nchebrov/azerothcore-wotlk
1
0
Fork 0
mirror of https://github.com/mod-playerbots/azerothcore-wotlk.git synced 2026-01-13 01:08:35 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' of https://github.com/azerothcore/azerothcore-wotlk into dir-restructure

Browse Source
This commit is contained in:
Yehonal
2017-12-21 11:26:43 +01:00
parent acd60005e8 a0d17509a2
commit 403ed2600f
445 changed files with 49192 additions and 15431 deletions
Download Patch File Download Diff File Expand all files Collapse all files

94
deps/jemalloc/CMakeLists.txt vendored
View File

@@ -1,4 +1,4 @@
# Copyright (C)
# Copyright (C) 2008-2017 TrinityCore <http://www.trinitycore.org/>
#
# This file is free software; as a special exception the author gives
# unlimited permission to copy and/or distribute it, with or without
@@ -8,14 +8,24 @@
# WITHOUT ANY WARRANTY, to the extent permitted by law; without even the
# implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# We need to generate the jemalloc_def.h header based on platform-specific settings
if (PLATFORM EQUAL 32)
set(JEM_SIZEDEF 2)
set(JEM_TLSMODEL)
else()
set(JEM_SIZEDEF 3)
set(JEM_TLSMODEL "__attribute__\(\(tls_model\(\"initial-exec\"\)\)\)")
endif()
# We need to generate the jemalloc_def.h header based on platform-specific settings
CHECK_SYMBOL_EXISTS(MADV_FREE "sys/mman.h" HAVE_MADV_FREE)
if (PLATFORM EQUAL 32)
set(JEM_SIZEDEF 2)
set(JEM_TLSMODEL)
set(JEM_VADDRBITS 32)
else()
set(JEM_SIZEDEF 3)
set(JEM_TLSMODEL "__attribute__\(\(tls_model\(\"initial-exec\"\)\)\)")
set(JEM_VADDRBITS 48)
endif()
if (HAVE_MADV_FREE)
set(JEM_MADFREE_DEF "#define")
else()
set(JEM_MADFREE_DEF "#undef")
endif()
# Create the header, so we can use it
configure_file(
@@ -24,31 +34,38 @@ configure_file(
@ONLY
)
# Done, let's continue
set(jemalloc_STAT_SRC
${CMAKE_CURRENT_SOURCE_DIR}/src/arena.c
${CMAKE_CURRENT_SOURCE_DIR}/src/atomic.c
${CMAKE_CURRENT_SOURCE_DIR}/src/base.c
${CMAKE_CURRENT_SOURCE_DIR}/src/bitmap.c
${CMAKE_CURRENT_SOURCE_DIR}/src/chunk.c
${CMAKE_CURRENT_SOURCE_DIR}/src/chunk_dss.c
${CMAKE_CURRENT_SOURCE_DIR}/src/chunk_mmap.c
${CMAKE_CURRENT_SOURCE_DIR}/src/ckh.c
${CMAKE_CURRENT_SOURCE_DIR}/src/ctl.c
${CMAKE_CURRENT_SOURCE_DIR}/src/extent.c
${CMAKE_CURRENT_SOURCE_DIR}/src/hash.c
${CMAKE_CURRENT_SOURCE_DIR}/src/huge.c
${CMAKE_CURRENT_SOURCE_DIR}/src/jemalloc.c
${CMAKE_CURRENT_SOURCE_DIR}/src/mb.c
${CMAKE_CURRENT_SOURCE_DIR}/src/mutex.c
${CMAKE_CURRENT_SOURCE_DIR}/src/prof.c
${CMAKE_CURRENT_SOURCE_DIR}/src/quarantine.c
${CMAKE_CURRENT_SOURCE_DIR}/src/rtree.c
${CMAKE_CURRENT_SOURCE_DIR}/src/stats.c
${CMAKE_CURRENT_SOURCE_DIR}/src/tcache.c
${CMAKE_CURRENT_SOURCE_DIR}/src/tsd.c
${CMAKE_CURRENT_SOURCE_DIR}/src/util.c
)
# Done, let's continue
set(jemalloc_STAT_SRC
${CMAKE_CURRENT_SOURCE_DIR}/src/arena.c
${CMAKE_CURRENT_SOURCE_DIR}/src/background_thread.c
${CMAKE_CURRENT_SOURCE_DIR}/src/base.c
${CMAKE_CURRENT_SOURCE_DIR}/src/bitmap.c
${CMAKE_CURRENT_SOURCE_DIR}/src/ckh.c
${CMAKE_CURRENT_SOURCE_DIR}/src/ctl.c
${CMAKE_CURRENT_SOURCE_DIR}/src/extent.c
${CMAKE_CURRENT_SOURCE_DIR}/src/extent_dss.c
${CMAKE_CURRENT_SOURCE_DIR}/src/extent_mmap.c
${CMAKE_CURRENT_SOURCE_DIR}/src/hash.c
${CMAKE_CURRENT_SOURCE_DIR}/src/hooks.c
${CMAKE_CURRENT_SOURCE_DIR}/src/jemalloc.c
${CMAKE_CURRENT_SOURCE_DIR}/src/jemalloc_cpp.cpp
${CMAKE_CURRENT_SOURCE_DIR}/src/large.c
${CMAKE_CURRENT_SOURCE_DIR}/src/malloc_io.c
${CMAKE_CURRENT_SOURCE_DIR}/src/mutex.c
${CMAKE_CURRENT_SOURCE_DIR}/src/mutex_pool.c
${CMAKE_CURRENT_SOURCE_DIR}/src/nstime.c
${CMAKE_CURRENT_SOURCE_DIR}/src/pages.c
${CMAKE_CURRENT_SOURCE_DIR}/src/prng.c
${CMAKE_CURRENT_SOURCE_DIR}/src/prof.c
${CMAKE_CURRENT_SOURCE_DIR}/src/rtree.c
${CMAKE_CURRENT_SOURCE_DIR}/src/spin.c
${CMAKE_CURRENT_SOURCE_DIR}/src/stats.c
${CMAKE_CURRENT_SOURCE_DIR}/src/sz.c
${CMAKE_CURRENT_SOURCE_DIR}/src/tcache.c
${CMAKE_CURRENT_SOURCE_DIR}/src/ticker.c
${CMAKE_CURRENT_SOURCE_DIR}/src/tsd.c
${CMAKE_CURRENT_SOURCE_DIR}/src/witness.c
)
include_directories(
${BUILDDIR}/
@@ -58,3 +75,12 @@ include_directories(
add_definitions(-D_GNU_SOURCE -D_REENTRANT)
add_library(jemalloc STATIC ${jemalloc_STAT_SRC})
target_link_libraries(jemalloc
PUBLIC
${CMAKE_DL_LIBS})
set_target_properties(jemalloc
PROPERTIES
FOLDER
"deps")

4
deps/jemalloc/COPYING vendored
View File

@@ -1,10 +1,10 @@
Unless otherwise specified, files in the jemalloc source distribution are
subject to the following license:
--------------------------------------------------------------------------------
Copyright (C) 2002-2014 Jason Evans <jasone@canonware.com>.
Copyright (C) 2002-2017 Jason Evans <jasone@canonware.com>.
All rights reserved.
Copyright (C) 2007-2012 Mozilla Foundation. All rights reserved.
Copyright (C) 2009-2014 Facebook, Inc. All rights reserved.
Copyright (C) 2009-2017 Facebook, Inc. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:

727
deps/jemalloc/ChangeLog vendored
View File

@@ -1,10 +1,727 @@
Following are change highlights associated with official releases. Important
bug fixes are all mentioned, but internal enhancements are omitted here for
brevity (even though they are more fun to write about). Much more detail can be
found in the git revision history:
bug fixes are all mentioned, but some internal enhancements are omitted here for
brevity. Much more detail can be found in the git revision history:
https://github.com/jemalloc/jemalloc
* 5.0.1 (July 1, 2017)
This bugfix release fixes several issues, most of which are obscure enough
that typical applications are not impacted.
Bug fixes:
- Update decay->nunpurged before purging, in order to avoid potential update
races and subsequent incorrect purging volume. (@interwq)
- Only abort on dlsym(3) error if the failure impacts an enabled feature (lazy
locking and/or background threads). This mitigates an initialization
failure bug for which we still do not have a clear reproduction test case.
(@interwq)
- Modify tsd management so that it neither crashes nor leaks if a thread's
only allocation activity is to call free() after TLS destructors have been
executed. This behavior was observed when operating with GNU libc, and is
unlikely to be an issue with other libc implementations. (@interwq)
- Mask signals during background thread creation. This prevents signals from
being inadvertently delivered to background threads. (@jasone,
@davidgoldblatt, @interwq)
- Avoid inactivity checks within background threads, in order to prevent
recursive mutex acquisition. (@interwq)
- Fix extent_grow_retained() to use the specified hooks when the
arena.<i>.extent_hooks mallctl is used to override the default hooks.
(@interwq)
- Add missing reentrancy support for custom extent hooks which allocate.
(@interwq)
- Post-fork(2), re-initialize the list of tcaches associated with each arena
to contain no tcaches except the forking thread's. (@interwq)
- Add missing post-fork(2) mutex reinitialization for extent_grow_mtx. This
fixes potential deadlocks after fork(2). (@interwq)
- Enforce minimum autoconf version (currently 2.68), since 2.63 is known to
generate corrupt configure scripts. (@jasone)
- Ensure that the configured page size (--with-lg-page) is no larger than the
configured huge page size (--with-lg-hugepage). (@jasone)
* 5.0.0 (June 13, 2017)
Unlike all previous jemalloc releases, this release does not use naturally
aligned "chunks" for virtual memory management, and instead uses page-aligned
"extents". This change has few externally visible effects, but the internal
impacts are... extensive. Many other internal changes combine to make this
the most cohesively designed version of jemalloc so far, with ample
opportunity for further enhancements.
Continuous integration is now an integral aspect of development thanks to the
efforts of @davidtgoldblatt, and the dev branch tends to remain reasonably
stable on the tested platforms (Linux, FreeBSD, macOS, and Windows). As a
side effect the official release frequency may decrease over time.
New features:
- Implement optional per-CPU arena support; threads choose which arena to use
based on current CPU rather than on fixed thread-->arena associations.
(@interwq)
- Implement two-phase decay of unused dirty pages. Pages transition from
dirty-->muzzy-->clean, where the first phase transition relies on
madvise(... MADV_FREE) semantics, and the second phase transition discards
pages such that they are replaced with demand-zeroed pages on next access.
(@jasone)
- Increase decay time resolution from seconds to milliseconds. (@jasone)
- Implement opt-in per CPU background threads, and use them for asynchronous
decay-driven unused dirty page purging. (@interwq)
- Add mutex profiling, which collects a variety of statistics useful for
diagnosing overhead/contention issues. (@interwq)
- Add C++ new/delete operator bindings. (@djwatson)
- Support manually created arena destruction, such that all data and metadata
are discarded. Add MALLCTL_ARENAS_DESTROYED for accessing merged stats
associated with destroyed arenas. (@jasone)
- Add MALLCTL_ARENAS_ALL as a fixed index for use in accessing
merged/destroyed arena statistics via mallctl. (@jasone)
- Add opt.abort_conf to optionally abort if invalid configuration options are
detected during initialization. (@interwq)
- Add opt.stats_print_opts, so that e.g. JSON output can be selected for the
stats dumped during exit if opt.stats_print is true. (@jasone)
- Add --with-version=VERSION for use when embedding jemalloc into another
project's git repository. (@jasone)
- Add --disable-thp to support cross compiling. (@jasone)
- Add --with-lg-hugepage to support cross compiling. (@jasone)
- Add mallctl interfaces (various authors):
+ background_thread
+ opt.abort_conf
+ opt.retain
+ opt.percpu_arena
+ opt.background_thread
+ opt.{dirty,muzzy}_decay_ms
+ opt.stats_print_opts
+ arena.<i>.initialized
+ arena.<i>.destroy
+ arena.<i>.{dirty,muzzy}_decay_ms
+ arena.<i>.extent_hooks
+ arenas.{dirty,muzzy}_decay_ms
+ arenas.bin.<i>.slab_size
+ arenas.nlextents
+ arenas.lextent.<i>.size
+ arenas.create
+ stats.background_thread.{num_threads,num_runs,run_interval}
+ stats.mutexes.{ctl,background_thread,prof,reset}.
{num_ops,num_spin_acq,num_wait,max_wait_time,total_wait_time,max_num_thds,
num_owner_switch}
+ stats.arenas.<i>.{dirty,muzzy}_decay_ms
+ stats.arenas.<i>.uptime
+ stats.arenas.<i>.{pmuzzy,base,internal,resident}
+ stats.arenas.<i>.{dirty,muzzy}_{npurge,nmadvise,purged}
+ stats.arenas.<i>.bins.<j>.{nslabs,reslabs,curslabs}
+ stats.arenas.<i>.bins.<j>.mutex.
{num_ops,num_spin_acq,num_wait,max_wait_time,total_wait_time,max_num_thds,
num_owner_switch}
+ stats.arenas.<i>.lextents.<j>.{nmalloc,ndalloc,nrequests,curlextents}
+ stats.arenas.i.mutexes.{large,extent_avail,extents_dirty,extents_muzzy,
extents_retained,decay_dirty,decay_muzzy,base,tcache_list}.
{num_ops,num_spin_acq,num_wait,max_wait_time,total_wait_time,max_num_thds,
num_owner_switch}
Portability improvements:
- Improve reentrant allocation support, such that deadlock is less likely if
e.g. a system library call in turn allocates memory. (@davidtgoldblatt,
@interwq)
- Support static linking of jemalloc with glibc. (@djwatson)
Optimizations and refactors:
- Organize virtual memory as "extents" of virtual memory pages, rather than as
naturally aligned "chunks", and store all metadata in arbitrarily distant
locations. This reduces virtual memory external fragmentation, and will
interact better with huge pages (not yet explicitly supported). (@jasone)
- Fold large and huge size classes together; only small and large size classes
remain. (@jasone)
- Unify the allocation paths, and merge most fast-path branching decisions.
(@davidtgoldblatt, @interwq)
- Embed per thread automatic tcache into thread-specific data, which reduces
conditional branches and dereferences. Also reorganize tcache to increase
fast-path data locality. (@interwq)
- Rewrite atomics to closely model the C11 API, convert various
synchronization from mutex-based to atomic, and use the explicit memory
ordering control to resolve various hypothetical races without increasing
synchronization overhead. (@davidtgoldblatt)
- Extensively optimize rtree via various methods:
+ Add multiple layers of rtree lookup caching, since rtree lookups are now
part of fast-path deallocation. (@interwq)
+ Determine rtree layout at compile time. (@jasone)
+ Make the tree shallower for common configurations. (@jasone)
+ Embed the root node in the top-level rtree data structure, thus avoiding
one level of indirection. (@jasone)
+ Further specialize leaf elements as compared to internal node elements,
and directly embed extent metadata needed for fast-path deallocation.
(@jasone)
+ Ignore leading always-zero address bits (architecture-specific).
(@jasone)
- Reorganize headers (ongoing work) to make them hermetic, and disentangle
various module dependencies. (@davidtgoldblatt)
- Convert various internal data structures such as size class metadata from
boot-time-initialized to compile-time-initialized. Propagate resulting data
structure simplifications, such as making arena metadata fixed-size.
(@jasone)
- Simplify size class lookups when constrained to size classes that are
multiples of the page size. This speeds lookups, but the primary benefit is
complexity reduction in code that was the source of numerous regressions.
(@jasone)
- Lock individual extents when possible for localized extent operations,
rather than relying on a top-level arena lock. (@davidtgoldblatt, @jasone)
- Use first fit layout policy instead of best fit, in order to improve
packing. (@jasone)
- If munmap(2) is not in use, use an exponential series to grow each arena's
virtual memory, so that the number of disjoint virtual memory mappings
remains low. (@jasone)
- Implement per arena base allocators, so that arenas never share any virtual
memory pages. (@jasone)
- Automatically generate private symbol name mangling macros. (@jasone)
Incompatible changes:
- Replace chunk hooks with an expanded/normalized set of extent hooks.
(@jasone)
- Remove ratio-based purging. (@jasone)
- Remove --disable-tcache. (@jasone)
- Remove --disable-tls. (@jasone)
- Remove --enable-ivsalloc. (@jasone)
- Remove --with-lg-size-class-group. (@jasone)
- Remove --with-lg-tiny-min. (@jasone)
- Remove --disable-cc-silence. (@jasone)
- Remove --enable-code-coverage. (@jasone)
- Remove --disable-munmap (replaced by opt.retain). (@jasone)
- Remove Valgrind support. (@jasone)
- Remove quarantine support. (@jasone)
- Remove redzone support. (@jasone)
- Remove mallctl interfaces (various authors):
+ config.munmap
+ config.tcache
+ config.tls
+ config.valgrind
+ opt.lg_chunk
+ opt.purge
+ opt.lg_dirty_mult
+ opt.decay_time
+ opt.quarantine
+ opt.redzone
+ opt.thp
+ arena.<i>.lg_dirty_mult
+ arena.<i>.decay_time
+ arena.<i>.chunk_hooks
+ arenas.initialized
+ arenas.lg_dirty_mult
+ arenas.decay_time
+ arenas.bin.<i>.run_size
+ arenas.nlruns
+ arenas.lrun.<i>.size
+ arenas.nhchunks
+ arenas.hchunk.<i>.size
+ arenas.extend
+ stats.cactive
+ stats.arenas.<i>.lg_dirty_mult
+ stats.arenas.<i>.decay_time
+ stats.arenas.<i>.metadata.{mapped,allocated}
+ stats.arenas.<i>.{npurge,nmadvise,purged}
+ stats.arenas.<i>.huge.{allocated,nmalloc,ndalloc,nrequests}
+ stats.arenas.<i>.bins.<j>.{nruns,reruns,curruns}
+ stats.arenas.<i>.lruns.<j>.{nmalloc,ndalloc,nrequests,curruns}
+ stats.arenas.<i>.hchunks.<j>.{nmalloc,ndalloc,nrequests,curhchunks}
Bug fixes:
- Improve interval-based profile dump triggering to dump only one profile when
a single allocation's size exceeds the interval. (@jasone)
- Use prefixed function names (as controlled by --with-jemalloc-prefix) when
pruning backtrace frames in jeprof. (@jasone)
* 4.5.0 (February 28, 2017)
This is the first release to benefit from much broader continuous integration
testing, thanks to @davidtgoldblatt. Had we had this testing infrastructure
in place for prior releases, it would have caught all of the most serious
regressions fixed by this release.
New features:
- Add --disable-thp and the opt.thp mallctl to provide opt-out mechanisms for
transparent huge page integration. (@jasone)
- Update zone allocator integration to work with macOS 10.12. (@glandium)
- Restructure *CFLAGS configuration, so that CFLAGS behaves typically, and
EXTRA_CFLAGS provides a way to specify e.g. -Werror during building, but not
during configuration. (@jasone, @ronawho)
Bug fixes:
- Fix DSS (sbrk(2)-based) allocation. This regression was first released in
4.3.0. (@jasone)
- Handle race in per size class utilization computation. This functionality
was first released in 4.0.0. (@interwq)
- Fix lock order reversal during gdump. (@jasone)
- Fix/refactor tcache synchronization. This regression was first released in
4.0.0. (@jasone)
- Fix various JSON-formatted malloc_stats_print() bugs. This functionality
was first released in 4.3.0. (@jasone)
- Fix huge-aligned allocation. This regression was first released in 4.4.0.
(@jasone)
- When transparent huge page integration is enabled, detect what state pages
start in according to the kernel's current operating mode, and only convert
arena chunks to non-huge during purging if that is not their initial state.
This functionality was first released in 4.4.0. (@jasone)
- Fix lg_chunk clamping for the --enable-cache-oblivious --disable-fill case.
This regression was first released in 4.0.0. (@jasone, @428desmo)
- Properly detect sparc64 when building for Linux. (@glaubitz)
* 4.4.0 (December 3, 2016)
New features:
- Add configure support for *-*-linux-android. (@cferris1000, @jasone)
- Add the --disable-syscall configure option, for use on systems that place
security-motivated limitations on syscall(2). (@jasone)
- Add support for Debian GNU/kFreeBSD. (@thesam)
Optimizations:
- Add extent serial numbers and use them where appropriate as a sort key that
is higher priority than address, so that the allocation policy prefers older
extents. This tends to improve locality (decrease fragmentation) when
memory grows downward. (@jasone)
- Refactor madvise(2) configuration so that MADV_FREE is detected and utilized
on Linux 4.5 and newer. (@jasone)
- Mark partially purged arena chunks as non-huge-page. This improves
interaction with Linux's transparent huge page functionality. (@jasone)
Bug fixes:
- Fix size class computations for edge conditions involving extremely large
allocations. This regression was first released in 4.0.0. (@jasone,
@ingvarha)
- Remove overly restrictive assertions related to the cactive statistic. This
regression was first released in 4.1.0. (@jasone)
- Implement a more reliable detection scheme for os_unfair_lock on macOS.
(@jszakmeister)
* 4.3.1 (November 7, 2016)
Bug fixes:
- Fix a severe virtual memory leak. This regression was first released in
4.3.0. (@interwq, @jasone)
- Refactor atomic and prng APIs to restore support for 32-bit platforms that
use pre-C11 toolchains, e.g. FreeBSD's mips. (@jasone)
* 4.3.0 (November 4, 2016)
This is the first release that passes the test suite for multiple Windows
configurations, thanks in large part to @glandium setting up continuous
integration via AppVeyor (and Travis CI for Linux and OS X).
New features:
- Add "J" (JSON) support to malloc_stats_print(). (@jasone)
- Add Cray compiler support. (@ronawho)
Optimizations:
- Add/use adaptive spinning for bootstrapping and radix tree node
initialization. (@jasone)
Bug fixes:
- Fix large allocation to search starting in the optimal size class heap,
which can substantially reduce virtual memory churn and fragmentation. This
regression was first released in 4.0.0. (@mjp41, @jasone)
- Fix stats.arenas.<i>.nthreads accounting. (@interwq)
- Fix and simplify decay-based purging. (@jasone)
- Make DSS (sbrk(2)-related) operations lockless, which resolves potential
deadlocks during thread exit. (@jasone)
- Fix over-sized allocation of radix tree leaf nodes. (@mjp41, @ogaun,
@jasone)
- Fix over-sized allocation of arena_t (plus associated stats) data
structures. (@jasone, @interwq)
- Fix EXTRA_CFLAGS to not affect configuration. (@jasone)
- Fix a Valgrind integration bug. (@ronawho)
- Disallow 0x5a junk filling when running in Valgrind. (@jasone)
- Fix a file descriptor leak on Linux. This regression was first released in
4.2.0. (@vsarunas, @jasone)
- Fix static linking of jemalloc with glibc. (@djwatson)
- Use syscall(2) rather than {open,read,close}(2) during boot on Linux. This
works around other libraries' system call wrappers performing reentrant
allocation. (@kspinka, @Whissi, @jasone)
- Fix OS X default zone replacement to work with OS X 10.12. (@glandium,
@jasone)
- Fix cached memory management to avoid needless commit/decommit operations
during purging, which resolves permanent virtual memory map fragmentation
issues on Windows. (@mjp41, @jasone)
- Fix TSD fetches to avoid (recursive) allocation. This is relevant to
non-TLS and Windows configurations. (@jasone)
- Fix malloc_conf overriding to work on Windows. (@jasone)
- Forcibly disable lazy-lock on Windows (was forcibly *enabled*). (@jasone)
* 4.2.1 (June 8, 2016)
Bug fixes:
- Fix bootstrapping issues for configurations that require allocation during
tsd initialization (e.g. --disable-tls). (@cferris1000, @jasone)
- Fix gettimeofday() version of nstime_update(). (@ronawho)
- Fix Valgrind regressions in calloc() and chunk_alloc_wrapper(). (@ronawho)
- Fix potential VM map fragmentation regression. (@jasone)
- Fix opt_zero-triggered in-place huge reallocation zeroing. (@jasone)
- Fix heap profiling context leaks in reallocation edge cases. (@jasone)
* 4.2.0 (May 12, 2016)
New features:
- Add the arena.<i>.reset mallctl, which makes it possible to discard all of
an arena's allocations in a single operation. (@jasone)
- Add the stats.retained and stats.arenas.<i>.retained statistics. (@jasone)
- Add the --with-version configure option. (@jasone)
- Support --with-lg-page values larger than actual page size. (@jasone)
Optimizations:
- Use pairing heaps rather than red-black trees for various hot data
structures. (@djwatson, @jasone)
- Streamline fast paths of rtree operations. (@jasone)
- Optimize the fast paths of calloc() and [m,d,sd]allocx(). (@jasone)
- Decommit unused virtual memory if the OS does not overcommit. (@jasone)
- Specify MAP_NORESERVE on Linux if [heuristic] overcommit is active, in order
to avoid unfortunate interactions during fork(2). (@jasone)
Bug fixes:
- Fix chunk accounting related to triggering gdump profiles. (@jasone)
- Link against librt for clock_gettime(2) if glibc < 2.17. (@jasone)
- Scale leak report summary according to sampling probability. (@jasone)
* 4.1.1 (May 3, 2016)
This bugfix release resolves a variety of mostly minor issues, though the
bitmap fix is critical for 64-bit Windows.
Bug fixes:
- Fix the linear scan version of bitmap_sfu() to shift by the proper amount
even when sizeof(long) is not the same as sizeof(void *), as on 64-bit
Windows. (@jasone)
- Fix hashing functions to avoid unaligned memory accesses (and resulting
crashes). This is relevant at least to some ARM-based platforms.
(@rkmisra)
- Fix fork()-related lock rank ordering reversals. These reversals were
unlikely to cause deadlocks in practice except when heap profiling was
enabled and active. (@jasone)
- Fix various chunk leaks in OOM code paths. (@jasone)
- Fix malloc_stats_print() to print opt.narenas correctly. (@jasone)
- Fix MSVC-specific build/test issues. (@rustyx, @yuslepukhin)
- Fix a variety of test failures that were due to test fragility rather than
core bugs. (@jasone)
* 4.1.0 (February 28, 2016)
This release is primarily about optimizations, but it also incorporates a lot
of portability-motivated refactoring and enhancements. Many people worked on
this release, to an extent that even with the omission here of minor changes
(see git revision history), and of the people who reported and diagnosed
issues, so much of the work was contributed that starting with this release,
changes are annotated with author credits to help reflect the collaborative
effort involved.
New features:
- Implement decay-based unused dirty page purging, a major optimization with
mallctl API impact. This is an alternative to the existing ratio-based
unused dirty page purging, and is intended to eventually become the sole
purging mechanism. New mallctls:
+ opt.purge
+ opt.decay_time
+ arena.<i>.decay
+ arena.<i>.decay_time
+ arenas.decay_time
+ stats.arenas.<i>.decay_time
(@jasone, @cevans87)
- Add --with-malloc-conf, which makes it possible to embed a default
options string during configuration. This was motivated by the desire to
specify --with-malloc-conf=purge:decay , since the default must remain
purge:ratio until the 5.0.0 release. (@jasone)
- Add MS Visual Studio 2015 support. (@rustyx, @yuslepukhin)
- Make *allocx() size class overflow behavior defined. The maximum
size class is now less than PTRDIFF_MAX to protect applications against
numerical overflow, and all allocation functions are guaranteed to indicate
errors rather than potentially crashing if the request size exceeds the
maximum size class. (@jasone)
- jeprof:
+ Add raw heap profile support. (@jasone)
+ Add --retain and --exclude for backtrace symbol filtering. (@jasone)
Optimizations:
- Optimize the fast path to combine various bootstrapping and configuration
checks and execute more streamlined code in the common case. (@interwq)
- Use linear scan for small bitmaps (used for small object tracking). In
addition to speeding up bitmap operations on 64-bit systems, this reduces
allocator metadata overhead by approximately 0.2%. (@djwatson)
- Separate arena_avail trees, which substantially speeds up run tree
operations. (@djwatson)
- Use memoization (boot-time-computed table) for run quantization. Separate
arena_avail trees reduced the importance of this optimization. (@jasone)
- Attempt mmap-based in-place huge reallocation. This can dramatically speed
up incremental huge reallocation. (@jasone)
Incompatible changes:
- Make opt.narenas unsigned rather than size_t. (@jasone)
Bug fixes:
- Fix stats.cactive accounting regression. (@rustyx, @jasone)
- Handle unaligned keys in hash(). This caused problems for some ARM systems.
(@jasone, @cferris1000)
- Refactor arenas array. In addition to fixing a fork-related deadlock, this
makes arena lookups faster and simpler. (@jasone)
- Move retained memory allocation out of the default chunk allocation
function, to a location that gets executed even if the application installs
a custom chunk allocation function. This resolves a virtual memory leak.
(@buchgr)
- Fix a potential tsd cleanup leak. (@cferris1000, @jasone)
- Fix run quantization. In practice this bug had no impact unless
applications requested memory with alignment exceeding one page.
(@jasone, @djwatson)
- Fix LinuxThreads-specific bootstrapping deadlock. (Cosmin Paraschiv)
- jeprof:
+ Don't discard curl options if timeout is not defined. (@djwatson)
+ Detect failed profile fetches. (@djwatson)
- Fix stats.arenas.<i>.{dss,lg_dirty_mult,decay_time,pactive,pdirty} for
--disable-stats case. (@jasone)
* 4.0.4 (October 24, 2015)
This bugfix release fixes another xallocx() regression. No other regressions
have come to light in over a month, so this is likely a good starting point
for people who prefer to wait for "dot one" releases with all the major issues
shaken out.
Bug fixes:
- Fix xallocx(..., MALLOCX_ZERO to zero the last full trailing page of large
allocations that have been randomly assigned an offset of 0 when
--enable-cache-oblivious configure option is enabled.
* 4.0.3 (September 24, 2015)
This bugfix release continues the trend of xallocx() and heap profiling fixes.
Bug fixes:
- Fix xallocx(..., MALLOCX_ZERO) to zero all trailing bytes of large
allocations when --enable-cache-oblivious configure option is enabled.
- Fix xallocx(..., MALLOCX_ZERO) to zero trailing bytes of huge allocations
when resizing from/to a size class that is not a multiple of the chunk size.
- Fix prof_tctx_dump_iter() to filter out nodes that were created after heap
profile dumping started.
- Work around a potentially bad thread-specific data initialization
interaction with NPTL (glibc's pthreads implementation).
* 4.0.2 (September 21, 2015)
This bugfix release addresses a few bugs specific to heap profiling.
Bug fixes:
- Fix ixallocx_prof_sample() to never modify nor create sampled small
allocations. xallocx() is in general incapable of moving small allocations,
so this fix removes buggy code without loss of generality.
- Fix irallocx_prof_sample() to always allocate large regions, even when
alignment is non-zero.
- Fix prof_alloc_rollback() to read tdata from thread-specific data rather
than dereferencing a potentially invalid tctx.
* 4.0.1 (September 15, 2015)
This is a bugfix release that is somewhat high risk due to the amount of
refactoring required to address deep xallocx() problems. As a side effect of
these fixes, xallocx() now tries harder to partially fulfill requests for
optional extra space. Note that a couple of minor heap profiling
optimizations are included, but these are better thought of as performance
fixes that were integral to disovering most of the other bugs.
Optimizations:
- Avoid a chunk metadata read in arena_prof_tctx_set(), since it is in the
fast path when heap profiling is enabled. Additionally, split a special
case out into arena_prof_tctx_reset(), which also avoids chunk metadata
reads.
- Optimize irallocx_prof() to optimistically update the sampler state. The
prior implementation appears to have been a holdover from when
rallocx()/xallocx() functionality was combined as rallocm().
Bug fixes:
- Fix TLS configuration such that it is enabled by default for platforms on
which it works correctly.
- Fix arenas_cache_cleanup() and arena_get_hard() to handle
allocation/deallocation within the application's thread-specific data
cleanup functions even after arenas_cache is torn down.
- Fix xallocx() bugs related to size+extra exceeding HUGE_MAXCLASS.
- Fix chunk purge hook calls for in-place huge shrinking reallocation to
specify the old chunk size rather than the new chunk size. This bug caused
no correctness issues for the default chunk purge function, but was
visible to custom functions set via the "arena.<i>.chunk_hooks" mallctl.
- Fix heap profiling bugs:
+ Fix heap profiling to distinguish among otherwise identical sample sites
with interposed resets (triggered via the "prof.reset" mallctl). This bug
could cause data structure corruption that would most likely result in a
segfault.
+ Fix irealloc_prof() to prof_alloc_rollback() on OOM.
+ Make one call to prof_active_get_unlocked() per allocation event, and use
the result throughout the relevant functions that handle an allocation
event. Also add a missing check in prof_realloc(). These fixes protect
allocation events against concurrent prof_active changes.
+ Fix ixallocx_prof() to pass usize_max and zero to ixallocx_prof_sample()
in the correct order.
+ Fix prof_realloc() to call prof_free_sampled_object() after calling
prof_malloc_sample_object(). Prior to this fix, if tctx and old_tctx were
the same, the tctx could have been prematurely destroyed.
- Fix portability bugs:
+ Don't bitshift by negative amounts when encoding/decoding run sizes in
chunk header maps. This affected systems with page sizes greater than 8
KiB.
+ Rename index_t to szind_t to avoid an existing type on Solaris.
+ Add JEMALLOC_CXX_THROW to the memalign() function prototype, in order to
match glibc and avoid compilation errors when including both
jemalloc/jemalloc.h and malloc.h in C++ code.
+ Don't assume that /bin/sh is appropriate when running size_classes.sh
during configuration.
+ Consider __sparcv9 a synonym for __sparc64__ when defining LG_QUANTUM.
+ Link tests to librt if it contains clock_gettime(2).
* 4.0.0 (August 17, 2015)
This version contains many speed and space optimizations, both minor and
major. The major themes are generalization, unification, and simplification.
Although many of these optimizations cause no visible behavior change, their
cumulative effect is substantial.
New features:
- Normalize size class spacing to be consistent across the complete size
range. By default there are four size classes per size doubling, but this
is now configurable via the --with-lg-size-class-group option. Also add the
--with-lg-page, --with-lg-page-sizes, --with-lg-quantum, and
--with-lg-tiny-min options, which can be used to tweak page and size class
settings. Impacts:
+ Worst case performance for incrementally growing/shrinking reallocation
is improved because there are far fewer size classes, and therefore
copying happens less often.
+ Internal fragmentation is limited to 20% for all but the smallest size
classes (those less than four times the quantum). (1B + 4 KiB)
and (1B + 4 MiB) previously suffered nearly 50% internal fragmentation.
+ Chunk fragmentation tends to be lower because there are fewer distinct run
sizes to pack.
- Add support for explicit tcaches. The "tcache.create", "tcache.flush", and
"tcache.destroy" mallctls control tcache lifetime and flushing, and the
MALLOCX_TCACHE(tc) and MALLOCX_TCACHE_NONE flags to the *allocx() API
control which tcache is used for each operation.
- Implement per thread heap profiling, as well as the ability to
enable/disable heap profiling on a per thread basis. Add the "prof.reset",
"prof.lg_sample", "thread.prof.name", "thread.prof.active",
"opt.prof_thread_active_init", "prof.thread_active_init", and
"thread.prof.active" mallctls.
- Add support for per arena application-specified chunk allocators, configured
via the "arena.<i>.chunk_hooks" mallctl.
- Refactor huge allocation to be managed by arenas, so that arenas now
function as general purpose independent allocators. This is important in
the context of user-specified chunk allocators, aside from the scalability
benefits. Related new statistics:
+ The "stats.arenas.<i>.huge.allocated", "stats.arenas.<i>.huge.nmalloc",
"stats.arenas.<i>.huge.ndalloc", and "stats.arenas.<i>.huge.nrequests"
mallctls provide high level per arena huge allocation statistics.
+ The "arenas.nhchunks", "arenas.hchunk.<i>.size",
"stats.arenas.<i>.hchunks.<j>.nmalloc",
"stats.arenas.<i>.hchunks.<j>.ndalloc",
"stats.arenas.<i>.hchunks.<j>.nrequests", and
"stats.arenas.<i>.hchunks.<j>.curhchunks" mallctls provide per size class
statistics.
- Add the 'util' column to malloc_stats_print() output, which reports the
proportion of available regions that are currently in use for each small
size class.
- Add "alloc" and "free" modes for for junk filling (see the "opt.junk"
mallctl), so that it is possible to separately enable junk filling for
allocation versus deallocation.
- Add the jemalloc-config script, which provides information about how
jemalloc was configured, and how to integrate it into application builds.
- Add metadata statistics, which are accessible via the "stats.metadata",
"stats.arenas.<i>.metadata.mapped", and
"stats.arenas.<i>.metadata.allocated" mallctls.
- Add the "stats.resident" mallctl, which reports the upper limit of
physically resident memory mapped by the allocator.
- Add per arena control over unused dirty page purging, via the
"arenas.lg_dirty_mult", "arena.<i>.lg_dirty_mult", and
"stats.arenas.<i>.lg_dirty_mult" mallctls.
- Add the "prof.gdump" mallctl, which makes it possible to toggle the gdump
feature on/off during program execution.
- Add sdallocx(), which implements sized deallocation. The primary
optimization over dallocx() is the removal of a metadata read, which often
suffers an L1 cache miss.
- Add missing header includes in jemalloc/jemalloc.h, so that applications
only have to #include <jemalloc/jemalloc.h>.
- Add support for additional platforms:
+ Bitrig
+ Cygwin
+ DragonFlyBSD
+ iOS
+ OpenBSD
+ OpenRISC/or1k
Optimizations:
- Maintain dirty runs in per arena LRUs rather than in per arena trees of
dirty-run-containing chunks. In practice this change significantly reduces
dirty page purging volume.
- Integrate whole chunks into the unused dirty page purging machinery. This
reduces the cost of repeated huge allocation/deallocation, because it
effectively introduces a cache of chunks.
- Split the arena chunk map into two separate arrays, in order to increase
cache locality for the frequently accessed bits.
- Move small run metadata out of runs, into arena chunk headers. This reduces
run fragmentation, smaller runs reduce external fragmentation for small size
classes, and packed (less uniformly aligned) metadata layout improves CPU
cache set distribution.
- Randomly distribute large allocation base pointer alignment relative to page
boundaries in order to more uniformly utilize CPU cache sets. This can be
disabled via the --disable-cache-oblivious configure option, and queried via
the "config.cache_oblivious" mallctl.
- Micro-optimize the fast paths for the public API functions.
- Refactor thread-specific data to reside in a single structure. This assures
that only a single TLS read is necessary per call into the public API.
- Implement in-place huge allocation growing and shrinking.
- Refactor rtree (radix tree for chunk lookups) to be lock-free, and make
additional optimizations that reduce maximum lookup depth to one or two
levels. This resolves what was a concurrency bottleneck for per arena huge
allocation, because a global data structure is critical for determining
which arenas own which huge allocations.
Incompatible changes:
- Replace --enable-cc-silence with --disable-cc-silence to suppress spurious
warnings by default.
- Assure that the constness of malloc_usable_size()'s return type matches that
of the system implementation.
- Change the heap profile dump format to support per thread heap profiling,
rename pprof to jeprof, and enhance it with the --thread=<n> option. As a
result, the bundled jeprof must now be used rather than the upstream
(gperftools) pprof.
- Disable "opt.prof_final" by default, in order to avoid atexit(3), which can
internally deadlock on some platforms.
- Change the "arenas.nlruns" mallctl type from size_t to unsigned.
- Replace the "stats.arenas.<i>.bins.<j>.allocated" mallctl with
"stats.arenas.<i>.bins.<j>.curregs".
- Ignore MALLOC_CONF in set{uid,gid,cap} binaries.
- Ignore MALLOCX_ARENA(a) in dallocx(), in favor of using the
MALLOCX_TCACHE(tc) and MALLOCX_TCACHE_NONE flags to control tcache usage.
Removed features:
- Remove the *allocm() API, which is superseded by the *allocx() API.
- Remove the --enable-dss options, and make dss non-optional on all platforms
which support sbrk(2).
- Remove the "arenas.purge" mallctl, which was obsoleted by the
"arena.<i>.purge" mallctl in 3.1.0.
- Remove the unnecessary "opt.valgrind" mallctl; jemalloc automatically
detects whether it is running inside Valgrind.
- Remove the "stats.huge.allocated", "stats.huge.nmalloc", and
"stats.huge.ndalloc" mallctls.
- Remove the --enable-mremap option.
- Remove the "stats.chunks.current", "stats.chunks.total", and
"stats.chunks.high" mallctls.
Bug fixes:
- Fix the cactive statistic to decrease (rather than increase) when active
memory decreases. This regression was first released in 3.5.0.
- Fix OOM handling in memalign() and valloc(). A variant of this bug existed
in all releases since 2.0.0, which introduced these functions.
- Fix an OOM-related regression in arena_tcache_fill_small(), which could
cause cache corruption on OOM. This regression was present in all releases
from 2.2.0 through 3.6.0.
- Fix size class overflow handling for malloc(), posix_memalign(), memalign(),
calloc(), and realloc() when profiling is enabled.
- Fix the "arena.<i>.dss" mallctl to return an error if "primary" or
"secondary" precedence is specified, but sbrk(2) is not supported.
- Fix fallback lg_floor() implementations to handle extremely large inputs.
- Ensure the default purgeable zone is after the default zone on OS X.
- Fix latent bugs in atomic_*().
- Fix the "arena.<i>.dss" mallctl to handle read-only calls.
- Fix tls_model configuration to enable the initial-exec model when possible.
- Mark malloc_conf as a weak symbol so that the application can override it.
- Correctly detect glibc's adaptive pthread mutexes.
- Fix the --without-export configure option.
* 3.6.0 (March 31, 2014)
This version contains a critical bug fix for a regression present in 3.5.0 and
@@ -21,7 +738,7 @@ found in the git revision history:
backtracing to be reliable.
- Use dss allocation precedence for huge allocations as well as small/large
allocations.
- Fix test assertion failure message formatting. This bug did not manifect on
- Fix test assertion failure message formatting. This bug did not manifest on
x86_64 systems because of implementation subtleties in va_list.
- Fix inconsequential test failures for hash and SFMT code.
@@ -516,7 +1233,7 @@ found in the git revision history:
- Make it possible for the application to manually flush a thread's cache, via
the "tcache.flush" mallctl.
- Base maximum dirty page count on proportion of active memory.
- Compute various addtional run-time statistics, including per size class
- Compute various additional run-time statistics, including per size class
statistics for large objects.
- Expose malloc_stats_print(), which can be called repeatedly by the
application.

