aboutsummaryrefslogtreecommitdiffstats
path: root/erts/emulator/beam/erl_alloc_util.h
AgeCommit message (Collapse)Author
2018-12-07Mark free blocks in pooled carriers as unused (MADV_FREE)John Högberg
This lets the OS reclaim the physical memory associated with these blocks which reduces the impact of long-lived awkward allocations. A small allocated block will still keep a huge carrier alive, but the unused part of the carrier will now be available to the OS. Co-authored-by: Dmytro Lytovchenko <[email protected]>
2018-06-28Allow cross-type carrier migrationJohn Högberg
2018-06-18Update copyright yearHenrik Nord
2018-05-22erts: Let allocator pooled_tree also use Age OrderSverker Eriksson
if configured for the allocator. This was not implemented and pooled_tree always used address order first fit. The "problem" was as a carrier can be part of both the pooled tree (of its owner) and the allocation tree (of a foreign instance) at the same time. Solved by duplicating 'birth_time' in both AOFF_RBTree_t nodes for each carrier. Blocks still cannot use age order and does not pay any memory cost for birth_time.
2018-04-23erts: Rewrite memory instrumentationJohn Högberg
This commit replaces the old memory instrumentation with a new implementation that scans carriers instead of wrapping erts_alloc/erts_free. The old implementation could not extract information without halting the emulator, had considerable runtime overhead, and the memory maps it produced were noisy and lacked critical information. Since the new implementation walks through existing data structures there's no longer a need to start the emulator with special flags to get information about carrier utilization/fragmentation. Memory fragmentation is also easier to diagnose as it's presented on a per-carrier basis which eliminates the need to account for "holes" between mmap segments. To help track allocations, each allocation can now be tagged with what it is and who allocated it at the cost of one extra word per allocation. This is controlled on a per-allocator basis with the +M<S>atags option, and is enabled by default for binary_alloc and driver_alloc (which is also used by NIFs).
2018-03-13Merge PR-1699 from sverker/hipe-amd64-high-code/OTP-14951Sverker Eriksson
Remove low memory need for HiPE on x86_64
2018-03-05erts: Remove hipe amd64 code super carrier (exec_mmap)Sverker Eriksson
2018-02-12Merge 'sverker/maint-20/alloc-n-migration/ERIERL-88'Sverker Eriksson
into 'sverker/master/alloc-n-migration/ERIERL-88'
2018-02-12Merge 'sverker/maint-19/alloc-n-migration/ERIERL-88'Sverker Eriksson
into 'sverker/maint-20/alloc-n-migration/ERIERL-88' OTP-14915 OTP-14916 OTP-14917 OTP-14918
2018-02-12Merge 'sverker/carrier-migration-improvements'Sverker Eriksson
into 'sverker/maint-19/alloc-n-migration/ERIERL-88'
2018-02-12erts: Add system_flags(erts_alloc,"+M?sbct *")Sverker Eriksson
to change sbct limit in runtime for chosen allocator type. With great power comes great responsibility.
2018-02-08erts: Add migration options "acnl" and "acfml"Sverker Eriksson
acnl: Abandon Carrier Nr Limit acfml: Abandon Carrier Free block Min Limit
2017-12-20erts: Add more stats for mbcs_poolSverker Eriksson
similar to the ones in OTP-19.2.3.1
2017-12-20erts: Improve carrier pool searchSverker Eriksson
* Give back carrier to owner when put in pool with use of dd-queue. * Replace pooled_list with pooled_tree for more efficient search of all owned pooled carriers. * Remove traitor_list as it does not serve much purpose anymore. * Add HOMECOMING bit flag in crr->allctr atomic to (1) avoid double enqueue into dd-enqueue. (2) trigger read barrier in get_used_allctr for newly received carriers.
2017-12-20erts: Refactor carrier dealloc migrationSverker Eriksson
by adding a dedicated 'homecoming_dd_block' to use in dd-queue. + Preparation for dd-queue-migration of non-empty carriers. + Get rid of ugly hack where blk->carrier pointer is overwritten by dd-queue and then have to be restored.
2017-09-11Merge branch 'maint'John Högberg
2017-09-07Replace ad-hoc MIN/MAX macros with common onesJohn Högberg
Besides being noisy, they were already defined by a global Unix- specific header, causing the Windows build to fail if one forgot to define them.
