of Allocators Based on alloc_util

Flags for Configuration of alloc_util

Special Flag for literal_alloc

Special Flag for exec_alloc

Instrumentation Flags

Other Flags

Notes

See Also