14
deps/jemalloc/README vendored
View File

@@ -3,12 +3,12 @@ fragmentation avoidance and scalable concurrency support. jemalloc first came
into use as the FreeBSD libc allocator in 2005, and since then it has found its
way into numerous applications that rely on its predictable behavior. In 2010
jemalloc development efforts broadened to include developer support features
such as heap profiling, Valgrind integration, and extensive monitoring/tuning
hooks. Modern jemalloc releases continue to be integrated back into FreeBSD,
and therefore versatility remains critical. Ongoing development efforts trend
toward making jemalloc among the best allocators for a broad range of demanding
applications, and eliminating/mitigating weaknesses that have practical
repercussions for real world applications.
such as heap profiling and extensive monitoring/tuning hooks. Modern jemalloc
releases continue to be integrated back into FreeBSD, and therefore versatility
remains critical. Ongoing development efforts trend toward making jemalloc
among the best allocators for a broad range of demanding applications, and
eliminating/mitigating weaknesses that have practical repercussions for real
world applications.
The COPYING file contains copyright and licensing information.
@@ -17,4 +17,4 @@ jemalloc.
The ChangeLog file contains a brief summary of changes for each release.
URL: http://www.canonware.com/jemalloc/
URL: http://jemalloc.net/

363
deps/jemalloc/include/jemalloc/internal/atomic.h vendored
View File

@@ -1,304 +1,77 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_ATOMIC_H
#define JEMALLOC_INTERNAL_ATOMIC_H
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#define ATOMIC_INLINE static inline
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#define atomic_read_uint64(p) atomic_add_uint64(p, 0)
#define atomic_read_uint32(p) atomic_add_uint32(p, 0)
#define atomic_read_z(p) atomic_add_z(p, 0)
#define atomic_read_u(p) atomic_add_u(p, 0)
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
uint64_t atomic_add_uint64(uint64_t *p, uint64_t x);
uint64_t atomic_sub_uint64(uint64_t *p, uint64_t x);
uint32_t atomic_add_uint32(uint32_t *p, uint32_t x);
uint32_t atomic_sub_uint32(uint32_t *p, uint32_t x);
size_t atomic_add_z(size_t *p, size_t x);
size_t atomic_sub_z(size_t *p, size_t x);
unsigned atomic_add_u(unsigned *p, unsigned x);
unsigned atomic_sub_u(unsigned *p, unsigned x);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_ATOMIC_C_))
/******************************************************************************/
/* 64-bit operations. */
#if (LG_SIZEOF_PTR == 3 || LG_SIZEOF_INT == 3)
# ifdef __GCC_HAVE_SYNC_COMPARE_AND_SWAP_8
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (__sync_sub_and_fetch(p, x));
}
#elif (defined(_MSC_VER))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (InterlockedExchangeAdd64(p, x));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (InterlockedExchangeAdd64(p, -((int64_t)x)));
}
#elif (defined(JEMALLOC_OSATOMIC))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (OSAtomicAdd64((int64_t)x, (int64_t *)p));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (OSAtomicAdd64(-((int64_t)x), (int64_t *)p));
}
# elif (defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
asm volatile (
"lock; xaddq %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
x = (uint64_t)(-(int64_t)x);
asm volatile (
"lock; xaddq %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
# elif (defined(JEMALLOC_ATOMIC9))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
/*
* atomic_fetchadd_64() doesn't exist, but we only ever use this
* function on LP64 systems, so atomic_fetchadd_long() will do.
*/
assert(sizeof(uint64_t) == sizeof(unsigned long));
return (atomic_fetchadd_long(p, (unsigned long)x) + x);
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
assert(sizeof(uint64_t) == sizeof(unsigned long));
return (atomic_fetchadd_long(p, (unsigned long)(-(long)x)) - x);
}
# elif (defined(JE_FORCE_SYNC_COMPARE_AND_SWAP_8))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (__sync_sub_and_fetch(p, x));
}
# else
# error "Missing implementation for 64-bit atomic operations"
# endif
#endif
/******************************************************************************/
/* 32-bit operations. */
#ifdef __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (__sync_sub_and_fetch(p, x));
}
#elif (defined(_MSC_VER))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (InterlockedExchangeAdd(p, x));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (InterlockedExchangeAdd(p, -((int32_t)x)));
}
#elif (defined(JEMALLOC_OSATOMIC))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (OSAtomicAdd32((int32_t)x, (int32_t *)p));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (OSAtomicAdd32(-((int32_t)x), (int32_t *)p));
}
#elif (defined(__i386__) || defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
asm volatile (
"lock; xaddl %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
x = (uint32_t)(-(int32_t)x);
asm volatile (
"lock; xaddl %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
#elif (defined(JEMALLOC_ATOMIC9))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (atomic_fetchadd_32(p, x) + x);
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (atomic_fetchadd_32(p, (uint32_t)(-(int32_t)x)) - x);
}
#elif (defined(JE_FORCE_SYNC_COMPARE_AND_SWAP_4))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (__sync_sub_and_fetch(p, x));
}
#if defined(JEMALLOC_GCC_ATOMIC_ATOMICS)
# include "jemalloc/internal/atomic_gcc_atomic.h"
#elif defined(JEMALLOC_GCC_SYNC_ATOMICS)
# include "jemalloc/internal/atomic_gcc_sync.h"
#elif defined(_MSC_VER)
# include "jemalloc/internal/atomic_msvc.h"
#elif defined(JEMALLOC_C11_ATOMICS)
# include "jemalloc/internal/atomic_c11.h"
#else
# error "Missing implementation for 32-bit atomic operations"
# error "Don't have atomics implemented on this platform."
#endif
/******************************************************************************/
/* size_t operations. */
JEMALLOC_INLINE size_t
atomic_add_z(size_t *p, size_t x)
{
/*
* This header gives more or less a backport of C11 atomics. The user can write
* JEMALLOC_GENERATE_ATOMICS(type, short_type, lg_sizeof_type); to generate
* counterparts of the C11 atomic functions for type, as so:
* JEMALLOC_GENERATE_ATOMICS(int *, pi, 3);
* and then write things like:
* int *some_ptr;
* atomic_pi_t atomic_ptr_to_int;
* atomic_store_pi(&atomic_ptr_to_int, some_ptr, ATOMIC_RELAXED);
* int *prev_value = atomic_exchange_pi(&ptr_to_int, NULL, ATOMIC_ACQ_REL);
* assert(some_ptr == prev_value);
* and expect things to work in the obvious way.
*
* Also included (with naming differences to avoid conflicts with the standard
* library):
* atomic_fence(atomic_memory_order_t) (mimics C11's atomic_thread_fence).
* ATOMIC_INIT (mimics C11's ATOMIC_VAR_INIT).
*/
#if (LG_SIZEOF_PTR == 3)
return ((size_t)atomic_add_uint64((uint64_t *)p, (uint64_t)x));
#elif (LG_SIZEOF_PTR == 2)
return ((size_t)atomic_add_uint32((uint32_t *)p, (uint32_t)x));
#endif
}
/*
* Pure convenience, so that we don't have to type "atomic_memory_order_"
* quite so often.
*/
#define ATOMIC_RELAXED atomic_memory_order_relaxed
#define ATOMIC_ACQUIRE atomic_memory_order_acquire
#define ATOMIC_RELEASE atomic_memory_order_release
#define ATOMIC_ACQ_REL atomic_memory_order_acq_rel
#define ATOMIC_SEQ_CST atomic_memory_order_seq_cst
JEMALLOC_INLINE size_t
atomic_sub_z(size_t *p, size_t x)
{
#if (LG_SIZEOF_PTR == 3)
return ((size_t)atomic_add_uint64((uint64_t *)p,
(uint64_t)-((int64_t)x)));
#elif (LG_SIZEOF_PTR == 2)
return ((size_t)atomic_add_uint32((uint32_t *)p,
(uint32_t)-((int32_t)x)));
#endif
}
/******************************************************************************/
/* unsigned operations. */
JEMALLOC_INLINE unsigned
atomic_add_u(unsigned *p, unsigned x)
{
#if (LG_SIZEOF_INT == 3)
return ((unsigned)atomic_add_uint64((uint64_t *)p, (uint64_t)x));
#elif (LG_SIZEOF_INT == 2)
return ((unsigned)atomic_add_uint32((uint32_t *)p, (uint32_t)x));
#endif
}
JEMALLOC_INLINE unsigned
atomic_sub_u(unsigned *p, unsigned x)
{
#if (LG_SIZEOF_INT == 3)
return ((unsigned)atomic_add_uint64((uint64_t *)p,
(uint64_t)-((int64_t)x)));
#elif (LG_SIZEOF_INT == 2)
return ((unsigned)atomic_add_uint32((uint32_t *)p,
(uint32_t)-((int32_t)x)));
#endif
}
/******************************************************************************/
/*
* Not all platforms have 64-bit atomics. If we do, this #define exposes that
* fact.
*/
#if (LG_SIZEOF_PTR == 3 || LG_SIZEOF_INT == 3)
# define JEMALLOC_ATOMIC_U64
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
JEMALLOC_GENERATE_ATOMICS(void *, p, LG_SIZEOF_PTR)
/*
* There's no actual guarantee that sizeof(bool) == 1, but it's true on the only
* platform that actually needs to know the size, MSVC.
*/
JEMALLOC_GENERATE_ATOMICS(bool, b, 0)
JEMALLOC_GENERATE_INT_ATOMICS(unsigned, u, LG_SIZEOF_INT)
JEMALLOC_GENERATE_INT_ATOMICS(size_t, zu, LG_SIZEOF_PTR)
JEMALLOC_GENERATE_INT_ATOMICS(ssize_t, zd, LG_SIZEOF_PTR)
JEMALLOC_GENERATE_INT_ATOMICS(uint32_t, u32, 2)
#ifdef JEMALLOC_ATOMIC_U64
JEMALLOC_GENERATE_INT_ATOMICS(uint64_t, u64, 3)
#endif
#undef ATOMIC_INLINE
#endif /* JEMALLOC_INTERNAL_ATOMIC_H */

349
deps/jemalloc/include/jemalloc/internal/bitmap.h vendored
View File

@@ -1,37 +1,159 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_BITMAP_H
#define JEMALLOC_INTERNAL_BITMAP_H
#include "jemalloc/internal/arena_types.h"
#include "jemalloc/internal/bit_util.h"
#include "jemalloc/internal/size_classes.h"
typedef unsigned long bitmap_t;
#define LG_SIZEOF_BITMAP LG_SIZEOF_LONG
/* Maximum bitmap bit count is 2^LG_BITMAP_MAXBITS. */
#define LG_BITMAP_MAXBITS LG_RUN_MAXREGS
typedef struct bitmap_level_s bitmap_level_t;
typedef struct bitmap_info_s bitmap_info_t;
typedef unsigned long bitmap_t;
#define LG_SIZEOF_BITMAP LG_SIZEOF_LONG
#if LG_SLAB_MAXREGS > LG_CEIL_NSIZES
/* Maximum bitmap bit count is determined by maximum regions per slab. */
# define LG_BITMAP_MAXBITS LG_SLAB_MAXREGS
#else
/* Maximum bitmap bit count is determined by number of extent size classes. */
# define LG_BITMAP_MAXBITS LG_CEIL_NSIZES
#endif
#define BITMAP_MAXBITS (ZU(1) << LG_BITMAP_MAXBITS)
/* Number of bits per group. */
#define LG_BITMAP_GROUP_NBITS (LG_SIZEOF_BITMAP + 3)
#define BITMAP_GROUP_NBITS (ZU(1) << LG_BITMAP_GROUP_NBITS)
#define BITMAP_GROUP_NBITS_MASK (BITMAP_GROUP_NBITS-1)
#define LG_BITMAP_GROUP_NBITS (LG_SIZEOF_BITMAP + 3)
#define BITMAP_GROUP_NBITS (1U << LG_BITMAP_GROUP_NBITS)
#define BITMAP_GROUP_NBITS_MASK (BITMAP_GROUP_NBITS-1)
/* Maximum number of levels possible. */
#define BITMAP_MAX_LEVELS \
(LG_BITMAP_MAXBITS / LG_SIZEOF_BITMAP) \
+ !!(LG_BITMAP_MAXBITS % LG_SIZEOF_BITMAP)
/*
* Do some analysis on how big the bitmap is before we use a tree. For a brute
* force linear search, if we would have to call ffs_lu() more than 2^3 times,
* use a tree instead.
*/
#if LG_BITMAP_MAXBITS - LG_BITMAP_GROUP_NBITS > 3
# define BITMAP_USE_TREE
#endif
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* Number of groups required to store a given number of bits. */
#define BITMAP_BITS2GROUPS(nbits) \
(((nbits) + BITMAP_GROUP_NBITS_MASK) >> LG_BITMAP_GROUP_NBITS)
struct bitmap_level_s {
/*
* Number of groups required at a particular level for a given number of bits.
*/
#define BITMAP_GROUPS_L0(nbits) \
BITMAP_BITS2GROUPS(nbits)
#define BITMAP_GROUPS_L1(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(nbits))
#define BITMAP_GROUPS_L2(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS((nbits))))
#define BITMAP_GROUPS_L3(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS( \
BITMAP_BITS2GROUPS((nbits)))))
#define BITMAP_GROUPS_L4(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS( \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS((nbits))))))
/*
* Assuming the number of levels, number of groups required for a given number
* of bits.
*/
#define BITMAP_GROUPS_1_LEVEL(nbits) \
BITMAP_GROUPS_L0(nbits)
#define BITMAP_GROUPS_2_LEVEL(nbits) \
(BITMAP_GROUPS_1_LEVEL(nbits) + BITMAP_GROUPS_L1(nbits))
#define BITMAP_GROUPS_3_LEVEL(nbits) \
(BITMAP_GROUPS_2_LEVEL(nbits) + BITMAP_GROUPS_L2(nbits))
#define BITMAP_GROUPS_4_LEVEL(nbits) \
(BITMAP_GROUPS_3_LEVEL(nbits) + BITMAP_GROUPS_L3(nbits))
#define BITMAP_GROUPS_5_LEVEL(nbits) \
(BITMAP_GROUPS_4_LEVEL(nbits) + BITMAP_GROUPS_L4(nbits))
/*
* Maximum number of groups required to support LG_BITMAP_MAXBITS.
*/
#ifdef BITMAP_USE_TREE
#if LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_1_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_1_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 2
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_2_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_2_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 3
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_3_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_3_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 4
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_4_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_4_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 5
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_5_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_5_LEVEL(BITMAP_MAXBITS)
#else
# error "Unsupported bitmap size"
#endif
/*
* Maximum number of levels possible. This could be statically computed based
* on LG_BITMAP_MAXBITS:
*
* #define BITMAP_MAX_LEVELS \
* (LG_BITMAP_MAXBITS / LG_SIZEOF_BITMAP) \
* + !!(LG_BITMAP_MAXBITS % LG_SIZEOF_BITMAP)
*
* However, that would not allow the generic BITMAP_INFO_INITIALIZER() macro, so
* instead hardcode BITMAP_MAX_LEVELS to the largest number supported by the
* various cascading macros. The only additional cost this incurs is some
* unused trailing entries in bitmap_info_t structures; the bitmaps themselves
* are not impacted.
*/
#define BITMAP_MAX_LEVELS 5
#define BITMAP_INFO_INITIALIZER(nbits) { \
/* nbits. */ \
nbits, \
/* nlevels. */ \
(BITMAP_GROUPS_L0(nbits) > BITMAP_GROUPS_L1(nbits)) + \
(BITMAP_GROUPS_L1(nbits) > BITMAP_GROUPS_L2(nbits)) + \
(BITMAP_GROUPS_L2(nbits) > BITMAP_GROUPS_L3(nbits)) + \
(BITMAP_GROUPS_L3(nbits) > BITMAP_GROUPS_L4(nbits)) + 1, \
/* levels. */ \
{ \
{0}, \
{BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L1(nbits) + BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L2(nbits) + BITMAP_GROUPS_L1(nbits) + \
BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L3(nbits) + BITMAP_GROUPS_L2(nbits) + \
BITMAP_GROUPS_L1(nbits) + BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L4(nbits) + BITMAP_GROUPS_L3(nbits) + \
BITMAP_GROUPS_L2(nbits) + BITMAP_GROUPS_L1(nbits) \
+ BITMAP_GROUPS_L0(nbits)} \
} \
}
#else /* BITMAP_USE_TREE */
#define BITMAP_GROUPS(nbits) BITMAP_BITS2GROUPS(nbits)
#define BITMAP_GROUPS_MAX BITMAP_BITS2GROUPS(BITMAP_MAXBITS)
#define BITMAP_INFO_INITIALIZER(nbits) { \
/* nbits. */ \
nbits, \
/* ngroups. */ \
BITMAP_BITS2GROUPS(nbits) \
}
#endif /* BITMAP_USE_TREE */
typedef struct bitmap_level_s {
/* Offset of this level's groups within the array of groups. */
size_t group_offset;
};
} bitmap_level_t;
struct bitmap_info_s {
typedef struct bitmap_info_s {
/* Logical number of bits in bitmap (stored at bottom level). */
size_t nbits;
#ifdef BITMAP_USE_TREE
/* Number of levels necessary for nbits. */
unsigned nlevels;
@@ -40,67 +162,62 @@ struct bitmap_info_s {
* bottom to top (e.g. the bottom level is stored in levels[0]).
*/
bitmap_level_t levels[BITMAP_MAX_LEVELS+1];
};
#else /* BITMAP_USE_TREE */
/* Number of groups necessary for nbits. */
size_t ngroups;
#endif /* BITMAP_USE_TREE */
} bitmap_info_t;
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
void bitmap_info_init(bitmap_info_t *binfo, size_t nbits);
void bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo, bool fill);
size_t bitmap_size(const bitmap_info_t *binfo);
void bitmap_info_init(bitmap_info_t *binfo, size_t nbits);
size_t bitmap_info_ngroups(const bitmap_info_t *binfo);
size_t bitmap_size(size_t nbits);
void bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
bool bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo);
bool bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit);
void bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit);
size_t bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo);
void bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_BITMAP_C_))
JEMALLOC_INLINE bool
bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo)
{
unsigned rgoff = binfo->levels[binfo->nlevels].group_offset - 1;
static inline bool
bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo) {
#ifdef BITMAP_USE_TREE
size_t rgoff = binfo->levels[binfo->nlevels].group_offset - 1;
bitmap_t rg = bitmap[rgoff];
/* The bitmap is full iff the root group is 0. */
return (rg == 0);
#else
size_t i;
for (i = 0; i < binfo->ngroups; i++) {
if (bitmap[i] != 0) {
return false;
}
}
return true;
#endif
}
JEMALLOC_INLINE bool
bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
{
static inline bool
bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) {
size_t goff;
bitmap_t g;
assert(bit < binfo->nbits);
goff = bit >> LG_BITMAP_GROUP_NBITS;
g = bitmap[goff];
return (!(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))));
return !(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)));
}
JEMALLOC_INLINE void
bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
{
static inline void
bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) {
size_t goff;
bitmap_t *gp;
bitmap_t g;
assert(bit < binfo->nbits);
assert(bitmap_get(bitmap, binfo, bit) == false);
assert(!bitmap_get(bitmap, binfo, bit));
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[goff];
g = *gp;
assert(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
assert(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
assert(bitmap_get(bitmap, binfo, bit));
#ifdef BITMAP_USE_TREE
/* Propagate group state transitions up the tree. */
if (g == 0) {
unsigned i;
@@ -109,45 +226,113 @@ bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[binfo->levels[i].group_offset + goff];
g = *gp;
assert(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
assert(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
if (g != 0)
if (g != 0) {
break;
}
}
}
#endif
}
/* ffu: find first unset >= bit. */
static inline size_t
bitmap_ffu(const bitmap_t *bitmap, const bitmap_info_t *binfo, size_t min_bit) {
assert(min_bit < binfo->nbits);
#ifdef BITMAP_USE_TREE
size_t bit = 0;
for (unsigned level = binfo->nlevels; level--;) {
size_t lg_bits_per_group = (LG_BITMAP_GROUP_NBITS * (level +
1));
bitmap_t group = bitmap[binfo->levels[level].group_offset + (bit
>> lg_bits_per_group)];
unsigned group_nmask = (unsigned)(((min_bit > bit) ? (min_bit -
bit) : 0) >> (lg_bits_per_group - LG_BITMAP_GROUP_NBITS));
assert(group_nmask <= BITMAP_GROUP_NBITS);
bitmap_t group_mask = ~((1LU << group_nmask) - 1);
bitmap_t group_masked = group & group_mask;
if (group_masked == 0LU) {
if (group == 0LU) {
return binfo->nbits;
}
/*
* min_bit was preceded by one or more unset bits in
* this group, but there are no other unset bits in this
* group. Try again starting at the first bit of the
* next sibling. This will recurse at most once per
* non-root level.
*/
size_t sib_base = bit + (ZU(1) << lg_bits_per_group);
assert(sib_base > min_bit);
assert(sib_base > bit);
if (sib_base >= binfo->nbits) {
return binfo->nbits;
}
return bitmap_ffu(bitmap, binfo, sib_base);
}
bit += ((size_t)(ffs_lu(group_masked) - 1)) <<
(lg_bits_per_group - LG_BITMAP_GROUP_NBITS);
}
assert(bit >= min_bit);
assert(bit < binfo->nbits);
return bit;
#else
size_t i = min_bit >> LG_BITMAP_GROUP_NBITS;
bitmap_t g = bitmap[i] & ~((1LU << (min_bit & BITMAP_GROUP_NBITS_MASK))
- 1);
size_t bit;
do {
bit = ffs_lu(g);
if (bit != 0) {
return (i << LG_BITMAP_GROUP_NBITS) + (bit - 1);
}
i++;
g = bitmap[i];
} while (i < binfo->ngroups);
return binfo->nbits;
#endif
}
/* sfu: set first unset. */
JEMALLOC_INLINE size_t
bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo)
{
static inline size_t
bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo) {
size_t bit;
bitmap_t g;
unsigned i;
assert(bitmap_full(bitmap, binfo) == false);
assert(!bitmap_full(bitmap, binfo));
#ifdef BITMAP_USE_TREE
i = binfo->nlevels - 1;
g = bitmap[binfo->levels[i].group_offset];
bit = ffsl(g) - 1;
bit = ffs_lu(g) - 1;
while (i > 0) {
i--;
g = bitmap[binfo->levels[i].group_offset + bit];
bit = (bit << LG_BITMAP_GROUP_NBITS) + (ffsl(g) - 1);
bit = (bit << LG_BITMAP_GROUP_NBITS) + (ffs_lu(g) - 1);
}
#else
i = 0;
g = bitmap[0];
while ((bit = ffs_lu(g)) == 0) {
i++;
g = bitmap[i];
}
bit = (i << LG_BITMAP_GROUP_NBITS) + (bit - 1);
#endif
bitmap_set(bitmap, binfo, bit);
return (bit);
return bit;
}
JEMALLOC_INLINE void
bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
{
static inline void
bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) {
size_t goff;
bitmap_t *gp;
bitmap_t g;
bool propagate;
UNUSED bool propagate;
assert(bit < binfo->nbits);
assert(bitmap_get(bitmap, binfo, bit));
@@ -155,10 +340,11 @@ bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
gp = &bitmap[goff];
g = *gp;
propagate = (g == 0);
assert((g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))) == 0);
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
assert((g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK))) == 0);
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
assert(bitmap_get(bitmap, binfo, bit) == false);
assert(!bitmap_get(bitmap, binfo, bit));
#ifdef BITMAP_USE_TREE
/* Propagate group state transitions up the tree. */
if (propagate) {
unsigned i;
@@ -168,17 +354,16 @@ bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
gp = &bitmap[binfo->levels[i].group_offset + goff];
g = *gp;
propagate = (g == 0);
assert((g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK)))
assert((g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)))
== 0);
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
if (propagate == false)
if (!propagate) {
break;
}
}
}
#endif /* BITMAP_USE_TREE */
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_INTERNAL_BITMAP_H */

117
deps/jemalloc/include/jemalloc/internal/ckh.h vendored
View File

@@ -1,88 +1,101 @@
#ifndef JEMALLOC_INTERNAL_CKH_H
#define JEMALLOC_INTERNAL_CKH_H
#include "jemalloc/internal/tsd.h"
/* Cuckoo hashing implementation. Skip to the end for the interface. */
/******************************************************************************/
/* INTERNAL DEFINITIONS -- IGNORE */
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct ckh_s ckh_t;
typedef struct ckhc_s ckhc_t;
/* Typedefs to allow easy function pointer passing. */
typedef void ckh_hash_t (const void *, size_t[2]);
typedef bool ckh_keycomp_t (const void *, const void *);
/* Maintain counters used to get an idea of performance. */
/* #define CKH_COUNT */
/* #define CKH_COUNT */
/* Print counter values in ckh_delete() (requires CKH_COUNT). */
/* #define CKH_VERBOSE */
/* #define CKH_VERBOSE */
/*
* There are 2^LG_CKH_BUCKET_CELLS cells in each hash table bucket. Try to fit
* one bucket per L1 cache line.
*/
#define LG_CKH_BUCKET_CELLS (LG_CACHELINE - LG_SIZEOF_PTR - 1)
#define LG_CKH_BUCKET_CELLS (LG_CACHELINE - LG_SIZEOF_PTR - 1)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* Typedefs to allow easy function pointer passing. */
typedef void ckh_hash_t (const void *, size_t[2]);
typedef bool ckh_keycomp_t (const void *, const void *);
/* Hash table cell. */
struct ckhc_s {
const void *key;
const void *data;
};
typedef struct {
const void *key;
const void *data;
} ckhc_t;
struct ckh_s {
/* The hash table itself. */
typedef struct {
#ifdef CKH_COUNT
/* Counters used to get an idea of performance. */
uint64_t ngrows;
uint64_t nshrinks;
uint64_t nshrinkfails;
uint64_t ninserts;
uint64_t nrelocs;
uint64_t ngrows;
uint64_t nshrinks;
uint64_t nshrinkfails;
uint64_t ninserts;
uint64_t nrelocs;
#endif
/* Used for pseudo-random number generation. */
#define CKH_A 1103515241
#define CKH_C 12347
uint32_t prng_state;
uint64_t prng_state;
/* Total number of items. */
size_t count;
size_t count;
/*
* Minimum and current number of hash table buckets. There are
* 2^LG_CKH_BUCKET_CELLS cells per bucket.
*/
unsigned lg_minbuckets;
unsigned lg_curbuckets;
unsigned lg_minbuckets;
unsigned lg_curbuckets;
/* Hash and comparison functions. */
ckh_hash_t *hash;
ckh_keycomp_t *keycomp;
ckh_hash_t *hash;
ckh_keycomp_t *keycomp;
/* Hash table with 2^lg_curbuckets buckets. */
ckhc_t *tab;
};
ckhc_t *tab;
} ckh_t;
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
/* BEGIN PUBLIC API */
/******************************************************************************/
bool ckh_new(ckh_t *ckh, size_t minitems, ckh_hash_t *hash,
/* Lifetime management. Minitems is the initial capacity. */
bool ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash,
ckh_keycomp_t *keycomp);
void ckh_delete(ckh_t *ckh);
size_t ckh_count(ckh_t *ckh);
bool ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data);
bool ckh_insert(ckh_t *ckh, const void *key, const void *data);
bool ckh_remove(ckh_t *ckh, const void *searchkey, void **key,
void ckh_delete(tsd_t *tsd, ckh_t *ckh);
/* Get the number of elements in the set. */
size_t ckh_count(ckh_t *ckh);
/*
* To iterate over the elements in the table, initialize *tabind to 0 and call
* this function until it returns true. Each call that returns false will
* update *key and *data to the next element in the table, assuming the pointers
* are non-NULL.
*/
bool ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data);
/*
* Basic hash table operations -- insert, removal, lookup. For ckh_remove and
* ckh_search, key or data can be NULL. The hash-table only stores pointers to
* the key and value, and doesn't do any lifetime management.
*/
bool ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data);
bool ckh_remove(tsd_t *tsd, ckh_t *ckh, const void *searchkey, void **key,
void **data);
bool ckh_search(ckh_t *ckh, const void *seachkey, void **key, void **data);
void ckh_string_hash(const void *key, size_t r_hash[2]);
bool ckh_string_keycomp(const void *k1, const void *k2);
void ckh_pointer_hash(const void *key, size_t r_hash[2]);
bool ckh_pointer_keycomp(const void *k1, const void *k2);
bool ckh_search(ckh_t *ckh, const void *searchkey, void **key, void **data);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
/* Some useful hash and comparison functions for strings and pointers. */
void ckh_string_hash(const void *key, size_t r_hash[2]);
bool ckh_string_keycomp(const void *k1, const void *k2);
void ckh_pointer_hash(const void *key, size_t r_hash[2]);
bool ckh_pointer_keycomp(const void *k1, const void *k2);
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_INTERNAL_CKH_H */

167
deps/jemalloc/include/jemalloc/internal/ctl.h vendored
View File

@@ -1,87 +1,106 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_CTL_H
#define JEMALLOC_INTERNAL_CTL_H
typedef struct ctl_node_s ctl_node_t;
typedef struct ctl_named_node_s ctl_named_node_t;
typedef struct ctl_indexed_node_s ctl_indexed_node_t;
typedef struct ctl_arena_stats_s ctl_arena_stats_t;
typedef struct ctl_stats_s ctl_stats_t;
#include "jemalloc/internal/jemalloc_internal_types.h"
#include "jemalloc/internal/malloc_io.h"
#include "jemalloc/internal/mutex_prof.h"
#include "jemalloc/internal/ql.h"
#include "jemalloc/internal/size_classes.h"
#include "jemalloc/internal/stats.h"
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* Maximum ctl tree depth. */
#define CTL_MAX_DEPTH 7
struct ctl_node_s {
bool named;
};
typedef struct ctl_node_s {
bool named;
} ctl_node_t;
struct ctl_named_node_s {
struct ctl_node_s node;
const char *name;
typedef struct ctl_named_node_s {
ctl_node_t node;
const char *name;
/* If (nchildren == 0), this is a terminal node. */
unsigned nchildren;
const ctl_node_t *children;
int (*ctl)(const size_t *, size_t, void *, size_t *,
void *, size_t);
};
size_t nchildren;
const ctl_node_t *children;
int (*ctl)(tsd_t *, const size_t *, size_t, void *, size_t *, void *,
size_t);
} ctl_named_node_t;
struct ctl_indexed_node_s {
struct ctl_node_s node;
const ctl_named_node_t *(*index)(const size_t *, size_t, size_t);
};
typedef struct ctl_indexed_node_s {
struct ctl_node_s node;
const ctl_named_node_t *(*index)(tsdn_t *, const size_t *, size_t,
size_t);
} ctl_indexed_node_t;
struct ctl_arena_stats_s {
bool initialized;
unsigned nthreads;
const char *dss;
size_t pactive;
size_t pdirty;
arena_stats_t astats;
typedef struct ctl_arena_stats_s {
arena_stats_t astats;
/* Aggregate stats for small size classes, based on bin stats. */
size_t allocated_small;
uint64_t nmalloc_small;
uint64_t ndalloc_small;
uint64_t nrequests_small;
size_t allocated_small;
uint64_t nmalloc_small;
uint64_t ndalloc_small;
uint64_t nrequests_small;
malloc_bin_stats_t bstats[NBINS];
malloc_large_stats_t *lstats; /* nlclasses elements. */
malloc_bin_stats_t bstats[NBINS];
malloc_large_stats_t lstats[NSIZES - NBINS];
} ctl_arena_stats_t;
typedef struct ctl_stats_s {
size_t allocated;
size_t active;
size_t metadata;
size_t resident;
size_t mapped;
size_t retained;
background_thread_stats_t background_thread;
mutex_prof_data_t mutex_prof_data[mutex_prof_num_global_mutexes];
} ctl_stats_t;
typedef struct ctl_arena_s ctl_arena_t;
struct ctl_arena_s {
unsigned arena_ind;
bool initialized;
ql_elm(ctl_arena_t) destroyed_link;
/* Basic stats, supported even if !config_stats. */
unsigned nthreads;
const char *dss;
ssize_t dirty_decay_ms;
ssize_t muzzy_decay_ms;
size_t pactive;
size_t pdirty;
size_t pmuzzy;
/* NULL if !config_stats. */
ctl_arena_stats_t *astats;
};
struct ctl_stats_s {
size_t allocated;
size_t active;
size_t mapped;
struct {
size_t current; /* stats_chunks.curchunks */
uint64_t total; /* stats_chunks.nchunks */
size_t high; /* stats_chunks.highchunks */
} chunks;
struct {
size_t allocated; /* huge_allocated */
uint64_t nmalloc; /* huge_nmalloc */
uint64_t ndalloc; /* huge_ndalloc */
} huge;
unsigned narenas;
ctl_arena_stats_t *arenas; /* (narenas + 1) elements. */
};
typedef struct ctl_arenas_s {
uint64_t epoch;
unsigned narenas;
ql_head(ctl_arena_t) destroyed;
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
/*
* Element 0 corresponds to merged stats for extant arenas (accessed via
* MALLCTL_ARENAS_ALL), element 1 corresponds to merged stats for
* destroyed arenas (accessed via MALLCTL_ARENAS_DESTROYED), and the
* remaining MALLOCX_ARENA_LIMIT elements correspond to arenas.
*/
ctl_arena_t *arenas[2 + MALLOCX_ARENA_LIMIT];
} ctl_arenas_t;
int ctl_byname(const char *name, void *oldp, size_t *oldlenp, void *newp,
size_t newlen);
int ctl_nametomib(const char *name, size_t *mibp, size_t *miblenp);
int ctl_bymib(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp,
int ctl_byname(tsd_t *tsd, const char *name, void *oldp, size_t *oldlenp,
void *newp, size_t newlen);
bool ctl_boot(void);
void ctl_prefork(void);
void ctl_postfork_parent(void);
void ctl_postfork_child(void);
int ctl_nametomib(tsd_t *tsd, const char *name, size_t *mibp, size_t *miblenp);
#define xmallctl(name, oldp, oldlenp, newp, newlen) do { \
int ctl_bymib(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp,
size_t *oldlenp, void *newp, size_t newlen);
bool ctl_boot(void);
void ctl_prefork(tsdn_t *tsdn);
void ctl_postfork_parent(tsdn_t *tsdn);
void ctl_postfork_child(tsdn_t *tsdn);
#define xmallctl(name, oldp, oldlenp, newp, newlen) do { \
if (je_mallctl(name, oldp, oldlenp, newp, newlen) \
!= 0) { \
malloc_printf( \
@@ -91,7 +110,7 @@ void ctl_postfork_child(void);
} \
} while (0)
#define xmallctlnametomib(name, mibp, miblenp) do { \
#define xmallctlnametomib(name, mibp, miblenp) do { \
if (je_mallctlnametomib(name, mibp, miblenp) != 0) { \
malloc_printf("<jemalloc>: Failure in " \
"xmallctlnametomib(\"%s\", ...)\n", name); \
@@ -99,7 +118,7 @@ void ctl_postfork_child(void);
} \
} while (0)
#define xmallctlbymib(mib, miblen, oldp, oldlenp, newp, newlen) do { \
#define xmallctlbymib(mib, miblen, oldp, oldlenp, newp, newlen) do { \
if (je_mallctlbymib(mib, miblen, oldp, oldlenp, newp, \
newlen) != 0) { \
malloc_write( \
@@ -108,10 +127,4 @@ void ctl_postfork_child(void);
} \
} while (0)
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_INTERNAL_CTL_H */

139
deps/jemalloc/include/jemalloc/internal/hash.h vendored
View File

@@ -1,92 +1,76 @@
#ifndef JEMALLOC_INTERNAL_HASH_H
#define JEMALLOC_INTERNAL_HASH_H
#include "jemalloc/internal/assert.h"
/*
* The following hash function is based on MurmurHash3, placed into the public
* domain by Austin Appleby. See http://code.google.com/p/smhasher/ for
* domain by Austin Appleby. See https://github.com/aappleby/smhasher for
* details.
*/
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
uint32_t hash_x86_32(const void *key, int len, uint32_t seed);
void hash_x86_128(const void *key, const int len, uint32_t seed,
uint64_t r_out[2]);
void hash_x64_128(const void *key, const int len, const uint32_t seed,
uint64_t r_out[2]);
void hash(const void *key, size_t len, const uint32_t seed,
size_t r_hash[2]);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_HASH_C_))
/******************************************************************************/
/* Internal implementation. */
JEMALLOC_INLINE uint32_t
hash_rotl_32(uint32_t x, int8_t r)
{
return (x << r) | (x >> (32 - r));
static inline uint32_t
hash_rotl_32(uint32_t x, int8_t r) {
return ((x << r) | (x >> (32 - r)));
}
JEMALLOC_INLINE uint64_t
hash_rotl_64(uint64_t x, int8_t r)
{
return (x << r) | (x >> (64 - r));
static inline uint64_t
hash_rotl_64(uint64_t x, int8_t r) {
return ((x << r) | (x >> (64 - r)));
}
JEMALLOC_INLINE uint32_t
hash_get_block_32(const uint32_t *p, int i)
{
static inline uint32_t
hash_get_block_32(const uint32_t *p, int i) {
/* Handle unaligned read. */
if (unlikely((uintptr_t)p & (sizeof(uint32_t)-1)) != 0) {
uint32_t ret;
return (p[i]);
memcpy(&ret, (uint8_t *)(p + i), sizeof(uint32_t));
return ret;
}
return p[i];
}
JEMALLOC_INLINE uint64_t
hash_get_block_64(const uint64_t *p, int i)
{
static inline uint64_t
hash_get_block_64(const uint64_t *p, int i) {
/* Handle unaligned read. */
if (unlikely((uintptr_t)p & (sizeof(uint64_t)-1)) != 0) {
uint64_t ret;
return (p[i]);
memcpy(&ret, (uint8_t *)(p + i), sizeof(uint64_t));
return ret;
}
return p[i];
}
JEMALLOC_INLINE uint32_t
hash_fmix_32(uint32_t h)
{
static inline uint32_t
hash_fmix_32(uint32_t h) {
h ^= h >> 16;
h *= 0x85ebca6b;
h ^= h >> 13;
h *= 0xc2b2ae35;
h ^= h >> 16;
return (h);
return h;
}
JEMALLOC_INLINE uint64_t
hash_fmix_64(uint64_t k)
{
static inline uint64_t
hash_fmix_64(uint64_t k) {
k ^= k >> 33;
k *= QU(0xff51afd7ed558ccdLLU);
k *= KQU(0xff51afd7ed558ccd);
k ^= k >> 33;
k *= QU(0xc4ceb9fe1a85ec53LLU);
k *= KQU(0xc4ceb9fe1a85ec53);
k ^= k >> 33;
return (k);
return k;
}
JEMALLOC_INLINE uint32_t
hash_x86_32(const void *key, int len, uint32_t seed)
{
static inline uint32_t
hash_x86_32(const void *key, int len, uint32_t seed) {
const uint8_t *data = (const uint8_t *) key;
const int nblocks = len / 4;
@@ -132,13 +116,12 @@ hash_x86_32(const void *key, int len, uint32_t seed)
h1 = hash_fmix_32(h1);
return (h1);
return h1;
}
UNUSED JEMALLOC_INLINE void
UNUSED static inline void
hash_x86_128(const void *key, const int len, uint32_t seed,
uint64_t r_out[2])
{
uint64_t r_out[2]) {
const uint8_t * data = (const uint8_t *) key;
const int nblocks = len / 16;
@@ -237,18 +220,17 @@ hash_x86_128(const void *key, const int len, uint32_t seed,
r_out[1] = (((uint64_t) h4) << 32) | h3;
}
UNUSED JEMALLOC_INLINE void
UNUSED static inline void
hash_x64_128(const void *key, const int len, const uint32_t seed,
uint64_t r_out[2])
{
uint64_t r_out[2]) {
const uint8_t *data = (const uint8_t *) key;
const int nblocks = len / 16;
uint64_t h1 = seed;
uint64_t h2 = seed;
const uint64_t c1 = QU(0x87c37b91114253d5LLU);
const uint64_t c2 = QU(0x4cf5ad432745937fLLU);
const uint64_t c1 = KQU(0x87c37b91114253d5);
const uint64_t c2 = KQU(0x4cf5ad432745937f);
/* body */
{
@@ -317,19 +299,20 @@ hash_x64_128(const void *key, const int len, const uint32_t seed,
/******************************************************************************/
/* API. */
JEMALLOC_INLINE void
hash(const void *key, size_t len, const uint32_t seed, size_t r_hash[2])
{
static inline void
hash(const void *key, size_t len, const uint32_t seed, size_t r_hash[2]) {
assert(len <= INT_MAX); /* Unfortunate implementation limitation. */
#if (LG_SIZEOF_PTR == 3 && !defined(JEMALLOC_BIG_ENDIAN))
hash_x64_128(key, len, seed, (uint64_t *)r_hash);
hash_x64_128(key, (int)len, seed, (uint64_t *)r_hash);
#else
uint64_t hashes[2];
hash_x86_128(key, len, seed, hashes);
r_hash[0] = (size_t)hashes[0];
r_hash[1] = (size_t)hashes[1];
{
uint64_t hashes[2];
hash_x86_128(key, (int)len, seed, hashes);
r_hash[0] = (size_t)hashes[0];
r_hash[1] = (size_t)hashes[1];
}
#endif
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_INTERNAL_HASH_H */