2017-07-17erts: Replace usage of all erts_smp prefixes to just ertsLukas Larsson
2017-07-17erts: Remove ERTS_SMP and USE_THREAD definesLukas Larsson
This refactor was done using the unifdef tool like this: for file in $(find erts/ -name *.[ch]); do unifdef -t -f defile -o $file $file; done where defile contained: #define ERTS_SMP 1 #define USE_THREADS 1 #define DDLL_SMP 1 #define ERTS_HAVE_SMP_EMU 1 #define SMP 1 #define ERL_BITS_REENTRANT 1 #define ERTS_USE_ASYNC_READY_Q 1 #define FDBLOCK 1 #undef ERTS_POLL_NEED_ASYNC_INTERRUPT_SUPPORT #define ERTS_POLL_ASYNC_INTERRUPT_SUPPORT 0 #define ERTS_POLL_USE_WAKEUP_PIPE 1 #define ERTS_POLL_USE_UPDATE_REQUESTS_QUEUE 1 #undef ERTS_HAVE_PLAIN_EMU #undef ERTS_SIGNAL_STATE
2017-07-06Merge branch 'john/erts/runtime-lcnt' into maintJohn Högberg
* john/erts/runtime-lcnt: Document rt_mask and add warnings about copy_save Add an emulator test suite for lock counting Break erts_debug:lock_counters/1 into separate BIFs Allow toggling lock counting at runtime Move lock flags to a common header Enable register_SUITE for lcnt builds Enable lcnt smoke test on all builds that have lcnt enabled Make lock counter info independent of the locks being counted OTP-14412 OTP-13170 OTP-14413
2017-07-06Allow toggling lock counting at runtimeJohn Högberg
The implementation is still hidden behind ERTS_ENABLE_LOCK_COUNT, and all categories are still enabled by default, but the actual counting can be toggled at will. OTP-13170
2017-06-14Update copyright yearHans Nilsson
2017-05-21erts: Make allocator functions staticSalikhov Dinislam
2016-11-22Merge branch 'maint'Sverker Eriksson
2016-11-17erts: Refactor crash dumping with cbprintfSverker Eriksson
Instead of passing around a file descriptor use a function pointer to facilitate more advanced backend write logic such as size limitation or compression.
2016-10-14erts: Enable exec_alloc for all hipe architecturesSverker Eriksson
For non-amd64 it's a "normal" allocator with a wrapper around mseg_alloc to call mprotect(PROT_EXEC).
2016-04-20Merge branch 'sverker/hipe-code-alloc'Sverker Eriksson
2016-04-20erts: Make sure literal MBCs have super aligned sizesSverker Eriksson
on 32-bit, as the granularity of the literal bit vector is super-alignment.
2016-04-15erts: Refactor callbacks for literal mseg allocSverker Eriksson
Make the callbacks more general to be usable for any allocator that that uses its own ErtsMemMapper.
2016-04-13Merge branch 'henrik/update-copyrightyear'Henrik Nord
* henrik/update-copyrightyear: update copyright-year
2016-04-06erts: Add lttng tracepoints for memory carriersBjörn-Egil Dahlberg
* carrier_create * carrier_destroy * carrier_pool_put * carrier_pool_get
2016-03-15update copyright-yearHenrik Nord
2015-11-27Merge branch 'maint'Sverker Eriksson
2015-11-12erts: Add support for fast erts_is_literal()Sverker Eriksson
2015-10-01erts: Fix confusion of callbacks destroying_mbc() vs remove_mbc()Sverker Eriksson
Problem #1 Goodfit was crippled by the fact that destroying_mbc() was called _before_ the carriers was unlinked from mbc_list. Problem #2 destroying_mbc() was called for carriers that later could be resurrected from dc_list without a matching call to creating_mbc(). This was mostly a practical problem for the new test case alloc_SUITE:migration that use the callbacks to create/destroy a mutex. Solution: destroying_mbc() is now only called just before a carrier is destroyed (deallocated or put in mseg cache). remove_mbc() is called both (like before) when inserted into cpool but now also when last block is freed and mbc is scheduled for destruction but may later be resurrected from dc_list.
2015-09-11erts: Fix strangeness in treatment of MSEG_ALIGN_BITSSverker Eriksson
And remove old case of using only page alignment (12 bits).
2015-06-24erts: Remove HALFWORD_HEAP definitionBjörn-Egil Dahlberg
2015-06-18Change license text to APLv2Bruce Yinhe
2014-10-03erts: Add pooled_list and traitor_listSverker Eriksson
2013-11-05Add switch for disabling sys_alloc carriersRickard Green
The switch "+Musac <boolean>" controls if sys_alloc carriers are allowed.
2013-09-30erts: Remove ASSERT_EXPR macroSverker Eriksson
2013-09-25Merge branch 'rickard/aligned-sys_alloc-carriers/OTP-11318' into ↵Rickard Green
rickard/aligned-sys_alloc-carriers_maint/OTP-11318 Conflicts: erts/emulator/beam/erl_alloc.c erts/emulator/beam/erl_alloc_util.c erts/emulator/beam/erl_alloc_util.h
2013-09-25Implement platform specific aligned sys_alloc and use when supportedRickard Green
erts_sys_aligned_alloc() is currently implemented using posix_memalign if it exist, or using _aligned_malloc on Windows. If erts_sys_aligned_alloc() exist allocators will create sys_alloc carriers similar to how this was done pre-R16.
2013-06-12Update copyright yearsBjörn-Egil Dahlberg
2013-06-04erts: Use carrier pool for migration of carriersRickard Green
2013-06-04erts: Implement carrier poolRickard Green
2013-06-04erts: Use Uint64 for call countsRickard Green
2013-06-03erts: Remove unnecessary flag arguments in allocatorsSverker Eriksson
A cleanup after SBMBC was removed.
2013-06-03erts: Remove SBMBC allocatorSverker Eriksson
2013-06-03erts: Change naive list to rb-tree of carriers in AOFF allocatorSverker Eriksson
and add new callbacks add_mbc(), remove_mbc() and largest_fblk_in_mbc() for carrier migration.