289
deps/jemalloc/include/jemalloc/internal/mutex.h vendored
View File

@@ -1,45 +1,123 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_MUTEX_H
#define JEMALLOC_INTERNAL_MUTEX_H
#include "jemalloc/internal/atomic.h"
#include "jemalloc/internal/mutex_prof.h"
#include "jemalloc/internal/tsd.h"
#include "jemalloc/internal/witness.h"
typedef enum {
/* Can only acquire one mutex of a given witness rank at a time. */
malloc_mutex_rank_exclusive,
/*
* Can acquire multiple mutexes of the same witness rank, but in
* address-ascending order only.
*/
malloc_mutex_address_ordered
} malloc_mutex_lock_order_t;
typedef struct malloc_mutex_s malloc_mutex_t;
#ifdef _WIN32
# define MALLOC_MUTEX_INITIALIZER
#elif (defined(JEMALLOC_OSSPIN))
# define MALLOC_MUTEX_INITIALIZER {0}
#elif (defined(JEMALLOC_MUTEX_INIT_CB))
# define MALLOC_MUTEX_INITIALIZER {PTHREAD_MUTEX_INITIALIZER, NULL}
#else
# if (defined(PTHREAD_MUTEX_ADAPTIVE_NP) && \
defined(PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP))
# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_ADAPTIVE_NP
# define MALLOC_MUTEX_INITIALIZER {PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP}
# else
# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_DEFAULT
# define MALLOC_MUTEX_INITIALIZER {PTHREAD_MUTEX_INITIALIZER}
# endif
#endif
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct malloc_mutex_s {
union {
struct {
/*
* prof_data is defined first to reduce cacheline
* bouncing: the data is not touched by the mutex holder
* during unlocking, while might be modified by
* contenders. Having it before the mutex itself could
* avoid prefetching a modified cacheline (for the
* unlocking thread).
*/
mutex_prof_data_t prof_data;
#ifdef _WIN32
CRITICAL_SECTION lock;
# if _WIN32_WINNT >= 0x0600
SRWLOCK lock;
# else
CRITICAL_SECTION lock;
# endif
#elif (defined(JEMALLOC_OS_UNFAIR_LOCK))
os_unfair_lock lock;
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLock lock;
OSSpinLock lock;
#elif (defined(JEMALLOC_MUTEX_INIT_CB))
pthread_mutex_t lock;
malloc_mutex_t *postponed_next;
pthread_mutex_t lock;
malloc_mutex_t *postponed_next;
#else
pthread_mutex_t lock;
pthread_mutex_t lock;
#endif
};
/*
* We only touch witness when configured w/ debug. However we
* keep the field in a union when !debug so that we don't have
* to pollute the code base with #ifdefs, while avoid paying the
* memory cost.
*/
#if !defined(JEMALLOC_DEBUG)
witness_t witness;
malloc_mutex_lock_order_t lock_order;
#endif
};
#if defined(JEMALLOC_DEBUG)
witness_t witness;
malloc_mutex_lock_order_t lock_order;
#endif
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
/*
* Based on benchmark results, a fixed spin with this amount of retries works
* well for our critical sections.
*/
#define MALLOC_MUTEX_MAX_SPIN 250
#ifdef _WIN32
# if _WIN32_WINNT >= 0x0600
# define MALLOC_MUTEX_LOCK(m) AcquireSRWLockExclusive(&(m)->lock)
# define MALLOC_MUTEX_UNLOCK(m) ReleaseSRWLockExclusive(&(m)->lock)
# define MALLOC_MUTEX_TRYLOCK(m) (!TryAcquireSRWLockExclusive(&(m)->lock))
# else
# define MALLOC_MUTEX_LOCK(m) EnterCriticalSection(&(m)->lock)
# define MALLOC_MUTEX_UNLOCK(m) LeaveCriticalSection(&(m)->lock)
# define MALLOC_MUTEX_TRYLOCK(m) (!TryEnterCriticalSection(&(m)->lock))
# endif
#elif (defined(JEMALLOC_OS_UNFAIR_LOCK))
# define MALLOC_MUTEX_LOCK(m) os_unfair_lock_lock(&(m)->lock)
# define MALLOC_MUTEX_UNLOCK(m) os_unfair_lock_unlock(&(m)->lock)
# define MALLOC_MUTEX_TRYLOCK(m) (!os_unfair_lock_trylock(&(m)->lock))
#elif (defined(JEMALLOC_OSSPIN))
# define MALLOC_MUTEX_LOCK(m) OSSpinLockLock(&(m)->lock)
# define MALLOC_MUTEX_UNLOCK(m) OSSpinLockUnlock(&(m)->lock)
# define MALLOC_MUTEX_TRYLOCK(m) (!OSSpinLockTry(&(m)->lock))
#else
# define MALLOC_MUTEX_LOCK(m) pthread_mutex_lock(&(m)->lock)
# define MALLOC_MUTEX_UNLOCK(m) pthread_mutex_unlock(&(m)->lock)
# define MALLOC_MUTEX_TRYLOCK(m) (pthread_mutex_trylock(&(m)->lock) != 0)
#endif
#define LOCK_PROF_DATA_INITIALIZER \
{NSTIME_ZERO_INITIALIZER, NSTIME_ZERO_INITIALIZER, 0, 0, 0, \
ATOMIC_INIT(0), 0, NULL, 0}
#ifdef _WIN32
# define MALLOC_MUTEX_INITIALIZER
#elif (defined(JEMALLOC_OS_UNFAIR_LOCK))
# define MALLOC_MUTEX_INITIALIZER \
{{{LOCK_PROF_DATA_INITIALIZER, OS_UNFAIR_LOCK_INIT}}, \
WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)}
#elif (defined(JEMALLOC_OSSPIN))
# define MALLOC_MUTEX_INITIALIZER \
{{{LOCK_PROF_DATA_INITIALIZER, 0}}, \
WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)}
#elif (defined(JEMALLOC_MUTEX_INIT_CB))
# define MALLOC_MUTEX_INITIALIZER \
{{{LOCK_PROF_DATA_INITIALIZER, PTHREAD_MUTEX_INITIALIZER, NULL}}, \
WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)}
#else
# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_DEFAULT
# define MALLOC_MUTEX_INITIALIZER \
{{{LOCK_PROF_DATA_INITIALIZER, PTHREAD_MUTEX_INITIALIZER}}, \
WITNESS_INITIALIZER("mutex", WITNESS_RANK_OMIT)}
#endif
#ifdef JEMALLOC_LAZY_LOCK
extern bool isthreaded;
@@ -48,52 +126,123 @@ extern bool isthreaded;
# define isthreaded true
#endif
bool malloc_mutex_init(malloc_mutex_t *mutex);
void malloc_mutex_prefork(malloc_mutex_t *mutex);
void malloc_mutex_postfork_parent(malloc_mutex_t *mutex);
void malloc_mutex_postfork_child(malloc_mutex_t *mutex);
bool mutex_boot(void);
bool malloc_mutex_init(malloc_mutex_t *mutex, const char *name,
witness_rank_t rank, malloc_mutex_lock_order_t lock_order);
void malloc_mutex_prefork(tsdn_t *tsdn, malloc_mutex_t *mutex);
void malloc_mutex_postfork_parent(tsdn_t *tsdn, malloc_mutex_t *mutex);
void malloc_mutex_postfork_child(tsdn_t *tsdn, malloc_mutex_t *mutex);
bool malloc_mutex_boot(void);
void malloc_mutex_prof_data_reset(tsdn_t *tsdn, malloc_mutex_t *mutex);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
void malloc_mutex_lock_slow(malloc_mutex_t *mutex);
#ifndef JEMALLOC_ENABLE_INLINE
void malloc_mutex_lock(malloc_mutex_t *mutex);
void malloc_mutex_unlock(malloc_mutex_t *mutex);
#endif
static inline void
malloc_mutex_lock_final(malloc_mutex_t *mutex) {
MALLOC_MUTEX_LOCK(mutex);
}
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_MUTEX_C_))
JEMALLOC_INLINE void
malloc_mutex_lock(malloc_mutex_t *mutex)
{
static inline bool
malloc_mutex_trylock_final(malloc_mutex_t *mutex) {
return MALLOC_MUTEX_TRYLOCK(mutex);
}
if (isthreaded) {
#ifdef _WIN32
EnterCriticalSection(&mutex->lock);
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLockLock(&mutex->lock);
#else
pthread_mutex_lock(&mutex->lock);
#endif
static inline void
mutex_owner_stats_update(tsdn_t *tsdn, malloc_mutex_t *mutex) {
if (config_stats) {
mutex_prof_data_t *data = &mutex->prof_data;
data->n_lock_ops++;
if (data->prev_owner != tsdn) {
data->prev_owner = tsdn;
data->n_owner_switches++;
}
}
}
JEMALLOC_INLINE void
malloc_mutex_unlock(malloc_mutex_t *mutex)
{
/* Trylock: return false if the lock is successfully acquired. */
static inline bool
malloc_mutex_trylock(tsdn_t *tsdn, malloc_mutex_t *mutex) {
witness_assert_not_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness);
if (isthreaded) {
#ifdef _WIN32
LeaveCriticalSection(&mutex->lock);
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLockUnlock(&mutex->lock);
#else
pthread_mutex_unlock(&mutex->lock);
#endif
if (malloc_mutex_trylock_final(mutex)) {
return true;
}
mutex_owner_stats_update(tsdn, mutex);
}
witness_lock(tsdn_witness_tsdp_get(tsdn), &mutex->witness);
return false;
}
/* Aggregate lock prof data. */
static inline void
malloc_mutex_prof_merge(mutex_prof_data_t *sum, mutex_prof_data_t *data) {
nstime_add(&sum->tot_wait_time, &data->tot_wait_time);
if (nstime_compare(&sum->max_wait_time, &data->max_wait_time) < 0) {
nstime_copy(&sum->max_wait_time, &data->max_wait_time);
}
sum->n_wait_times += data->n_wait_times;
sum->n_spin_acquired += data->n_spin_acquired;
if (sum->max_n_thds < data->max_n_thds) {
sum->max_n_thds = data->max_n_thds;
}
uint32_t cur_n_waiting_thds = atomic_load_u32(&sum->n_waiting_thds,
ATOMIC_RELAXED);
uint32_t new_n_waiting_thds = cur_n_waiting_thds + atomic_load_u32(
&data->n_waiting_thds, ATOMIC_RELAXED);
atomic_store_u32(&sum->n_waiting_thds, new_n_waiting_thds,
ATOMIC_RELAXED);
sum->n_owner_switches += data->n_owner_switches;
sum->n_lock_ops += data->n_lock_ops;
}
static inline void
malloc_mutex_lock(tsdn_t *tsdn, malloc_mutex_t *mutex) {
witness_assert_not_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness);
if (isthreaded) {
if (malloc_mutex_trylock_final(mutex)) {
malloc_mutex_lock_slow(mutex);
}
mutex_owner_stats_update(tsdn, mutex);
}
witness_lock(tsdn_witness_tsdp_get(tsdn), &mutex->witness);
}
static inline void
malloc_mutex_unlock(tsdn_t *tsdn, malloc_mutex_t *mutex) {
witness_unlock(tsdn_witness_tsdp_get(tsdn), &mutex->witness);
if (isthreaded) {
MALLOC_MUTEX_UNLOCK(mutex);
}
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
static inline void
malloc_mutex_assert_owner(tsdn_t *tsdn, malloc_mutex_t *mutex) {
witness_assert_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness);
}
static inline void
malloc_mutex_assert_not_owner(tsdn_t *tsdn, malloc_mutex_t *mutex) {
witness_assert_not_owner(tsdn_witness_tsdp_get(tsdn), &mutex->witness);
}
/* Copy the prof data from mutex for processing. */
static inline void
malloc_mutex_prof_read(tsdn_t *tsdn, mutex_prof_data_t *data,
malloc_mutex_t *mutex) {
mutex_prof_data_t *source = &mutex->prof_data;
/* Can only read holding the mutex. */
malloc_mutex_assert_owner(tsdn, mutex);
/*
* Not *really* allowed (we shouldn't be doing non-atomic loads of
* atomic data), but the mutex protection makes this safe, and writing
* a member-for-member copy is tedious for this situation.
*/
*data = *source;
/* n_wait_thds is not reported (modified w/o locking). */
atomic_store_u32(&data->n_waiting_thds, 0, ATOMIC_RELAXED);
}
#endif /* JEMALLOC_INTERNAL_MUTEX_H */

538
deps/jemalloc/include/jemalloc/internal/private_namespace.h vendored
View File

@@ -1,147 +1,199 @@
#define a0calloc JEMALLOC_N(a0calloc)
#define a0free JEMALLOC_N(a0free)
#define a0dalloc JEMALLOC_N(a0dalloc)
#define a0get JEMALLOC_N(a0get)
#define a0malloc JEMALLOC_N(a0malloc)
#define arena_aalloc JEMALLOC_N(arena_aalloc)
#define arena_alloc_junk_small JEMALLOC_N(arena_alloc_junk_small)
#define arena_basic_stats_merge JEMALLOC_N(arena_basic_stats_merge)
#define arena_bin_index JEMALLOC_N(arena_bin_index)
#define arena_bin_info JEMALLOC_N(arena_bin_info)
#define arena_bitselm_get_const JEMALLOC_N(arena_bitselm_get_const)
#define arena_bitselm_get_mutable JEMALLOC_N(arena_bitselm_get_mutable)
#define arena_boot JEMALLOC_N(arena_boot)
#define arena_choose JEMALLOC_N(arena_choose)
#define arena_choose_hard JEMALLOC_N(arena_choose_hard)
#define arena_choose_impl JEMALLOC_N(arena_choose_impl)
#define arena_chunk_alloc_huge JEMALLOC_N(arena_chunk_alloc_huge)
#define arena_chunk_cache_maybe_insert JEMALLOC_N(arena_chunk_cache_maybe_insert)
#define arena_chunk_cache_maybe_remove JEMALLOC_N(arena_chunk_cache_maybe_remove)
#define arena_chunk_dalloc_huge JEMALLOC_N(arena_chunk_dalloc_huge)
#define arena_chunk_ralloc_huge_expand JEMALLOC_N(arena_chunk_ralloc_huge_expand)
#define arena_chunk_ralloc_huge_shrink JEMALLOC_N(arena_chunk_ralloc_huge_shrink)
#define arena_chunk_ralloc_huge_similar JEMALLOC_N(arena_chunk_ralloc_huge_similar)
#define arena_cleanup JEMALLOC_N(arena_cleanup)
#define arena_dalloc JEMALLOC_N(arena_dalloc)
#define arena_dalloc_bin JEMALLOC_N(arena_dalloc_bin)
#define arena_dalloc_bin_locked JEMALLOC_N(arena_dalloc_bin_locked)
#define arena_dalloc_bin_junked_locked JEMALLOC_N(arena_dalloc_bin_junked_locked)
#define arena_dalloc_junk_large JEMALLOC_N(arena_dalloc_junk_large)
#define arena_dalloc_junk_small JEMALLOC_N(arena_dalloc_junk_small)
#define arena_dalloc_large JEMALLOC_N(arena_dalloc_large)
#define arena_dalloc_large_locked JEMALLOC_N(arena_dalloc_large_locked)
#define arena_dalloc_large_junked_locked JEMALLOC_N(arena_dalloc_large_junked_locked)
#define arena_dalloc_small JEMALLOC_N(arena_dalloc_small)
#define arena_decay_tick JEMALLOC_N(arena_decay_tick)
#define arena_decay_ticks JEMALLOC_N(arena_decay_ticks)
#define arena_decay_time_default_get JEMALLOC_N(arena_decay_time_default_get)
#define arena_decay_time_default_set JEMALLOC_N(arena_decay_time_default_set)
#define arena_decay_time_get JEMALLOC_N(arena_decay_time_get)
#define arena_decay_time_set JEMALLOC_N(arena_decay_time_set)
#define arena_dss_prec_get JEMALLOC_N(arena_dss_prec_get)
#define arena_dss_prec_set JEMALLOC_N(arena_dss_prec_set)
#define arena_extent_sn_next JEMALLOC_N(arena_extent_sn_next)
#define arena_get JEMALLOC_N(arena_get)
#define arena_ichoose JEMALLOC_N(arena_ichoose)
#define arena_init JEMALLOC_N(arena_init)
#define arena_lg_dirty_mult_default_get JEMALLOC_N(arena_lg_dirty_mult_default_get)
#define arena_lg_dirty_mult_default_set JEMALLOC_N(arena_lg_dirty_mult_default_set)
#define arena_lg_dirty_mult_get JEMALLOC_N(arena_lg_dirty_mult_get)
#define arena_lg_dirty_mult_set JEMALLOC_N(arena_lg_dirty_mult_set)
#define arena_malloc JEMALLOC_N(arena_malloc)
#define arena_malloc_hard JEMALLOC_N(arena_malloc_hard)
#define arena_malloc_large JEMALLOC_N(arena_malloc_large)
#define arena_malloc_small JEMALLOC_N(arena_malloc_small)
#define arena_mapbits_allocated_get JEMALLOC_N(arena_mapbits_allocated_get)
#define arena_mapbits_binind_get JEMALLOC_N(arena_mapbits_binind_get)
#define arena_mapbits_decommitted_get JEMALLOC_N(arena_mapbits_decommitted_get)
#define arena_mapbits_dirty_get JEMALLOC_N(arena_mapbits_dirty_get)
#define arena_mapbits_get JEMALLOC_N(arena_mapbits_get)
#define arena_mapbits_internal_set JEMALLOC_N(arena_mapbits_internal_set)
#define arena_mapbits_large_binind_set JEMALLOC_N(arena_mapbits_large_binind_set)
#define arena_mapbits_large_get JEMALLOC_N(arena_mapbits_large_get)
#define arena_mapbits_large_set JEMALLOC_N(arena_mapbits_large_set)
#define arena_mapbits_large_size_get JEMALLOC_N(arena_mapbits_large_size_get)
#define arena_mapbits_size_decode JEMALLOC_N(arena_mapbits_size_decode)
#define arena_mapbits_size_encode JEMALLOC_N(arena_mapbits_size_encode)
#define arena_mapbits_small_runind_get JEMALLOC_N(arena_mapbits_small_runind_get)
#define arena_mapbits_small_set JEMALLOC_N(arena_mapbits_small_set)
#define arena_mapbits_unallocated_set JEMALLOC_N(arena_mapbits_unallocated_set)
#define arena_mapbits_unallocated_size_get JEMALLOC_N(arena_mapbits_unallocated_size_get)
#define arena_mapbits_unallocated_size_set JEMALLOC_N(arena_mapbits_unallocated_size_set)
#define arena_mapbits_unzeroed_get JEMALLOC_N(arena_mapbits_unzeroed_get)
#define arena_mapbits_unzeroed_set JEMALLOC_N(arena_mapbits_unzeroed_set)
#define arena_mapbitsp_get JEMALLOC_N(arena_mapbitsp_get)
#define arena_mapbitsp_get_const JEMALLOC_N(arena_mapbitsp_get_const)
#define arena_mapbitsp_get_mutable JEMALLOC_N(arena_mapbitsp_get_mutable)
#define arena_mapbitsp_read JEMALLOC_N(arena_mapbitsp_read)
#define arena_mapbitsp_write JEMALLOC_N(arena_mapbitsp_write)
#define arena_mapp_get JEMALLOC_N(arena_mapp_get)
#define arena_maxclass JEMALLOC_N(arena_maxclass)
#define arena_maxrun JEMALLOC_N(arena_maxrun)
#define arena_maybe_purge JEMALLOC_N(arena_maybe_purge)
#define arena_metadata_allocated_add JEMALLOC_N(arena_metadata_allocated_add)
#define arena_metadata_allocated_get JEMALLOC_N(arena_metadata_allocated_get)
#define arena_metadata_allocated_sub JEMALLOC_N(arena_metadata_allocated_sub)
#define arena_migrate JEMALLOC_N(arena_migrate)
#define arena_miscelm_get_const JEMALLOC_N(arena_miscelm_get_const)
#define arena_miscelm_get_mutable JEMALLOC_N(arena_miscelm_get_mutable)
#define arena_miscelm_to_pageind JEMALLOC_N(arena_miscelm_to_pageind)
#define arena_miscelm_to_rpages JEMALLOC_N(arena_miscelm_to_rpages)
#define arena_new JEMALLOC_N(arena_new)
#define arena_node_alloc JEMALLOC_N(arena_node_alloc)
#define arena_node_dalloc JEMALLOC_N(arena_node_dalloc)
#define arena_nthreads_dec JEMALLOC_N(arena_nthreads_dec)
#define arena_nthreads_get JEMALLOC_N(arena_nthreads_get)
#define arena_nthreads_inc JEMALLOC_N(arena_nthreads_inc)
#define arena_palloc JEMALLOC_N(arena_palloc)
#define arena_postfork_child JEMALLOC_N(arena_postfork_child)
#define arena_postfork_parent JEMALLOC_N(arena_postfork_parent)
#define arena_prefork JEMALLOC_N(arena_prefork)
#define arena_prefork0 JEMALLOC_N(arena_prefork0)
#define arena_prefork1 JEMALLOC_N(arena_prefork1)
#define arena_prefork2 JEMALLOC_N(arena_prefork2)
#define arena_prefork3 JEMALLOC_N(arena_prefork3)
#define arena_prof_accum JEMALLOC_N(arena_prof_accum)
#define arena_prof_accum_impl JEMALLOC_N(arena_prof_accum_impl)
#define arena_prof_accum_locked JEMALLOC_N(arena_prof_accum_locked)
#define arena_prof_ctx_get JEMALLOC_N(arena_prof_ctx_get)
#define arena_prof_ctx_set JEMALLOC_N(arena_prof_ctx_set)
#define arena_prof_promoted JEMALLOC_N(arena_prof_promoted)
#define arena_prof_tctx_get JEMALLOC_N(arena_prof_tctx_get)
#define arena_prof_tctx_reset JEMALLOC_N(arena_prof_tctx_reset)
#define arena_prof_tctx_set JEMALLOC_N(arena_prof_tctx_set)
#define arena_ptr_small_binind_get JEMALLOC_N(arena_ptr_small_binind_get)
#define arena_purge_all JEMALLOC_N(arena_purge_all)
#define arena_purge JEMALLOC_N(arena_purge)
#define arena_quarantine_junk_small JEMALLOC_N(arena_quarantine_junk_small)
#define arena_ralloc JEMALLOC_N(arena_ralloc)
#define arena_ralloc_junk_large JEMALLOC_N(arena_ralloc_junk_large)
#define arena_ralloc_no_move JEMALLOC_N(arena_ralloc_no_move)
#define arena_rd_to_miscelm JEMALLOC_N(arena_rd_to_miscelm)
#define arena_redzone_corruption JEMALLOC_N(arena_redzone_corruption)
#define arena_reset JEMALLOC_N(arena_reset)
#define arena_run_regind JEMALLOC_N(arena_run_regind)
#define arena_run_to_miscelm JEMALLOC_N(arena_run_to_miscelm)
#define arena_salloc JEMALLOC_N(arena_salloc)
#define arena_sdalloc JEMALLOC_N(arena_sdalloc)
#define arena_stats_merge JEMALLOC_N(arena_stats_merge)
#define arena_tcache_fill_small JEMALLOC_N(arena_tcache_fill_small)
#define arena_tdata_get JEMALLOC_N(arena_tdata_get)
#define arena_tdata_get_hard JEMALLOC_N(arena_tdata_get_hard)
#define arenas JEMALLOC_N(arenas)
#define arenas_booted JEMALLOC_N(arenas_booted)
#define arenas_cleanup JEMALLOC_N(arenas_cleanup)
#define arenas_extend JEMALLOC_N(arenas_extend)
#define arenas_initialized JEMALLOC_N(arenas_initialized)
#define arenas_lock JEMALLOC_N(arenas_lock)
#define arenas_tls JEMALLOC_N(arenas_tls)
#define arenas_tsd JEMALLOC_N(arenas_tsd)
#define arenas_tsd_boot JEMALLOC_N(arenas_tsd_boot)
#define arenas_tsd_cleanup_wrapper JEMALLOC_N(arenas_tsd_cleanup_wrapper)
#define arenas_tsd_get JEMALLOC_N(arenas_tsd_get)
#define arenas_tsd_get_wrapper JEMALLOC_N(arenas_tsd_get_wrapper)
#define arenas_tsd_init_head JEMALLOC_N(arenas_tsd_init_head)
#define arenas_tsd_set JEMALLOC_N(arenas_tsd_set)
#define arenas_tdata_bypass_cleanup JEMALLOC_N(arenas_tdata_bypass_cleanup)
#define arenas_tdata_cleanup JEMALLOC_N(arenas_tdata_cleanup)
#define atomic_add_p JEMALLOC_N(atomic_add_p)
#define atomic_add_u JEMALLOC_N(atomic_add_u)
#define atomic_add_uint32 JEMALLOC_N(atomic_add_uint32)
#define atomic_add_uint64 JEMALLOC_N(atomic_add_uint64)
#define atomic_add_z JEMALLOC_N(atomic_add_z)
#define atomic_cas_p JEMALLOC_N(atomic_cas_p)
#define atomic_cas_u JEMALLOC_N(atomic_cas_u)
#define atomic_cas_uint32 JEMALLOC_N(atomic_cas_uint32)
#define atomic_cas_uint64 JEMALLOC_N(atomic_cas_uint64)
#define atomic_cas_z JEMALLOC_N(atomic_cas_z)
#define atomic_sub_p JEMALLOC_N(atomic_sub_p)
#define atomic_sub_u JEMALLOC_N(atomic_sub_u)
#define atomic_sub_uint32 JEMALLOC_N(atomic_sub_uint32)
#define atomic_sub_uint64 JEMALLOC_N(atomic_sub_uint64)
#define atomic_sub_z JEMALLOC_N(atomic_sub_z)
#define atomic_write_p JEMALLOC_N(atomic_write_p)
#define atomic_write_u JEMALLOC_N(atomic_write_u)
#define atomic_write_uint32 JEMALLOC_N(atomic_write_uint32)
#define atomic_write_uint64 JEMALLOC_N(atomic_write_uint64)
#define atomic_write_z JEMALLOC_N(atomic_write_z)
#define base_alloc JEMALLOC_N(base_alloc)
#define base_boot JEMALLOC_N(base_boot)
#define base_calloc JEMALLOC_N(base_calloc)
#define base_node_alloc JEMALLOC_N(base_node_alloc)
#define base_node_dealloc JEMALLOC_N(base_node_dealloc)
#define base_postfork_child JEMALLOC_N(base_postfork_child)
#define base_postfork_parent JEMALLOC_N(base_postfork_parent)
#define base_prefork JEMALLOC_N(base_prefork)
#define base_stats_get JEMALLOC_N(base_stats_get)
#define bitmap_full JEMALLOC_N(bitmap_full)
#define bitmap_get JEMALLOC_N(bitmap_get)
#define bitmap_info_init JEMALLOC_N(bitmap_info_init)
#define bitmap_info_ngroups JEMALLOC_N(bitmap_info_ngroups)
#define bitmap_init JEMALLOC_N(bitmap_init)
#define bitmap_set JEMALLOC_N(bitmap_set)
#define bitmap_sfu JEMALLOC_N(bitmap_sfu)
#define bitmap_size JEMALLOC_N(bitmap_size)
#define bitmap_unset JEMALLOC_N(bitmap_unset)
#define bootstrap_calloc JEMALLOC_N(bootstrap_calloc)
#define bootstrap_free JEMALLOC_N(bootstrap_free)
#define bootstrap_malloc JEMALLOC_N(bootstrap_malloc)
#define bt_init JEMALLOC_N(bt_init)
#define buferror JEMALLOC_N(buferror)
#define choose_arena JEMALLOC_N(choose_arena)
#define choose_arena_hard JEMALLOC_N(choose_arena_hard)
#define chunk_alloc JEMALLOC_N(chunk_alloc)
#define chunk_alloc_base JEMALLOC_N(chunk_alloc_base)
#define chunk_alloc_cache JEMALLOC_N(chunk_alloc_cache)
#define chunk_alloc_dss JEMALLOC_N(chunk_alloc_dss)
#define chunk_alloc_mmap JEMALLOC_N(chunk_alloc_mmap)
#define chunk_alloc_wrapper JEMALLOC_N(chunk_alloc_wrapper)
#define chunk_boot JEMALLOC_N(chunk_boot)
#define chunk_dealloc JEMALLOC_N(chunk_dealloc)
#define chunk_dealloc_mmap JEMALLOC_N(chunk_dealloc_mmap)
#define chunk_dalloc_cache JEMALLOC_N(chunk_dalloc_cache)
#define chunk_dalloc_mmap JEMALLOC_N(chunk_dalloc_mmap)
#define chunk_dalloc_wrapper JEMALLOC_N(chunk_dalloc_wrapper)
#define chunk_deregister JEMALLOC_N(chunk_deregister)
#define chunk_dss_boot JEMALLOC_N(chunk_dss_boot)
#define chunk_dss_postfork_child JEMALLOC_N(chunk_dss_postfork_child)
#define chunk_dss_postfork_parent JEMALLOC_N(chunk_dss_postfork_parent)
#define chunk_dss_mergeable JEMALLOC_N(chunk_dss_mergeable)
#define chunk_dss_prec_get JEMALLOC_N(chunk_dss_prec_get)
#define chunk_dss_prec_set JEMALLOC_N(chunk_dss_prec_set)
#define chunk_dss_prefork JEMALLOC_N(chunk_dss_prefork)
#define chunk_hooks_default JEMALLOC_N(chunk_hooks_default)
#define chunk_hooks_get JEMALLOC_N(chunk_hooks_get)
#define chunk_hooks_set JEMALLOC_N(chunk_hooks_set)
#define chunk_in_dss JEMALLOC_N(chunk_in_dss)
#define chunk_lookup JEMALLOC_N(chunk_lookup)
#define chunk_npages JEMALLOC_N(chunk_npages)
#define chunk_postfork_child JEMALLOC_N(chunk_postfork_child)
#define chunk_postfork_parent JEMALLOC_N(chunk_postfork_parent)
#define chunk_prefork JEMALLOC_N(chunk_prefork)
#define chunk_unmap JEMALLOC_N(chunk_unmap)
#define chunks_mtx JEMALLOC_N(chunks_mtx)
#define chunk_purge_wrapper JEMALLOC_N(chunk_purge_wrapper)
#define chunk_register JEMALLOC_N(chunk_register)
#define chunks_rtree JEMALLOC_N(chunks_rtree)
#define chunksize JEMALLOC_N(chunksize)
#define chunksize_mask JEMALLOC_N(chunksize_mask)
#define ckh_bucket_search JEMALLOC_N(ckh_bucket_search)
#define ckh_count JEMALLOC_N(ckh_count)
#define ckh_delete JEMALLOC_N(ckh_delete)
#define ckh_evict_reloc_insert JEMALLOC_N(ckh_evict_reloc_insert)
#define ckh_insert JEMALLOC_N(ckh_insert)
#define ckh_isearch JEMALLOC_N(ckh_isearch)
#define ckh_iter JEMALLOC_N(ckh_iter)
#define ckh_new JEMALLOC_N(ckh_new)
#define ckh_pointer_hash JEMALLOC_N(ckh_pointer_hash)
#define ckh_pointer_keycomp JEMALLOC_N(ckh_pointer_keycomp)
#define ckh_rebuild JEMALLOC_N(ckh_rebuild)
#define ckh_remove JEMALLOC_N(ckh_remove)
#define ckh_search JEMALLOC_N(ckh_search)
#define ckh_string_hash JEMALLOC_N(ckh_string_hash)
#define ckh_string_keycomp JEMALLOC_N(ckh_string_keycomp)
#define ckh_try_bucket_insert JEMALLOC_N(ckh_try_bucket_insert)
#define ckh_try_insert JEMALLOC_N(ckh_try_insert)
#define ctl_boot JEMALLOC_N(ctl_boot)
#define ctl_bymib JEMALLOC_N(ctl_bymib)
#define ctl_byname JEMALLOC_N(ctl_byname)
@@ -149,7 +201,33 @@
#define ctl_postfork_child JEMALLOC_N(ctl_postfork_child)
#define ctl_postfork_parent JEMALLOC_N(ctl_postfork_parent)
#define ctl_prefork JEMALLOC_N(ctl_prefork)
#define decay_ticker_get JEMALLOC_N(decay_ticker_get)
#define dss_prec_names JEMALLOC_N(dss_prec_names)
#define extent_node_achunk_get JEMALLOC_N(extent_node_achunk_get)
#define extent_node_achunk_set JEMALLOC_N(extent_node_achunk_set)
#define extent_node_addr_get JEMALLOC_N(extent_node_addr_get)
#define extent_node_addr_set JEMALLOC_N(extent_node_addr_set)
#define extent_node_arena_get JEMALLOC_N(extent_node_arena_get)
#define extent_node_arena_set JEMALLOC_N(extent_node_arena_set)
#define extent_node_committed_get JEMALLOC_N(extent_node_committed_get)
#define extent_node_committed_set JEMALLOC_N(extent_node_committed_set)
#define extent_node_dirty_insert JEMALLOC_N(extent_node_dirty_insert)
#define extent_node_dirty_linkage_init JEMALLOC_N(extent_node_dirty_linkage_init)
#define extent_node_dirty_remove JEMALLOC_N(extent_node_dirty_remove)
#define extent_node_init JEMALLOC_N(extent_node_init)
#define extent_node_prof_tctx_get JEMALLOC_N(extent_node_prof_tctx_get)
#define extent_node_prof_tctx_set JEMALLOC_N(extent_node_prof_tctx_set)
#define extent_node_size_get JEMALLOC_N(extent_node_size_get)
#define extent_node_size_set JEMALLOC_N(extent_node_size_set)
#define extent_node_sn_get JEMALLOC_N(extent_node_sn_get)
#define extent_node_sn_set JEMALLOC_N(extent_node_sn_set)
#define extent_node_zeroed_get JEMALLOC_N(extent_node_zeroed_get)
#define extent_node_zeroed_set JEMALLOC_N(extent_node_zeroed_set)
#define extent_size_quantize_ceil JEMALLOC_N(extent_size_quantize_ceil)
#define extent_size_quantize_floor JEMALLOC_N(extent_size_quantize_floor)
#define extent_tree_ad_destroy JEMALLOC_N(extent_tree_ad_destroy)
#define extent_tree_ad_destroy_recurse JEMALLOC_N(extent_tree_ad_destroy_recurse)
#define extent_tree_ad_empty JEMALLOC_N(extent_tree_ad_empty)
#define extent_tree_ad_first JEMALLOC_N(extent_tree_ad_first)
#define extent_tree_ad_insert JEMALLOC_N(extent_tree_ad_insert)
#define extent_tree_ad_iter JEMALLOC_N(extent_tree_ad_iter)
@@ -166,22 +244,31 @@
#define extent_tree_ad_reverse_iter_recurse JEMALLOC_N(extent_tree_ad_reverse_iter_recurse)
#define extent_tree_ad_reverse_iter_start JEMALLOC_N(extent_tree_ad_reverse_iter_start)
#define extent_tree_ad_search JEMALLOC_N(extent_tree_ad_search)
#define extent_tree_szad_first JEMALLOC_N(extent_tree_szad_first)
#define extent_tree_szad_insert JEMALLOC_N(extent_tree_szad_insert)
#define extent_tree_szad_iter JEMALLOC_N(extent_tree_szad_iter)
#define extent_tree_szad_iter_recurse JEMALLOC_N(extent_tree_szad_iter_recurse)
#define extent_tree_szad_iter_start JEMALLOC_N(extent_tree_szad_iter_start)
#define extent_tree_szad_last JEMALLOC_N(extent_tree_szad_last)
#define extent_tree_szad_new JEMALLOC_N(extent_tree_szad_new)
#define extent_tree_szad_next JEMALLOC_N(extent_tree_szad_next)
#define extent_tree_szad_nsearch JEMALLOC_N(extent_tree_szad_nsearch)
#define extent_tree_szad_prev JEMALLOC_N(extent_tree_szad_prev)
#define extent_tree_szad_psearch JEMALLOC_N(extent_tree_szad_psearch)
#define extent_tree_szad_remove JEMALLOC_N(extent_tree_szad_remove)
#define extent_tree_szad_reverse_iter JEMALLOC_N(extent_tree_szad_reverse_iter)
#define extent_tree_szad_reverse_iter_recurse JEMALLOC_N(extent_tree_szad_reverse_iter_recurse)
#define extent_tree_szad_reverse_iter_start JEMALLOC_N(extent_tree_szad_reverse_iter_start)
#define extent_tree_szad_search JEMALLOC_N(extent_tree_szad_search)
#define extent_tree_szsnad_destroy JEMALLOC_N(extent_tree_szsnad_destroy)
#define extent_tree_szsnad_destroy_recurse JEMALLOC_N(extent_tree_szsnad_destroy_recurse)
#define extent_tree_szsnad_empty JEMALLOC_N(extent_tree_szsnad_empty)
#define extent_tree_szsnad_first JEMALLOC_N(extent_tree_szsnad_first)
#define extent_tree_szsnad_insert JEMALLOC_N(extent_tree_szsnad_insert)
#define extent_tree_szsnad_iter JEMALLOC_N(extent_tree_szsnad_iter)
#define extent_tree_szsnad_iter_recurse JEMALLOC_N(extent_tree_szsnad_iter_recurse)
#define extent_tree_szsnad_iter_start JEMALLOC_N(extent_tree_szsnad_iter_start)
#define extent_tree_szsnad_last JEMALLOC_N(extent_tree_szsnad_last)
#define extent_tree_szsnad_new JEMALLOC_N(extent_tree_szsnad_new)
#define extent_tree_szsnad_next JEMALLOC_N(extent_tree_szsnad_next)
#define extent_tree_szsnad_nsearch JEMALLOC_N(extent_tree_szsnad_nsearch)
#define extent_tree_szsnad_prev JEMALLOC_N(extent_tree_szsnad_prev)
#define extent_tree_szsnad_psearch JEMALLOC_N(extent_tree_szsnad_psearch)
#define extent_tree_szsnad_remove JEMALLOC_N(extent_tree_szsnad_remove)
#define extent_tree_szsnad_reverse_iter JEMALLOC_N(extent_tree_szsnad_reverse_iter)
#define extent_tree_szsnad_reverse_iter_recurse JEMALLOC_N(extent_tree_szsnad_reverse_iter_recurse)
#define extent_tree_szsnad_reverse_iter_start JEMALLOC_N(extent_tree_szsnad_reverse_iter_start)
#define extent_tree_szsnad_search JEMALLOC_N(extent_tree_szsnad_search)
#define ffs_llu JEMALLOC_N(ffs_llu)
#define ffs_lu JEMALLOC_N(ffs_lu)
#define ffs_u JEMALLOC_N(ffs_u)
#define ffs_u32 JEMALLOC_N(ffs_u32)
#define ffs_u64 JEMALLOC_N(ffs_u64)
#define ffs_zu JEMALLOC_N(ffs_zu)
#define get_errno JEMALLOC_N(get_errno)
#define hash JEMALLOC_N(hash)
#define hash_fmix_32 JEMALLOC_N(hash_fmix_32)
@@ -193,46 +280,51 @@
#define hash_x64_128 JEMALLOC_N(hash_x64_128)
#define hash_x86_128 JEMALLOC_N(hash_x86_128)
#define hash_x86_32 JEMALLOC_N(hash_x86_32)
#define huge_allocated JEMALLOC_N(huge_allocated)
#define huge_boot JEMALLOC_N(huge_boot)
#define huge_aalloc JEMALLOC_N(huge_aalloc)
#define huge_dalloc JEMALLOC_N(huge_dalloc)
#define huge_dalloc_junk JEMALLOC_N(huge_dalloc_junk)
#define huge_dss_prec_get JEMALLOC_N(huge_dss_prec_get)
#define huge_malloc JEMALLOC_N(huge_malloc)
#define huge_mtx JEMALLOC_N(huge_mtx)
#define huge_ndalloc JEMALLOC_N(huge_ndalloc)
#define huge_nmalloc JEMALLOC_N(huge_nmalloc)
#define huge_palloc JEMALLOC_N(huge_palloc)
#define huge_postfork_child JEMALLOC_N(huge_postfork_child)
#define huge_postfork_parent JEMALLOC_N(huge_postfork_parent)
#define huge_prefork JEMALLOC_N(huge_prefork)
#define huge_prof_ctx_get JEMALLOC_N(huge_prof_ctx_get)
#define huge_prof_ctx_set JEMALLOC_N(huge_prof_ctx_set)
#define huge_prof_tctx_get JEMALLOC_N(huge_prof_tctx_get)
#define huge_prof_tctx_reset JEMALLOC_N(huge_prof_tctx_reset)
#define huge_prof_tctx_set JEMALLOC_N(huge_prof_tctx_set)
#define huge_ralloc JEMALLOC_N(huge_ralloc)
#define huge_ralloc_no_move JEMALLOC_N(huge_ralloc_no_move)
#define huge_salloc JEMALLOC_N(huge_salloc)
#define iallocm JEMALLOC_N(iallocm)
#define icalloc JEMALLOC_N(icalloc)
#define icalloct JEMALLOC_N(icalloct)
#define iaalloc JEMALLOC_N(iaalloc)
#define ialloc JEMALLOC_N(ialloc)
#define iallocztm JEMALLOC_N(iallocztm)
#define iarena_cleanup JEMALLOC_N(iarena_cleanup)
#define idalloc JEMALLOC_N(idalloc)
#define idalloct JEMALLOC_N(idalloct)
#define imalloc JEMALLOC_N(imalloc)
#define imalloct JEMALLOC_N(imalloct)
#define idalloctm JEMALLOC_N(idalloctm)
#define in_valgrind JEMALLOC_N(in_valgrind)
#define index2size JEMALLOC_N(index2size)
#define index2size_compute JEMALLOC_N(index2size_compute)
#define index2size_lookup JEMALLOC_N(index2size_lookup)
#define index2size_tab JEMALLOC_N(index2size_tab)
#define ipalloc JEMALLOC_N(ipalloc)
#define ipalloct JEMALLOC_N(ipalloct)
#define ipallocztm JEMALLOC_N(ipallocztm)
#define iqalloc JEMALLOC_N(iqalloc)
#define iqalloct JEMALLOC_N(iqalloct)
#define iralloc JEMALLOC_N(iralloc)
#define iralloct JEMALLOC_N(iralloct)
#define iralloct_realign JEMALLOC_N(iralloct_realign)
#define isalloc JEMALLOC_N(isalloc)
#define isdalloct JEMALLOC_N(isdalloct)
#define isqalloc JEMALLOC_N(isqalloc)
#define isthreaded JEMALLOC_N(isthreaded)
#define ivsalloc JEMALLOC_N(ivsalloc)
#define ixalloc JEMALLOC_N(ixalloc)
#define jemalloc_postfork_child JEMALLOC_N(jemalloc_postfork_child)
#define jemalloc_postfork_parent JEMALLOC_N(jemalloc_postfork_parent)
#define jemalloc_prefork JEMALLOC_N(jemalloc_prefork)
#define large_maxclass JEMALLOC_N(large_maxclass)
#define lg_floor JEMALLOC_N(lg_floor)
#define lg_prof_sample JEMALLOC_N(lg_prof_sample)
#define malloc_cprintf JEMALLOC_N(malloc_cprintf)
#define malloc_mutex_assert_not_owner JEMALLOC_N(malloc_mutex_assert_not_owner)
#define malloc_mutex_assert_owner JEMALLOC_N(malloc_mutex_assert_owner)
#define malloc_mutex_boot JEMALLOC_N(malloc_mutex_boot)
#define malloc_mutex_init JEMALLOC_N(malloc_mutex_init)
#define malloc_mutex_lock JEMALLOC_N(malloc_mutex_lock)
#define malloc_mutex_postfork_child JEMALLOC_N(malloc_mutex_postfork_child)
@@ -242,7 +334,8 @@
#define malloc_printf JEMALLOC_N(malloc_printf)
#define malloc_snprintf JEMALLOC_N(malloc_snprintf)
#define malloc_strtoumax JEMALLOC_N(malloc_strtoumax)
#define malloc_tsd_boot JEMALLOC_N(malloc_tsd_boot)
#define malloc_tsd_boot0 JEMALLOC_N(malloc_tsd_boot0)
#define malloc_tsd_boot1 JEMALLOC_N(malloc_tsd_boot1)
#define malloc_tsd_cleanup_register JEMALLOC_N(malloc_tsd_cleanup_register)
#define malloc_tsd_dalloc JEMALLOC_N(malloc_tsd_dalloc)
#define malloc_tsd_malloc JEMALLOC_N(malloc_tsd_malloc)
@@ -251,16 +344,35 @@
#define malloc_vsnprintf JEMALLOC_N(malloc_vsnprintf)
#define malloc_write JEMALLOC_N(malloc_write)
#define map_bias JEMALLOC_N(map_bias)
#define map_misc_offset JEMALLOC_N(map_misc_offset)
#define mb_write JEMALLOC_N(mb_write)
#define mutex_boot JEMALLOC_N(mutex_boot)
#define narenas_auto JEMALLOC_N(narenas_auto)
#define narenas_total JEMALLOC_N(narenas_total)
#define narenas_tdata_cleanup JEMALLOC_N(narenas_tdata_cleanup)
#define narenas_total_get JEMALLOC_N(narenas_total_get)
#define ncpus JEMALLOC_N(ncpus)
#define nhbins JEMALLOC_N(nhbins)
#define nhclasses JEMALLOC_N(nhclasses)
#define nlclasses JEMALLOC_N(nlclasses)
#define nstime_add JEMALLOC_N(nstime_add)
#define nstime_compare JEMALLOC_N(nstime_compare)
#define nstime_copy JEMALLOC_N(nstime_copy)
#define nstime_divide JEMALLOC_N(nstime_divide)
#define nstime_idivide JEMALLOC_N(nstime_idivide)
#define nstime_imultiply JEMALLOC_N(nstime_imultiply)
#define nstime_init JEMALLOC_N(nstime_init)
#define nstime_init2 JEMALLOC_N(nstime_init2)
#define nstime_monotonic JEMALLOC_N(nstime_monotonic)
#define nstime_ns JEMALLOC_N(nstime_ns)
#define nstime_nsec JEMALLOC_N(nstime_nsec)
#define nstime_sec JEMALLOC_N(nstime_sec)
#define nstime_subtract JEMALLOC_N(nstime_subtract)
#define nstime_update JEMALLOC_N(nstime_update)
#define opt_abort JEMALLOC_N(opt_abort)
#define opt_decay_time JEMALLOC_N(opt_decay_time)
#define opt_dss JEMALLOC_N(opt_dss)
#define opt_junk JEMALLOC_N(opt_junk)
#define opt_junk_alloc JEMALLOC_N(opt_junk_alloc)
#define opt_junk_free JEMALLOC_N(opt_junk_free)
#define opt_lg_chunk JEMALLOC_N(opt_lg_chunk)
#define opt_lg_dirty_mult JEMALLOC_N(opt_lg_dirty_mult)
#define opt_lg_prof_interval JEMALLOC_N(opt_lg_prof_interval)
@@ -274,140 +386,254 @@
#define opt_prof_gdump JEMALLOC_N(opt_prof_gdump)
#define opt_prof_leak JEMALLOC_N(opt_prof_leak)
#define opt_prof_prefix JEMALLOC_N(opt_prof_prefix)
#define opt_prof_thread_active_init JEMALLOC_N(opt_prof_thread_active_init)
#define opt_purge JEMALLOC_N(opt_purge)
#define opt_quarantine JEMALLOC_N(opt_quarantine)
#define opt_redzone JEMALLOC_N(opt_redzone)
#define opt_stats_print JEMALLOC_N(opt_stats_print)
#define opt_tcache JEMALLOC_N(opt_tcache)
#define opt_thp JEMALLOC_N(opt_thp)
#define opt_utrace JEMALLOC_N(opt_utrace)
#define opt_valgrind JEMALLOC_N(opt_valgrind)
#define opt_xmalloc JEMALLOC_N(opt_xmalloc)
#define opt_zero JEMALLOC_N(opt_zero)
#define p2rz JEMALLOC_N(p2rz)
#define pages_boot JEMALLOC_N(pages_boot)
#define pages_commit JEMALLOC_N(pages_commit)
#define pages_decommit JEMALLOC_N(pages_decommit)
#define pages_huge JEMALLOC_N(pages_huge)
#define pages_map JEMALLOC_N(pages_map)
#define pages_nohuge JEMALLOC_N(pages_nohuge)
#define pages_purge JEMALLOC_N(pages_purge)
#define pow2_ceil JEMALLOC_N(pow2_ceil)
#define pages_trim JEMALLOC_N(pages_trim)
#define pages_unmap JEMALLOC_N(pages_unmap)
#define pind2sz JEMALLOC_N(pind2sz)
#define pind2sz_compute JEMALLOC_N(pind2sz_compute)
#define pind2sz_lookup JEMALLOC_N(pind2sz_lookup)
#define pind2sz_tab JEMALLOC_N(pind2sz_tab)
#define pow2_ceil_u32 JEMALLOC_N(pow2_ceil_u32)
#define pow2_ceil_u64 JEMALLOC_N(pow2_ceil_u64)
#define pow2_ceil_zu JEMALLOC_N(pow2_ceil_zu)
#define prng_lg_range_u32 JEMALLOC_N(prng_lg_range_u32)
#define prng_lg_range_u64 JEMALLOC_N(prng_lg_range_u64)
#define prng_lg_range_zu JEMALLOC_N(prng_lg_range_zu)
#define prng_range_u32 JEMALLOC_N(prng_range_u32)
#define prng_range_u64 JEMALLOC_N(prng_range_u64)
#define prng_range_zu JEMALLOC_N(prng_range_zu)
#define prng_state_next_u32 JEMALLOC_N(prng_state_next_u32)
#define prng_state_next_u64 JEMALLOC_N(prng_state_next_u64)
#define prng_state_next_zu JEMALLOC_N(prng_state_next_zu)
#define prof_active JEMALLOC_N(prof_active)
#define prof_active_get JEMALLOC_N(prof_active_get)
#define prof_active_get_unlocked JEMALLOC_N(prof_active_get_unlocked)
#define prof_active_set JEMALLOC_N(prof_active_set)
#define prof_alloc_prep JEMALLOC_N(prof_alloc_prep)
#define prof_alloc_rollback JEMALLOC_N(prof_alloc_rollback)
#define prof_backtrace JEMALLOC_N(prof_backtrace)
#define prof_boot0 JEMALLOC_N(prof_boot0)
#define prof_boot1 JEMALLOC_N(prof_boot1)
#define prof_boot2 JEMALLOC_N(prof_boot2)
#define prof_bt_count JEMALLOC_N(prof_bt_count)
#define prof_ctx_get JEMALLOC_N(prof_ctx_get)
#define prof_ctx_set JEMALLOC_N(prof_ctx_set)
#define prof_dump_header JEMALLOC_N(prof_dump_header)
#define prof_dump_open JEMALLOC_N(prof_dump_open)
#define prof_free JEMALLOC_N(prof_free)
#define prof_free_sampled_object JEMALLOC_N(prof_free_sampled_object)
#define prof_gdump JEMALLOC_N(prof_gdump)
#define prof_gdump_get JEMALLOC_N(prof_gdump_get)
#define prof_gdump_get_unlocked JEMALLOC_N(prof_gdump_get_unlocked)
#define prof_gdump_set JEMALLOC_N(prof_gdump_set)
#define prof_gdump_val JEMALLOC_N(prof_gdump_val)
#define prof_idump JEMALLOC_N(prof_idump)
#define prof_interval JEMALLOC_N(prof_interval)
#define prof_lookup JEMALLOC_N(prof_lookup)
#define prof_malloc JEMALLOC_N(prof_malloc)
#define prof_malloc_sample_object JEMALLOC_N(prof_malloc_sample_object)
#define prof_mdump JEMALLOC_N(prof_mdump)
#define prof_postfork_child JEMALLOC_N(prof_postfork_child)
#define prof_postfork_parent JEMALLOC_N(prof_postfork_parent)
#define prof_prefork JEMALLOC_N(prof_prefork)
#define prof_promote JEMALLOC_N(prof_promote)
#define prof_prefork0 JEMALLOC_N(prof_prefork0)
#define prof_prefork1 JEMALLOC_N(prof_prefork1)
#define prof_realloc JEMALLOC_N(prof_realloc)
#define prof_reset JEMALLOC_N(prof_reset)
#define prof_sample_accum_update JEMALLOC_N(prof_sample_accum_update)
#define prof_sample_threshold_update JEMALLOC_N(prof_sample_threshold_update)
#define prof_tdata_booted JEMALLOC_N(prof_tdata_booted)
#define prof_tctx_get JEMALLOC_N(prof_tctx_get)
#define prof_tctx_reset JEMALLOC_N(prof_tctx_reset)
#define prof_tctx_set JEMALLOC_N(prof_tctx_set)
#define prof_tdata_cleanup JEMALLOC_N(prof_tdata_cleanup)
#define prof_tdata_count JEMALLOC_N(prof_tdata_count)
#define prof_tdata_get JEMALLOC_N(prof_tdata_get)
#define prof_tdata_init JEMALLOC_N(prof_tdata_init)
#define prof_tdata_initialized JEMALLOC_N(prof_tdata_initialized)
#define prof_tdata_tls JEMALLOC_N(prof_tdata_tls)
#define prof_tdata_tsd JEMALLOC_N(prof_tdata_tsd)
#define prof_tdata_tsd_boot JEMALLOC_N(prof_tdata_tsd_boot)
#define prof_tdata_tsd_cleanup_wrapper JEMALLOC_N(prof_tdata_tsd_cleanup_wrapper)
#define prof_tdata_tsd_get JEMALLOC_N(prof_tdata_tsd_get)
#define prof_tdata_tsd_get_wrapper JEMALLOC_N(prof_tdata_tsd_get_wrapper)
#define prof_tdata_tsd_init_head JEMALLOC_N(prof_tdata_tsd_init_head)
#define prof_tdata_tsd_set JEMALLOC_N(prof_tdata_tsd_set)
#define prof_tdata_reinit JEMALLOC_N(prof_tdata_reinit)
#define prof_thread_active_get JEMALLOC_N(prof_thread_active_get)
#define prof_thread_active_init_get JEMALLOC_N(prof_thread_active_init_get)
#define prof_thread_active_init_set JEMALLOC_N(prof_thread_active_init_set)
#define prof_thread_active_set JEMALLOC_N(prof_thread_active_set)
#define prof_thread_name_get JEMALLOC_N(prof_thread_name_get)
#define prof_thread_name_set JEMALLOC_N(prof_thread_name_set)
#define psz2ind JEMALLOC_N(psz2ind)
#define psz2u JEMALLOC_N(psz2u)
#define purge_mode_names JEMALLOC_N(purge_mode_names)
#define quarantine JEMALLOC_N(quarantine)
#define quarantine_alloc_hook JEMALLOC_N(quarantine_alloc_hook)
#define quarantine_boot JEMALLOC_N(quarantine_boot)
#define quarantine_booted JEMALLOC_N(quarantine_booted)
#define quarantine_alloc_hook_work JEMALLOC_N(quarantine_alloc_hook_work)
#define quarantine_cleanup JEMALLOC_N(quarantine_cleanup)
#define quarantine_init JEMALLOC_N(quarantine_init)
#define quarantine_tls JEMALLOC_N(quarantine_tls)
#define quarantine_tsd JEMALLOC_N(quarantine_tsd)
#define quarantine_tsd_boot JEMALLOC_N(quarantine_tsd_boot)
#define quarantine_tsd_cleanup_wrapper JEMALLOC_N(quarantine_tsd_cleanup_wrapper)
#define quarantine_tsd_get JEMALLOC_N(quarantine_tsd_get)
#define quarantine_tsd_get_wrapper JEMALLOC_N(quarantine_tsd_get_wrapper)
#define quarantine_tsd_init_head JEMALLOC_N(quarantine_tsd_init_head)
#define quarantine_tsd_set JEMALLOC_N(quarantine_tsd_set)
#define register_zone JEMALLOC_N(register_zone)
#define rtree_child_read JEMALLOC_N(rtree_child_read)
#define rtree_child_read_hard JEMALLOC_N(rtree_child_read_hard)
#define rtree_child_tryread JEMALLOC_N(rtree_child_tryread)
#define rtree_delete JEMALLOC_N(rtree_delete)
#define rtree_get JEMALLOC_N(rtree_get)
#define rtree_get_locked JEMALLOC_N(rtree_get_locked)
#define rtree_new JEMALLOC_N(rtree_new)
#define rtree_postfork_child JEMALLOC_N(rtree_postfork_child)
#define rtree_postfork_parent JEMALLOC_N(rtree_postfork_parent)
#define rtree_prefork JEMALLOC_N(rtree_prefork)
#define rtree_node_valid JEMALLOC_N(rtree_node_valid)
#define rtree_set JEMALLOC_N(rtree_set)
#define rtree_start_level JEMALLOC_N(rtree_start_level)
#define rtree_subkey JEMALLOC_N(rtree_subkey)
#define rtree_subtree_read JEMALLOC_N(rtree_subtree_read)
#define rtree_subtree_read_hard JEMALLOC_N(rtree_subtree_read_hard)
#define rtree_subtree_tryread JEMALLOC_N(rtree_subtree_tryread)
#define rtree_val_read JEMALLOC_N(rtree_val_read)
#define rtree_val_write JEMALLOC_N(rtree_val_write)
#define run_quantize_ceil JEMALLOC_N(run_quantize_ceil)
#define run_quantize_floor JEMALLOC_N(run_quantize_floor)
#define s2u JEMALLOC_N(s2u)
#define s2u_compute JEMALLOC_N(s2u_compute)
#define s2u_lookup JEMALLOC_N(s2u_lookup)
#define sa2u JEMALLOC_N(sa2u)
#define set_errno JEMALLOC_N(set_errno)
#define small_size2bin JEMALLOC_N(small_size2bin)
#define size2index JEMALLOC_N(size2index)
#define size2index_compute JEMALLOC_N(size2index_compute)
#define size2index_lookup JEMALLOC_N(size2index_lookup)
#define size2index_tab JEMALLOC_N(size2index_tab)
#define spin_adaptive JEMALLOC_N(spin_adaptive)
#define spin_init JEMALLOC_N(spin_init)
#define stats_cactive JEMALLOC_N(stats_cactive)
#define stats_cactive_add JEMALLOC_N(stats_cactive_add)
#define stats_cactive_get JEMALLOC_N(stats_cactive_get)
#define stats_cactive_sub JEMALLOC_N(stats_cactive_sub)
#define stats_chunks JEMALLOC_N(stats_chunks)
#define stats_print JEMALLOC_N(stats_print)
#define tcache_alloc_easy JEMALLOC_N(tcache_alloc_easy)
#define tcache_alloc_large JEMALLOC_N(tcache_alloc_large)
#define tcache_alloc_small JEMALLOC_N(tcache_alloc_small)
#define tcache_alloc_small_hard JEMALLOC_N(tcache_alloc_small_hard)
#define tcache_arena_associate JEMALLOC_N(tcache_arena_associate)
#define tcache_arena_dissociate JEMALLOC_N(tcache_arena_dissociate)
#define tcache_arena_reassociate JEMALLOC_N(tcache_arena_reassociate)
#define tcache_bin_flush_large JEMALLOC_N(tcache_bin_flush_large)
#define tcache_bin_flush_small JEMALLOC_N(tcache_bin_flush_small)
#define tcache_bin_info JEMALLOC_N(tcache_bin_info)
#define tcache_boot0 JEMALLOC_N(tcache_boot0)
#define tcache_boot1 JEMALLOC_N(tcache_boot1)
#define tcache_booted JEMALLOC_N(tcache_booted)
#define tcache_boot JEMALLOC_N(tcache_boot)
#define tcache_cleanup JEMALLOC_N(tcache_cleanup)
#define tcache_create JEMALLOC_N(tcache_create)
#define tcache_dalloc_large JEMALLOC_N(tcache_dalloc_large)
#define tcache_dalloc_small JEMALLOC_N(tcache_dalloc_small)
#define tcache_destroy JEMALLOC_N(tcache_destroy)
#define tcache_enabled_booted JEMALLOC_N(tcache_enabled_booted)
#define tcache_enabled_cleanup JEMALLOC_N(tcache_enabled_cleanup)
#define tcache_enabled_get JEMALLOC_N(tcache_enabled_get)
#define tcache_enabled_initialized JEMALLOC_N(tcache_enabled_initialized)
#define tcache_enabled_set JEMALLOC_N(tcache_enabled_set)
#define tcache_enabled_tls JEMALLOC_N(tcache_enabled_tls)
#define tcache_enabled_tsd JEMALLOC_N(tcache_enabled_tsd)
#define tcache_enabled_tsd_boot JEMALLOC_N(tcache_enabled_tsd_boot)
#define tcache_enabled_tsd_cleanup_wrapper JEMALLOC_N(tcache_enabled_tsd_cleanup_wrapper)
#define tcache_enabled_tsd_get JEMALLOC_N(tcache_enabled_tsd_get)
#define tcache_enabled_tsd_get_wrapper JEMALLOC_N(tcache_enabled_tsd_get_wrapper)
#define tcache_enabled_tsd_init_head JEMALLOC_N(tcache_enabled_tsd_init_head)
#define tcache_enabled_tsd_set JEMALLOC_N(tcache_enabled_tsd_set)
#define tcache_event JEMALLOC_N(tcache_event)
#define tcache_event_hard JEMALLOC_N(tcache_event_hard)
#define tcache_flush JEMALLOC_N(tcache_flush)
#define tcache_get JEMALLOC_N(tcache_get)
#define tcache_initialized JEMALLOC_N(tcache_initialized)
#define tcache_get_hard JEMALLOC_N(tcache_get_hard)
#define tcache_maxclass JEMALLOC_N(tcache_maxclass)
#define tcache_postfork_child JEMALLOC_N(tcache_postfork_child)
#define tcache_postfork_parent JEMALLOC_N(tcache_postfork_parent)
#define tcache_prefork JEMALLOC_N(tcache_prefork)
#define tcache_salloc JEMALLOC_N(tcache_salloc)
#define tcache_stats_merge JEMALLOC_N(tcache_stats_merge)
#define tcache_thread_cleanup JEMALLOC_N(tcache_thread_cleanup)
#define tcache_tls JEMALLOC_N(tcache_tls)
#define tcache_tsd JEMALLOC_N(tcache_tsd)
#define tcache_tsd_boot JEMALLOC_N(tcache_tsd_boot)
#define tcache_tsd_cleanup_wrapper JEMALLOC_N(tcache_tsd_cleanup_wrapper)
#define tcache_tsd_get JEMALLOC_N(tcache_tsd_get)
#define tcache_tsd_get_wrapper JEMALLOC_N(tcache_tsd_get_wrapper)
#define tcache_tsd_init_head JEMALLOC_N(tcache_tsd_init_head)
#define tcache_tsd_set JEMALLOC_N(tcache_tsd_set)
#define thread_allocated_booted JEMALLOC_N(thread_allocated_booted)
#define thread_allocated_initialized JEMALLOC_N(thread_allocated_initialized)
#define thread_allocated_tls JEMALLOC_N(thread_allocated_tls)
#define thread_allocated_tsd JEMALLOC_N(thread_allocated_tsd)
#define thread_allocated_tsd_boot JEMALLOC_N(thread_allocated_tsd_boot)
#define thread_allocated_tsd_cleanup_wrapper JEMALLOC_N(thread_allocated_tsd_cleanup_wrapper)
#define thread_allocated_tsd_get JEMALLOC_N(thread_allocated_tsd_get)
#define thread_allocated_tsd_get_wrapper JEMALLOC_N(thread_allocated_tsd_get_wrapper)
#define thread_allocated_tsd_init_head JEMALLOC_N(thread_allocated_tsd_init_head)
#define thread_allocated_tsd_set JEMALLOC_N(thread_allocated_tsd_set)
#define tcaches JEMALLOC_N(tcaches)
#define tcaches_create JEMALLOC_N(tcaches_create)
#define tcaches_destroy JEMALLOC_N(tcaches_destroy)
#define tcaches_flush JEMALLOC_N(tcaches_flush)
#define tcaches_get JEMALLOC_N(tcaches_get)
#define thread_allocated_cleanup JEMALLOC_N(thread_allocated_cleanup)
#define thread_deallocated_cleanup JEMALLOC_N(thread_deallocated_cleanup)
#define ticker_copy JEMALLOC_N(ticker_copy)
#define ticker_init JEMALLOC_N(ticker_init)
#define ticker_read JEMALLOC_N(ticker_read)
#define ticker_tick JEMALLOC_N(ticker_tick)
#define ticker_ticks JEMALLOC_N(ticker_ticks)
#define tsd_arena_get JEMALLOC_N(tsd_arena_get)
#define tsd_arena_set JEMALLOC_N(tsd_arena_set)
#define tsd_arenap_get JEMALLOC_N(tsd_arenap_get)
#define tsd_arenas_tdata_bypass_get JEMALLOC_N(tsd_arenas_tdata_bypass_get)
#define tsd_arenas_tdata_bypass_set JEMALLOC_N(tsd_arenas_tdata_bypass_set)
#define tsd_arenas_tdata_bypassp_get JEMALLOC_N(tsd_arenas_tdata_bypassp_get)
#define tsd_arenas_tdata_get JEMALLOC_N(tsd_arenas_tdata_get)
#define tsd_arenas_tdata_set JEMALLOC_N(tsd_arenas_tdata_set)
#define tsd_arenas_tdatap_get JEMALLOC_N(tsd_arenas_tdatap_get)
#define tsd_boot JEMALLOC_N(tsd_boot)
#define tsd_boot0 JEMALLOC_N(tsd_boot0)
#define tsd_boot1 JEMALLOC_N(tsd_boot1)
#define tsd_booted JEMALLOC_N(tsd_booted)
#define tsd_booted_get JEMALLOC_N(tsd_booted_get)
#define tsd_cleanup JEMALLOC_N(tsd_cleanup)
#define tsd_cleanup_wrapper JEMALLOC_N(tsd_cleanup_wrapper)
#define tsd_fetch JEMALLOC_N(tsd_fetch)
#define tsd_fetch_impl JEMALLOC_N(tsd_fetch_impl)
#define tsd_get JEMALLOC_N(tsd_get)
#define tsd_get_allocates JEMALLOC_N(tsd_get_allocates)
#define tsd_iarena_get JEMALLOC_N(tsd_iarena_get)
#define tsd_iarena_set JEMALLOC_N(tsd_iarena_set)
#define tsd_iarenap_get JEMALLOC_N(tsd_iarenap_get)
#define tsd_initialized JEMALLOC_N(tsd_initialized)
#define tsd_init_check_recursion JEMALLOC_N(tsd_init_check_recursion)
#define tsd_init_finish JEMALLOC_N(tsd_init_finish)
#define tsd_init_head JEMALLOC_N(tsd_init_head)
#define tsd_narenas_tdata_get JEMALLOC_N(tsd_narenas_tdata_get)
#define tsd_narenas_tdata_set JEMALLOC_N(tsd_narenas_tdata_set)
#define tsd_narenas_tdatap_get JEMALLOC_N(tsd_narenas_tdatap_get)
#define tsd_wrapper_get JEMALLOC_N(tsd_wrapper_get)
#define tsd_wrapper_set JEMALLOC_N(tsd_wrapper_set)
#define tsd_nominal JEMALLOC_N(tsd_nominal)
#define tsd_prof_tdata_get JEMALLOC_N(tsd_prof_tdata_get)
#define tsd_prof_tdata_set JEMALLOC_N(tsd_prof_tdata_set)
#define tsd_prof_tdatap_get JEMALLOC_N(tsd_prof_tdatap_get)
#define tsd_quarantine_get JEMALLOC_N(tsd_quarantine_get)
#define tsd_quarantine_set JEMALLOC_N(tsd_quarantine_set)
#define tsd_quarantinep_get JEMALLOC_N(tsd_quarantinep_get)
#define tsd_set JEMALLOC_N(tsd_set)
#define tsd_tcache_enabled_get JEMALLOC_N(tsd_tcache_enabled_get)
#define tsd_tcache_enabled_set JEMALLOC_N(tsd_tcache_enabled_set)
#define tsd_tcache_enabledp_get JEMALLOC_N(tsd_tcache_enabledp_get)
#define tsd_tcache_get JEMALLOC_N(tsd_tcache_get)
#define tsd_tcache_set JEMALLOC_N(tsd_tcache_set)
#define tsd_tcachep_get JEMALLOC_N(tsd_tcachep_get)
#define tsd_thread_allocated_get JEMALLOC_N(tsd_thread_allocated_get)
#define tsd_thread_allocated_set JEMALLOC_N(tsd_thread_allocated_set)
#define tsd_thread_allocatedp_get JEMALLOC_N(tsd_thread_allocatedp_get)
#define tsd_thread_deallocated_get JEMALLOC_N(tsd_thread_deallocated_get)
#define tsd_thread_deallocated_set JEMALLOC_N(tsd_thread_deallocated_set)
#define tsd_thread_deallocatedp_get JEMALLOC_N(tsd_thread_deallocatedp_get)
#define tsd_tls JEMALLOC_N(tsd_tls)
#define tsd_tsd JEMALLOC_N(tsd_tsd)
#define tsd_tsdn JEMALLOC_N(tsd_tsdn)
#define tsd_witness_fork_get JEMALLOC_N(tsd_witness_fork_get)
#define tsd_witness_fork_set JEMALLOC_N(tsd_witness_fork_set)
#define tsd_witness_forkp_get JEMALLOC_N(tsd_witness_forkp_get)
#define tsd_witnesses_get JEMALLOC_N(tsd_witnesses_get)
#define tsd_witnesses_set JEMALLOC_N(tsd_witnesses_set)
#define tsd_witnessesp_get JEMALLOC_N(tsd_witnessesp_get)
#define tsdn_fetch JEMALLOC_N(tsdn_fetch)
#define tsdn_null JEMALLOC_N(tsdn_null)
#define tsdn_tsd JEMALLOC_N(tsdn_tsd)
#define u2rz JEMALLOC_N(u2rz)
#define valgrind_freelike_block JEMALLOC_N(valgrind_freelike_block)
#define valgrind_make_mem_defined JEMALLOC_N(valgrind_make_mem_defined)
#define valgrind_make_mem_noaccess JEMALLOC_N(valgrind_make_mem_noaccess)
#define valgrind_make_mem_undefined JEMALLOC_N(valgrind_make_mem_undefined)
#define witness_assert_depth JEMALLOC_N(witness_assert_depth)
#define witness_assert_depth_to_rank JEMALLOC_N(witness_assert_depth_to_rank)
#define witness_assert_lockless JEMALLOC_N(witness_assert_lockless)
#define witness_assert_not_owner JEMALLOC_N(witness_assert_not_owner)
#define witness_assert_owner JEMALLOC_N(witness_assert_owner)
#define witness_depth_error JEMALLOC_N(witness_depth_error)
#define witness_fork_cleanup JEMALLOC_N(witness_fork_cleanup)
#define witness_init JEMALLOC_N(witness_init)
#define witness_lock JEMALLOC_N(witness_lock)
#define witness_lock_error JEMALLOC_N(witness_lock_error)
#define witness_not_owner_error JEMALLOC_N(witness_not_owner_error)
#define witness_owner JEMALLOC_N(witness_owner)
#define witness_owner_error JEMALLOC_N(witness_owner_error)
#define witness_postfork_child JEMALLOC_N(witness_postfork_child)
#define witness_postfork_parent JEMALLOC_N(witness_postfork_parent)
#define witness_prefork JEMALLOC_N(witness_prefork)
#define witness_unlock JEMALLOC_N(witness_unlock)
#define witnesses_cleanup JEMALLOC_N(witnesses_cleanup)
#define zone_register JEMALLOC_N(zone_register)

197
deps/jemalloc/include/jemalloc/internal/prng.h vendored
View File

@@ -1,5 +1,8 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_PRNG_H
#define JEMALLOC_INTERNAL_PRNG_H
#include "jemalloc/internal/atomic.h"
#include "jemalloc/internal/bit_util.h"
/*
* Simple linear congruential pseudo-random number generator:
@@ -15,46 +18,168 @@
* See Knuth's TAOCP 3rd Ed., Vol. 2, pg. 17 for details on these constraints.
*
* This choice of m has the disadvantage that the quality of the bits is
* proportional to bit position. For example. the lowest bit has a cycle of 2,
* proportional to bit position. For example, the lowest bit has a cycle of 2,
* the next has a cycle of 4, etc. For this reason, we prefer to use the upper
* bits.
*
* Macro parameters:
* uint32_t r : Result.
* unsigned lg_range : (0..32], number of least significant bits to return.
* uint32_t state : Seed value.
* const uint32_t a, c : See above discussion.
*/
#define prng32(r, lg_range, state, a, c) do { \
assert(lg_range > 0); \
assert(lg_range <= 32); \
\
r = (state * (a)) + (c); \
state = r; \
r >>= (32 - lg_range); \
} while (false)
/* Same as prng32(), but 64 bits of pseudo-randomness, using uint64_t. */
#define prng64(r, lg_range, state, a, c) do { \
assert(lg_range > 0); \
assert(lg_range <= 64); \
\
r = (state * (a)) + (c); \
state = r; \
r >>= (64 - lg_range); \
} while (false)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/* INTERNAL DEFINITIONS -- IGNORE */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#define PRNG_A_32 UINT32_C(1103515241)
#define PRNG_C_32 UINT32_C(12347)
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#define PRNG_A_64 UINT64_C(6364136223846793005)
#define PRNG_C_64 UINT64_C(1442695040888963407)
JEMALLOC_ALWAYS_INLINE uint32_t
prng_state_next_u32(uint32_t state) {
return (state * PRNG_A_32) + PRNG_C_32;
}
JEMALLOC_ALWAYS_INLINE uint64_t
prng_state_next_u64(uint64_t state) {
return (state * PRNG_A_64) + PRNG_C_64;
}
JEMALLOC_ALWAYS_INLINE size_t
prng_state_next_zu(size_t state) {
#if LG_SIZEOF_PTR == 2
return (state * PRNG_A_32) + PRNG_C_32;
#elif LG_SIZEOF_PTR == 3
return (state * PRNG_A_64) + PRNG_C_64;
#else
#error Unsupported pointer size
#endif
}
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
/* BEGIN PUBLIC API */
/******************************************************************************/
/*
* The prng_lg_range functions give a uniform int in the half-open range [0,
* 2**lg_range). If atomic is true, they do so safely from multiple threads.
* Multithreaded 64-bit prngs aren't supported.
*/
JEMALLOC_ALWAYS_INLINE uint32_t
prng_lg_range_u32(atomic_u32_t *state, unsigned lg_range, bool atomic) {
uint32_t ret, state0, state1;
assert(lg_range > 0);
assert(lg_range <= 32);
state0 = atomic_load_u32(state, ATOMIC_RELAXED);
if (atomic) {
do {
state1 = prng_state_next_u32(state0);
} while (!atomic_compare_exchange_weak_u32(state, &state0,
state1, ATOMIC_RELAXED, ATOMIC_RELAXED));
} else {
state1 = prng_state_next_u32(state0);
atomic_store_u32(state, state1, ATOMIC_RELAXED);
}
ret = state1 >> (32 - lg_range);
return ret;
}
JEMALLOC_ALWAYS_INLINE uint64_t
prng_lg_range_u64(uint64_t *state, unsigned lg_range) {
uint64_t ret, state1;
assert(lg_range > 0);
assert(lg_range <= 64);
state1 = prng_state_next_u64(*state);
*state = state1;
ret = state1 >> (64 - lg_range);
return ret;
}
JEMALLOC_ALWAYS_INLINE size_t
prng_lg_range_zu(atomic_zu_t *state, unsigned lg_range, bool atomic) {
size_t ret, state0, state1;
assert(lg_range > 0);
assert(lg_range <= ZU(1) << (3 + LG_SIZEOF_PTR));
state0 = atomic_load_zu(state, ATOMIC_RELAXED);
if (atomic) {
do {
state1 = prng_state_next_zu(state0);
} while (atomic_compare_exchange_weak_zu(state, &state0,
state1, ATOMIC_RELAXED, ATOMIC_RELAXED));
} else {
state1 = prng_state_next_zu(state0);
atomic_store_zu(state, state1, ATOMIC_RELAXED);
}
ret = state1 >> ((ZU(1) << (3 + LG_SIZEOF_PTR)) - lg_range);
return ret;
}
/*
* The prng_range functions behave like the prng_lg_range, but return a result
* in [0, range) instead of [0, 2**lg_range).
*/
JEMALLOC_ALWAYS_INLINE uint32_t
prng_range_u32(atomic_u32_t *state, uint32_t range, bool atomic) {
uint32_t ret;
unsigned lg_range;
assert(range > 1);
/* Compute the ceiling of lg(range). */
lg_range = ffs_u32(pow2_ceil_u32(range)) - 1;
/* Generate a result in [0..range) via repeated trial. */
do {
ret = prng_lg_range_u32(state, lg_range, atomic);
} while (ret >= range);
return ret;
}
JEMALLOC_ALWAYS_INLINE uint64_t
prng_range_u64(uint64_t *state, uint64_t range) {
uint64_t ret;
unsigned lg_range;
assert(range > 1);
/* Compute the ceiling of lg(range). */
lg_range = ffs_u64(pow2_ceil_u64(range)) - 1;
/* Generate a result in [0..range) via repeated trial. */
do {
ret = prng_lg_range_u64(state, lg_range);
} while (ret >= range);
return ret;
}
JEMALLOC_ALWAYS_INLINE size_t
prng_range_zu(atomic_zu_t *state, size_t range, bool atomic) {
size_t ret;
unsigned lg_range;
assert(range > 1);
/* Compute the ceiling of lg(range). */
lg_range = ffs_u64(pow2_ceil_u64(range)) - 1;
/* Generate a result in [0..range) via repeated trial. */
do {
ret = prng_lg_range_zu(state, lg_range, atomic);
} while (ret >= range);
return ret;
}
#endif /* JEMALLOC_INTERNAL_PRNG_H */

47
deps/jemalloc/include/jemalloc/internal/ql.h vendored
View File

@@ -1,61 +1,64 @@
/*
* List definitions.
*/
#define ql_head(a_type) \
#ifndef JEMALLOC_INTERNAL_QL_H
#define JEMALLOC_INTERNAL_QL_H
#include "jemalloc/internal/qr.h"
/* List definitions. */
#define ql_head(a_type) \
struct { \
a_type *qlh_first; \
}
#define ql_head_initializer(a_head) {NULL}
#define ql_head_initializer(a_head) {NULL}
#define ql_elm(a_type) qr(a_type)
#define ql_elm(a_type) qr(a_type)
/* List functions. */
#define ql_new(a_head) do { \
#define ql_new(a_head) do { \
(a_head)->qlh_first = NULL; \
} while (0)
#define ql_elm_new(a_elm, a_field) qr_new((a_elm), a_field)
#define ql_elm_new(a_elm, a_field) qr_new((a_elm), a_field)
#define ql_first(a_head) ((a_head)->qlh_first)
#define ql_first(a_head) ((a_head)->qlh_first)
#define ql_last(a_head, a_field) \
#define ql_last(a_head, a_field) \
((ql_first(a_head) != NULL) \
? qr_prev(ql_first(a_head), a_field) : NULL)
#define ql_next(a_head, a_elm, a_field) \
#define ql_next(a_head, a_elm, a_field) \
((ql_last(a_head, a_field) != (a_elm)) \
? qr_next((a_elm), a_field) : NULL)
#define ql_prev(a_head, a_elm, a_field) \
#define ql_prev(a_head, a_elm, a_field) \
((ql_first(a_head) != (a_elm)) ? qr_prev((a_elm), a_field) \
: NULL)
#define ql_before_insert(a_head, a_qlelm, a_elm, a_field) do { \
#define ql_before_insert(a_head, a_qlelm, a_elm, a_field) do { \
qr_before_insert((a_qlelm), (a_elm), a_field); \
if (ql_first(a_head) == (a_qlelm)) { \
ql_first(a_head) = (a_elm); \
} \
} while (0)
#define ql_after_insert(a_qlelm, a_elm, a_field) \
#define ql_after_insert(a_qlelm, a_elm, a_field) \
qr_after_insert((a_qlelm), (a_elm), a_field)
#define ql_head_insert(a_head, a_elm, a_field) do { \
#define ql_head_insert(a_head, a_elm, a_field) do { \
if (ql_first(a_head) != NULL) { \
qr_before_insert(ql_first(a_head), (a_elm), a_field); \
} \
ql_first(a_head) = (a_elm); \
} while (0)
#define ql_tail_insert(a_head, a_elm, a_field) do { \
#define ql_tail_insert(a_head, a_elm, a_field) do { \
if (ql_first(a_head) != NULL) { \
qr_before_insert(ql_first(a_head), (a_elm), a_field); \
} \
ql_first(a_head) = qr_next((a_elm), a_field); \
} while (0)
#define ql_remove(a_head, a_elm, a_field) do { \
#define ql_remove(a_head, a_elm, a_field) do { \
if (ql_first(a_head) == (a_elm)) { \
ql_first(a_head) = qr_next(ql_first(a_head), a_field); \
} \
@@ -66,18 +69,20 @@ struct { \
} \
} while (0)
#define ql_head_remove(a_head, a_type, a_field) do { \
#define ql_head_remove(a_head, a_type, a_field) do { \
a_type *t = ql_first(a_head); \
ql_remove((a_head), t, a_field); \
} while (0)
#define ql_tail_remove(a_head, a_type, a_field) do { \
#define ql_tail_remove(a_head, a_type, a_field) do { \
a_type *t = ql_last(a_head, a_field); \
ql_remove((a_head), t, a_field); \
} while (0)
#define ql_foreach(a_var, a_head, a_field) \
#define ql_foreach(a_var, a_head, a_field) \
qr_foreach((a_var), ql_first(a_head), a_field)
#define ql_reverse_foreach(a_var, a_head, a_field) \
#define ql_reverse_foreach(a_var, a_head, a_field) \
qr_reverse_foreach((a_var), ql_first(a_head), a_field)
#endif /* JEMALLOC_INTERNAL_QL_H */

41
deps/jemalloc/include/jemalloc/internal/qr.h vendored
View File

@@ -1,38 +1,39 @@
#ifndef JEMALLOC_INTERNAL_QR_H
#define JEMALLOC_INTERNAL_QR_H
/* Ring definitions. */
#define qr(a_type) \
#define qr(a_type) \
struct { \
a_type *qre_next; \
a_type *qre_prev; \
}
/* Ring functions. */
#define qr_new(a_qr, a_field) do { \
#define qr_new(a_qr, a_field) do { \
(a_qr)->a_field.qre_next = (a_qr); \
(a_qr)->a_field.qre_prev = (a_qr); \
} while (0)
#define qr_next(a_qr, a_field) ((a_qr)->a_field.qre_next)
#define qr_next(a_qr, a_field) ((a_qr)->a_field.qre_next)
#define qr_prev(a_qr, a_field) ((a_qr)->a_field.qre_prev)
#define qr_prev(a_qr, a_field) ((a_qr)->a_field.qre_prev)
#define qr_before_insert(a_qrelm, a_qr, a_field) do { \
#define qr_before_insert(a_qrelm, a_qr, a_field) do { \
(a_qr)->a_field.qre_prev = (a_qrelm)->a_field.qre_prev; \
(a_qr)->a_field.qre_next = (a_qrelm); \
(a_qr)->a_field.qre_prev->a_field.qre_next = (a_qr); \
(a_qrelm)->a_field.qre_prev = (a_qr); \
} while (0)
#define qr_after_insert(a_qrelm, a_qr, a_field) \
do \
{ \
#define qr_after_insert(a_qrelm, a_qr, a_field) do { \
(a_qr)->a_field.qre_next = (a_qrelm)->a_field.qre_next; \
(a_qr)->a_field.qre_prev = (a_qrelm); \
(a_qr)->a_field.qre_next->a_field.qre_prev = (a_qr); \
(a_qrelm)->a_field.qre_next = (a_qr); \
} while (0)
} while (0)
#define qr_meld(a_qr_a, a_qr_b, a_field) do { \
void *t; \
#define qr_meld(a_qr_a, a_qr_b, a_type, a_field) do { \
a_type *t; \
(a_qr_a)->a_field.qre_prev->a_field.qre_next = (a_qr_b); \
(a_qr_b)->a_field.qre_prev->a_field.qre_next = (a_qr_a); \
t = (a_qr_a)->a_field.qre_prev; \
@@ -40,12 +41,14 @@ struct { \
(a_qr_b)->a_field.qre_prev = t; \
} while (0)
/* qr_meld() and qr_split() are functionally equivalent, so there's no need to
* have two copies of the code. */
#define qr_split(a_qr_a, a_qr_b, a_field) \
qr_meld((a_qr_a), (a_qr_b), a_field)
/*
* qr_meld() and qr_split() are functionally equivalent, so there's no need to
* have two copies of the code.
*/
#define qr_split(a_qr_a, a_qr_b, a_type, a_field) \
qr_meld((a_qr_a), (a_qr_b), a_type, a_field)
#define qr_remove(a_qr, a_field) do { \
#define qr_remove(a_qr, a_field) do { \
(a_qr)->a_field.qre_prev->a_field.qre_next \
= (a_qr)->a_field.qre_next; \
(a_qr)->a_field.qre_next->a_field.qre_prev \
@@ -54,14 +57,16 @@ struct { \
(a_qr)->a_field.qre_prev = (a_qr); \
} while (0)
#define qr_foreach(var, a_qr, a_field) \
#define qr_foreach(var, a_qr, a_field) \
for ((var) = (a_qr); \
(var) != NULL; \
(var) = (((var)->a_field.qre_next != (a_qr)) \
? (var)->a_field.qre_next : NULL))
#define qr_reverse_foreach(var, a_qr, a_field) \
#define qr_reverse_foreach(var, a_qr, a_field) \
for ((var) = ((a_qr) != NULL) ? qr_prev(a_qr, a_field) : NULL; \
(var) != NULL; \
(var) = (((var) != (a_qr)) \
? (var)->a_field.qre_prev : NULL))
#endif /* JEMALLOC_INTERNAL_QR_H */

351
deps/jemalloc/include/jemalloc/internal/rb.h vendored
View File

@@ -20,17 +20,21 @@
*/
#ifndef RB_H_
#define RB_H_
#define RB_H_
#ifndef __PGI
#define RB_COMPACT
#endif
#ifdef RB_COMPACT
/* Node structure. */
#define rb_node(a_type) \
#define rb_node(a_type) \
struct { \
a_type *rbn_left; \
a_type *rbn_right_red; \
}
#else
#define rb_node(a_type) \
#define rb_node(a_type) \
struct { \
a_type *rbn_left; \
a_type *rbn_right; \
@@ -39,111 +43,116 @@ struct { \
#endif
/* Root structure. */
#define rb_tree(a_type) \
#define rb_tree(a_type) \
struct { \
a_type *rbt_root; \
a_type rbt_nil; \
}
/* Left accessors. */
#define rbtn_left_get(a_type, a_field, a_node) \
#define rbtn_left_get(a_type, a_field, a_node) \
((a_node)->a_field.rbn_left)
#define rbtn_left_set(a_type, a_field, a_node, a_left) do { \
#define rbtn_left_set(a_type, a_field, a_node, a_left) do { \
(a_node)->a_field.rbn_left = a_left; \
} while (0)
#ifdef RB_COMPACT
/* Right accessors. */
#define rbtn_right_get(a_type, a_field, a_node) \
#define rbtn_right_get(a_type, a_field, a_node) \
((a_type *) (((intptr_t) (a_node)->a_field.rbn_right_red) \
& ((ssize_t)-2)))
#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \
#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \
(a_node)->a_field.rbn_right_red = (a_type *) (((uintptr_t) a_right) \
| (((uintptr_t) (a_node)->a_field.rbn_right_red) & ((size_t)1))); \
} while (0)
/* Color accessors. */
#define rbtn_red_get(a_type, a_field, a_node) \
#define rbtn_red_get(a_type, a_field, a_node) \
((bool) (((uintptr_t) (a_node)->a_field.rbn_right_red) \
& ((size_t)1)))
#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \
#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \
(a_node)->a_field.rbn_right_red = (a_type *) ((((intptr_t) \
(a_node)->a_field.rbn_right_red) & ((ssize_t)-2)) \
| ((ssize_t)a_red)); \
} while (0)
#define rbtn_red_set(a_type, a_field, a_node) do { \
#define rbtn_red_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_right_red = (a_type *) (((uintptr_t) \
(a_node)->a_field.rbn_right_red) | ((size_t)1)); \
} while (0)
#define rbtn_black_set(a_type, a_field, a_node) do { \
#define rbtn_black_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_right_red = (a_type *) (((intptr_t) \
(a_node)->a_field.rbn_right_red) & ((ssize_t)-2)); \
} while (0)
/* Node initializer. */
#define rbt_node_new(a_type, a_field, a_rbt, a_node) do { \
/* Bookkeeping bit cannot be used by node pointer. */ \
assert(((uintptr_t)(a_node) & 0x1) == 0); \
rbtn_left_set(a_type, a_field, (a_node), NULL); \
rbtn_right_set(a_type, a_field, (a_node), NULL); \
rbtn_red_set(a_type, a_field, (a_node)); \
} while (0)
#else
/* Right accessors. */
#define rbtn_right_get(a_type, a_field, a_node) \
#define rbtn_right_get(a_type, a_field, a_node) \
((a_node)->a_field.rbn_right)
#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \
#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \
(a_node)->a_field.rbn_right = a_right; \
} while (0)
/* Color accessors. */
#define rbtn_red_get(a_type, a_field, a_node) \
#define rbtn_red_get(a_type, a_field, a_node) \
((a_node)->a_field.rbn_red)
#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \
#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \
(a_node)->a_field.rbn_red = (a_red); \
} while (0)
#define rbtn_red_set(a_type, a_field, a_node) do { \
#define rbtn_red_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_red = true; \
} while (0)
#define rbtn_black_set(a_type, a_field, a_node) do { \
#define rbtn_black_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_red = false; \
} while (0)
/* Node initializer. */
#define rbt_node_new(a_type, a_field, a_rbt, a_node) do { \
rbtn_left_set(a_type, a_field, (a_node), NULL); \
rbtn_right_set(a_type, a_field, (a_node), NULL); \
rbtn_red_set(a_type, a_field, (a_node)); \
} while (0)
#endif
/* Node initializer. */
#define rbt_node_new(a_type, a_field, a_rbt, a_node) do { \
rbtn_left_set(a_type, a_field, (a_node), &(a_rbt)->rbt_nil); \
rbtn_right_set(a_type, a_field, (a_node), &(a_rbt)->rbt_nil); \
rbtn_red_set(a_type, a_field, (a_node)); \
} while (0)
/* Tree initializer. */
#define rb_new(a_type, a_field, a_rbt) do { \
(a_rbt)->rbt_root = &(a_rbt)->rbt_nil; \
rbt_node_new(a_type, a_field, a_rbt, &(a_rbt)->rbt_nil); \
rbtn_black_set(a_type, a_field, &(a_rbt)->rbt_nil); \
#define rb_new(a_type, a_field, a_rbt) do { \
(a_rbt)->rbt_root = NULL; \
} while (0)
/* Internal utility macros. */
#define rbtn_first(a_type, a_field, a_rbt, a_root, r_node) do { \
#define rbtn_first(a_type, a_field, a_rbt, a_root, r_node) do { \
(r_node) = (a_root); \
if ((r_node) != &(a_rbt)->rbt_nil) { \
if ((r_node) != NULL) { \
for (; \
rbtn_left_get(a_type, a_field, (r_node)) != &(a_rbt)->rbt_nil;\
rbtn_left_get(a_type, a_field, (r_node)) != NULL; \
(r_node) = rbtn_left_get(a_type, a_field, (r_node))) { \
} \
} \
} while (0)
#define rbtn_last(a_type, a_field, a_rbt, a_root, r_node) do { \
#define rbtn_last(a_type, a_field, a_rbt, a_root, r_node) do { \
(r_node) = (a_root); \
if ((r_node) != &(a_rbt)->rbt_nil) { \
for (; rbtn_right_get(a_type, a_field, (r_node)) != \
&(a_rbt)->rbt_nil; (r_node) = rbtn_right_get(a_type, a_field, \
(r_node))) { \
if ((r_node) != NULL) { \
for (; rbtn_right_get(a_type, a_field, (r_node)) != NULL; \
(r_node) = rbtn_right_get(a_type, a_field, (r_node))) { \
} \
} \
} while (0)
#define rbtn_rotate_left(a_type, a_field, a_node, r_node) do { \
#define rbtn_rotate_left(a_type, a_field, a_node, r_node) do { \
(r_node) = rbtn_right_get(a_type, a_field, (a_node)); \
rbtn_right_set(a_type, a_field, (a_node), \
rbtn_left_get(a_type, a_field, (r_node))); \
rbtn_left_set(a_type, a_field, (r_node), (a_node)); \
} while (0)
#define rbtn_rotate_right(a_type, a_field, a_node, r_node) do { \
#define rbtn_rotate_right(a_type, a_field, a_node, r_node) do { \
(r_node) = rbtn_left_get(a_type, a_field, (a_node)); \
rbtn_left_set(a_type, a_field, (a_node), \
rbtn_right_get(a_type, a_field, (r_node))); \
@@ -155,9 +164,11 @@ struct { \
* functions generated by an equivalently parameterized call to rb_gen().
*/
#define rb_proto(a_attr, a_prefix, a_rbt_type, a_type) \
#define rb_proto(a_attr, a_prefix, a_rbt_type, a_type) \
a_attr void \
a_prefix##new(a_rbt_type *rbtree); \
a_attr bool \
a_prefix##empty(a_rbt_type *rbtree); \
a_attr a_type * \
a_prefix##first(a_rbt_type *rbtree); \
a_attr a_type * \
@@ -167,11 +178,11 @@ a_prefix##next(a_rbt_type *rbtree, a_type *node); \
a_attr a_type * \
a_prefix##prev(a_rbt_type *rbtree, a_type *node); \
a_attr a_type * \
a_prefix##search(a_rbt_type *rbtree, a_type *key); \
a_prefix##search(a_rbt_type *rbtree, const a_type *key); \
a_attr a_type * \
a_prefix##nsearch(a_rbt_type *rbtree, a_type *key); \
a_prefix##nsearch(a_rbt_type *rbtree, const a_type *key); \
a_attr a_type * \
a_prefix##psearch(a_rbt_type *rbtree, a_type *key); \
a_prefix##psearch(a_rbt_type *rbtree, const a_type *key); \
a_attr void \
a_prefix##insert(a_rbt_type *rbtree, a_type *node); \
a_attr void \
@@ -181,7 +192,10 @@ a_prefix##iter(a_rbt_type *rbtree, a_type *start, a_type *(*cb)( \
a_rbt_type *, a_type *, void *), void *arg); \
a_attr a_type * \
a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg);
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg); \
a_attr void \
a_prefix##destroy(a_rbt_type *rbtree, void (*cb)(a_type *, void *), \
void *arg);
/*
* The rb_gen() macro generates a type-specific red-black tree implementation,
@@ -198,7 +212,7 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
* int (a_cmp *)(a_type *a_node, a_type *a_other);
* ^^^^^^
* or a_key
* Interpretation of comparision function return values:
* Interpretation of comparison function return values:
* -1 : a_node < a_other
* 0 : a_node == a_other
* 1 : a_node > a_other
@@ -224,6 +238,13 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
* Args:
* tree: Pointer to an uninitialized red-black tree object.
*
* static bool
* ex_empty(ex_t *tree);
* Description: Determine whether tree is empty.
* Args:
* tree: Pointer to an initialized red-black tree object.
* Ret: True if tree is empty, false otherwise.
*
* static ex_node_t *
* ex_first(ex_t *tree);
* static ex_node_t *
@@ -245,7 +266,7 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
* last/first.
*
* static ex_node_t *
* ex_search(ex_t *tree, ex_node_t *key);
* ex_search(ex_t *tree, const ex_node_t *key);
* Description: Search for node that matches key.
* Args:
* tree: Pointer to an initialized red-black tree object.
@@ -253,9 +274,9 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
* Ret: Node in tree that matches key, or NULL if no match.
*
* static ex_node_t *
* ex_nsearch(ex_t *tree, ex_node_t *key);
* ex_nsearch(ex_t *tree, const ex_node_t *key);
* static ex_node_t *
* ex_psearch(ex_t *tree, ex_node_t *key);
* ex_psearch(ex_t *tree, const ex_node_t *key);
* Description: Search for node that matches key. If no match is found,
* return what would be key's successor/predecessor, were
* key in tree.
@@ -303,40 +324,52 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
* arg : Opaque pointer passed to cb().
* Ret: NULL if iteration completed, or the non-NULL callback return value
* that caused termination of the iteration.
*
* static void
* ex_destroy(ex_t *tree, void (*cb)(ex_node_t *, void *), void *arg);
* Description: Iterate over the tree with post-order traversal, remove
* each node, and run the callback if non-null. This is
* used for destroying a tree without paying the cost to
* rebalance it. The tree must not be otherwise altered
* during traversal.
* Args:
* tree: Pointer to an initialized red-black tree object.
* cb : Callback function, which, if non-null, is called for each node
* during iteration. There is no way to stop iteration once it
* has begun.
* arg : Opaque pointer passed to cb().
*/
#define rb_gen(a_attr, a_prefix, a_rbt_type, a_type, a_field, a_cmp) \
#define rb_gen(a_attr, a_prefix, a_rbt_type, a_type, a_field, a_cmp) \
a_attr void \
a_prefix##new(a_rbt_type *rbtree) { \
rb_new(a_type, a_field, rbtree); \
} \
a_attr bool \
a_prefix##empty(a_rbt_type *rbtree) { \
return (rbtree->rbt_root == NULL); \
} \
a_attr a_type * \
a_prefix##first(a_rbt_type *rbtree) { \
a_type *ret; \
rbtn_first(a_type, a_field, rbtree, rbtree->rbt_root, ret); \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
} \
return (ret); \
return ret; \
} \
a_attr a_type * \
a_prefix##last(a_rbt_type *rbtree) { \
a_type *ret; \
rbtn_last(a_type, a_field, rbtree, rbtree->rbt_root, ret); \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
} \
return (ret); \
return ret; \
} \
a_attr a_type * \
a_prefix##next(a_rbt_type *rbtree, a_type *node) { \
a_type *ret; \
if (rbtn_right_get(a_type, a_field, node) != &rbtree->rbt_nil) { \
if (rbtn_right_get(a_type, a_field, node) != NULL) { \
rbtn_first(a_type, a_field, rbtree, rbtn_right_get(a_type, \
a_field, node), ret); \
} else { \
a_type *tnode = rbtree->rbt_root; \
assert(tnode != &rbtree->rbt_nil); \
ret = &rbtree->rbt_nil; \
assert(tnode != NULL); \
ret = NULL; \
while (true) { \
int cmp = (a_cmp)(node, tnode); \
if (cmp < 0) { \
@@ -347,24 +380,21 @@ a_prefix##next(a_rbt_type *rbtree, a_type *node) { \
} else { \
break; \
} \
assert(tnode != &rbtree->rbt_nil); \
assert(tnode != NULL); \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
return ret; \
} \
a_attr a_type * \
a_prefix##prev(a_rbt_type *rbtree, a_type *node) { \
a_type *ret; \
if (rbtn_left_get(a_type, a_field, node) != &rbtree->rbt_nil) { \
if (rbtn_left_get(a_type, a_field, node) != NULL) { \
rbtn_last(a_type, a_field, rbtree, rbtn_left_get(a_type, \
a_field, node), ret); \
} else { \
a_type *tnode = rbtree->rbt_root; \
assert(tnode != &rbtree->rbt_nil); \
ret = &rbtree->rbt_nil; \
assert(tnode != NULL); \
ret = NULL; \
while (true) { \
int cmp = (a_cmp)(node, tnode); \
if (cmp < 0) { \
@@ -375,20 +405,17 @@ a_prefix##prev(a_rbt_type *rbtree, a_type *node) { \
} else { \
break; \
} \
assert(tnode != &rbtree->rbt_nil); \
assert(tnode != NULL); \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
return ret; \
} \
a_attr a_type * \
a_prefix##search(a_rbt_type *rbtree, a_type *key) { \
a_prefix##search(a_rbt_type *rbtree, const a_type *key) { \
a_type *ret; \
int cmp; \
ret = rbtree->rbt_root; \
while (ret != &rbtree->rbt_nil \
while (ret != NULL \
&& (cmp = (a_cmp)(key, ret)) != 0) { \
if (cmp < 0) { \
ret = rbtn_left_get(a_type, a_field, ret); \
@@ -396,17 +423,14 @@ a_prefix##search(a_rbt_type *rbtree, a_type *key) { \
ret = rbtn_right_get(a_type, a_field, ret); \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
return ret; \
} \
a_attr a_type * \
a_prefix##nsearch(a_rbt_type *rbtree, a_type *key) { \
a_prefix##nsearch(a_rbt_type *rbtree, const a_type *key) { \
a_type *ret; \
a_type *tnode = rbtree->rbt_root; \
ret = &rbtree->rbt_nil; \
while (tnode != &rbtree->rbt_nil) { \
ret = NULL; \
while (tnode != NULL) { \
int cmp = (a_cmp)(key, tnode); \
if (cmp < 0) { \
ret = tnode; \
@@ -418,17 +442,14 @@ a_prefix##nsearch(a_rbt_type *rbtree, a_type *key) { \
break; \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
return ret; \
} \
a_attr a_type * \
a_prefix##psearch(a_rbt_type *rbtree, a_type *key) { \
a_prefix##psearch(a_rbt_type *rbtree, const a_type *key) { \
a_type *ret; \
a_type *tnode = rbtree->rbt_root; \
ret = &rbtree->rbt_nil; \
while (tnode != &rbtree->rbt_nil) { \
ret = NULL; \
while (tnode != NULL) { \
int cmp = (a_cmp)(key, tnode); \
if (cmp < 0) { \
tnode = rbtn_left_get(a_type, a_field, tnode); \
@@ -440,10 +461,7 @@ a_prefix##psearch(a_rbt_type *rbtree, a_type *key) { \
break; \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
return ret; \
} \
a_attr void \
a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \
@@ -454,7 +472,7 @@ a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \
rbt_node_new(a_type, a_field, rbtree, node); \
/* Wind. */ \
path->node = rbtree->rbt_root; \
for (pathp = path; pathp->node != &rbtree->rbt_nil; pathp++) { \
for (pathp = path; pathp->node != NULL; pathp++) { \
int cmp = pathp->cmp = a_cmp(node, pathp->node); \
assert(cmp != 0); \
if (cmp < 0) { \
@@ -474,7 +492,8 @@ a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \
rbtn_left_set(a_type, a_field, cnode, left); \
if (rbtn_red_get(a_type, a_field, left)) { \
a_type *leftleft = rbtn_left_get(a_type, a_field, left);\
if (rbtn_red_get(a_type, a_field, leftleft)) { \
if (leftleft != NULL && rbtn_red_get(a_type, a_field, \
leftleft)) { \
/* Fix up 4-node. */ \
a_type *tnode; \
rbtn_black_set(a_type, a_field, leftleft); \
@@ -489,7 +508,8 @@ a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \
rbtn_right_set(a_type, a_field, cnode, right); \
if (rbtn_red_get(a_type, a_field, right)) { \
a_type *left = rbtn_left_get(a_type, a_field, cnode); \
if (rbtn_red_get(a_type, a_field, left)) { \
if (left != NULL && rbtn_red_get(a_type, a_field, \
left)) { \
/* Split 4-node. */ \
rbtn_black_set(a_type, a_field, left); \
rbtn_black_set(a_type, a_field, right); \
@@ -522,7 +542,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
/* Wind. */ \
nodep = NULL; /* Silence compiler warning. */ \
path->node = rbtree->rbt_root; \
for (pathp = path; pathp->node != &rbtree->rbt_nil; pathp++) { \
for (pathp = path; pathp->node != NULL; pathp++) { \
int cmp = pathp->cmp = a_cmp(node, pathp->node); \
if (cmp < 0) { \
pathp[1].node = rbtn_left_get(a_type, a_field, \
@@ -534,8 +554,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
/* Find node's successor, in preparation for swap. */ \
pathp->cmp = 1; \
nodep = pathp; \
for (pathp++; pathp->node != &rbtree->rbt_nil; \
pathp++) { \
for (pathp++; pathp->node != NULL; pathp++) { \
pathp->cmp = -1; \
pathp[1].node = rbtn_left_get(a_type, a_field, \
pathp->node); \
@@ -577,10 +596,10 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
} \
} else { \
a_type *left = rbtn_left_get(a_type, a_field, node); \
if (left != &rbtree->rbt_nil) { \
if (left != NULL) { \
/* node has no successor, but it has a left child. */\
/* Splice node out, without losing the left child. */\
assert(rbtn_red_get(a_type, a_field, node) == false); \
assert(!rbtn_red_get(a_type, a_field, node)); \
assert(rbtn_red_get(a_type, a_field, left)); \
rbtn_black_set(a_type, a_field, left); \
if (pathp == path) { \
@@ -597,34 +616,32 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
return; \
} else if (pathp == path) { \
/* The tree only contained one node. */ \
rbtree->rbt_root = &rbtree->rbt_nil; \
rbtree->rbt_root = NULL; \
return; \
} \
} \
if (rbtn_red_get(a_type, a_field, pathp->node)) { \
/* Prune red node, which requires no fixup. */ \
assert(pathp[-1].cmp < 0); \
rbtn_left_set(a_type, a_field, pathp[-1].node, \
&rbtree->rbt_nil); \
rbtn_left_set(a_type, a_field, pathp[-1].node, NULL); \
return; \
} \
/* The node to be pruned is black, so unwind until balance is */\
/* restored. */\
pathp->node = &rbtree->rbt_nil; \
pathp->node = NULL; \
for (pathp--; (uintptr_t)pathp >= (uintptr_t)path; pathp--) { \
assert(pathp->cmp != 0); \
if (pathp->cmp < 0) { \
rbtn_left_set(a_type, a_field, pathp->node, \
pathp[1].node); \
assert(rbtn_red_get(a_type, a_field, pathp[1].node) \
== false); \
if (rbtn_red_get(a_type, a_field, pathp->node)) { \
a_type *right = rbtn_right_get(a_type, a_field, \
pathp->node); \
a_type *rightleft = rbtn_left_get(a_type, a_field, \
right); \
a_type *tnode; \
if (rbtn_red_get(a_type, a_field, rightleft)) { \
if (rightleft != NULL && rbtn_red_get(a_type, a_field, \
rightleft)) { \
/* In the following diagrams, ||, //, and \\ */\
/* indicate the path to the removed node. */\
/* */\
@@ -667,7 +684,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
pathp->node); \
a_type *rightleft = rbtn_left_get(a_type, a_field, \
right); \
if (rbtn_red_get(a_type, a_field, rightleft)) { \
if (rightleft != NULL && rbtn_red_get(a_type, a_field, \
rightleft)) { \
/* || */\
/* pathp(b) */\
/* // \ */\
@@ -681,7 +699,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
rbtn_rotate_left(a_type, a_field, pathp->node, \
tnode); \
/* Balance restored, but rotation modified */\
/* subree root, which may actually be the tree */\
/* subtree root, which may actually be the tree */\
/* root. */\
if (pathp == path) { \
/* Set root. */ \
@@ -721,7 +739,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
left); \
a_type *leftrightleft = rbtn_left_get(a_type, a_field, \
leftright); \
if (rbtn_red_get(a_type, a_field, leftrightleft)) { \
if (leftrightleft != NULL && rbtn_red_get(a_type, \
a_field, leftrightleft)) { \
/* || */\
/* pathp(b) */\
/* / \\ */\
@@ -747,7 +766,7 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
/* (b) */\
/* / */\
/* (b) */\
assert(leftright != &rbtree->rbt_nil); \
assert(leftright != NULL); \
rbtn_red_set(a_type, a_field, leftright); \
rbtn_rotate_right(a_type, a_field, pathp->node, \
tnode); \
@@ -770,7 +789,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
return; \
} else if (rbtn_red_get(a_type, a_field, pathp->node)) { \
a_type *leftleft = rbtn_left_get(a_type, a_field, left);\
if (rbtn_red_get(a_type, a_field, leftleft)) { \
if (leftleft != NULL && rbtn_red_get(a_type, a_field, \
leftleft)) { \
/* || */\
/* pathp(r) */\
/* / \\ */\
@@ -808,7 +828,8 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
} \
} else { \
a_type *leftleft = rbtn_left_get(a_type, a_field, left);\
if (rbtn_red_get(a_type, a_field, leftleft)) { \
if (leftleft != NULL && rbtn_red_get(a_type, a_field, \
leftleft)) { \
/* || */\
/* pathp(b) */\
/* / \\ */\
@@ -849,22 +870,22 @@ a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
} \
/* Set root. */ \
rbtree->rbt_root = path->node; \
assert(rbtn_red_get(a_type, a_field, rbtree->rbt_root) == false); \
assert(!rbtn_red_get(a_type, a_field, rbtree->rbt_root)); \
} \
a_attr a_type * \
a_prefix##iter_recurse(a_rbt_type *rbtree, a_type *node, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \
if (node == &rbtree->rbt_nil) { \
return (&rbtree->rbt_nil); \
if (node == NULL) { \
return NULL; \
} else { \
a_type *ret; \
if ((ret = a_prefix##iter_recurse(rbtree, rbtn_left_get(a_type, \
a_field, node), cb, arg)) != &rbtree->rbt_nil \
|| (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
a_field, node), cb, arg)) != NULL || (ret = cb(rbtree, node, \
arg)) != NULL) { \
return ret; \
} \
return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg)); \
return a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg); \
} \
} \
a_attr a_type * \
@@ -874,22 +895,22 @@ a_prefix##iter_start(a_rbt_type *rbtree, a_type *start, a_type *node, \
if (cmp < 0) { \
a_type *ret; \
if ((ret = a_prefix##iter_start(rbtree, start, \
rbtn_left_get(a_type, a_field, node), cb, arg)) != \
&rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
rbtn_left_get(a_type, a_field, node), cb, arg)) != NULL || \
(ret = cb(rbtree, node, arg)) != NULL) { \
return ret; \
} \
return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg)); \
return a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg); \
} else if (cmp > 0) { \
return (a_prefix##iter_start(rbtree, start, \
rbtn_right_get(a_type, a_field, node), cb, arg)); \
return a_prefix##iter_start(rbtree, start, \
rbtn_right_get(a_type, a_field, node), cb, arg); \
} else { \
a_type *ret; \
if ((ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
return ret; \
} \
return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg)); \
return a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg); \
} \
} \
a_attr a_type * \
@@ -902,25 +923,22 @@ a_prefix##iter(a_rbt_type *rbtree, a_type *start, a_type *(*cb)( \
} else { \
ret = a_prefix##iter_recurse(rbtree, rbtree->rbt_root, cb, arg);\
} \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
} \
return (ret); \
return ret; \
} \
a_attr a_type * \
a_prefix##reverse_iter_recurse(a_rbt_type *rbtree, a_type *node, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \
if (node == &rbtree->rbt_nil) { \
return (&rbtree->rbt_nil); \
if (node == NULL) { \
return NULL; \
} else { \
a_type *ret; \
if ((ret = a_prefix##reverse_iter_recurse(rbtree, \
rbtn_right_get(a_type, a_field, node), cb, arg)) != \
&rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
rbtn_right_get(a_type, a_field, node), cb, arg)) != NULL || \
(ret = cb(rbtree, node, arg)) != NULL) { \
return ret; \
} \
return (a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
return a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg); \
} \
} \
a_attr a_type * \
@@ -931,22 +949,22 @@ a_prefix##reverse_iter_start(a_rbt_type *rbtree, a_type *start, \
if (cmp > 0) { \
a_type *ret; \
if ((ret = a_prefix##reverse_iter_start(rbtree, start, \
rbtn_right_get(a_type, a_field, node), cb, arg)) != \
&rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
rbtn_right_get(a_type, a_field, node), cb, arg)) != NULL || \
(ret = cb(rbtree, node, arg)) != NULL) { \
return ret; \
} \
return (a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
return a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg); \
} else if (cmp < 0) { \
return (a_prefix##reverse_iter_start(rbtree, start, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
return a_prefix##reverse_iter_start(rbtree, start, \
rbtn_left_get(a_type, a_field, node), cb, arg); \
} else { \
a_type *ret; \
if ((ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
return ret; \
} \
return (a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
return a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg); \
} \
} \
a_attr a_type * \
@@ -960,10 +978,29 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
ret = a_prefix##reverse_iter_recurse(rbtree, rbtree->rbt_root, \
cb, arg); \
} \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
return ret; \
} \
a_attr void \
a_prefix##destroy_recurse(a_rbt_type *rbtree, a_type *node, void (*cb)( \
a_type *, void *), void *arg) { \
if (node == NULL) { \
return; \
} \
return (ret); \
a_prefix##destroy_recurse(rbtree, rbtn_left_get(a_type, a_field, \
node), cb, arg); \
rbtn_left_set(a_type, a_field, (node), NULL); \
a_prefix##destroy_recurse(rbtree, rbtn_right_get(a_type, a_field, \
node), cb, arg); \
rbtn_right_set(a_type, a_field, (node), NULL); \
if (cb) { \
cb(node, arg); \
} \
} \
a_attr void \
a_prefix##destroy(a_rbt_type *rbtree, void (*cb)(a_type *, void *), \
void *arg) { \
a_prefix##destroy_recurse(rbtree, rbtree->rbt_root, cb, arg); \
rbtree->rbt_root = NULL; \
}
#endif /* RB_H_ */

610
deps/jemalloc/include/jemalloc/internal/rtree.h vendored
View File

@@ -1,172 +1,474 @@
#ifndef JEMALLOC_INTERNAL_RTREE_H
#define JEMALLOC_INTERNAL_RTREE_H
#include "jemalloc/internal/atomic.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/rtree_tsd.h"
#include "jemalloc/internal/size_classes.h"
#include "jemalloc/internal/tsd.h"
/*
* This radix tree implementation is tailored to the singular purpose of
* tracking which chunks are currently owned by jemalloc. This functionality
* is mandatory for OS X, where jemalloc must be able to respond to object
* ownership queries.
* associating metadata with extents that are currently owned by jemalloc.
*
*******************************************************************************
*/
#ifdef JEMALLOC_H_TYPES
typedef struct rtree_s rtree_t;
/* Number of high insignificant bits. */
#define RTREE_NHIB ((1U << (LG_SIZEOF_PTR+3)) - LG_VADDR)
/* Number of low insigificant bits. */
#define RTREE_NLIB LG_PAGE
/* Number of significant bits. */
#define RTREE_NSB (LG_VADDR - RTREE_NLIB)
/* Number of levels in radix tree. */
#if RTREE_NSB <= 10
# define RTREE_HEIGHT 1
#elif RTREE_NSB <= 36
# define RTREE_HEIGHT 2
#elif RTREE_NSB <= 52
# define RTREE_HEIGHT 3
#else
# error Unsupported number of significant virtual address bits
#endif
/* Use compact leaf representation if virtual address encoding allows. */
#if RTREE_NHIB >= LG_CEIL_NSIZES
# define RTREE_LEAF_COMPACT
#endif
/*
* Size of each radix tree node (must be a power of 2). This impacts tree
* depth.
*/
#define RTREE_NODESIZE (1U << 16)
/* Needed for initialization only. */
#define RTREE_LEAFKEY_INVALID ((uintptr_t)1)
typedef void *(rtree_alloc_t)(size_t);
typedef void (rtree_dalloc_t)(void *);
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct rtree_s {
rtree_alloc_t *alloc;
rtree_dalloc_t *dalloc;
malloc_mutex_t mutex;
void **root;
unsigned height;
unsigned level2bits[1]; /* Dynamically sized. */
typedef struct rtree_node_elm_s rtree_node_elm_t;
struct rtree_node_elm_s {
atomic_p_t child; /* (rtree_{node,leaf}_elm_t *) */
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
rtree_t *rtree_new(unsigned bits, rtree_alloc_t *alloc, rtree_dalloc_t *dalloc);
void rtree_delete(rtree_t *rtree);
void rtree_prefork(rtree_t *rtree);
void rtree_postfork_parent(rtree_t *rtree);
void rtree_postfork_child(rtree_t *rtree);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
#ifdef JEMALLOC_DEBUG
uint8_t rtree_get_locked(rtree_t *rtree, uintptr_t key);
#endif
uint8_t rtree_get(rtree_t *rtree, uintptr_t key);
bool rtree_set(rtree_t *rtree, uintptr_t key, uint8_t val);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_RTREE_C_))
#define RTREE_GET_GENERATE(f) \
/* The least significant bits of the key are ignored. */ \
JEMALLOC_INLINE uint8_t \
f(rtree_t *rtree, uintptr_t key) \
{ \
uint8_t ret; \
uintptr_t subkey; \
unsigned i, lshift, height, bits; \
void **node, **child; \
\
RTREE_LOCK(&rtree->mutex); \
for (i = lshift = 0, height = rtree->height, node = rtree->root;\
i < height - 1; \
i++, lshift += bits, node = child) { \
bits = rtree->level2bits[i]; \
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR + \
3)) - bits); \
child = (void**)node[subkey]; \
if (child == NULL) { \
RTREE_UNLOCK(&rtree->mutex); \
return (0); \
} \
} \
\
/* \
* node is a leaf, so it contains values rather than node \
* pointers. \
*/ \
bits = rtree->level2bits[i]; \
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR+3)) - \
bits); \
{ \
uint8_t *leaf = (uint8_t *)node; \
ret = leaf[subkey]; \
} \
RTREE_UNLOCK(&rtree->mutex); \
\
RTREE_GET_VALIDATE \
return (ret); \
}
#ifdef JEMALLOC_DEBUG
# define RTREE_LOCK(l) malloc_mutex_lock(l)
# define RTREE_UNLOCK(l) malloc_mutex_unlock(l)
# define RTREE_GET_VALIDATE
RTREE_GET_GENERATE(rtree_get_locked)
# undef RTREE_LOCK
# undef RTREE_UNLOCK
# undef RTREE_GET_VALIDATE
#endif
#define RTREE_LOCK(l)
#define RTREE_UNLOCK(l)
#ifdef JEMALLOC_DEBUG
/*
* Suppose that it were possible for a jemalloc-allocated chunk to be
* munmap()ped, followed by a different allocator in another thread re-using
* overlapping virtual memory, all without invalidating the cached rtree
* value. The result would be a false positive (the rtree would claim that
* jemalloc owns memory that it had actually discarded). This scenario
* seems impossible, but the following assertion is a prudent sanity check.
*/
# define RTREE_GET_VALIDATE \
assert(rtree_get_locked(rtree, key) == ret);
struct rtree_leaf_elm_s {
#ifdef RTREE_LEAF_COMPACT
/*
* Single pointer-width field containing all three leaf element fields.
* For example, on a 64-bit x64 system with 48 significant virtual
* memory address bits, the index, extent, and slab fields are packed as
* such:
*
* x: index
* e: extent
* b: slab
*
* 00000000 xxxxxxxx eeeeeeee [...] eeeeeeee eeee000b
*/
atomic_p_t le_bits;
#else
# define RTREE_GET_VALIDATE
atomic_p_t le_extent; /* (extent_t *) */
atomic_u_t le_szind; /* (szind_t) */
atomic_b_t le_slab; /* (bool) */
#endif
RTREE_GET_GENERATE(rtree_get)
#undef RTREE_LOCK
#undef RTREE_UNLOCK
#undef RTREE_GET_VALIDATE
};
JEMALLOC_INLINE bool
rtree_set(rtree_t *rtree, uintptr_t key, uint8_t val)
{
uintptr_t subkey;
unsigned i, lshift, height, bits;
void **node, **child;
typedef struct rtree_level_s rtree_level_t;
struct rtree_level_s {
/* Number of key bits distinguished by this level. */
unsigned bits;
/*
* Cumulative number of key bits distinguished by traversing to
* corresponding tree level.
*/
unsigned cumbits;
};
malloc_mutex_lock(&rtree->mutex);
for (i = lshift = 0, height = rtree->height, node = rtree->root;
i < height - 1;
i++, lshift += bits, node = child) {
bits = rtree->level2bits[i];
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR+3)) -
bits);
child = (void**)node[subkey];
if (child == NULL) {
size_t size = ((i + 1 < height - 1) ? sizeof(void *)
: (sizeof(uint8_t))) << rtree->level2bits[i+1];
child = (void**)rtree->alloc(size);
if (child == NULL) {
malloc_mutex_unlock(&rtree->mutex);
return (true);
}
memset(child, 0, size);
node[subkey] = child;
}
}
typedef struct rtree_s rtree_t;
struct rtree_s {
malloc_mutex_t init_lock;
/* Number of elements based on rtree_levels[0].bits. */
#if RTREE_HEIGHT > 1
rtree_node_elm_t root[1U << (RTREE_NSB/RTREE_HEIGHT)];
#else
rtree_leaf_elm_t root[1U << (RTREE_NSB/RTREE_HEIGHT)];
#endif
};
/* node is a leaf, so it contains values rather than node pointers. */
bits = rtree->level2bits[i];
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR+3)) - bits);
{
uint8_t *leaf = (uint8_t *)node;
leaf[subkey] = val;
}
malloc_mutex_unlock(&rtree->mutex);
/*
* Split the bits into one to three partitions depending on number of
* significant bits. It the number of bits does not divide evenly into the
* number of levels, place one remainder bit per level starting at the leaf
* level.
*/
static const rtree_level_t rtree_levels[] = {
#if RTREE_HEIGHT == 1
{RTREE_NSB, RTREE_NHIB + RTREE_NSB}
#elif RTREE_HEIGHT == 2
{RTREE_NSB/2, RTREE_NHIB + RTREE_NSB/2},
{RTREE_NSB/2 + RTREE_NSB%2, RTREE_NHIB + RTREE_NSB}
#elif RTREE_HEIGHT == 3
{RTREE_NSB/3, RTREE_NHIB + RTREE_NSB/3},
{RTREE_NSB/3 + RTREE_NSB%3/2,
RTREE_NHIB + RTREE_NSB/3*2 + RTREE_NSB%3/2},
{RTREE_NSB/3 + RTREE_NSB%3 - RTREE_NSB%3/2, RTREE_NHIB + RTREE_NSB}
#else
# error Unsupported rtree height
#endif
};
return (false);
bool rtree_new(rtree_t *rtree, bool zeroed);
typedef rtree_node_elm_t *(rtree_node_alloc_t)(tsdn_t *, rtree_t *, size_t);
extern rtree_node_alloc_t *JET_MUTABLE rtree_node_alloc;
typedef rtree_leaf_elm_t *(rtree_leaf_alloc_t)(tsdn_t *, rtree_t *, size_t);
extern rtree_leaf_alloc_t *JET_MUTABLE rtree_leaf_alloc;
typedef void (rtree_node_dalloc_t)(tsdn_t *, rtree_t *, rtree_node_elm_t *);
extern rtree_node_dalloc_t *JET_MUTABLE rtree_node_dalloc;
typedef void (rtree_leaf_dalloc_t)(tsdn_t *, rtree_t *, rtree_leaf_elm_t *);
extern rtree_leaf_dalloc_t *JET_MUTABLE rtree_leaf_dalloc;
#ifdef JEMALLOC_JET
void rtree_delete(tsdn_t *tsdn, rtree_t *rtree);
#endif
rtree_leaf_elm_t *rtree_leaf_elm_lookup_hard(tsdn_t *tsdn, rtree_t *rtree,
rtree_ctx_t *rtree_ctx, uintptr_t key, bool dependent, bool init_missing);
JEMALLOC_ALWAYS_INLINE uintptr_t
rtree_leafkey(uintptr_t key) {
unsigned ptrbits = ZU(1) << (LG_SIZEOF_PTR+3);
unsigned cumbits = (rtree_levels[RTREE_HEIGHT-1].cumbits -
rtree_levels[RTREE_HEIGHT-1].bits);
unsigned maskbits = ptrbits - cumbits;
uintptr_t mask = ~((ZU(1) << maskbits) - 1);
return (key & mask);
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
JEMALLOC_ALWAYS_INLINE size_t
rtree_cache_direct_map(uintptr_t key) {
unsigned ptrbits = ZU(1) << (LG_SIZEOF_PTR+3);
unsigned cumbits = (rtree_levels[RTREE_HEIGHT-1].cumbits -
rtree_levels[RTREE_HEIGHT-1].bits);
unsigned maskbits = ptrbits - cumbits;
return (size_t)((key >> maskbits) & (RTREE_CTX_NCACHE - 1));
}
JEMALLOC_ALWAYS_INLINE uintptr_t
rtree_subkey(uintptr_t key, unsigned level) {
unsigned ptrbits = ZU(1) << (LG_SIZEOF_PTR+3);
unsigned cumbits = rtree_levels[level].cumbits;
unsigned shiftbits = ptrbits - cumbits;
unsigned maskbits = rtree_levels[level].bits;
uintptr_t mask = (ZU(1) << maskbits) - 1;
return ((key >> shiftbits) & mask);
}
/*
* Atomic getters.
*
* dependent: Reading a value on behalf of a pointer to a valid allocation
* is guaranteed to be a clean read even without synchronization,
* because the rtree update became visible in memory before the
* pointer came into existence.
* !dependent: An arbitrary read, e.g. on behalf of ivsalloc(), may not be
* dependent on a previous rtree write, which means a stale read
* could result if synchronization were omitted here.
*/
# ifdef RTREE_LEAF_COMPACT
JEMALLOC_ALWAYS_INLINE uintptr_t
rtree_leaf_elm_bits_read(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
bool dependent) {
return (uintptr_t)atomic_load_p(&elm->le_bits, dependent
? ATOMIC_RELAXED : ATOMIC_ACQUIRE);
}
JEMALLOC_ALWAYS_INLINE extent_t *
rtree_leaf_elm_bits_extent_get(uintptr_t bits) {
/* Restore sign-extended high bits, mask slab bit. */
return (extent_t *)((uintptr_t)((intptr_t)(bits << RTREE_NHIB) >>
RTREE_NHIB) & ~((uintptr_t)0x1));
}
JEMALLOC_ALWAYS_INLINE szind_t
rtree_leaf_elm_bits_szind_get(uintptr_t bits) {
return (szind_t)(bits >> LG_VADDR);
}
JEMALLOC_ALWAYS_INLINE bool
rtree_leaf_elm_bits_slab_get(uintptr_t bits) {
return (bool)(bits & (uintptr_t)0x1);
}
# endif
JEMALLOC_ALWAYS_INLINE extent_t *
rtree_leaf_elm_extent_read(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
bool dependent) {
#ifdef RTREE_LEAF_COMPACT
uintptr_t bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, dependent);
return rtree_leaf_elm_bits_extent_get(bits);
#else
extent_t *extent = (extent_t *)atomic_load_p(&elm->le_extent, dependent
? ATOMIC_RELAXED : ATOMIC_ACQUIRE);
return extent;
#endif
}
JEMALLOC_ALWAYS_INLINE szind_t
rtree_leaf_elm_szind_read(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
bool dependent) {
#ifdef RTREE_LEAF_COMPACT
uintptr_t bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, dependent);
return rtree_leaf_elm_bits_szind_get(bits);
#else
return (szind_t)atomic_load_u(&elm->le_szind, dependent ? ATOMIC_RELAXED
: ATOMIC_ACQUIRE);
#endif
}
JEMALLOC_ALWAYS_INLINE bool
rtree_leaf_elm_slab_read(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
bool dependent) {
#ifdef RTREE_LEAF_COMPACT
uintptr_t bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, dependent);
return rtree_leaf_elm_bits_slab_get(bits);
#else
return atomic_load_b(&elm->le_slab, dependent ? ATOMIC_RELAXED :
ATOMIC_ACQUIRE);
#endif
}
static inline void
rtree_leaf_elm_extent_write(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
extent_t *extent) {
#ifdef RTREE_LEAF_COMPACT
uintptr_t old_bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm, true);
uintptr_t bits = ((uintptr_t)rtree_leaf_elm_bits_szind_get(old_bits) <<
LG_VADDR) | ((uintptr_t)extent & (((uintptr_t)0x1 << LG_VADDR) - 1))
| ((uintptr_t)rtree_leaf_elm_bits_slab_get(old_bits));
atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE);
#else
atomic_store_p(&elm->le_extent, extent, ATOMIC_RELEASE);
#endif
}
static inline void
rtree_leaf_elm_szind_write(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
szind_t szind) {
assert(szind <= NSIZES);
#ifdef RTREE_LEAF_COMPACT
uintptr_t old_bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm,
true);
uintptr_t bits = ((uintptr_t)szind << LG_VADDR) |
((uintptr_t)rtree_leaf_elm_bits_extent_get(old_bits) &
(((uintptr_t)0x1 << LG_VADDR) - 1)) |
((uintptr_t)rtree_leaf_elm_bits_slab_get(old_bits));
atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE);
#else
atomic_store_u(&elm->le_szind, szind, ATOMIC_RELEASE);
#endif
}
static inline void
rtree_leaf_elm_slab_write(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
bool slab) {
#ifdef RTREE_LEAF_COMPACT
uintptr_t old_bits = rtree_leaf_elm_bits_read(tsdn, rtree, elm,
true);
uintptr_t bits = ((uintptr_t)rtree_leaf_elm_bits_szind_get(old_bits) <<
LG_VADDR) | ((uintptr_t)rtree_leaf_elm_bits_extent_get(old_bits) &
(((uintptr_t)0x1 << LG_VADDR) - 1)) | ((uintptr_t)slab);
atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE);
#else
atomic_store_b(&elm->le_slab, slab, ATOMIC_RELEASE);
#endif
}
static inline void
rtree_leaf_elm_write(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
extent_t *extent, szind_t szind, bool slab) {
#ifdef RTREE_LEAF_COMPACT
uintptr_t bits = ((uintptr_t)szind << LG_VADDR) |
((uintptr_t)extent & (((uintptr_t)0x1 << LG_VADDR) - 1)) |
((uintptr_t)slab);
atomic_store_p(&elm->le_bits, (void *)bits, ATOMIC_RELEASE);
#else
rtree_leaf_elm_slab_write(tsdn, rtree, elm, slab);
rtree_leaf_elm_szind_write(tsdn, rtree, elm, szind);
/*
* Write extent last, since the element is atomically considered valid
* as soon as the extent field is non-NULL.
*/
rtree_leaf_elm_extent_write(tsdn, rtree, elm, extent);
#endif
}
static inline void
rtree_leaf_elm_szind_slab_update(tsdn_t *tsdn, rtree_t *rtree,
rtree_leaf_elm_t *elm, szind_t szind, bool slab) {
assert(!slab || szind < NBINS);
/*
* The caller implicitly assures that it is the only writer to the szind
* and slab fields, and that the extent field cannot currently change.
*/
rtree_leaf_elm_slab_write(tsdn, rtree, elm, slab);
rtree_leaf_elm_szind_write(tsdn, rtree, elm, szind);
}
JEMALLOC_ALWAYS_INLINE rtree_leaf_elm_t *
rtree_leaf_elm_lookup(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key, bool dependent, bool init_missing) {
assert(key != 0);
assert(!dependent || !init_missing);
size_t slot = rtree_cache_direct_map(key);
uintptr_t leafkey = rtree_leafkey(key);
assert(leafkey != RTREE_LEAFKEY_INVALID);
/* Fast path: L1 direct mapped cache. */
if (likely(rtree_ctx->cache[slot].leafkey == leafkey)) {
rtree_leaf_elm_t *leaf = rtree_ctx->cache[slot].leaf;
assert(leaf != NULL);
uintptr_t subkey = rtree_subkey(key, RTREE_HEIGHT-1);
return &leaf[subkey];
}
/*
* Search the L2 LRU cache. On hit, swap the matching element into the
* slot in L1 cache, and move the position in L2 up by 1.
*/
#define RTREE_CACHE_CHECK_L2(i) do { \
if (likely(rtree_ctx->l2_cache[i].leafkey == leafkey)) { \
rtree_leaf_elm_t *leaf = rtree_ctx->l2_cache[i].leaf; \
assert(leaf != NULL); \
if (i > 0) { \
/* Bubble up by one. */ \
rtree_ctx->l2_cache[i].leafkey = \
rtree_ctx->l2_cache[i - 1].leafkey; \
rtree_ctx->l2_cache[i].leaf = \
rtree_ctx->l2_cache[i - 1].leaf; \
rtree_ctx->l2_cache[i - 1].leafkey = \
rtree_ctx->cache[slot].leafkey; \
rtree_ctx->l2_cache[i - 1].leaf = \
rtree_ctx->cache[slot].leaf; \
} else { \
rtree_ctx->l2_cache[0].leafkey = \
rtree_ctx->cache[slot].leafkey; \
rtree_ctx->l2_cache[0].leaf = \
rtree_ctx->cache[slot].leaf; \
} \
rtree_ctx->cache[slot].leafkey = leafkey; \
rtree_ctx->cache[slot].leaf = leaf; \
uintptr_t subkey = rtree_subkey(key, RTREE_HEIGHT-1); \
return &leaf[subkey]; \
} \
} while (0)
/* Check the first cache entry. */
RTREE_CACHE_CHECK_L2(0);
/* Search the remaining cache elements. */
for (unsigned i = 1; i < RTREE_CTX_NCACHE_L2; i++) {
RTREE_CACHE_CHECK_L2(i);
}
#undef RTREE_CACHE_CHECK_L2
return rtree_leaf_elm_lookup_hard(tsdn, rtree, rtree_ctx, key,
dependent, init_missing);
}
static inline bool
rtree_write(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, uintptr_t key,
extent_t *extent, szind_t szind, bool slab) {
/* Use rtree_clear() to set the extent to NULL. */
assert(extent != NULL);
rtree_leaf_elm_t *elm = rtree_leaf_elm_lookup(tsdn, rtree, rtree_ctx,
key, false, true);
if (elm == NULL) {
return true;
}
assert(rtree_leaf_elm_extent_read(tsdn, rtree, elm, false) == NULL);
rtree_leaf_elm_write(tsdn, rtree, elm, extent, szind, slab);
return false;
}
JEMALLOC_ALWAYS_INLINE rtree_leaf_elm_t *
rtree_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx, uintptr_t key,
bool dependent) {
rtree_leaf_elm_t *elm = rtree_leaf_elm_lookup(tsdn, rtree, rtree_ctx,
key, dependent, false);
if (!dependent && elm == NULL) {
return NULL;
}
assert(elm != NULL);
return elm;
}
JEMALLOC_ALWAYS_INLINE extent_t *
rtree_extent_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key, bool dependent) {
rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key,
dependent);
if (!dependent && elm == NULL) {
return NULL;
}
return rtree_leaf_elm_extent_read(tsdn, rtree, elm, dependent);
}
JEMALLOC_ALWAYS_INLINE szind_t
rtree_szind_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key, bool dependent) {
rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key,
dependent);
if (!dependent && elm == NULL) {
return NSIZES;
}
return rtree_leaf_elm_szind_read(tsdn, rtree, elm, dependent);
}
/*
* rtree_slab_read() is intentionally omitted because slab is always read in
* conjunction with szind, which makes rtree_szind_slab_read() a better choice.
*/
JEMALLOC_ALWAYS_INLINE bool
rtree_extent_szind_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key, bool dependent, extent_t **r_extent, szind_t *r_szind) {
rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key,
dependent);
if (!dependent && elm == NULL) {
return true;
}
*r_extent = rtree_leaf_elm_extent_read(tsdn, rtree, elm, dependent);
*r_szind = rtree_leaf_elm_szind_read(tsdn, rtree, elm, dependent);
return false;
}
JEMALLOC_ALWAYS_INLINE bool
rtree_szind_slab_read(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key, bool dependent, szind_t *r_szind, bool *r_slab) {
rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key,
dependent);
if (!dependent && elm == NULL) {
return true;
}
*r_szind = rtree_leaf_elm_szind_read(tsdn, rtree, elm, dependent);
*r_slab = rtree_leaf_elm_slab_read(tsdn, rtree, elm, dependent);
return false;
}
static inline void
rtree_szind_slab_update(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key, szind_t szind, bool slab) {
assert(!slab || szind < NBINS);
rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, true);
rtree_leaf_elm_szind_slab_update(tsdn, rtree, elm, szind, slab);
}
static inline void
rtree_clear(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key) {
rtree_leaf_elm_t *elm = rtree_read(tsdn, rtree, rtree_ctx, key, true);
assert(rtree_leaf_elm_extent_read(tsdn, rtree, elm, false) !=
NULL);
rtree_leaf_elm_write(tsdn, rtree, elm, NULL, NSIZES, false);
}
#endif /* JEMALLOC_INTERNAL_RTREE_H */

2091
deps/jemalloc/include/jemalloc/internal/size_classes.h vendored
View File

File diff suppressed because it is too large Load Diff

231
deps/jemalloc/include/jemalloc/internal/stats.h vendored
View File

@@ -1,31 +1,51 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_STATS_H
#define JEMALLOC_INTERNAL_STATS_H
typedef struct tcache_bin_stats_s tcache_bin_stats_t;
typedef struct malloc_bin_stats_s malloc_bin_stats_t;
typedef struct malloc_large_stats_s malloc_large_stats_t;
typedef struct arena_stats_s arena_stats_t;
typedef struct chunk_stats_s chunk_stats_t;
#include "jemalloc/internal/atomic.h"
#include "jemalloc/internal/mutex_prof.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/size_classes.h"
#include "jemalloc/internal/stats_tsd.h"
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* OPTION(opt, var_name, default, set_value_to) */
#define STATS_PRINT_OPTIONS \
OPTION('J', json, false, true) \
OPTION('g', general, true, false) \
OPTION('m', merged, config_stats, false) \
OPTION('d', destroyed, config_stats, false) \
OPTION('a', unmerged, config_stats, false) \
OPTION('b', bins, true, false) \
OPTION('l', large, true, false) \
OPTION('x', mutex, true, false)
struct tcache_bin_stats_s {
/*
* Number of allocation requests that corresponded to the size of this
* bin.
*/
uint64_t nrequests;
enum {
#define OPTION(o, v, d, s) stats_print_option_num_##v,
STATS_PRINT_OPTIONS
#undef OPTION
stats_print_tot_num_options
};
struct malloc_bin_stats_s {
/*
* Current number of bytes allocated, including objects currently
* cached by tcache.
*/
size_t allocated;
/* Options for stats_print. */
extern bool opt_stats_print;
extern char opt_stats_print_opts[stats_print_tot_num_options+1];
/* Implements je_malloc_stats_print. */
void stats_print(void (*write_cb)(void *, const char *), void *cbopaque,
const char *opts);
/*
* In those architectures that support 64-bit atomics, we use atomic updates for
* our 64-bit values. Otherwise, we use a plain uint64_t and synchronize
* externally.
*/
#ifdef JEMALLOC_ATOMIC_U64
typedef atomic_u64_t arena_stats_u64_t;
#else
/* Must hold the arena stats mutex while reading atomically. */
typedef uint64_t arena_stats_u64_t;
#endif
typedef struct malloc_bin_stats_s {
/*
* Total number of allocation/deallocation requests served directly by
* the bin. Note that tcache may allocate an object, then recycle it
@@ -42,132 +62,103 @@ struct malloc_bin_stats_s {
*/
uint64_t nrequests;
/*
* Current number of regions of this size class, including regions
* currently cached by tcache.
*/
size_t curregs;
/* Number of tcache fills from this bin. */
uint64_t nfills;
/* Number of tcache flushes to this bin. */
uint64_t nflushes;
/* Total number of runs created for this bin's size class. */
uint64_t nruns;
/* Total number of slabs created for this bin's size class. */
uint64_t nslabs;
/*
* Total number of runs reused by extracting them from the runs tree for
* this bin's size class.
* Total number of slabs reused by extracting them from the slabs heap
* for this bin's size class.
*/
uint64_t reruns;
uint64_t reslabs;
/* Current number of runs in this bin. */
size_t curruns;
};
/* Current number of slabs in this bin. */
size_t curslabs;
struct malloc_large_stats_s {
mutex_prof_data_t mutex_data;
} malloc_bin_stats_t;
typedef struct malloc_large_stats_s {
/*
* Total number of allocation/deallocation requests served directly by
* the arena. Note that tcache may allocate an object, then recycle it
* many times, resulting many increments to nrequests, but only one
* each to nmalloc and ndalloc.
* the arena.
*/
uint64_t nmalloc;
uint64_t ndalloc;
arena_stats_u64_t nmalloc;
arena_stats_u64_t ndalloc;
/*
* Number of allocation requests that correspond to this size class.
* This includes requests served by tcache, though tcache only
* periodically merges into this counter.
*/
uint64_t nrequests;
arena_stats_u64_t nrequests; /* Partially derived. */
/* Current number of runs of this size class. */
size_t curruns;
};
/* Current number of allocations of this size class. */
size_t curlextents; /* Derived. */
} malloc_large_stats_t;
struct arena_stats_s {
/* Number of bytes currently mapped. */
size_t mapped;
typedef struct decay_stats_s {
/* Total number of purge sweeps. */
arena_stats_u64_t npurge;
/* Total number of madvise calls made. */
arena_stats_u64_t nmadvise;
/* Total number of pages purged. */
arena_stats_u64_t purged;
} decay_stats_t;
/*
* Total number of purge sweeps, total number of madvise calls made,
* and total pages purged in order to keep dirty unused memory under
* control.
*/
uint64_t npurge;
uint64_t nmadvise;
uint64_t purged;
/* Per-size-category statistics. */
size_t allocated_large;
uint64_t nmalloc_large;
uint64_t ndalloc_large;
uint64_t nrequests_large;
/*
* One element for each possible size class, including sizes that
* overlap with bin size classes. This is necessary because ipalloc()
* sometimes has to use such large objects in order to assure proper
* alignment.
*/
malloc_large_stats_t *lstats;
};
struct chunk_stats_s {
/* Number of chunks that were allocated. */
uint64_t nchunks;
/* High-water mark for number of chunks allocated. */
size_t highchunks;
/*
* Current number of chunks allocated. This value isn't maintained for
* any other purpose, so keep track of it in order to be able to set
* highchunks.
*/
size_t curchunks;
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern bool opt_stats_print;
extern size_t stats_cactive;
void stats_print(void (*write)(void *, const char *), void *cbopaque,
const char *opts);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
size_t stats_cactive_get(void);
void stats_cactive_add(size_t size);
void stats_cactive_sub(size_t size);
/*
* Arena stats. Note that fields marked "derived" are not directly maintained
* within the arena code; rather their values are derived during stats merge
* requests.
*/
typedef struct arena_stats_s {
#ifndef JEMALLOC_ATOMIC_U64
malloc_mutex_t mtx;
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_STATS_C_))
JEMALLOC_INLINE size_t
stats_cactive_get(void)
{
/* Number of bytes currently mapped, excluding retained memory. */
atomic_zu_t mapped; /* Partially derived. */
return (atomic_read_z(&stats_cactive));
}
/*
* Number of unused virtual memory bytes currently retained. Retained
* bytes are technically mapped (though always decommitted or purged),
* but they are excluded from the mapped statistic (above).
*/
atomic_zu_t retained; /* Derived. */
JEMALLOC_INLINE void
stats_cactive_add(size_t size)
{
decay_stats_t decay_dirty;
decay_stats_t decay_muzzy;
atomic_add_z(&stats_cactive, size);
}
atomic_zu_t base; /* Derived. */
atomic_zu_t internal;
atomic_zu_t resident; /* Derived. */
JEMALLOC_INLINE void
stats_cactive_sub(size_t size)
{
atomic_zu_t allocated_large; /* Derived. */
arena_stats_u64_t nmalloc_large; /* Derived. */
arena_stats_u64_t ndalloc_large; /* Derived. */
arena_stats_u64_t nrequests_large; /* Derived. */
atomic_sub_z(&stats_cactive, size);
}
#endif
/* Number of bytes cached in tcache associated with this arena. */
atomic_zu_t tcache_bytes; /* Derived. */
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
mutex_prof_data_t mutex_prof_data[mutex_prof_num_arena_mutexes];
/* One element for each large size class. */
malloc_large_stats_t lstats[NSIZES - NBINS];
/* Arena uptime. */
nstime_t uptime;
} arena_stats_t;
#endif /* JEMALLOC_INTERNAL_STATS_H */

722
deps/jemalloc/include/jemalloc/internal/tsd.h vendored
View File

@@ -1,434 +1,324 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_TSD_H
#define JEMALLOC_INTERNAL_TSD_H
/* Maximum number of malloc_tsd users with cleanup functions. */
#define MALLOC_TSD_CLEANUPS_MAX 8
#include "jemalloc/internal/arena_types.h"
#include "jemalloc/internal/assert.h"
#include "jemalloc/internal/jemalloc_internal_externs.h"
#include "jemalloc/internal/prof_types.h"
#include "jemalloc/internal/ql.h"
#include "jemalloc/internal/rtree_tsd.h"
#include "jemalloc/internal/tcache_types.h"
#include "jemalloc/internal/tcache_structs.h"
#include "jemalloc/internal/util.h"
#include "jemalloc/internal/witness.h"
typedef bool (*malloc_tsd_cleanup_t)(void);
/*
* Thread-Specific-Data layout
* --- data accessed on tcache fast path: state, rtree_ctx, stats, prof ---
* s: state
* e: tcache_enabled
* m: thread_allocated (config_stats)
* f: thread_deallocated (config_stats)
* p: prof_tdata (config_prof)
* c: rtree_ctx (rtree cache accessed on deallocation)
* t: tcache
* --- data not accessed on tcache fast path: arena-related fields ---
* d: arenas_tdata_bypass
* r: reentrancy_level
* x: narenas_tdata
* i: iarena
* a: arena
* o: arenas_tdata
* Loading TSD data is on the critical path of basically all malloc operations.
* In particular, tcache and rtree_ctx rely on hot CPU cache to be effective.
* Use a compact layout to reduce cache footprint.
* +--- 64-bit and 64B cacheline; 1B each letter; First byte on the left. ---+
* |---------------------------- 1st cacheline ----------------------------|
* | sedrxxxx mmmmmmmm ffffffff pppppppp [c * 32 ........ ........ .......] |
* |---------------------------- 2nd cacheline ----------------------------|
* | [c * 64 ........ ........ ........ ........ ........ ........ .......] |
* |---------------------------- 3nd cacheline ----------------------------|
* | [c * 32 ........ ........ .......] iiiiiiii aaaaaaaa oooooooo [t...... |
* +-------------------------------------------------------------------------+
* Note: the entire tcache is embedded into TSD and spans multiple cachelines.
*
* The last 3 members (i, a and o) before tcache isn't really needed on tcache
* fast path. However we have a number of unused tcache bins and witnesses
* (never touched unless config_debug) at the end of tcache, so we place them
* there to avoid breaking the cachelines and possibly paging in an extra page.
*/
#ifdef JEMALLOC_JET
typedef void (*test_callback_t)(int *);
# define MALLOC_TSD_TEST_DATA_INIT 0x72b65c10
# define MALLOC_TEST_TSD \
O(test_data, int, int) \
O(test_callback, test_callback_t, int)
# define MALLOC_TEST_TSD_INITIALIZER , MALLOC_TSD_TEST_DATA_INIT, NULL
#else
# define MALLOC_TEST_TSD
# define MALLOC_TEST_TSD_INITIALIZER
#endif
#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
!defined(_WIN32))
typedef struct tsd_init_block_s tsd_init_block_t;
typedef struct tsd_init_head_s tsd_init_head_t;
/* O(name, type, nullable type */
#define MALLOC_TSD \
O(tcache_enabled, bool, bool) \
O(arenas_tdata_bypass, bool, bool) \
O(reentrancy_level, int8_t, int8_t) \
O(narenas_tdata, uint32_t, uint32_t) \
O(thread_allocated, uint64_t, uint64_t) \
O(thread_deallocated, uint64_t, uint64_t) \
O(prof_tdata, prof_tdata_t *, prof_tdata_t *) \
O(rtree_ctx, rtree_ctx_t, rtree_ctx_t) \
O(iarena, arena_t *, arena_t *) \
O(arena, arena_t *, arena_t *) \
O(arenas_tdata, arena_tdata_t *, arena_tdata_t *)\
O(tcache, tcache_t, tcache_t) \
O(witness_tsd, witness_tsd_t, witness_tsdn_t) \
MALLOC_TEST_TSD
#define TSD_INITIALIZER { \
tsd_state_uninitialized, \
TCACHE_ENABLED_ZERO_INITIALIZER, \
false, \
0, \
0, \
0, \
0, \
NULL, \
RTREE_CTX_ZERO_INITIALIZER, \
NULL, \
NULL, \
NULL, \
TCACHE_ZERO_INITIALIZER, \
WITNESS_TSD_INITIALIZER \
MALLOC_TEST_TSD_INITIALIZER \
}
enum {
tsd_state_nominal = 0, /* Common case --> jnz. */
tsd_state_nominal_slow = 1, /* Initialized but on slow path. */
/* the above 2 nominal states should be lower values. */
tsd_state_nominal_max = 1, /* used for comparison only. */
tsd_state_minimal_initialized = 2,
tsd_state_purgatory = 3,
tsd_state_reincarnated = 4,
tsd_state_uninitialized = 5
};
/* Manually limit tsd_state_t to a single byte. */
typedef uint8_t tsd_state_t;
/* The actual tsd. */
struct tsd_s {
/*
* The contents should be treated as totally opaque outside the tsd
* module. Access any thread-local state through the getters and
* setters below.
*/
tsd_state_t state;
#define O(n, t, nt) \
t use_a_getter_or_setter_instead_##n;
MALLOC_TSD
#undef O
};
/*
* Wrapper around tsd_t that makes it possible to avoid implicit conversion
* between tsd_t and tsdn_t, where tsdn_t is "nullable" and has to be
* explicitly converted to tsd_t, which is non-nullable.
*/
struct tsdn_s {
tsd_t tsd;
};
#define TSDN_NULL ((tsdn_t *)0)
JEMALLOC_ALWAYS_INLINE tsdn_t *
tsd_tsdn(tsd_t *tsd) {
return (tsdn_t *)tsd;
}
JEMALLOC_ALWAYS_INLINE bool
tsdn_null(const tsdn_t *tsdn) {
return tsdn == NULL;
}
JEMALLOC_ALWAYS_INLINE tsd_t *
tsdn_tsd(tsdn_t *tsdn) {
assert(!tsdn_null(tsdn));
return &tsdn->tsd;
}
void *malloc_tsd_malloc(size_t size);
void malloc_tsd_dalloc(void *wrapper);
void malloc_tsd_cleanup_register(bool (*f)(void));
tsd_t *malloc_tsd_boot0(void);
void malloc_tsd_boot1(void);
void tsd_cleanup(void *arg);
tsd_t *tsd_fetch_slow(tsd_t *tsd, bool internal);
void tsd_slow_update(tsd_t *tsd);
/*
* We put the platform-specific data declarations and inlines into their own
* header files to avoid cluttering this file. They define tsd_boot0,
* tsd_boot1, tsd_boot, tsd_booted_get, tsd_get_allocates, tsd_get, and tsd_set.
*/
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
#include "jemalloc/internal/tsd_malloc_thread_cleanup.h"
#elif (defined(JEMALLOC_TLS))
#include "jemalloc/internal/tsd_tls.h"
#elif (defined(_WIN32))
#include "jemalloc/internal/tsd_win.h"
#else
#include "jemalloc/internal/tsd_generic.h"
#endif
/*
* TLS/TSD-agnostic macro-based implementation of thread-specific data. There
* are four macros that support (at least) three use cases: file-private,
* library-private, and library-private inlined. Following is an example
* library-private tsd variable:
*
* In example.h:
* typedef struct {
* int x;
* int y;
* } example_t;
* #define EX_INITIALIZER JEMALLOC_CONCAT({0, 0})
* malloc_tsd_protos(, example, example_t *)
* malloc_tsd_externs(example, example_t *)
* In example.c:
* malloc_tsd_data(, example, example_t *, EX_INITIALIZER)
* malloc_tsd_funcs(, example, example_t *, EX_INITIALIZER,
* example_tsd_cleanup)
*
* The result is a set of generated functions, e.g.:
*
* bool example_tsd_boot(void) {...}
* example_t **example_tsd_get() {...}
* void example_tsd_set(example_t **val) {...}
*
* Note that all of the functions deal in terms of (a_type *) rather than
* (a_type) so that it is possible to support non-pointer types (unlike
* pthreads TSD). example_tsd_cleanup() is passed an (a_type *) pointer that is
* cast to (void *). This means that the cleanup function needs to cast *and*
* dereference the function argument, e.g.:
*
* void
* example_tsd_cleanup(void *arg)
* {
* example_t *example = *(example_t **)arg;
*
* [...]
* if ([want the cleanup function to be called again]) {
* example_tsd_set(&example);
* }
* }
*
* If example_tsd_set() is called within example_tsd_cleanup(), it will be
* called again. This is similar to how pthreads TSD destruction works, except
* that pthreads only calls the cleanup function again if the value was set to
* non-NULL.
* tsd_foop_get_unsafe(tsd) returns a pointer to the thread-local instance of
* foo. This omits some safety checks, and so can be used during tsd
* initialization and cleanup.
*/
/* malloc_tsd_protos(). */
#define malloc_tsd_protos(a_attr, a_name, a_type) \
a_attr bool \
a_name##_tsd_boot(void); \
a_attr a_type * \
a_name##_tsd_get(void); \
a_attr void \
a_name##_tsd_set(a_type *val);
/* malloc_tsd_externs(). */
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
#define malloc_tsd_externs(a_name, a_type) \
extern __thread a_type a_name##_tls; \
extern __thread bool a_name##_initialized; \
extern bool a_name##_booted;
#elif (defined(JEMALLOC_TLS))
#define malloc_tsd_externs(a_name, a_type) \
extern __thread a_type a_name##_tls; \
extern pthread_key_t a_name##_tsd; \
extern bool a_name##_booted;
#elif (defined(_WIN32))
#define malloc_tsd_externs(a_name, a_type) \
extern DWORD a_name##_tsd; \
extern bool a_name##_booted;
#else
#define malloc_tsd_externs(a_name, a_type) \
extern pthread_key_t a_name##_tsd; \
extern tsd_init_head_t a_name##_tsd_init_head; \
extern bool a_name##_booted;
#endif
/* malloc_tsd_data(). */
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr __thread a_type JEMALLOC_TLS_MODEL \
a_name##_tls = a_initializer; \
a_attr __thread bool JEMALLOC_TLS_MODEL \
a_name##_initialized = false; \
a_attr bool a_name##_booted = false;
#elif (defined(JEMALLOC_TLS))
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr __thread a_type JEMALLOC_TLS_MODEL \
a_name##_tls = a_initializer; \
a_attr pthread_key_t a_name##_tsd; \
a_attr bool a_name##_booted = false;
#elif (defined(_WIN32))
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr DWORD a_name##_tsd; \
a_attr bool a_name##_booted = false;
#else
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr pthread_key_t a_name##_tsd; \
a_attr tsd_init_head_t a_name##_tsd_init_head = { \
ql_head_initializer(blocks), \
MALLOC_MUTEX_INITIALIZER \
}; \
a_attr bool a_name##_booted = false;
#endif
/* malloc_tsd_funcs(). */
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Initialization/cleanup. */ \
a_attr bool \
a_name##_tsd_cleanup_wrapper(void) \
{ \
\
if (a_name##_initialized) { \
a_name##_initialized = false; \
a_cleanup(&a_name##_tls); \
} \
return (a_name##_initialized); \
} \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
if (a_cleanup != malloc_tsd_no_cleanup) { \
malloc_tsd_cleanup_register( \
&a_name##_tsd_cleanup_wrapper); \
} \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
\
assert(a_name##_booted); \
return (&a_name##_tls); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
\
assert(a_name##_booted); \
a_name##_tls = (*val); \
if (a_cleanup != malloc_tsd_no_cleanup) \
a_name##_initialized = true; \
#define O(n, t, nt) \
JEMALLOC_ALWAYS_INLINE t * \
tsd_##n##p_get_unsafe(tsd_t *tsd) { \
return &tsd->use_a_getter_or_setter_instead_##n; \
}
#elif (defined(JEMALLOC_TLS))
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Initialization/cleanup. */ \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
if (a_cleanup != malloc_tsd_no_cleanup) { \
if (pthread_key_create(&a_name##_tsd, a_cleanup) != 0) \
return (true); \
} \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
\
assert(a_name##_booted); \
return (&a_name##_tls); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
\
assert(a_name##_booted); \
a_name##_tls = (*val); \
if (a_cleanup != malloc_tsd_no_cleanup) { \
if (pthread_setspecific(a_name##_tsd, \
(void *)(&a_name##_tls))) { \
malloc_write("<jemalloc>: Error" \
" setting TSD for "#a_name"\n"); \
if (opt_abort) \
abort(); \
} \
} \
MALLOC_TSD
#undef O
/* tsd_foop_get(tsd) returns a pointer to the thread-local instance of foo. */
#define O(n, t, nt) \
JEMALLOC_ALWAYS_INLINE t * \
tsd_##n##p_get(tsd_t *tsd) { \
assert(tsd->state == tsd_state_nominal || \
tsd->state == tsd_state_nominal_slow || \
tsd->state == tsd_state_reincarnated || \
tsd->state == tsd_state_minimal_initialized); \
return tsd_##n##p_get_unsafe(tsd); \
}
#elif (defined(_WIN32))
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Data structure. */ \
typedef struct { \
bool initialized; \
a_type val; \
} a_name##_tsd_wrapper_t; \
/* Initialization/cleanup. */ \
a_attr bool \
a_name##_tsd_cleanup_wrapper(void) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
wrapper = (a_name##_tsd_wrapper_t *) TlsGetValue(a_name##_tsd); \
if (wrapper == NULL) \
return (false); \
if (a_cleanup != malloc_tsd_no_cleanup && \
wrapper->initialized) { \
a_type val = wrapper->val; \
a_type tsd_static_data = a_initializer; \
wrapper->initialized = false; \
wrapper->val = tsd_static_data; \
a_cleanup(&val); \
if (wrapper->initialized) { \
/* Trigger another cleanup round. */ \
return (true); \
} \
MALLOC_TSD
#undef O
/*
* tsdn_foop_get(tsdn) returns either the thread-local instance of foo (if tsdn
* isn't NULL), or NULL (if tsdn is NULL), cast to the nullable pointer type.
*/
#define O(n, t, nt) \
JEMALLOC_ALWAYS_INLINE nt * \
tsdn_##n##p_get(tsdn_t *tsdn) { \
if (tsdn_null(tsdn)) { \
return NULL; \
} \
malloc_tsd_dalloc(wrapper); \
return (false); \
} \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
a_name##_tsd = TlsAlloc(); \
if (a_name##_tsd == TLS_OUT_OF_INDEXES) \
return (true); \
if (a_cleanup != malloc_tsd_no_cleanup) { \
malloc_tsd_cleanup_register( \
&a_name##_tsd_cleanup_wrapper); \
} \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_name##_tsd_wrapper_t * \
a_name##_tsd_get_wrapper(void) \
{ \
a_name##_tsd_wrapper_t *wrapper = (a_name##_tsd_wrapper_t *) \
TlsGetValue(a_name##_tsd); \
\
if (wrapper == NULL) { \
wrapper = (a_name##_tsd_wrapper_t *) \
malloc_tsd_malloc(sizeof(a_name##_tsd_wrapper_t)); \
if (wrapper == NULL) { \
malloc_write("<jemalloc>: Error allocating" \
" TSD for "#a_name"\n"); \
abort(); \
} else { \
static a_type tsd_static_data = a_initializer; \
wrapper->initialized = false; \
wrapper->val = tsd_static_data; \
} \
if (!TlsSetValue(a_name##_tsd, (void *)wrapper)) { \
malloc_write("<jemalloc>: Error setting" \
" TSD for "#a_name"\n"); \
abort(); \
} \
} \
return (wrapper); \
} \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
return (&wrapper->val); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
wrapper->val = *(val); \
if (a_cleanup != malloc_tsd_no_cleanup) \
wrapper->initialized = true; \
tsd_t *tsd = tsdn_tsd(tsdn); \
return (nt *)tsd_##n##p_get(tsd); \
}
#else
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Data structure. */ \
typedef struct { \
bool initialized; \
a_type val; \
} a_name##_tsd_wrapper_t; \
/* Initialization/cleanup. */ \
a_attr void \
a_name##_tsd_cleanup_wrapper(void *arg) \
{ \
a_name##_tsd_wrapper_t *wrapper = (a_name##_tsd_wrapper_t *)arg;\
\
if (a_cleanup != malloc_tsd_no_cleanup && \
wrapper->initialized) { \
wrapper->initialized = false; \
a_cleanup(&wrapper->val); \
if (wrapper->initialized) { \
/* Trigger another cleanup round. */ \
if (pthread_setspecific(a_name##_tsd, \
(void *)wrapper)) { \
malloc_write("<jemalloc>: Error" \
" setting TSD for "#a_name"\n"); \
if (opt_abort) \
abort(); \
} \
return; \
} \
} \
malloc_tsd_dalloc(wrapper); \
} \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
if (pthread_key_create(&a_name##_tsd, \
a_name##_tsd_cleanup_wrapper) != 0) \
return (true); \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_name##_tsd_wrapper_t * \
a_name##_tsd_get_wrapper(void) \
{ \
a_name##_tsd_wrapper_t *wrapper = (a_name##_tsd_wrapper_t *) \
pthread_getspecific(a_name##_tsd); \
\
if (wrapper == NULL) { \
tsd_init_block_t block; \
wrapper = tsd_init_check_recursion( \
&a_name##_tsd_init_head, &block); \
if (wrapper) \
return (wrapper); \
wrapper = (a_name##_tsd_wrapper_t *) \
malloc_tsd_malloc(sizeof(a_name##_tsd_wrapper_t)); \
block.data = wrapper; \
if (wrapper == NULL) { \
malloc_write("<jemalloc>: Error allocating" \
" TSD for "#a_name"\n"); \
abort(); \
} else { \
static a_type tsd_static_data = a_initializer; \
wrapper->initialized = false; \
wrapper->val = tsd_static_data; \
} \
if (pthread_setspecific(a_name##_tsd, \
(void *)wrapper)) { \
malloc_write("<jemalloc>: Error setting" \
" TSD for "#a_name"\n"); \
abort(); \
} \
tsd_init_finish(&a_name##_tsd_init_head, &block); \
} \
return (wrapper); \
} \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
return (&wrapper->val); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
wrapper->val = *(val); \
if (a_cleanup != malloc_tsd_no_cleanup) \
wrapper->initialized = true; \
MALLOC_TSD
#undef O
/* tsd_foo_get(tsd) returns the value of the thread-local instance of foo. */
#define O(n, t, nt) \
JEMALLOC_ALWAYS_INLINE t \
tsd_##n##_get(tsd_t *tsd) { \
return *tsd_##n##p_get(tsd); \
}
#endif
MALLOC_TSD
#undef O
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* tsd_foo_set(tsd, val) updates the thread-local instance of foo to be val. */
#define O(n, t, nt) \
JEMALLOC_ALWAYS_INLINE void \
tsd_##n##_set(tsd_t *tsd, t val) { \
assert(tsd->state != tsd_state_reincarnated && \
tsd->state != tsd_state_minimal_initialized); \
*tsd_##n##p_get(tsd) = val; \
}
MALLOC_TSD
#undef O
#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
!defined(_WIN32))
struct tsd_init_block_s {
ql_elm(tsd_init_block_t) link;
pthread_t thread;
void *data;
};
struct tsd_init_head_s {
ql_head(tsd_init_block_t) blocks;
malloc_mutex_t lock;
};
#endif
JEMALLOC_ALWAYS_INLINE void
tsd_assert_fast(tsd_t *tsd) {
assert(!malloc_slow && tsd_tcache_enabled_get(tsd) &&
tsd_reentrancy_level_get(tsd) == 0);
}
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
JEMALLOC_ALWAYS_INLINE bool
tsd_fast(tsd_t *tsd) {
bool fast = (tsd->state == tsd_state_nominal);
if (fast) {
tsd_assert_fast(tsd);
}
void *malloc_tsd_malloc(size_t size);
void malloc_tsd_dalloc(void *wrapper);
void malloc_tsd_no_cleanup(void *);
void malloc_tsd_cleanup_register(bool (*f)(void));
void malloc_tsd_boot(void);
#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
!defined(_WIN32))
void *tsd_init_check_recursion(tsd_init_head_t *head,
tsd_init_block_t *block);
void tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block);
#endif
return fast;
}
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
JEMALLOC_ALWAYS_INLINE tsd_t *
tsd_fetch_impl(bool init, bool minimal) {
tsd_t *tsd = tsd_get(init);
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
if (!init && tsd_get_allocates() && tsd == NULL) {
return NULL;
}
assert(tsd != NULL);
if (unlikely(tsd->state != tsd_state_nominal)) {
return tsd_fetch_slow(tsd, minimal);
}
assert(tsd_fast(tsd));
tsd_assert_fast(tsd);
return tsd;
}
/* Get a minimal TSD that requires no cleanup. See comments in free(). */
JEMALLOC_ALWAYS_INLINE tsd_t *
tsd_fetch_min(void) {
return tsd_fetch_impl(true, true);
}
/* For internal background threads use only. */
JEMALLOC_ALWAYS_INLINE tsd_t *
tsd_internal_fetch(void) {
tsd_t *tsd = tsd_fetch_min();
/* Use reincarnated state to prevent full initialization. */
tsd->state = tsd_state_reincarnated;
return tsd;
}
JEMALLOC_ALWAYS_INLINE tsd_t *
tsd_fetch(void) {
return tsd_fetch_impl(true, false);
}
static inline bool
tsd_nominal(tsd_t *tsd) {
return (tsd->state <= tsd_state_nominal_max);
}
JEMALLOC_ALWAYS_INLINE tsdn_t *
tsdn_fetch(void) {
if (!tsd_booted_get()) {
return NULL;
}
return tsd_tsdn(tsd_fetch_impl(false, false));
}
JEMALLOC_ALWAYS_INLINE rtree_ctx_t *
tsd_rtree_ctx(tsd_t *tsd) {
return tsd_rtree_ctxp_get(tsd);
}
JEMALLOC_ALWAYS_INLINE rtree_ctx_t *
tsdn_rtree_ctx(tsdn_t *tsdn, rtree_ctx_t *fallback) {
/*
* If tsd cannot be accessed, initialize the fallback rtree_ctx and
* return a pointer to it.
*/
if (unlikely(tsdn_null(tsdn))) {
rtree_ctx_data_init(fallback);
return fallback;
}
return tsd_rtree_ctx(tsdn_tsd(tsdn));
}
#endif /* JEMALLOC_INTERNAL_TSD_H */

167
deps/jemalloc/include/jemalloc/internal/util.h vendored
View File

@@ -1,143 +1,50 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifndef JEMALLOC_INTERNAL_UTIL_H
#define JEMALLOC_INTERNAL_UTIL_H
/* Size of stack-allocated buffer passed to buferror(). */
#define BUFERROR_BUF 64
#define UTIL_INLINE static inline
/*
* Size of stack-allocated buffer used by malloc_{,v,vc}printf(). This must be
* large enough for all possible uses within jemalloc.
*/
#define MALLOC_PRINTF_BUFSIZE 4096
/* Junk fill patterns. */
#ifndef JEMALLOC_ALLOC_JUNK
# define JEMALLOC_ALLOC_JUNK ((uint8_t)0xa5)
#endif
#ifndef JEMALLOC_FREE_JUNK
# define JEMALLOC_FREE_JUNK ((uint8_t)0x5a)
#endif
/*
* Wrap a cpp argument that contains commas such that it isn't broken up into
* multiple arguments.
*/
#define JEMALLOC_ARG_CONCAT(...) __VA_ARGS__
#define JEMALLOC_ARG_CONCAT(...) __VA_ARGS__
/* cpp macro definition stringification. */
#define STRINGIFY_HELPER(x) #x
#define STRINGIFY(x) STRINGIFY_HELPER(x)
/*
* Silence compiler warnings due to uninitialized values. This is used
* wherever the compiler fails to recognize that the variable is never used
* uninitialized.
*/
#ifdef JEMALLOC_CC_SILENCE
# define JEMALLOC_CC_SILENCE_INIT(v) = v
#define JEMALLOC_CC_SILENCE_INIT(v) = v
#ifdef __GNUC__
# define likely(x) __builtin_expect(!!(x), 1)
# define unlikely(x) __builtin_expect(!!(x), 0)
#else
# define JEMALLOC_CC_SILENCE_INIT(v)
# define likely(x) !!(x)
# define unlikely(x) !!(x)
#endif
/*
* Define a custom assert() in order to reduce the chances of deadlock during
* assertion failure.
*/
#ifndef assert
#define assert(e) do { \
if (config_debug && !(e)) { \
malloc_printf( \
"<jemalloc>: %s:%d: Failed assertion: \"%s\"\n", \
__FILE__, __LINE__, #e); \
abort(); \
} \
} while (0)
#if !defined(JEMALLOC_INTERNAL_UNREACHABLE)
# error JEMALLOC_INTERNAL_UNREACHABLE should have been defined by configure
#endif
#ifndef not_reached
#define not_reached() do { \
if (config_debug) { \
malloc_printf( \
"<jemalloc>: %s:%d: Unreachable code reached\n", \
__FILE__, __LINE__); \
abort(); \
} \
} while (0)
#endif
#ifndef not_implemented
#define not_implemented() do { \
if (config_debug) { \
malloc_printf("<jemalloc>: %s:%d: Not implemented\n", \
__FILE__, __LINE__); \
abort(); \
} \
} while (0)
#endif
#ifndef assert_not_implemented
#define assert_not_implemented(e) do { \
if (config_debug && !(e)) \
not_implemented(); \
} while (0)
#endif
/* Use to assert a particular configuration, e.g., cassert(config_debug). */
#define cassert(c) do { \
if ((c) == false) \
not_reached(); \
} while (0)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
int buferror(int err, char *buf, size_t buflen);
uintmax_t malloc_strtoumax(const char *restrict nptr,
char **restrict endptr, int base);
void malloc_write(const char *s);
/*
* malloc_vsnprintf() supports a subset of snprintf(3) that avoids floating
* point math.
*/
int malloc_vsnprintf(char *str, size_t size, const char *format,
va_list ap);
int malloc_snprintf(char *str, size_t size, const char *format, ...)
JEMALLOC_ATTR(format(printf, 3, 4));
void malloc_vcprintf(void (*write_cb)(void *, const char *), void *cbopaque,
const char *format, va_list ap);
void malloc_cprintf(void (*write)(void *, const char *), void *cbopaque,
const char *format, ...) JEMALLOC_ATTR(format(printf, 3, 4));
void malloc_printf(const char *format, ...)
JEMALLOC_ATTR(format(printf, 1, 2));
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
size_t pow2_ceil(size_t x);
void set_errno(int errnum);
int get_errno(void);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_UTIL_C_))
/* Compute the smallest power of 2 that is >= x. */
JEMALLOC_INLINE size_t
pow2_ceil(size_t x)
{
x--;
x |= x >> 1;
x |= x >> 2;
x |= x >> 4;
x |= x >> 8;
x |= x >> 16;
#if (LG_SIZEOF_PTR == 3)
x |= x >> 32;
#endif
x++;
return (x);
}
/* Sets error code */
JEMALLOC_INLINE void
set_errno(int errnum)
{
#define unreachable() JEMALLOC_INTERNAL_UNREACHABLE()
/* Set error code. */
UTIL_INLINE void
set_errno(int errnum) {
#ifdef _WIN32
SetLastError(errnum);
#else
@@ -145,18 +52,16 @@ set_errno(int errnum)
#endif
}
/* Get last error code */
JEMALLOC_INLINE int
get_errno(void)
{
/* Get last error code. */
UTIL_INLINE int
get_errno(void) {
#ifdef _WIN32
return (GetLastError());
return GetLastError();
#else
return (errno);
return errno;
#endif
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#undef UTIL_INLINE
#endif /* JEMALLOC_INTERNAL_UTIL_H */

456
deps/jemalloc/include/jemalloc/jemalloc.h vendored
View File

@@ -1,45 +1,203 @@
#ifndef JEMALLOC_H_
#define JEMALLOC_H_
#define JEMALLOC_H_
#ifdef __cplusplus
extern "C" {
#endif
/* Defined if __attribute__((...)) syntax is supported. */
#define JEMALLOC_HAVE_ATTR
/* Defined if alloc_size attribute is supported. */
#define JEMALLOC_HAVE_ATTR_ALLOC_SIZE
/* Defined if format(gnu_printf, ...) attribute is supported. */
#define JEMALLOC_HAVE_ATTR_FORMAT_GNU_PRINTF
/* Defined if format(printf, ...) attribute is supported. */
#define JEMALLOC_HAVE_ATTR_FORMAT_PRINTF
/*
* Define overrides for non-standard allocator-related functions if they are
* present on the system.
*/
#define JEMALLOC_OVERRIDE_MEMALIGN
#define JEMALLOC_OVERRIDE_VALLOC
/*
* At least Linux omits the "const" in:
*
* size_t malloc_usable_size(const void *ptr);
*
* Match the operating system's prototype.
*/
#define JEMALLOC_USABLE_SIZE_CONST
/*
* If defined, specify throw() for the public function prototypes when compiling
* with C++. The only justification for this is to match the prototypes that
* glibc defines.
*/
#define JEMALLOC_USE_CXX_THROW
#ifdef _MSC_VER
# ifdef _WIN64
# define LG_SIZEOF_PTR_WIN 3
# else
# define LG_SIZEOF_PTR_WIN 2
# endif
#endif
/*
* Name mangling for public symbols is controlled by --with-mangling and
* --with-jemalloc-prefix. With default settings the je_ prefix is stripped by
* these macro definitions.
*/
#ifndef JEMALLOC_NO_RENAME
# define je_aligned_alloc aligned_alloc
# define je_calloc calloc
# define je_dallocx dallocx
# define je_free free
# define je_mallctl mallctl
# define je_mallctlbymib mallctlbymib
# define je_mallctlnametomib mallctlnametomib
# define je_malloc malloc
# define je_malloc_conf malloc_conf
# define je_malloc_message malloc_message
# define je_malloc_stats_print malloc_stats_print
# define je_malloc_usable_size malloc_usable_size
# define je_mallocx mallocx
# define je_nallocx nallocx
# define je_posix_memalign posix_memalign
# define je_rallocx rallocx
# define je_realloc realloc
# define je_sallocx sallocx
# define je_sdallocx sdallocx
# define je_xallocx xallocx
# define je_memalign memalign
# define je_valloc valloc
#endif
#include <stdlib.h>
#include <stdbool.h>
#include <stdint.h>
#include <limits.h>
#include <strings.h>
#define JEMALLOC_VERSION "3.6.0-0-g46c0af68bd248b04df75e4f92d5fb804c3d75340"
#define JEMALLOC_VERSION_MAJOR 3
#define JEMALLOC_VERSION_MINOR 6
#define JEMALLOC_VERSION_BUGFIX 0
#define JEMALLOC_VERSION_NREV 0
#define JEMALLOC_VERSION_GID "46c0af68bd248b04df75e4f92d5fb804c3d75340"
#define JEMALLOC_VERSION "5.0.1-0-g896ed3a8b3f41998d4fb4d625d30ac63ef2d51fb"
#define JEMALLOC_VERSION_MAJOR 5
#define JEMALLOC_VERSION_MINOR 0
#define JEMALLOC_VERSION_BUGFIX 1
#define JEMALLOC_VERSION_NREV 0
#define JEMALLOC_VERSION_GID "896ed3a8b3f41998d4fb4d625d30ac63ef2d51fb"
# define MALLOCX_LG_ALIGN(la) (la)
# if LG_SIZEOF_PTR == 2
# define MALLOCX_ALIGN(a) (ffs(a)-1)
# else
# define MALLOCX_ALIGN(a) \
((a < (size_t)INT_MAX) ? ffs(a)-1 : ffs(a>>32)+31)
# endif
# define MALLOCX_ZERO ((int)0x40)
/* Bias arena index bits so that 0 encodes "MALLOCX_ARENA() unspecified". */
# define MALLOCX_ARENA(a) ((int)(((a)+1) << 8))
#define MALLOCX_LG_ALIGN(la) ((int)(la))
#if LG_SIZEOF_PTR == 2
# define MALLOCX_ALIGN(a) ((int)(ffs((int)(a))-1))
#else
# define MALLOCX_ALIGN(a) \
((int)(((size_t)(a) < (size_t)INT_MAX) ? ffs((int)(a))-1 : \
ffs((int)(((size_t)(a))>>32))+31))
#endif
#define MALLOCX_ZERO ((int)0x40)
/*
* Bias tcache index bits so that 0 encodes "automatic tcache management", and 1
* encodes MALLOCX_TCACHE_NONE.
*/
#define MALLOCX_TCACHE(tc) ((int)(((tc)+2) << 8))
#define MALLOCX_TCACHE_NONE MALLOCX_TCACHE(-1)
/*
* Bias arena index bits so that 0 encodes "use an automatically chosen arena".
*/
#define MALLOCX_ARENA(a) ((((int)(a))+1) << 20)
#ifdef JEMALLOC_EXPERIMENTAL
# define ALLOCM_LG_ALIGN(la) (la)
# if LG_SIZEOF_PTR == 2
# define ALLOCM_ALIGN(a) (ffs(a)-1)
# else
# define ALLOCM_ALIGN(a) \
((a < (size_t)INT_MAX) ? ffs(a)-1 : ffs(a>>32)+31)
/*
* Use as arena index in "arena.<i>.{purge,decay,dss}" and
* "stats.arenas.<i>.*" mallctl interfaces to select all arenas. This
* definition is intentionally specified in raw decimal format to support
* cpp-based string concatenation, e.g.
*
* #define STRINGIFY_HELPER(x) #x
* #define STRINGIFY(x) STRINGIFY_HELPER(x)
*
* mallctl("arena." STRINGIFY(MALLCTL_ARENAS_ALL) ".purge", NULL, NULL, NULL,
* 0);
*/
#define MALLCTL_ARENAS_ALL 4096
/*
* Use as arena index in "stats.arenas.<i>.*" mallctl interfaces to select
* destroyed arenas.
*/
#define MALLCTL_ARENAS_DESTROYED 4097
#if defined(__cplusplus) && defined(JEMALLOC_USE_CXX_THROW)
# define JEMALLOC_CXX_THROW throw()
#else
# define JEMALLOC_CXX_THROW
#endif
#if defined(_MSC_VER)
# define JEMALLOC_ATTR(s)
# define JEMALLOC_ALIGNED(s) __declspec(align(s))
# define JEMALLOC_ALLOC_SIZE(s)
# define JEMALLOC_ALLOC_SIZE2(s1, s2)
# ifndef JEMALLOC_EXPORT
# ifdef DLLEXPORT
# define JEMALLOC_EXPORT __declspec(dllexport)
# else
# define JEMALLOC_EXPORT __declspec(dllimport)
# endif
# endif
# define ALLOCM_ZERO ((int)0x40)
# define ALLOCM_NO_MOVE ((int)0x80)
/* Bias arena index bits so that 0 encodes "ALLOCM_ARENA() unspecified". */
# define ALLOCM_ARENA(a) ((int)(((a)+1) << 8))
# define ALLOCM_SUCCESS 0
# define ALLOCM_ERR_OOM 1
# define ALLOCM_ERR_NOT_MOVED 2
# define JEMALLOC_FORMAT_PRINTF(s, i)
# define JEMALLOC_NOINLINE __declspec(noinline)
# ifdef __cplusplus
# define JEMALLOC_NOTHROW __declspec(nothrow)
# else
# define JEMALLOC_NOTHROW
# endif
# define JEMALLOC_SECTION(s) __declspec(allocate(s))
# define JEMALLOC_RESTRICT_RETURN __declspec(restrict)
# if _MSC_VER >= 1900 && !defined(__EDG__)
# define JEMALLOC_ALLOCATOR __declspec(allocator)
# else
# define JEMALLOC_ALLOCATOR
# endif
#elif defined(JEMALLOC_HAVE_ATTR)
# define JEMALLOC_ATTR(s) __attribute__((s))
# define JEMALLOC_ALIGNED(s) JEMALLOC_ATTR(aligned(s))
# ifdef JEMALLOC_HAVE_ATTR_ALLOC_SIZE
# define JEMALLOC_ALLOC_SIZE(s) JEMALLOC_ATTR(alloc_size(s))
# define JEMALLOC_ALLOC_SIZE2(s1, s2) JEMALLOC_ATTR(alloc_size(s1, s2))
# else
# define JEMALLOC_ALLOC_SIZE(s)
# define JEMALLOC_ALLOC_SIZE2(s1, s2)
# endif
# ifndef JEMALLOC_EXPORT
# define JEMALLOC_EXPORT JEMALLOC_ATTR(visibility("default"))
# endif
# ifdef JEMALLOC_HAVE_ATTR_FORMAT_GNU_PRINTF
# define JEMALLOC_FORMAT_PRINTF(s, i) JEMALLOC_ATTR(format(gnu_printf, s, i))
# elif defined(JEMALLOC_HAVE_ATTR_FORMAT_PRINTF)
# define JEMALLOC_FORMAT_PRINTF(s, i) JEMALLOC_ATTR(format(printf, s, i))
# else
# define JEMALLOC_FORMAT_PRINTF(s, i)
# endif
# define JEMALLOC_NOINLINE JEMALLOC_ATTR(noinline)
# define JEMALLOC_NOTHROW JEMALLOC_ATTR(nothrow)
# define JEMALLOC_SECTION(s) JEMALLOC_ATTR(section(s))
# define JEMALLOC_RESTRICT_RETURN
# define JEMALLOC_ALLOCATOR
#else
# define JEMALLOC_ATTR(s)
# define JEMALLOC_ALIGNED(s)
# define JEMALLOC_ALLOC_SIZE(s)
# define JEMALLOC_ALLOC_SIZE2(s1, s2)
# define JEMALLOC_EXPORT
# define JEMALLOC_FORMAT_PRINTF(s, i)
# define JEMALLOC_NOINLINE
# define JEMALLOC_NOTHROW
# define JEMALLOC_SECTION(s)
# define JEMALLOC_RESTRICT_RETURN
# define JEMALLOC_ALLOCATOR
#endif
/*
@@ -51,55 +209,141 @@ extern JEMALLOC_EXPORT const char *je_malloc_conf;
extern JEMALLOC_EXPORT void (*je_malloc_message)(void *cbopaque,
const char *s);
JEMALLOC_EXPORT void *je_malloc(size_t size) JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT void *je_calloc(size_t num, size_t size)
JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT int je_posix_memalign(void **memptr, size_t alignment,
size_t size) JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT void *je_aligned_alloc(size_t alignment, size_t size)
JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT void *je_realloc(void *ptr, size_t size);
JEMALLOC_EXPORT void je_free(void *ptr);
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_malloc(size_t size)
JEMALLOC_CXX_THROW JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE(1);
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_calloc(size_t num, size_t size)
JEMALLOC_CXX_THROW JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE2(1, 2);
JEMALLOC_EXPORT int JEMALLOC_NOTHROW je_posix_memalign(void **memptr,
size_t alignment, size_t size) JEMALLOC_CXX_THROW JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_aligned_alloc(size_t alignment,
size_t size) JEMALLOC_CXX_THROW JEMALLOC_ATTR(malloc)
JEMALLOC_ALLOC_SIZE(2);
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_realloc(void *ptr, size_t size)
JEMALLOC_CXX_THROW JEMALLOC_ALLOC_SIZE(2);
JEMALLOC_EXPORT void JEMALLOC_NOTHROW je_free(void *ptr)
JEMALLOC_CXX_THROW;
JEMALLOC_EXPORT void *je_mallocx(size_t size, int flags);
JEMALLOC_EXPORT void *je_rallocx(void *ptr, size_t size, int flags);
JEMALLOC_EXPORT size_t je_xallocx(void *ptr, size_t size, size_t extra,
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_mallocx(size_t size, int flags)
JEMALLOC_ATTR(malloc) JEMALLOC_ALLOC_SIZE(1);
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_rallocx(void *ptr, size_t size,
int flags) JEMALLOC_ALLOC_SIZE(2);
JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW je_xallocx(void *ptr, size_t size,
size_t extra, int flags);
JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW je_sallocx(const void *ptr,
int flags) JEMALLOC_ATTR(pure);
JEMALLOC_EXPORT void JEMALLOC_NOTHROW je_dallocx(void *ptr, int flags);
JEMALLOC_EXPORT void JEMALLOC_NOTHROW je_sdallocx(void *ptr, size_t size,
int flags);
JEMALLOC_EXPORT size_t je_sallocx(const void *ptr, int flags);
JEMALLOC_EXPORT void je_dallocx(void *ptr, int flags);
JEMALLOC_EXPORT size_t je_nallocx(size_t size, int flags);
JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW je_nallocx(size_t size, int flags)
JEMALLOC_ATTR(pure);
JEMALLOC_EXPORT int je_mallctl(const char *name, void *oldp,
size_t *oldlenp, void *newp, size_t newlen);
JEMALLOC_EXPORT int je_mallctlnametomib(const char *name, size_t *mibp,
size_t *miblenp);
JEMALLOC_EXPORT int je_mallctlbymib(const size_t *mib, size_t miblen,
JEMALLOC_EXPORT int JEMALLOC_NOTHROW je_mallctl(const char *name,
void *oldp, size_t *oldlenp, void *newp, size_t newlen);
JEMALLOC_EXPORT void je_malloc_stats_print(void (*write_cb)(void *,
const char *), void *je_cbopaque, const char *opts);
JEMALLOC_EXPORT size_t je_malloc_usable_size(
JEMALLOC_USABLE_SIZE_CONST void *ptr);
JEMALLOC_EXPORT int JEMALLOC_NOTHROW je_mallctlnametomib(const char *name,
size_t *mibp, size_t *miblenp);
JEMALLOC_EXPORT int JEMALLOC_NOTHROW je_mallctlbymib(const size_t *mib,
size_t miblen, void *oldp, size_t *oldlenp, void *newp, size_t newlen);
JEMALLOC_EXPORT void JEMALLOC_NOTHROW je_malloc_stats_print(
void (*write_cb)(void *, const char *), void *je_cbopaque,
const char *opts);
JEMALLOC_EXPORT size_t JEMALLOC_NOTHROW je_malloc_usable_size(
JEMALLOC_USABLE_SIZE_CONST void *ptr) JEMALLOC_CXX_THROW;
#ifdef JEMALLOC_OVERRIDE_MEMALIGN
JEMALLOC_EXPORT void * je_memalign(size_t alignment, size_t size)
JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_memalign(size_t alignment, size_t size)
JEMALLOC_CXX_THROW JEMALLOC_ATTR(malloc);
#endif
#ifdef JEMALLOC_OVERRIDE_VALLOC
JEMALLOC_EXPORT void * je_valloc(size_t size) JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT JEMALLOC_ALLOCATOR JEMALLOC_RESTRICT_RETURN
void JEMALLOC_NOTHROW *je_valloc(size_t size) JEMALLOC_CXX_THROW
JEMALLOC_ATTR(malloc);
#endif
#ifdef JEMALLOC_EXPERIMENTAL
JEMALLOC_EXPORT int je_allocm(void **ptr, size_t *rsize, size_t size,
int flags) JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_rallocm(void **ptr, size_t *rsize, size_t size,
size_t extra, int flags) JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_sallocm(const void *ptr, size_t *rsize, int flags)
JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_dallocm(void *ptr, int flags)
JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_nallocm(size_t *rsize, size_t size, int flags);
#endif
typedef struct extent_hooks_s extent_hooks_t;
/*
* void *
* extent_alloc(extent_hooks_t *extent_hooks, void *new_addr, size_t size,
* size_t alignment, bool *zero, bool *commit, unsigned arena_ind);
*/
typedef void *(extent_alloc_t)(extent_hooks_t *, void *, size_t, size_t, bool *,
bool *, unsigned);
/*
* bool
* extent_dalloc(extent_hooks_t *extent_hooks, void *addr, size_t size,
* bool committed, unsigned arena_ind);
*/
typedef bool (extent_dalloc_t)(extent_hooks_t *, void *, size_t, bool,
unsigned);
/*
* void
* extent_destroy(extent_hooks_t *extent_hooks, void *addr, size_t size,
* bool committed, unsigned arena_ind);
*/
typedef void (extent_destroy_t)(extent_hooks_t *, void *, size_t, bool,
unsigned);
/*
* bool
* extent_commit(extent_hooks_t *extent_hooks, void *addr, size_t size,
* size_t offset, size_t length, unsigned arena_ind);
*/
typedef bool (extent_commit_t)(extent_hooks_t *, void *, size_t, size_t, size_t,
unsigned);
/*
* bool
* extent_decommit(extent_hooks_t *extent_hooks, void *addr, size_t size,
* size_t offset, size_t length, unsigned arena_ind);
*/
typedef bool (extent_decommit_t)(extent_hooks_t *, void *, size_t, size_t,
size_t, unsigned);
/*
* bool
* extent_purge(extent_hooks_t *extent_hooks, void *addr, size_t size,
* size_t offset, size_t length, unsigned arena_ind);
*/
typedef bool (extent_purge_t)(extent_hooks_t *, void *, size_t, size_t, size_t,
unsigned);
/*
* bool
* extent_split(extent_hooks_t *extent_hooks, void *addr, size_t size,
* size_t size_a, size_t size_b, bool committed, unsigned arena_ind);
*/
typedef bool (extent_split_t)(extent_hooks_t *, void *, size_t, size_t, size_t,
bool, unsigned);
/*
* bool
* extent_merge(extent_hooks_t *extent_hooks, void *addr_a, size_t size_a,
* void *addr_b, size_t size_b, bool committed, unsigned arena_ind);
*/
typedef bool (extent_merge_t)(extent_hooks_t *, void *, size_t, void *, size_t,
bool, unsigned);
struct extent_hooks_s {
extent_alloc_t *alloc;
extent_dalloc_t *dalloc;
extent_destroy_t *destroy;
extent_commit_t *commit;
extent_decommit_t *decommit;
extent_purge_t *purge_lazy;
extent_purge_t *purge_forced;
extent_split_t *split;
extent_merge_t *merge;
};
/*
* By default application code must explicitly refer to mangled symbol names,
@@ -112,32 +356,28 @@ JEMALLOC_EXPORT int je_nallocm(size_t *rsize, size_t size, int flags);
# ifndef JEMALLOC_NO_DEMANGLE
# define JEMALLOC_NO_DEMANGLE
# endif
# define aligned_alloc je_aligned_alloc
# define calloc je_calloc
# define dallocx je_dallocx
# define free je_free
# define mallctl je_mallctl
# define mallctlbymib je_mallctlbymib
# define mallctlnametomib je_mallctlnametomib
# define malloc je_malloc
# define malloc_conf je_malloc_conf
# define malloc_message je_malloc_message
# define malloc je_malloc
# define calloc je_calloc
# define posix_memalign je_posix_memalign
# define aligned_alloc je_aligned_alloc
# define realloc je_realloc
# define free je_free
# define mallocx je_mallocx
# define rallocx je_rallocx
# define xallocx je_xallocx
# define sallocx je_sallocx
# define dallocx je_dallocx
# define nallocx je_nallocx
# define mallctl je_mallctl
# define mallctlnametomib je_mallctlnametomib
# define mallctlbymib je_mallctlbymib
# define malloc_stats_print je_malloc_stats_print
# define malloc_usable_size je_malloc_usable_size
# define mallocx je_mallocx
# define nallocx je_nallocx
# define posix_memalign je_posix_memalign
# define rallocx je_rallocx
# define realloc je_realloc
# define sallocx je_sallocx
# define sdallocx je_sdallocx
# define xallocx je_xallocx
# define memalign je_memalign
# define valloc je_valloc
# define allocm je_allocm
# define dallocm je_dallocm
# define nallocm je_nallocm
# define rallocm je_rallocm
# define sallocm je_sallocm
#endif
/*
@@ -148,35 +388,31 @@ JEMALLOC_EXPORT int je_nallocm(size_t *rsize, size_t size, int flags);
* and/or --with-jemalloc-prefix.
*/
#ifndef JEMALLOC_NO_DEMANGLE
# undef je_aligned_alloc
# undef je_calloc
# undef je_dallocx
# undef je_free
# undef je_mallctl
# undef je_mallctlbymib
# undef je_mallctlnametomib
# undef je_malloc
# undef je_malloc_conf
# undef je_malloc_message
# undef je_malloc
# undef je_calloc
# undef je_posix_memalign
# undef je_aligned_alloc
# undef je_realloc
# undef je_free
# undef je_mallocx
# undef je_rallocx
# undef je_xallocx
# undef je_sallocx
# undef je_dallocx
# undef je_nallocx
# undef je_mallctl
# undef je_mallctlnametomib
# undef je_mallctlbymib
# undef je_malloc_stats_print
# undef je_malloc_usable_size
# undef je_mallocx
# undef je_nallocx
# undef je_posix_memalign
# undef je_rallocx
# undef je_realloc
# undef je_sallocx
# undef je_sdallocx
# undef je_xallocx
# undef je_memalign
# undef je_valloc
# undef je_allocm
# undef je_dallocm
# undef je_nallocm
# undef je_rallocm
# undef je_sallocm
#endif
#ifdef __cplusplus
};
}
#endif
#endif /* JEMALLOC_H_ */

57
deps/jemalloc/include/msvc_compat/strings.h vendored
View File

@@ -3,21 +3,56 @@
/* MSVC doesn't define ffs/ffsl. This dummy strings.h header is provided
* for both */
#include <intrin.h>
#pragma intrinsic(_BitScanForward)
static __forceinline int ffsl(long x)
{
#ifdef _MSC_VER
# include <intrin.h>
# pragma intrinsic(_BitScanForward)
static __forceinline int ffsl(long x) {
unsigned long i;
if (_BitScanForward(&i, x))
return (i + 1);
return (0);
if (_BitScanForward(&i, x)) {
return i + 1;
}
return 0;
}
static __forceinline int ffs(int x)
{
return (ffsl(x));
static __forceinline int ffs(int x) {
return ffsl(x);
}
# ifdef _M_X64
# pragma intrinsic(_BitScanForward64)
# endif
static __forceinline int ffsll(unsigned __int64 x) {
unsigned long i;
#ifdef _M_X64
if (_BitScanForward64(&i, x)) {
return i + 1;
}
return 0;
#else
// Fallback for 32-bit build where 64-bit version not available
// assuming little endian
union {
unsigned __int64 ll;
unsigned long l[2];
} s;
s.ll = x;
if (_BitScanForward(&i, s.l[0])) {
return i + 1;
} else if(_BitScanForward(&i, s.l[1])) {
return i + 33;
}
return 0;
#endif
}
#else
# define ffsll(x) __builtin_ffsll(x)
# define ffsl(x) __builtin_ffsl(x)
# define ffs(x) __builtin_ffs(x)
#endif
#endif /* strings_h */

311
deps/jemalloc/jemalloc_defs.h.in.cmake vendored
View File

@@ -1,4 +1,6 @@
/* include/jemalloc/jemalloc_defs.h. Generated from jemalloc_defs.h.in by configure. */
/* include/jemalloc/internal/jemalloc_internal_defs.h. Generated from jemalloc_internal_defs.h.in by configure. */
#ifndef JEMALLOC_INTERNAL_DEFS_H_
#define JEMALLOC_INTERNAL_DEFS_H_
/*
* If JEMALLOC_PREFIX is defined via --with-jemalloc-prefix, it will cause all
* public APIs to be prefixed. This makes it possible, with some care, to use
@@ -8,30 +10,16 @@
/* #undef JEMALLOC_CPREFIX */
/*
* Name mangling for public symbols is controlled by --with-mangling and
* --with-jemalloc-prefix. With default settings the je_ prefix is stripped by
* these macro definitions.
* Define overrides for non-standard allocator-related functions if they are
* present on the system.
*/
#define je_malloc_conf malloc_conf
#define je_malloc_message malloc_message
#define je_malloc malloc
#define je_calloc calloc
#define je_posix_memalign posix_memalign
#define je_aligned_alloc aligned_alloc
#define je_realloc realloc
#define je_free free
#define je_malloc_usable_size malloc_usable_size
#define je_malloc_stats_print malloc_stats_print
#define je_mallctl mallctl
#define je_mallctlnametomib mallctlnametomib
#define je_mallctlbymib mallctlbymib
#define je_memalign memalign
#define je_valloc valloc
#define je_allocm allocm
#define je_rallocm rallocm
#define je_sallocm sallocm
#define je_dallocm dallocm
#define je_nallocm nallocm
#define JEMALLOC_OVERRIDE___LIBC_CALLOC
#define JEMALLOC_OVERRIDE___LIBC_FREE
#define JEMALLOC_OVERRIDE___LIBC_MALLOC
#define JEMALLOC_OVERRIDE___LIBC_MEMALIGN
#define JEMALLOC_OVERRIDE___LIBC_REALLOC
#define JEMALLOC_OVERRIDE___LIBC_VALLOC
/* #undef JEMALLOC_OVERRIDE___POSIX_MEMALIGN */
/*
* JEMALLOC_PRIVATE_NAMESPACE is used as a prefix for all library-private APIs.
@@ -39,8 +27,7 @@
* from being exported, but for static libraries, naming collisions are a real
* possibility.
*/
#define JEMALLOC_PRIVATE_NAMESPACE ""
#define JEMALLOC_N(string_that_no_one_should_want_to_use_as_a_jemalloc_private_namespace_prefix) string_that_no_one_should_want_to_use_as_a_jemalloc_private_namespace_prefix
#define JEMALLOC_PRIVATE_NAMESPACE je_
/*
* Hyper-threaded CPUs may need a special instruction inside spin loops in
@@ -48,20 +35,27 @@
*/
#define CPU_SPINWAIT __asm__ volatile("pause")
/* Defined if the equivalent of FreeBSD's atomic(9) functions are available. */
/* #undef JEMALLOC_ATOMIC9 */
/*
* Defined if OSAtomic*() functions are available, as provided by Darwin, and
* documented in the atomic(3) manual page.
* Number of significant bits in virtual addresses. This may be less than the
* total number of bits in a pointer, e.g. on x64, for which the uppermost 16
* bits are the same as bit 47.
*/
/* #undef JEMALLOC_OSATOMIC */
#define LG_VADDR @JEM_VADDRBITS@
/* Defined if C11 atomics are available. */
#define JEMALLOC_C11_ATOMICS 1
/* Defined if GCC __atomic atomics are available. */
#define JEMALLOC_GCC_ATOMIC_ATOMICS 1
/* Defined if GCC __sync atomics are available. */
#define JEMALLOC_GCC_SYNC_ATOMICS 1
/*
* Defined if __sync_add_and_fetch(uint32_t *, uint32_t) and
* __sync_sub_and_fetch(uint32_t *, uint32_t) are available, despite
* __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 not being defined (which means the
* functions are defined in libgcc instead of being inlines)
* functions are defined in libgcc instead of being inlines).
*/
/* #undef JE_FORCE_SYNC_COMPARE_AND_SWAP_4 */
@@ -69,16 +63,60 @@
* Defined if __sync_add_and_fetch(uint64_t *, uint64_t) and
* __sync_sub_and_fetch(uint64_t *, uint64_t) are available, despite
* __GCC_HAVE_SYNC_COMPARE_AND_SWAP_8 not being defined (which means the
* functions are defined in libgcc instead of being inlines)
* functions are defined in libgcc instead of being inlines).
*/
/* #undef JE_FORCE_SYNC_COMPARE_AND_SWAP_8 */
/*
* Defined if __builtin_clz() and __builtin_clzl() are available.
*/
#define JEMALLOC_HAVE_BUILTIN_CLZ
/*
* Defined if os_unfair_lock_*() functions are available, as provided by Darwin.
*/
/* #undef JEMALLOC_OS_UNFAIR_LOCK */
/*
* Defined if OSSpin*() functions are available, as provided by Darwin, and
* documented in the spinlock(3) manual page.
*/
/* #undef JEMALLOC_OSSPIN */
/* Defined if syscall(2) is usable. */
#define JEMALLOC_USE_SYSCALL
/*
* Defined if secure_getenv(3) is available.
*/
#define JEMALLOC_HAVE_SECURE_GETENV
/*
* Defined if issetugid(2) is available.
*/
/* #undef JEMALLOC_HAVE_ISSETUGID */
/* Defined if pthread_atfork(3) is available. */
#define JEMALLOC_HAVE_PTHREAD_ATFORK
/* Defined if pthread_setname_np(3) is available. */
#define JEMALLOC_HAVE_PTHREAD_SETNAME_NP
/*
* Defined if clock_gettime(CLOCK_MONOTONIC_COARSE, ...) is available.
*/
#define JEMALLOC_HAVE_CLOCK_MONOTONIC_COARSE 1
/*
* Defined if clock_gettime(CLOCK_MONOTONIC, ...) is available.
*/
#define JEMALLOC_HAVE_CLOCK_MONOTONIC 1
/*
* Defined if mach_absolute_time() is available.
*/
/* #undef JEMALLOC_HAVE_MACH_ABSOLUTE_TIME */
/*
* Defined if _malloc_thread_cleanup() exists. At least in the case of
* FreeBSD, pthread_key_create() allocates, which if used during malloc
@@ -102,41 +140,9 @@
*/
/* #undef JEMALLOC_MUTEX_INIT_CB */
/* Defined if __attribute__((...)) syntax is supported. */
#define JEMALLOC_HAVE_ATTR
#ifdef JEMALLOC_HAVE_ATTR
# define JEMALLOC_ATTR(s) __attribute__((s))
# define JEMALLOC_EXPORT JEMALLOC_ATTR(visibility("default"))
# define JEMALLOC_ALIGNED(s) JEMALLOC_ATTR(aligned(s))
# define JEMALLOC_SECTION(s) JEMALLOC_ATTR(section(s))
# define JEMALLOC_NOINLINE JEMALLOC_ATTR(noinline)
#elif _MSC_VER
# define JEMALLOC_ATTR(s)
# ifdef DLLEXPORT
# define JEMALLOC_EXPORT __declspec(dllexport)
# else
# define JEMALLOC_EXPORT __declspec(dllimport)
# endif
# define JEMALLOC_ALIGNED(s) __declspec(align(s))
# define JEMALLOC_SECTION(s) __declspec(allocate(s))
# define JEMALLOC_NOINLINE __declspec(noinline)
#else
# define JEMALLOC_ATTR(s)
# define JEMALLOC_EXPORT
# define JEMALLOC_ALIGNED(s)
# define JEMALLOC_SECTION(s)
# define JEMALLOC_NOINLINE
#endif
/* Defined if sbrk() is supported. */
#define JEMALLOC_HAVE_SBRK
/* Non-empty if the tls_model attribute is supported. */
#define JEMALLOC_TLS_MODEL @JEM_TLSMODEL@
/* JEMALLOC_CC_SILENCE enables code that silences unuseful compiler warnings. */
/* #undef JEMALLOC_CC_SILENCE */
/*
* JEMALLOC_DEBUG enables assertions and other sanity checks, and disables
* inline functions.
@@ -144,7 +150,7 @@
/* #undef JEMALLOC_DEBUG */
/* JEMALLOC_STATS enables statistics calculation. */
#define JEMALLOC_STATS
/* #undef JEMALLOC_STATS */
/* JEMALLOC_PROF enables allocation profiling. */
/* #undef JEMALLOC_PROF */
@@ -159,104 +165,127 @@
/* #undef JEMALLOC_PROF_GCC */
/*
* JEMALLOC_TCACHE enables a thread-specific caching layer for small objects.
* This makes it possible to allocate/deallocate objects without any locking
* when the cache is in the steady state.
*/
#define JEMALLOC_TCACHE
/*
* JEMALLOC_DSS enables use of sbrk(2) to allocate chunks from the data storage
* JEMALLOC_DSS enables use of sbrk(2) to allocate extents from the data storage
* segment (DSS).
*/
/* #undef JEMALLOC_DSS */
#define JEMALLOC_DSS
/* Support memory filling (junk/zero/quarantine/redzone). */
/* Support memory filling (junk/zero). */
#define JEMALLOC_FILL
/* Support the experimental API. */
#define JEMALLOC_EXPERIMENTAL
/* Support utrace(2)-based tracing. */
/* #undef JEMALLOC_UTRACE */
/* Support Valgrind. */
/* #undef JEMALLOC_VALGRIND */
/* Support optional abort() on OOM. */
/* #undef JEMALLOC_XMALLOC */
/* Support lazy locking (avoid locking unless a second thread is launched). */
/* #undef JEMALLOC_LAZY_LOCK */
/* One page is 2^STATIC_PAGE_SHIFT bytes. */
#define STATIC_PAGE_SHIFT 12
/*
* Minimum allocation alignment is 2^LG_QUANTUM bytes (ignoring tiny size
* classes).
*/
/* #undef LG_QUANTUM */
/* One page is 2^LG_PAGE bytes. */
#define LG_PAGE 12
/*
* If defined, use munmap() to unmap freed chunks, rather than storing them for
* later reuse. This is disabled by default on Linux because common sequences
* of mmap()/munmap() calls will cause virtual memory map holes.
* One huge page is 2^LG_HUGEPAGE bytes. Note that this is defined even if the
* system does not explicitly support huge pages; system calls that require
* explicit huge page support are separately configured.
*/
/* #undef JEMALLOC_MUNMAP */
#define LG_HUGEPAGE 21
/*
* If defined, use mremap(...MREMAP_FIXED...) for huge realloc(). This is
* disabled by default because it is Linux-specific and it will cause virtual
* memory map holes, much like munmap(2) does.
* If defined, adjacent virtual memory mappings with identical attributes
* automatically coalesce, and they fragment when changes are made to subranges.
* This is the normal order of things for mmap()/munmap(), but on Windows
* VirtualAlloc()/VirtualFree() operations must be precisely matched, i.e.
* mappings do *not* coalesce/fragment.
*/
/* #undef JEMALLOC_MREMAP */
#define JEMALLOC_MAPS_COALESCE
/*
* If defined, retain memory for later reuse by default rather than using e.g.
* munmap() to unmap freed extents. This is enabled on 64-bit Linux because
* common sequences of mmap()/munmap() calls will cause virtual memory map
* holes.
*/
#define JEMALLOC_RETAIN
/* TLS is used to map arenas and magazine caches to threads. */
#define JEMALLOC_TLS
/*
* JEMALLOC_IVSALLOC enables ivsalloc(), which verifies that pointers reside
* within jemalloc-owned chunks before dereferencing them.
* Used to mark unreachable code to quiet "end of non-void" compiler warnings.
* Don't use this directly; instead use unreachable() from util.h
*/
/* #undef JEMALLOC_IVSALLOC */
#define JEMALLOC_INTERNAL_UNREACHABLE __builtin_unreachable
/*
* Define overrides for non-standard allocator-related functions if they
* are present on the system.
* ffs*() functions to use for bitmapping. Don't use these directly; instead,
* use ffs_*() from util.h.
*/
#define JEMALLOC_OVERRIDE_MEMALIGN
#define JEMALLOC_OVERRIDE_VALLOC
#define JEMALLOC_INTERNAL_FFSLL __builtin_ffsll
#define JEMALLOC_INTERNAL_FFSL __builtin_ffsl
#define JEMALLOC_INTERNAL_FFS __builtin_ffs
/*
* At least Linux omits the "const" in:
*
* size_t malloc_usable_size(const void *ptr);
*
* Match the operating system's prototype.
* If defined, explicitly attempt to more uniformly distribute large allocation
* pointer alignments across all cache indices.
*/
#define JEMALLOC_USABLE_SIZE_CONST
#define JEMALLOC_CACHE_OBLIVIOUS
/*
* Darwin (OS X) uses zones to work around Mach-O symbol override shortcomings.
*/
/* #undef JEMALLOC_ZONE */
/* #undef JEMALLOC_ZONE_VERSION */
/*
* Methods for determining whether the OS overcommits.
* JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY: Linux's
* /proc/sys/vm.overcommit_memory file.
* JEMALLOC_SYSCTL_VM_OVERCOMMIT: FreeBSD's vm.overcommit sysctl.
*/
/* #undef JEMALLOC_SYSCTL_VM_OVERCOMMIT */
#define JEMALLOC_PROC_SYS_VM_OVERCOMMIT_MEMORY
/* Defined if madvise(2) is available. */
#define JEMALLOC_HAVE_MADVISE
/*
* Methods for purging unused pages differ between operating systems.
*
* madvise(..., MADV_DONTNEED) : On Linux, this immediately discards pages,
* madvise(..., MADV_FREE) : This marks pages as being unused, such that they
* will be discarded rather than swapped out.
* madvise(..., MADV_DONTNEED) : If JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS is
* defined, this immediately discards pages,
* such that new pages will be demand-zeroed if
* the address region is later touched.
* madvise(..., MADV_FREE) : On FreeBSD and Darwin, this marks pages as being
* unused, such that they will be discarded rather
* than swapped out.
* the address region is later touched;
* otherwise this behaves similarly to
* MADV_FREE, though typically with higher
* system overhead.
*/
@JEM_MADFREE_DEF@ JEMALLOC_PURGE_MADVISE_FREE
#define JEMALLOC_PURGE_MADVISE_DONTNEED
/* #undef JEMALLOC_PURGE_MADVISE_FREE */
#define JEMALLOC_PURGE_MADVISE_DONTNEED_ZEROS 1
/*
* Define if operating system has alloca.h header.
* Defined if transparent huge pages (THPs) are supported via the
* MADV_[NO]HUGEPAGE arguments to madvise(2), and THP support is enabled.
*/
#define JEMALLOC_HAS_ALLOCA_H
#define JEMALLOC_THP
/* sizeof(void *) == 2^LG_SIZEOF_PTR. */
#define LG_SIZEOF_PTR @JEM_SIZEDEF@
/* Define if operating system has alloca.h header. */
#define JEMALLOC_HAS_ALLOCA_H 1
/* C99 restrict keyword supported. */
#define JEMALLOC_HAS_RESTRICT 1
/* For use by hash code. */
/* #undef JEMALLOC_BIG_ENDIAN */
/* sizeof(int) == 2^LG_SIZEOF_INT. */
#define LG_SIZEOF_INT 2
@@ -264,11 +293,51 @@
/* sizeof(long) == 2^LG_SIZEOF_LONG. */
#define LG_SIZEOF_LONG @JEM_SIZEDEF@
/* sizeof(long long) == 2^LG_SIZEOF_LONG_LONG. */
#define LG_SIZEOF_LONG_LONG 3
/* sizeof(intmax_t) == 2^LG_SIZEOF_INTMAX_T. */
#define LG_SIZEOF_INTMAX_T 3
/* C99 restrict keyword supported. */
/*#define JEMALLOC_HAS_RESTRICT*/
/* glibc malloc hooks (__malloc_hook, __realloc_hook, __free_hook). */
#define JEMALLOC_GLIBC_MALLOC_HOOK
/* JEMALLOC_CODE_COVERAGE enables test code coverage analysis. */
#undef JEMALLOC_CODE_COVERAGE
/* glibc memalign hook. */
#define JEMALLOC_GLIBC_MEMALIGN_HOOK
/* pthread support */
#define JEMALLOC_HAVE_PTHREAD
/* dlsym() support */
#define JEMALLOC_HAVE_DLSYM
/* Adaptive mutex support in pthreads. */
#define JEMALLOC_HAVE_PTHREAD_MUTEX_ADAPTIVE_NP
/* GNU specific sched_getcpu support */
#define JEMALLOC_HAVE_SCHED_GETCPU
/* GNU specific sched_setaffinity support */
#define JEMALLOC_HAVE_SCHED_SETAFFINITY
/*
* If defined, all the features necessary for background threads are present.
*/
#define JEMALLOC_BACKGROUND_THREAD 1
/*
* If defined, jemalloc symbols are not exported (doesn't work when
* JEMALLOC_PREFIX is not defined).
*/
/* #undef JEMALLOC_EXPORT */
/* config.malloc_conf options string. */
#define JEMALLOC_CONFIG_MALLOC_CONF ""
/* If defined, jemalloc takes the malloc/free/etc. symbol names. */
#define JEMALLOC_IS_MALLOC 1
/* sizeof(void *) == 2^LG_SIZEOF_PTR. */
#define LG_SIZEOF_PTR @JEM_SIZEDEF@
#endif /* JEMALLOC_INTERNAL_DEFS_H_ */

4280
deps/jemalloc/src/arena.c vendored
View File

File diff suppressed because it is too large Load Diff

496
deps/jemalloc/src/base.c vendored
View File

@@ -1,142 +1,402 @@
#define JEMALLOC_BASE_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_BASE_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
#include "jemalloc/internal/assert.h"
#include "jemalloc/internal/extent_mmap.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/sz.h"
/******************************************************************************/
/* Data. */
static malloc_mutex_t base_mtx;
static base_t *b0;
/******************************************************************************/
static void *
base_map(tsdn_t *tsdn, extent_hooks_t *extent_hooks, unsigned ind, size_t size) {
void *addr;
bool zero = true;
bool commit = true;
assert(size == HUGEPAGE_CEILING(size));
if (extent_hooks == &extent_hooks_default) {
addr = extent_alloc_mmap(NULL, size, PAGE, &zero, &commit);
} else {
/* No arena context as we are creating new arenas. */
tsd_t *tsd = tsdn_null(tsdn) ? tsd_fetch() : tsdn_tsd(tsdn);
pre_reentrancy(tsd, NULL);
addr = extent_hooks->alloc(extent_hooks, NULL, size, PAGE,
&zero, &commit, ind);
post_reentrancy(tsd);
}
return addr;
}
static void
base_unmap(tsdn_t *tsdn, extent_hooks_t *extent_hooks, unsigned ind, void *addr,
size_t size) {
/*
* Cascade through dalloc, decommit, purge_forced, and purge_lazy,
* stopping at first success. This cascade is performed for consistency
* with the cascade in extent_dalloc_wrapper() because an application's
* custom hooks may not support e.g. dalloc. This function is only ever
* called as a side effect of arena destruction, so although it might
* seem pointless to do anything besides dalloc here, the application
* may in fact want the end state of all associated virtual memory to be
* in some consistent-but-allocated state.
*/
if (extent_hooks == &extent_hooks_default) {
if (!extent_dalloc_mmap(addr, size)) {
return;
}
if (!pages_decommit(addr, size)) {
return;
}
if (!pages_purge_forced(addr, size)) {
return;
}
if (!pages_purge_lazy(addr, size)) {
return;
}
/* Nothing worked. This should never happen. */
not_reached();
} else {
tsd_t *tsd = tsdn_null(tsdn) ? tsd_fetch() : tsdn_tsd(tsdn);
pre_reentrancy(tsd, NULL);
if (extent_hooks->dalloc != NULL &&
!extent_hooks->dalloc(extent_hooks, addr, size, true,
ind)) {
goto label_done;
}
if (extent_hooks->decommit != NULL &&
!extent_hooks->decommit(extent_hooks, addr, size, 0, size,
ind)) {
goto label_done;
}
if (extent_hooks->purge_forced != NULL &&
!extent_hooks->purge_forced(extent_hooks, addr, size, 0,
size, ind)) {
goto label_done;
}
if (extent_hooks->purge_lazy != NULL &&
!extent_hooks->purge_lazy(extent_hooks, addr, size, 0, size,
ind)) {
goto label_done;
}
/* Nothing worked. That's the application's problem. */
label_done:
post_reentrancy(tsd);
return;
}
}
static void
base_extent_init(size_t *extent_sn_next, extent_t *extent, void *addr,
size_t size) {
size_t sn;
sn = *extent_sn_next;
(*extent_sn_next)++;
extent_binit(extent, addr, size, sn);
}
static void *
base_extent_bump_alloc_helper(extent_t *extent, size_t *gap_size, size_t size,
size_t alignment) {
void *ret;
assert(alignment == ALIGNMENT_CEILING(alignment, QUANTUM));
assert(size == ALIGNMENT_CEILING(size, alignment));
*gap_size = ALIGNMENT_CEILING((uintptr_t)extent_addr_get(extent),
alignment) - (uintptr_t)extent_addr_get(extent);
ret = (void *)((uintptr_t)extent_addr_get(extent) + *gap_size);
assert(extent_bsize_get(extent) >= *gap_size + size);
extent_binit(extent, (void *)((uintptr_t)extent_addr_get(extent) +
*gap_size + size), extent_bsize_get(extent) - *gap_size - size,
extent_sn_get(extent));
return ret;
}
static void
base_extent_bump_alloc_post(tsdn_t *tsdn, base_t *base, extent_t *extent,
size_t gap_size, void *addr, size_t size) {
if (extent_bsize_get(extent) > 0) {
/*
* Compute the index for the largest size class that does not
* exceed extent's size.
*/
szind_t index_floor =
sz_size2index(extent_bsize_get(extent) + 1) - 1;
extent_heap_insert(&base->avail[index_floor], extent);
}
if (config_stats) {
base->allocated += size;
/*
* Add one PAGE to base_resident for every page boundary that is
* crossed by the new allocation.
*/
base->resident += PAGE_CEILING((uintptr_t)addr + size) -
PAGE_CEILING((uintptr_t)addr - gap_size);
assert(base->allocated <= base->resident);
assert(base->resident <= base->mapped);
}
}
static void *
base_extent_bump_alloc(tsdn_t *tsdn, base_t *base, extent_t *extent,
size_t size, size_t alignment) {
void *ret;
size_t gap_size;
ret = base_extent_bump_alloc_helper(extent, &gap_size, size, alignment);
base_extent_bump_alloc_post(tsdn, base, extent, gap_size, ret, size);
return ret;
}
/*
* Current pages that are being used for internal memory allocations. These
* pages are carved up in cacheline-size quanta, so that there is no chance of
* false cache line sharing.
* Allocate a block of virtual memory that is large enough to start with a
* base_block_t header, followed by an object of specified size and alignment.
* On success a pointer to the initialized base_block_t header is returned.
*/
static void *base_pages;
static void *base_next_addr;
static void *base_past_addr; /* Addr immediately past base_pages. */
static extent_node_t *base_nodes;
/******************************************************************************/
/* Function prototypes for non-inline static functions. */
static bool base_pages_alloc(size_t minsize);
/******************************************************************************/
static bool
base_pages_alloc(size_t minsize)
{
size_t csize;
bool zero;
assert(minsize != 0);
csize = CHUNK_CEILING(minsize);
zero = false;
base_pages = chunk_alloc(csize, chunksize, true, &zero,
chunk_dss_prec_get());
if (base_pages == NULL)
return (true);
base_next_addr = base_pages;
base_past_addr = (void *)((uintptr_t)base_pages + csize);
return (false);
}
void *
base_alloc(size_t size)
{
void *ret;
size_t csize;
/* Round size up to nearest multiple of the cacheline size. */
csize = CACHELINE_CEILING(size);
malloc_mutex_lock(&base_mtx);
/* Make sure there's enough space for the allocation. */
if ((uintptr_t)base_next_addr + csize > (uintptr_t)base_past_addr) {
if (base_pages_alloc(csize)) {
malloc_mutex_unlock(&base_mtx);
return (NULL);
}
static base_block_t *
base_block_alloc(tsdn_t *tsdn, extent_hooks_t *extent_hooks, unsigned ind,
pszind_t *pind_last, size_t *extent_sn_next, size_t size,
size_t alignment) {
alignment = ALIGNMENT_CEILING(alignment, QUANTUM);
size_t usize = ALIGNMENT_CEILING(size, alignment);
size_t header_size = sizeof(base_block_t);
size_t gap_size = ALIGNMENT_CEILING(header_size, alignment) -
header_size;
/*
* Create increasingly larger blocks in order to limit the total number
* of disjoint virtual memory ranges. Choose the next size in the page
* size class series (skipping size classes that are not a multiple of
* HUGEPAGE), or a size large enough to satisfy the requested size and
* alignment, whichever is larger.
*/
size_t min_block_size = HUGEPAGE_CEILING(sz_psz2u(header_size + gap_size
+ usize));
pszind_t pind_next = (*pind_last + 1 < NPSIZES) ? *pind_last + 1 :
*pind_last;
size_t next_block_size = HUGEPAGE_CEILING(sz_pind2sz(pind_next));
size_t block_size = (min_block_size > next_block_size) ? min_block_size
: next_block_size;
base_block_t *block = (base_block_t *)base_map(tsdn, extent_hooks, ind,
block_size);
if (block == NULL) {
return NULL;
}
/* Allocate. */
ret = base_next_addr;
base_next_addr = (void *)((uintptr_t)base_next_addr + csize);
malloc_mutex_unlock(&base_mtx);
VALGRIND_MAKE_MEM_UNDEFINED(ret, csize);
return (ret);
*pind_last = sz_psz2ind(block_size);
block->size = block_size;
block->next = NULL;
assert(block_size >= header_size);
base_extent_init(extent_sn_next, &block->extent,
(void *)((uintptr_t)block + header_size), block_size - header_size);
return block;
}
void *
base_calloc(size_t number, size_t size)
{
void *ret = base_alloc(number * size);
/*
* Allocate an extent that is at least as large as specified size, with
* specified alignment.
*/
static extent_t *
base_extent_alloc(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment) {
malloc_mutex_assert_owner(tsdn, &base->mtx);
if (ret != NULL)
memset(ret, 0, number * size);
return (ret);
extent_hooks_t *extent_hooks = base_extent_hooks_get(base);
/*
* Drop mutex during base_block_alloc(), because an extent hook will be
* called.
*/
malloc_mutex_unlock(tsdn, &base->mtx);
base_block_t *block = base_block_alloc(tsdn, extent_hooks,
base_ind_get(base), &base->pind_last, &base->extent_sn_next, size,
alignment);
malloc_mutex_lock(tsdn, &base->mtx);
if (block == NULL) {
return NULL;
}
block->next = base->blocks;
base->blocks = block;
if (config_stats) {
base->allocated += sizeof(base_block_t);
base->resident += PAGE_CEILING(sizeof(base_block_t));
base->mapped += block->size;
assert(base->allocated <= base->resident);
assert(base->resident <= base->mapped);
}
return &block->extent;
}
extent_node_t *
base_node_alloc(void)
{
extent_node_t *ret;
base_t *
b0get(void) {
return b0;
}
malloc_mutex_lock(&base_mtx);
if (base_nodes != NULL) {
ret = base_nodes;
base_nodes = *(extent_node_t **)ret;
malloc_mutex_unlock(&base_mtx);
VALGRIND_MAKE_MEM_UNDEFINED(ret, sizeof(extent_node_t));
} else {
malloc_mutex_unlock(&base_mtx);
ret = (extent_node_t *)base_alloc(sizeof(extent_node_t));
base_t *
base_new(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks) {
pszind_t pind_last = 0;
size_t extent_sn_next = 0;
base_block_t *block = base_block_alloc(tsdn, extent_hooks, ind,
&pind_last, &extent_sn_next, sizeof(base_t), QUANTUM);
if (block == NULL) {
return NULL;
}
return (ret);
size_t gap_size;
size_t base_alignment = CACHELINE;
size_t base_size = ALIGNMENT_CEILING(sizeof(base_t), base_alignment);
base_t *base = (base_t *)base_extent_bump_alloc_helper(&block->extent,
&gap_size, base_size, base_alignment);
base->ind = ind;
atomic_store_p(&base->extent_hooks, extent_hooks, ATOMIC_RELAXED);
if (malloc_mutex_init(&base->mtx, "base", WITNESS_RANK_BASE,
malloc_mutex_rank_exclusive)) {
base_unmap(tsdn, extent_hooks, ind, block, block->size);
return NULL;
}
base->pind_last = pind_last;
base->extent_sn_next = extent_sn_next;
base->blocks = block;
for (szind_t i = 0; i < NSIZES; i++) {
extent_heap_new(&base->avail[i]);
}
if (config_stats) {
base->allocated = sizeof(base_block_t);
base->resident = PAGE_CEILING(sizeof(base_block_t));
base->mapped = block->size;
assert(base->allocated <= base->resident);
assert(base->resident <= base->mapped);
}
base_extent_bump_alloc_post(tsdn, base, &block->extent, gap_size, base,
base_size);
return base;
}
void
base_node_dealloc(extent_node_t *node)
{
base_delete(tsdn_t *tsdn, base_t *base) {
extent_hooks_t *extent_hooks = base_extent_hooks_get(base);
base_block_t *next = base->blocks;
do {
base_block_t *block = next;
next = block->next;
base_unmap(tsdn, extent_hooks, base_ind_get(base), block,
block->size);
} while (next != NULL);
}
VALGRIND_MAKE_MEM_UNDEFINED(node, sizeof(extent_node_t));
malloc_mutex_lock(&base_mtx);
*(extent_node_t **)node = base_nodes;
base_nodes = node;
malloc_mutex_unlock(&base_mtx);
extent_hooks_t *
base_extent_hooks_get(base_t *base) {
return (extent_hooks_t *)atomic_load_p(&base->extent_hooks,
ATOMIC_ACQUIRE);
}
extent_hooks_t *
base_extent_hooks_set(base_t *base, extent_hooks_t *extent_hooks) {
extent_hooks_t *old_extent_hooks = base_extent_hooks_get(base);
atomic_store_p(&base->extent_hooks, extent_hooks, ATOMIC_RELEASE);
return old_extent_hooks;
}
static void *
base_alloc_impl(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment,
size_t *esn) {
alignment = QUANTUM_CEILING(alignment);
size_t usize = ALIGNMENT_CEILING(size, alignment);
size_t asize = usize + alignment - QUANTUM;
extent_t *extent = NULL;
malloc_mutex_lock(tsdn, &base->mtx);
for (szind_t i = sz_size2index(asize); i < NSIZES; i++) {
extent = extent_heap_remove_first(&base->avail[i]);
if (extent != NULL) {
/* Use existing space. */
break;
}
}
if (extent == NULL) {
/* Try to allocate more space. */
extent = base_extent_alloc(tsdn, base, usize, alignment);
}
void *ret;
if (extent == NULL) {
ret = NULL;
goto label_return;
}
ret = base_extent_bump_alloc(tsdn, base, extent, usize, alignment);
if (esn != NULL) {
*esn = extent_sn_get(extent);
}
label_return:
malloc_mutex_unlock(tsdn, &base->mtx);
return ret;
}
/*
* base_alloc() returns zeroed memory, which is always demand-zeroed for the
* auto arenas, in order to make multi-page sparse data structures such as radix
* tree nodes efficient with respect to physical memory usage. Upon success a
* pointer to at least size bytes with specified alignment is returned. Note
* that size is rounded up to the nearest multiple of alignment to avoid false
* sharing.
*/
void *
base_alloc(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment) {
return base_alloc_impl(tsdn, base, size, alignment, NULL);
}
extent_t *
base_alloc_extent(tsdn_t *tsdn, base_t *base) {
size_t esn;
extent_t *extent = base_alloc_impl(tsdn, base, sizeof(extent_t),
CACHELINE, &esn);
if (extent == NULL) {
return NULL;
}
extent_esn_set(extent, esn);
return extent;
}
void
base_stats_get(tsdn_t *tsdn, base_t *base, size_t *allocated, size_t *resident,
size_t *mapped) {
cassert(config_stats);
malloc_mutex_lock(tsdn, &base->mtx);
assert(base->allocated <= base->resident);
assert(base->resident <= base->mapped);
*allocated = base->allocated;
*resident = base->resident;
*mapped = base->mapped;
malloc_mutex_unlock(tsdn, &base->mtx);
}
void
base_prefork(tsdn_t *tsdn, base_t *base) {
malloc_mutex_prefork(tsdn, &base->mtx);
}
void
base_postfork_parent(tsdn_t *tsdn, base_t *base) {
malloc_mutex_postfork_parent(tsdn, &base->mtx);
}
void
base_postfork_child(tsdn_t *tsdn, base_t *base) {
malloc_mutex_postfork_child(tsdn, &base->mtx);
}
bool
base_boot(void)
{
base_nodes = NULL;
if (malloc_mutex_init(&base_mtx))
return (true);
return (false);
}
void
base_prefork(void)
{
malloc_mutex_prefork(&base_mtx);
}
void
base_postfork_parent(void)
{
malloc_mutex_postfork_parent(&base_mtx);
}
void
base_postfork_child(void)
{
malloc_mutex_postfork_child(&base_mtx);
base_boot(tsdn_t *tsdn) {
b0 = base_new(tsdn, 0, (extent_hooks_t *)&extent_hooks_default);
return (b0 == NULL);
}

113
deps/jemalloc/src/bitmap.c vendored
View File

@@ -1,24 +1,15 @@
#define JEMALLOC_BITMAP_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_BITMAP_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
/******************************************************************************/
/* Function prototypes for non-inline static functions. */
static size_t bits2groups(size_t nbits);
#include "jemalloc/internal/assert.h"
/******************************************************************************/
static size_t
bits2groups(size_t nbits)
{
return ((nbits >> LG_BITMAP_GROUP_NBITS) +
!!(nbits & BITMAP_GROUP_NBITS_MASK));
}
#ifdef BITMAP_USE_TREE
void
bitmap_info_init(bitmap_info_t *binfo, size_t nbits)
{
bitmap_info_init(bitmap_info_t *binfo, size_t nbits) {
unsigned i;
size_t group_count;
@@ -31,60 +22,100 @@ bitmap_info_init(bitmap_info_t *binfo, size_t nbits)
* that requires only one group.
*/
binfo->levels[0].group_offset = 0;
group_count = bits2groups(nbits);
group_count = BITMAP_BITS2GROUPS(nbits);
for (i = 1; group_count > 1; i++) {
assert(i < BITMAP_MAX_LEVELS);
binfo->levels[i].group_offset = binfo->levels[i-1].group_offset
+ group_count;
group_count = bits2groups(group_count);
group_count = BITMAP_BITS2GROUPS(group_count);
}
binfo->levels[i].group_offset = binfo->levels[i-1].group_offset
+ group_count;
assert(binfo->levels[i].group_offset <= BITMAP_GROUPS_MAX);
binfo->nlevels = i;
binfo->nbits = nbits;
}
size_t
bitmap_info_ngroups(const bitmap_info_t *binfo)
{
return (binfo->levels[binfo->nlevels].group_offset << LG_SIZEOF_BITMAP);
}
size_t
bitmap_size(size_t nbits)
{
bitmap_info_t binfo;
bitmap_info_init(&binfo, nbits);
return (bitmap_info_ngroups(&binfo));
static size_t
bitmap_info_ngroups(const bitmap_info_t *binfo) {
return binfo->levels[binfo->nlevels].group_offset;
}
void
bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo)
{
bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo, bool fill) {
size_t extra;
unsigned i;
/*
* Bits are actually inverted with regard to the external bitmap
* interface, so the bitmap starts out with all 1 bits, except for
* trailing unused bits (if any). Note that each group uses bit 0 to
* correspond to the first logical bit in the group, so extra bits
* are the most significant bits of the last group.
* interface.
*/
memset(bitmap, 0xffU, binfo->levels[binfo->nlevels].group_offset <<
LG_SIZEOF_BITMAP);
if (fill) {
/* The "filled" bitmap starts out with all 0 bits. */
memset(bitmap, 0, bitmap_size(binfo));
return;
}
/*
* The "empty" bitmap starts out with all 1 bits, except for trailing
* unused bits (if any). Note that each group uses bit 0 to correspond
* to the first logical bit in the group, so extra bits are the most
* significant bits of the last group.
*/
memset(bitmap, 0xffU, bitmap_size(binfo));
extra = (BITMAP_GROUP_NBITS - (binfo->nbits & BITMAP_GROUP_NBITS_MASK))
& BITMAP_GROUP_NBITS_MASK;
if (extra != 0)
if (extra != 0) {
bitmap[binfo->levels[1].group_offset - 1] >>= extra;
}
for (i = 1; i < binfo->nlevels; i++) {
size_t group_count = binfo->levels[i].group_offset -
binfo->levels[i-1].group_offset;
extra = (BITMAP_GROUP_NBITS - (group_count &
BITMAP_GROUP_NBITS_MASK)) & BITMAP_GROUP_NBITS_MASK;
if (extra != 0)
if (extra != 0) {
bitmap[binfo->levels[i+1].group_offset - 1] >>= extra;
}
}
}
#else /* BITMAP_USE_TREE */
void
bitmap_info_init(bitmap_info_t *binfo, size_t nbits) {
assert(nbits > 0);
assert(nbits <= (ZU(1) << LG_BITMAP_MAXBITS));
binfo->ngroups = BITMAP_BITS2GROUPS(nbits);
binfo->nbits = nbits;
}
static size_t
bitmap_info_ngroups(const bitmap_info_t *binfo) {
return binfo->ngroups;
}
void
bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo, bool fill) {
size_t extra;
if (fill) {
memset(bitmap, 0, bitmap_size(binfo));
return;
}
memset(bitmap, 0xffU, bitmap_size(binfo));
extra = (BITMAP_GROUP_NBITS - (binfo->nbits & BITMAP_GROUP_NBITS_MASK))
& BITMAP_GROUP_NBITS_MASK;
if (extra != 0) {
bitmap[binfo->ngroups - 1] >>= extra;
}
}
#endif /* BITMAP_USE_TREE */
size_t
bitmap_size(const bitmap_info_t *binfo) {
return (bitmap_info_ngroups(binfo) << LG_SIZEOF_BITMAP);
}

254
deps/jemalloc/src/ckh.c vendored
View File

@@ -34,14 +34,24 @@
* respectively.
*
******************************************************************************/
#define JEMALLOC_CKH_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_CKH_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/ckh.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
#include "jemalloc/internal/assert.h"
#include "jemalloc/internal/hash.h"
#include "jemalloc/internal/malloc_io.h"
#include "jemalloc/internal/prng.h"
#include "jemalloc/internal/util.h"
/******************************************************************************/
/* Function prototypes for non-inline static functions. */
static bool ckh_grow(ckh_t *ckh);
static void ckh_shrink(ckh_t *ckh);
static bool ckh_grow(tsd_t *tsd, ckh_t *ckh);
static void ckh_shrink(tsd_t *tsd, ckh_t *ckh);
/******************************************************************************/
@@ -49,27 +59,26 @@ static void ckh_shrink(ckh_t *ckh);
* Search bucket for key and return the cell number if found; SIZE_T_MAX
* otherwise.
*/
JEMALLOC_INLINE_C size_t
ckh_bucket_search(ckh_t *ckh, size_t bucket, const void *key)
{
static size_t
ckh_bucket_search(ckh_t *ckh, size_t bucket, const void *key) {
ckhc_t *cell;
unsigned i;
for (i = 0; i < (ZU(1) << LG_CKH_BUCKET_CELLS); i++) {
cell = &ckh->tab[(bucket << LG_CKH_BUCKET_CELLS) + i];
if (cell->key != NULL && ckh->keycomp(key, cell->key))
return ((bucket << LG_CKH_BUCKET_CELLS) + i);
if (cell->key != NULL && ckh->keycomp(key, cell->key)) {
return (bucket << LG_CKH_BUCKET_CELLS) + i;
}
}
return (SIZE_T_MAX);
return SIZE_T_MAX;
}
/*
* Search table for key and return cell number if found; SIZE_T_MAX otherwise.
*/
JEMALLOC_INLINE_C size_t
ckh_isearch(ckh_t *ckh, const void *key)
{
static size_t
ckh_isearch(ckh_t *ckh, const void *key) {
size_t hashes[2], bucket, cell;
assert(ckh != NULL);
@@ -79,19 +88,19 @@ ckh_isearch(ckh_t *ckh, const void *key)
/* Search primary bucket. */
bucket = hashes[0] & ((ZU(1) << ckh->lg_curbuckets) - 1);
cell = ckh_bucket_search(ckh, bucket, key);
if (cell != SIZE_T_MAX)
return (cell);
if (cell != SIZE_T_MAX) {
return cell;
}
/* Search secondary bucket. */
bucket = hashes[1] & ((ZU(1) << ckh->lg_curbuckets) - 1);
cell = ckh_bucket_search(ckh, bucket, key);
return (cell);
return cell;
}
JEMALLOC_INLINE_C bool
static bool
ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key,
const void *data)
{
const void *data) {
ckhc_t *cell;
unsigned offset, i;
@@ -99,7 +108,8 @@ ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key,
* Cycle through the cells in the bucket, starting at a random position.
* The randomness avoids worst-case search overhead as buckets fill up.
*/
prng32(offset, LG_CKH_BUCKET_CELLS, ckh->prng_state, CKH_A, CKH_C);
offset = (unsigned)prng_lg_range_u64(&ckh->prng_state,
LG_CKH_BUCKET_CELLS);
for (i = 0; i < (ZU(1) << LG_CKH_BUCKET_CELLS); i++) {
cell = &ckh->tab[(bucket << LG_CKH_BUCKET_CELLS) +
((i + offset) & ((ZU(1) << LG_CKH_BUCKET_CELLS) - 1))];
@@ -107,11 +117,11 @@ ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key,
cell->key = key;
cell->data = data;
ckh->count++;
return (false);
return false;
}
}
return (true);
return true;
}
/*
@@ -120,10 +130,9 @@ ckh_try_bucket_insert(ckh_t *ckh, size_t bucket, const void *key,
* eviction/relocation procedure until either success or detection of an
* eviction/relocation bucket cycle.
*/
JEMALLOC_INLINE_C bool
static bool
ckh_evict_reloc_insert(ckh_t *ckh, size_t argbucket, void const **argkey,
void const **argdata)
{
void const **argdata) {
const void *key, *data, *tkey, *tdata;
ckhc_t *cell;
size_t hashes[2], bucket, tbucket;
@@ -141,7 +150,8 @@ ckh_evict_reloc_insert(ckh_t *ckh, size_t argbucket, void const **argkey,
* were an item for which both hashes indicated the same
* bucket.
*/
prng32(i, LG_CKH_BUCKET_CELLS, ckh->prng_state, CKH_A, CKH_C);
i = (unsigned)prng_lg_range_u64(&ckh->prng_state,
LG_CKH_BUCKET_CELLS);
cell = &ckh->tab[(bucket << LG_CKH_BUCKET_CELLS) + i];
assert(cell->key != NULL);
@@ -181,18 +191,18 @@ ckh_evict_reloc_insert(ckh_t *ckh, size_t argbucket, void const **argkey,
if (tbucket == argbucket) {
*argkey = key;
*argdata = data;
return (true);
return true;
}
bucket = tbucket;
if (ckh_try_bucket_insert(ckh, bucket, key, data) == false)
return (false);
if (!ckh_try_bucket_insert(ckh, bucket, key, data)) {
return false;
}
}
}
JEMALLOC_INLINE_C bool
ckh_try_insert(ckh_t *ckh, void const**argkey, void const**argdata)
{
static bool
ckh_try_insert(ckh_t *ckh, void const**argkey, void const**argdata) {
size_t hashes[2], bucket;
const void *key = *argkey;
const void *data = *argdata;
@@ -201,27 +211,28 @@ ckh_try_insert(ckh_t *ckh, void const**argkey, void const**argdata)
/* Try to insert in primary bucket. */
bucket = hashes[0] & ((ZU(1) << ckh->lg_curbuckets) - 1);
if (ckh_try_bucket_insert(ckh, bucket, key, data) == false)
return (false);
if (!ckh_try_bucket_insert(ckh, bucket, key, data)) {
return false;
}
/* Try to insert in secondary bucket. */
bucket = hashes[1] & ((ZU(1) << ckh->lg_curbuckets) - 1);
if (ckh_try_bucket_insert(ckh, bucket, key, data) == false)
return (false);
if (!ckh_try_bucket_insert(ckh, bucket, key, data)) {
return false;
}
/*
* Try to find a place for this item via iterative eviction/relocation.
*/
return (ckh_evict_reloc_insert(ckh, bucket, argkey, argdata));
return ckh_evict_reloc_insert(ckh, bucket, argkey, argdata);
}
/*
* Try to rebuild the hash table from scratch by inserting all items from the
* old table into the new.
*/
JEMALLOC_INLINE_C bool
ckh_rebuild(ckh_t *ckh, ckhc_t *aTab)
{
static bool
ckh_rebuild(ckh_t *ckh, ckhc_t *aTab) {
size_t count, i, nins;
const void *key, *data;
@@ -233,22 +244,20 @@ ckh_rebuild(ckh_t *ckh, ckhc_t *aTab)
data = aTab[i].data;
if (ckh_try_insert(ckh, &key, &data)) {
ckh->count = count;
return (true);
return true;
}
nins++;
}
}
return (false);
return false;
}
static bool
ckh_grow(ckh_t *ckh)
{
ckh_grow(tsd_t *tsd, ckh_t *ckh) {
bool ret;
ckhc_t *tab, *ttab;
size_t lg_curcells;
unsigned lg_prevbuckets;
unsigned lg_prevbuckets, lg_curcells;
#ifdef CKH_COUNT
ckh->ngrows++;
@@ -265,12 +274,13 @@ ckh_grow(ckh_t *ckh)
size_t usize;
lg_curcells++;
usize = sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE);
if (usize == 0) {
usize = sz_sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE);
if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) {
ret = true;
goto label_return;
}
tab = (ckhc_t *)ipalloc(usize, CACHELINE, true);
tab = (ckhc_t *)ipallocztm(tsd_tsdn(tsd), usize, CACHELINE,
true, NULL, true, arena_ichoose(tsd, NULL));
if (tab == NULL) {
ret = true;
goto label_return;
@@ -281,28 +291,27 @@ ckh_grow(ckh_t *ckh)
tab = ttab;
ckh->lg_curbuckets = lg_curcells - LG_CKH_BUCKET_CELLS;
if (ckh_rebuild(ckh, tab) == false) {
idalloc(tab);
if (!ckh_rebuild(ckh, tab)) {
idalloctm(tsd_tsdn(tsd), tab, NULL, NULL, true, true);
break;
}
/* Rebuilding failed, so back out partially rebuilt table. */
idalloc(ckh->tab);
idalloctm(tsd_tsdn(tsd), ckh->tab, NULL, NULL, true, true);
ckh->tab = tab;
ckh->lg_curbuckets = lg_prevbuckets;
}
ret = false;
label_return:
return (ret);
return ret;
}
static void
ckh_shrink(ckh_t *ckh)
{
ckh_shrink(tsd_t *tsd, ckh_t *ckh) {
ckhc_t *tab, *ttab;
size_t lg_curcells, usize;
unsigned lg_prevbuckets;
size_t usize;
unsigned lg_prevbuckets, lg_curcells;
/*
* It is possible (though unlikely, given well behaved hashes) that the
@@ -310,10 +319,12 @@ ckh_shrink(ckh_t *ckh)
*/
lg_prevbuckets = ckh->lg_curbuckets;
lg_curcells = ckh->lg_curbuckets + LG_CKH_BUCKET_CELLS - 1;
usize = sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE);
if (usize == 0)
usize = sz_sa2u(sizeof(ckhc_t) << lg_curcells, CACHELINE);
if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) {
return;
tab = (ckhc_t *)ipalloc(usize, CACHELINE, true);
}
tab = (ckhc_t *)ipallocztm(tsd_tsdn(tsd), usize, CACHELINE, true, NULL,
true, arena_ichoose(tsd, NULL));
if (tab == NULL) {
/*
* An OOM error isn't worth propagating, since it doesn't
@@ -327,8 +338,8 @@ ckh_shrink(ckh_t *ckh)
tab = ttab;
ckh->lg_curbuckets = lg_curcells - LG_CKH_BUCKET_CELLS;
if (ckh_rebuild(ckh, tab) == false) {
idalloc(tab);
if (!ckh_rebuild(ckh, tab)) {
idalloctm(tsd_tsdn(tsd), tab, NULL, NULL, true, true);
#ifdef CKH_COUNT
ckh->nshrinks++;
#endif
@@ -336,7 +347,7 @@ ckh_shrink(ckh_t *ckh)
}
/* Rebuilding failed, so back out partially rebuilt table. */
idalloc(ckh->tab);
idalloctm(tsd_tsdn(tsd), ckh->tab, NULL, NULL, true, true);
ckh->tab = tab;
ckh->lg_curbuckets = lg_prevbuckets;
#ifdef CKH_COUNT
@@ -345,8 +356,8 @@ ckh_shrink(ckh_t *ckh)
}
bool
ckh_new(ckh_t *ckh, size_t minitems, ckh_hash_t *hash, ckh_keycomp_t *keycomp)
{
ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash,
ckh_keycomp_t *keycomp) {
bool ret;
size_t mincells, usize;
unsigned lg_mincells;
@@ -366,29 +377,31 @@ ckh_new(ckh_t *ckh, size_t minitems, ckh_hash_t *hash, ckh_keycomp_t *keycomp)
ckh->count = 0;
/*
* Find the minimum power of 2 that is large enough to fit aBaseCount
* Find the minimum power of 2 that is large enough to fit minitems
* entries. We are using (2+,2) cuckoo hashing, which has an expected
* maximum load factor of at least ~0.86, so 0.75 is a conservative load
* factor that will typically allow 2^aLgMinItems to fit without ever
* factor that will typically allow mincells items to fit without ever
* growing the table.
*/
assert(LG_CKH_BUCKET_CELLS > 0);
mincells = ((minitems + (3 - (minitems % 3))) / 3) << 2;
for (lg_mincells = LG_CKH_BUCKET_CELLS;
(ZU(1) << lg_mincells) < mincells;
lg_mincells++)
; /* Do nothing. */
lg_mincells++) {
/* Do nothing. */
}
ckh->lg_minbuckets = lg_mincells - LG_CKH_BUCKET_CELLS;
ckh->lg_curbuckets = lg_mincells - LG_CKH_BUCKET_CELLS;
ckh->hash = hash;
ckh->keycomp = keycomp;
usize = sa2u(sizeof(ckhc_t) << lg_mincells, CACHELINE);
if (usize == 0) {
usize = sz_sa2u(sizeof(ckhc_t) << lg_mincells, CACHELINE);
if (unlikely(usize == 0 || usize > LARGE_MAXCLASS)) {
ret = true;
goto label_return;
}
ckh->tab = (ckhc_t *)ipalloc(usize, CACHELINE, true);
ckh->tab = (ckhc_t *)ipallocztm(tsd_tsdn(tsd), usize, CACHELINE, true,
NULL, true, arena_ichoose(tsd, NULL));
if (ckh->tab == NULL) {
ret = true;
goto label_return;
@@ -396,20 +409,18 @@ ckh_new(ckh_t *ckh, size_t minitems, ckh_hash_t *hash, ckh_keycomp_t *keycomp)
ret = false;
label_return:
return (ret);
return ret;
}
void
ckh_delete(ckh_t *ckh)
{
ckh_delete(tsd_t *tsd, ckh_t *ckh) {
assert(ckh != NULL);
#ifdef CKH_VERBOSE
malloc_printf(
"%s(%p): ngrows: %"PRIu64", nshrinks: %"PRIu64","
" nshrinkfails: %"PRIu64", ninserts: %"PRIu64","
" nrelocs: %"PRIu64"\n", __func__, ckh,
"%s(%p): ngrows: %"FMTu64", nshrinks: %"FMTu64","
" nshrinkfails: %"FMTu64", ninserts: %"FMTu64","
" nrelocs: %"FMTu64"\n", __func__, ckh,
(unsigned long long)ckh->ngrows,
(unsigned long long)ckh->nshrinks,
(unsigned long long)ckh->nshrinkfails,
@@ -417,43 +428,42 @@ ckh_delete(ckh_t *ckh)
(unsigned long long)ckh->nrelocs);
#endif
idalloc(ckh->tab);
if (config_debug)
memset(ckh, 0x5a, sizeof(ckh_t));
idalloctm(tsd_tsdn(tsd), ckh->tab, NULL, NULL, true, true);
if (config_debug) {
memset(ckh, JEMALLOC_FREE_JUNK, sizeof(ckh_t));
}
}
size_t
ckh_count(ckh_t *ckh)
{
ckh_count(ckh_t *ckh) {
assert(ckh != NULL);
return (ckh->count);
return ckh->count;
}
bool
ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data)
{
ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data) {
size_t i, ncells;
for (i = *tabind, ncells = (ZU(1) << (ckh->lg_curbuckets +
LG_CKH_BUCKET_CELLS)); i < ncells; i++) {
if (ckh->tab[i].key != NULL) {
if (key != NULL)
if (key != NULL) {
*key = (void *)ckh->tab[i].key;
if (data != NULL)
}
if (data != NULL) {
*data = (void *)ckh->tab[i].data;
}
*tabind = i + 1;
return (false);
return false;
}
}
return (true);
return true;
}
bool
ckh_insert(ckh_t *ckh, const void *key, const void *data)
{
ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data) {
bool ret;
assert(ckh != NULL);
@@ -464,7 +474,7 @@ ckh_insert(ckh_t *ckh, const void *key, const void *data)
#endif
while (ckh_try_insert(ckh, &key, &data)) {
if (ckh_grow(ckh)) {
if (ckh_grow(tsd, ckh)) {
ret = true;
goto label_return;
}
@@ -472,22 +482,24 @@ ckh_insert(ckh_t *ckh, const void *key, const void *data)
ret = false;
label_return:
return (ret);
return ret;
}
bool
ckh_remove(ckh_t *ckh, const void *searchkey, void **key, void **data)
{
ckh_remove(tsd_t *tsd, ckh_t *ckh, const void *searchkey, void **key,
void **data) {
size_t cell;
assert(ckh != NULL);
cell = ckh_isearch(ckh, searchkey);
if (cell != SIZE_T_MAX) {
if (key != NULL)
if (key != NULL) {
*key = (void *)ckh->tab[cell].key;
if (data != NULL)
}
if (data != NULL) {
*data = (void *)ckh->tab[cell].data;
}
ckh->tab[cell].key = NULL;
ckh->tab[cell].data = NULL; /* Not necessary. */
@@ -497,54 +509,50 @@ ckh_remove(ckh_t *ckh, const void *searchkey, void **key, void **data)
+ LG_CKH_BUCKET_CELLS - 2)) && ckh->lg_curbuckets
> ckh->lg_minbuckets) {
/* Ignore error due to OOM. */
ckh_shrink(ckh);
ckh_shrink(tsd, ckh);
}
return (false);
return false;
}
return (true);
return true;
}
bool
ckh_search(ckh_t *ckh, const void *searchkey, void **key, void **data)
{
ckh_search(ckh_t *ckh, const void *searchkey, void **key, void **data) {
size_t cell;
assert(ckh != NULL);
cell = ckh_isearch(ckh, searchkey);
if (cell != SIZE_T_MAX) {
if (key != NULL)
if (key != NULL) {
*key = (void *)ckh->tab[cell].key;
if (data != NULL)
}
if (data != NULL) {
*data = (void *)ckh->tab[cell].data;
return (false);
}
return false;
}
return (true);
return true;
}
void
ckh_string_hash(const void *key, size_t r_hash[2])
{
ckh_string_hash(const void *key, size_t r_hash[2]) {
hash(key, strlen((const char *)key), 0x94122f33U, r_hash);
}
bool
ckh_string_keycomp(const void *k1, const void *k2)
{
ckh_string_keycomp(const void *k1, const void *k2) {
assert(k1 != NULL);
assert(k2 != NULL);
assert(k1 != NULL);
assert(k2 != NULL);
return (strcmp((char *)k1, (char *)k2) ? false : true);
return !strcmp((char *)k1, (char *)k2);
}
void
ckh_pointer_hash(const void *key, size_t r_hash[2])
{
ckh_pointer_hash(const void *key, size_t r_hash[2]) {
union {
const void *v;
size_t i;
@@ -556,8 +564,6 @@ ckh_pointer_hash(const void *key, size_t r_hash[2])
}
bool
ckh_pointer_keycomp(const void *k1, const void *k2)
{
return ((k1 == k2) ? true : false);
ckh_pointer_keycomp(const void *k1, const void *k2) {
return (k1 == k2);
}

2770
deps/jemalloc/src/ctl.c vendored
View File

File diff suppressed because it is too large Load Diff

2004
deps/jemalloc/src/extent.c vendored
View File

File diff suppressed because it is too large Load Diff

5
deps/jemalloc/src/hash.c vendored
View File

@@ -1,2 +1,3 @@
#define JEMALLOC_HASH_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_HASH_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"

3649
deps/jemalloc/src/jemalloc.c vendored
View File

File diff suppressed because it is too large Load Diff

206
deps/jemalloc/src/mutex.c vendored
View File

@@ -1,12 +1,12 @@
#define JEMALLOC_MUTEX_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_MUTEX_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
#if defined(JEMALLOC_LAZY_LOCK) && !defined(_WIN32)
#include <dlfcn.h>
#endif
#include "jemalloc/internal/assert.h"
#include "jemalloc/internal/malloc_io.h"
#ifndef _CRT_SPINCOUNT
#define _CRT_SPINCOUNT 4000
#define _CRT_SPINCOUNT 4000
#endif
/******************************************************************************/
@@ -20,10 +20,6 @@ static bool postpone_init = true;
static malloc_mutex_t *postponed_mutexes = NULL;
#endif
#if defined(JEMALLOC_LAZY_LOCK) && !defined(_WIN32)
static void pthread_create_once(void);
#endif
/******************************************************************************/
/*
* We intercept pthread_create() calls in order to toggle isthreaded if the
@@ -31,33 +27,11 @@ static void pthread_create_once(void);
*/
#if defined(JEMALLOC_LAZY_LOCK) && !defined(_WIN32)
static int (*pthread_create_fptr)(pthread_t *__restrict, const pthread_attr_t *,
void *(*)(void *), void *__restrict);
static void
pthread_create_once(void)
{
pthread_create_fptr = dlsym(RTLD_NEXT, "pthread_create");
if (pthread_create_fptr == NULL) {
malloc_write("<jemalloc>: Error in dlsym(RTLD_NEXT, "
"\"pthread_create\")\n");
abort();
}
isthreaded = true;
}
JEMALLOC_EXPORT int
pthread_create(pthread_t *__restrict thread,
const pthread_attr_t *__restrict attr, void *(*start_routine)(void *),
void *__restrict arg)
{
static pthread_once_t once_control = PTHREAD_ONCE_INIT;
pthread_once(&once_control, pthread_create_once);
return (pthread_create_fptr(thread, attr, start_routine, arg));
void *__restrict arg) {
return pthread_create_wrapper(thread, attr, start_routine, arg);
}
#endif
@@ -68,14 +42,108 @@ JEMALLOC_EXPORT int _pthread_mutex_init_calloc_cb(pthread_mutex_t *mutex,
void *(calloc_cb)(size_t, size_t));
#endif
bool
malloc_mutex_init(malloc_mutex_t *mutex)
{
void
malloc_mutex_lock_slow(malloc_mutex_t *mutex) {
mutex_prof_data_t *data = &mutex->prof_data;
UNUSED nstime_t before = NSTIME_ZERO_INITIALIZER;
if (ncpus == 1) {
goto label_spin_done;
}
int cnt = 0, max_cnt = MALLOC_MUTEX_MAX_SPIN;
do {
CPU_SPINWAIT;
if (!malloc_mutex_trylock_final(mutex)) {
data->n_spin_acquired++;
return;
}
} while (cnt++ < max_cnt);
if (!config_stats) {
/* Only spin is useful when stats is off. */
malloc_mutex_lock_final(mutex);
return;
}
label_spin_done:
nstime_update(&before);
/* Copy before to after to avoid clock skews. */
nstime_t after;
nstime_copy(&after, &before);
uint32_t n_thds = atomic_fetch_add_u32(&data->n_waiting_thds, 1,
ATOMIC_RELAXED) + 1;
/* One last try as above two calls may take quite some cycles. */
if (!malloc_mutex_trylock_final(mutex)) {
atomic_fetch_sub_u32(&data->n_waiting_thds, 1, ATOMIC_RELAXED);
data->n_spin_acquired++;
return;
}
/* True slow path. */
malloc_mutex_lock_final(mutex);
/* Update more slow-path only counters. */
atomic_fetch_sub_u32(&data->n_waiting_thds, 1, ATOMIC_RELAXED);
nstime_update(&after);
nstime_t delta;
nstime_copy(&delta, &after);
nstime_subtract(&delta, &before);
data->n_wait_times++;
nstime_add(&data->tot_wait_time, &delta);
if (nstime_compare(&data->max_wait_time, &delta) < 0) {
nstime_copy(&data->max_wait_time, &delta);
}
if (n_thds > data->max_n_thds) {
data->max_n_thds = n_thds;
}
}
static void
mutex_prof_data_init(mutex_prof_data_t *data) {
memset(data, 0, sizeof(mutex_prof_data_t));
nstime_init(&data->max_wait_time, 0);
nstime_init(&data->tot_wait_time, 0);
data->prev_owner = NULL;
}
void
malloc_mutex_prof_data_reset(tsdn_t *tsdn, malloc_mutex_t *mutex) {
malloc_mutex_assert_owner(tsdn, mutex);
mutex_prof_data_init(&mutex->prof_data);
}
static int
mutex_addr_comp(const witness_t *witness1, void *mutex1,
const witness_t *witness2, void *mutex2) {
assert(mutex1 != NULL);
assert(mutex2 != NULL);
uintptr_t mu1int = (uintptr_t)mutex1;
uintptr_t mu2int = (uintptr_t)mutex2;
if (mu1int < mu2int) {
return -1;
} else if (mu1int == mu2int) {
return 0;
} else {
return 1;
}
}
bool
malloc_mutex_init(malloc_mutex_t *mutex, const char *name,
witness_rank_t rank, malloc_mutex_lock_order_t lock_order) {
mutex_prof_data_init(&mutex->prof_data);
#ifdef _WIN32
# if _WIN32_WINNT >= 0x0600
InitializeSRWLock(&mutex->lock);
# else
if (!InitializeCriticalSectionAndSpinCount(&mutex->lock,
_CRT_SPINCOUNT))
return (true);
_CRT_SPINCOUNT)) {
return true;
}
# endif
#elif (defined(JEMALLOC_OS_UNFAIR_LOCK))
mutex->lock = OS_UNFAIR_LOCK_INIT;
#elif (defined(JEMALLOC_OSSPIN))
mutex->lock = 0;
#elif (defined(JEMALLOC_MUTEX_INIT_CB))
@@ -83,67 +151,73 @@ malloc_mutex_init(malloc_mutex_t *mutex)
mutex->postponed_next = postponed_mutexes;
postponed_mutexes = mutex;
} else {
if (_pthread_mutex_init_calloc_cb(&mutex->lock, base_calloc) !=
0)
return (true);
if (_pthread_mutex_init_calloc_cb(&mutex->lock,
bootstrap_calloc) != 0) {
return true;
}
}
#else
pthread_mutexattr_t attr;
if (pthread_mutexattr_init(&attr) != 0)
return (true);
if (pthread_mutexattr_init(&attr) != 0) {
return true;
}
pthread_mutexattr_settype(&attr, MALLOC_MUTEX_TYPE);
if (pthread_mutex_init(&mutex->lock, &attr) != 0) {
pthread_mutexattr_destroy(&attr);
return (true);
return true;
}
pthread_mutexattr_destroy(&attr);
#endif
return (false);
if (config_debug) {
mutex->lock_order = lock_order;
if (lock_order == malloc_mutex_address_ordered) {
witness_init(&mutex->witness, name, rank,
mutex_addr_comp, &mutex);
} else {
witness_init(&mutex->witness, name, rank, NULL, NULL);
}
}
return false;
}
void
malloc_mutex_prefork(malloc_mutex_t *mutex)
{
malloc_mutex_lock(mutex);
malloc_mutex_prefork(tsdn_t *tsdn, malloc_mutex_t *mutex) {
malloc_mutex_lock(tsdn, mutex);
}
void
malloc_mutex_postfork_parent(malloc_mutex_t *mutex)
{
malloc_mutex_unlock(mutex);
malloc_mutex_postfork_parent(tsdn_t *tsdn, malloc_mutex_t *mutex) {
malloc_mutex_unlock(tsdn, mutex);
}
void
malloc_mutex_postfork_child(malloc_mutex_t *mutex)
{
malloc_mutex_postfork_child(tsdn_t *tsdn, malloc_mutex_t *mutex) {
#ifdef JEMALLOC_MUTEX_INIT_CB
malloc_mutex_unlock(mutex);
malloc_mutex_unlock(tsdn, mutex);
#else
if (malloc_mutex_init(mutex)) {
if (malloc_mutex_init(mutex, mutex->witness.name,
mutex->witness.rank, mutex->lock_order)) {
malloc_printf("<jemalloc>: Error re-initializing mutex in "
"child\n");
if (opt_abort)
if (opt_abort) {
abort();
}
}
#endif
}
bool
mutex_boot(void)
{
malloc_mutex_boot(void) {
#ifdef JEMALLOC_MUTEX_INIT_CB
postpone_init = false;
while (postponed_mutexes != NULL) {
if (_pthread_mutex_init_calloc_cb(&postponed_mutexes->lock,
base_calloc) != 0)
return (true);
bootstrap_calloc) != 0) {
return true;
}
postponed_mutexes = postponed_mutexes->postponed_next;
}
#endif
return (false);
return false;
}

2488
deps/jemalloc/src/prof.c vendored
View File

File diff suppressed because it is too large Load Diff

391
deps/jemalloc/src/rtree.c vendored
View File

@@ -1,105 +1,320 @@
#define JEMALLOC_RTREE_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_RTREE_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
rtree_t *
rtree_new(unsigned bits, rtree_alloc_t *alloc, rtree_dalloc_t *dalloc)
{
rtree_t *ret;
unsigned bits_per_level, bits_in_leaf, height, i;
#include "jemalloc/internal/assert.h"
#include "jemalloc/internal/mutex.h"
assert(bits > 0 && bits <= (sizeof(uintptr_t) << 3));
bits_per_level = ffs(pow2_ceil((RTREE_NODESIZE / sizeof(void *)))) - 1;
bits_in_leaf = ffs(pow2_ceil((RTREE_NODESIZE / sizeof(uint8_t)))) - 1;
if (bits > bits_in_leaf) {
height = 1 + (bits - bits_in_leaf) / bits_per_level;
if ((height-1) * bits_per_level + bits_in_leaf != bits)
height++;
} else {
height = 1;
/*
* Only the most significant bits of keys passed to rtree_{read,write}() are
* used.
*/
bool
rtree_new(rtree_t *rtree, bool zeroed) {
#ifdef JEMALLOC_JET
if (!zeroed) {
memset(rtree, 0, sizeof(rtree_t)); /* Clear root. */
}
assert((height-1) * bits_per_level + bits_in_leaf >= bits);
#else
assert(zeroed);
#endif
ret = (rtree_t*)alloc(offsetof(rtree_t, level2bits) +
(sizeof(unsigned) * height));
if (ret == NULL)
return (NULL);
memset(ret, 0, offsetof(rtree_t, level2bits) + (sizeof(unsigned) *
height));
ret->alloc = alloc;
ret->dalloc = dalloc;
if (malloc_mutex_init(&ret->mutex)) {
if (dalloc != NULL)
dalloc(ret);
return (NULL);
if (malloc_mutex_init(&rtree->init_lock, "rtree", WITNESS_RANK_RTREE,
malloc_mutex_rank_exclusive)) {
return true;
}
ret->height = height;
if (height > 1) {
if ((height-1) * bits_per_level + bits_in_leaf > bits) {
ret->level2bits[0] = (bits - bits_in_leaf) %
bits_per_level;
} else
ret->level2bits[0] = bits_per_level;
for (i = 1; i < height-1; i++)
ret->level2bits[i] = bits_per_level;
ret->level2bits[height-1] = bits_in_leaf;
} else
ret->level2bits[0] = bits;
ret->root = (void**)alloc(sizeof(void *) << ret->level2bits[0]);
if (ret->root == NULL) {
if (dalloc != NULL)
dalloc(ret);
return (NULL);
}
memset(ret->root, 0, sizeof(void *) << ret->level2bits[0]);
return (ret);
return false;
}
static rtree_node_elm_t *
rtree_node_alloc_impl(tsdn_t *tsdn, rtree_t *rtree, size_t nelms) {
return (rtree_node_elm_t *)base_alloc(tsdn, b0get(), nelms *
sizeof(rtree_node_elm_t), CACHELINE);
}
rtree_node_alloc_t *JET_MUTABLE rtree_node_alloc = rtree_node_alloc_impl;
static void
rtree_delete_subtree(rtree_t *rtree, void **node, unsigned level)
{
rtree_node_dalloc_impl(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *node) {
/* Nodes are never deleted during normal operation. */
not_reached();
}
UNUSED rtree_node_dalloc_t *JET_MUTABLE rtree_node_dalloc =
rtree_node_dalloc_impl;
if (level < rtree->height - 1) {
size_t nchildren, i;
static rtree_leaf_elm_t *
rtree_leaf_alloc_impl(tsdn_t *tsdn, rtree_t *rtree, size_t nelms) {
return (rtree_leaf_elm_t *)base_alloc(tsdn, b0get(), nelms *
sizeof(rtree_leaf_elm_t), CACHELINE);
}
rtree_leaf_alloc_t *JET_MUTABLE rtree_leaf_alloc = rtree_leaf_alloc_impl;
nchildren = ZU(1) << rtree->level2bits[level];
for (i = 0; i < nchildren; i++) {
void **child = (void **)node[i];
if (child != NULL)
rtree_delete_subtree(rtree, child, level + 1);
static void
rtree_leaf_dalloc_impl(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *leaf) {
/* Leaves are never deleted during normal operation. */
not_reached();
}
UNUSED rtree_leaf_dalloc_t *JET_MUTABLE rtree_leaf_dalloc =
rtree_leaf_dalloc_impl;
#ifdef JEMALLOC_JET
# if RTREE_HEIGHT > 1
static void
rtree_delete_subtree(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *subtree,
unsigned level) {
size_t nchildren = ZU(1) << rtree_levels[level].bits;
if (level + 2 < RTREE_HEIGHT) {
for (size_t i = 0; i < nchildren; i++) {
rtree_node_elm_t *node =
(rtree_node_elm_t *)atomic_load_p(&subtree[i].child,
ATOMIC_RELAXED);
if (node != NULL) {
rtree_delete_subtree(tsdn, rtree, node, level +
1);
}
}
} else {
for (size_t i = 0; i < nchildren; i++) {
rtree_leaf_elm_t *leaf =
(rtree_leaf_elm_t *)atomic_load_p(&subtree[i].child,
ATOMIC_RELAXED);
if (leaf != NULL) {
rtree_leaf_dalloc(tsdn, rtree, leaf);
}
}
}
rtree->dalloc(node);
if (subtree != rtree->root) {
rtree_node_dalloc(tsdn, rtree, subtree);
}
}
# endif
void
rtree_delete(tsdn_t *tsdn, rtree_t *rtree) {
# if RTREE_HEIGHT > 1
rtree_delete_subtree(tsdn, rtree, rtree->root, 0);
# endif
}
#endif
static rtree_node_elm_t *
rtree_node_init(tsdn_t *tsdn, rtree_t *rtree, unsigned level,
atomic_p_t *elmp) {
malloc_mutex_lock(tsdn, &rtree->init_lock);
/*
* If *elmp is non-null, then it was initialized with the init lock
* held, so we can get by with 'relaxed' here.
*/
rtree_node_elm_t *node = atomic_load_p(elmp, ATOMIC_RELAXED);
if (node == NULL) {
node = rtree_node_alloc(tsdn, rtree, ZU(1) <<
rtree_levels[level].bits);
if (node == NULL) {
malloc_mutex_unlock(tsdn, &rtree->init_lock);
return NULL;
}
/*
* Even though we hold the lock, a later reader might not; we
* need release semantics.
*/
atomic_store_p(elmp, node, ATOMIC_RELEASE);
}
malloc_mutex_unlock(tsdn, &rtree->init_lock);
return node;
}
static rtree_leaf_elm_t *
rtree_leaf_init(tsdn_t *tsdn, rtree_t *rtree, atomic_p_t *elmp) {
malloc_mutex_lock(tsdn, &rtree->init_lock);
/*
* If *elmp is non-null, then it was initialized with the init lock
* held, so we can get by with 'relaxed' here.
*/
rtree_leaf_elm_t *leaf = atomic_load_p(elmp, ATOMIC_RELAXED);
if (leaf == NULL) {
leaf = rtree_leaf_alloc(tsdn, rtree, ZU(1) <<
rtree_levels[RTREE_HEIGHT-1].bits);
if (leaf == NULL) {
malloc_mutex_unlock(tsdn, &rtree->init_lock);
return NULL;
}
/*
* Even though we hold the lock, a later reader might not; we
* need release semantics.
*/
atomic_store_p(elmp, leaf, ATOMIC_RELEASE);
}
malloc_mutex_unlock(tsdn, &rtree->init_lock);
return leaf;
}
static bool
rtree_node_valid(rtree_node_elm_t *node) {
return ((uintptr_t)node != (uintptr_t)0);
}
static bool
rtree_leaf_valid(rtree_leaf_elm_t *leaf) {
return ((uintptr_t)leaf != (uintptr_t)0);
}
static rtree_node_elm_t *
rtree_child_node_tryread(rtree_node_elm_t *elm, bool dependent) {
rtree_node_elm_t *node;
if (dependent) {
node = (rtree_node_elm_t *)atomic_load_p(&elm->child,
ATOMIC_RELAXED);
} else {
node = (rtree_node_elm_t *)atomic_load_p(&elm->child,
ATOMIC_ACQUIRE);
}
assert(!dependent || node != NULL);
return node;
}
static rtree_node_elm_t *
rtree_child_node_read(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *elm,
unsigned level, bool dependent) {
rtree_node_elm_t *node;
node = rtree_child_node_tryread(elm, dependent);
if (!dependent && unlikely(!rtree_node_valid(node))) {
node = rtree_node_init(tsdn, rtree, level + 1, &elm->child);
}
assert(!dependent || node != NULL);
return node;
}
static rtree_leaf_elm_t *
rtree_child_leaf_tryread(rtree_node_elm_t *elm, bool dependent) {
rtree_leaf_elm_t *leaf;
if (dependent) {
leaf = (rtree_leaf_elm_t *)atomic_load_p(&elm->child,
ATOMIC_RELAXED);
} else {
leaf = (rtree_leaf_elm_t *)atomic_load_p(&elm->child,
ATOMIC_ACQUIRE);
}
assert(!dependent || leaf != NULL);
return leaf;
}
static rtree_leaf_elm_t *
rtree_child_leaf_read(tsdn_t *tsdn, rtree_t *rtree, rtree_node_elm_t *elm,
unsigned level, bool dependent) {
rtree_leaf_elm_t *leaf;
leaf = rtree_child_leaf_tryread(elm, dependent);
if (!dependent && unlikely(!rtree_leaf_valid(leaf))) {
leaf = rtree_leaf_init(tsdn, rtree, &elm->child);
}
assert(!dependent || leaf != NULL);
return leaf;
}
rtree_leaf_elm_t *
rtree_leaf_elm_lookup_hard(tsdn_t *tsdn, rtree_t *rtree, rtree_ctx_t *rtree_ctx,
uintptr_t key, bool dependent, bool init_missing) {
rtree_node_elm_t *node;
rtree_leaf_elm_t *leaf;
#if RTREE_HEIGHT > 1
node = rtree->root;
#else
leaf = rtree->root;
#endif
if (config_debug) {
uintptr_t leafkey = rtree_leafkey(key);
for (unsigned i = 0; i < RTREE_CTX_NCACHE; i++) {
assert(rtree_ctx->cache[i].leafkey != leafkey);
}
for (unsigned i = 0; i < RTREE_CTX_NCACHE_L2; i++) {
assert(rtree_ctx->l2_cache[i].leafkey != leafkey);
}
}
#define RTREE_GET_CHILD(level) { \
assert(level < RTREE_HEIGHT-1); \
if (level != 0 && !dependent && \
unlikely(!rtree_node_valid(node))) { \
return NULL; \
} \
uintptr_t subkey = rtree_subkey(key, level); \
if (level + 2 < RTREE_HEIGHT) { \
node = init_missing ? \
rtree_child_node_read(tsdn, rtree, \
&node[subkey], level, dependent) : \
rtree_child_node_tryread(&node[subkey], \
dependent); \
} else { \
leaf = init_missing ? \
rtree_child_leaf_read(tsdn, rtree, \
&node[subkey], level, dependent) : \
rtree_child_leaf_tryread(&node[subkey], \
dependent); \
} \
}
/*
* Cache replacement upon hard lookup (i.e. L1 & L2 rtree cache miss):
* (1) evict last entry in L2 cache; (2) move the collision slot from L1
* cache down to L2; and 3) fill L1.
*/
#define RTREE_GET_LEAF(level) { \
assert(level == RTREE_HEIGHT-1); \
if (!dependent && unlikely(!rtree_leaf_valid(leaf))) { \
return NULL; \
} \
if (RTREE_CTX_NCACHE_L2 > 1) { \
memmove(&rtree_ctx->l2_cache[1], \
&rtree_ctx->l2_cache[0], \
sizeof(rtree_ctx_cache_elm_t) * \
(RTREE_CTX_NCACHE_L2 - 1)); \
} \
size_t slot = rtree_cache_direct_map(key); \
rtree_ctx->l2_cache[0].leafkey = \
rtree_ctx->cache[slot].leafkey; \
rtree_ctx->l2_cache[0].leaf = \
rtree_ctx->cache[slot].leaf; \
uintptr_t leafkey = rtree_leafkey(key); \
rtree_ctx->cache[slot].leafkey = leafkey; \
rtree_ctx->cache[slot].leaf = leaf; \
uintptr_t subkey = rtree_subkey(key, level); \
return &leaf[subkey]; \
}
if (RTREE_HEIGHT > 1) {
RTREE_GET_CHILD(0)
}
if (RTREE_HEIGHT > 2) {
RTREE_GET_CHILD(1)
}
if (RTREE_HEIGHT > 3) {
for (unsigned i = 2; i < RTREE_HEIGHT-1; i++) {
RTREE_GET_CHILD(i)
}
}
RTREE_GET_LEAF(RTREE_HEIGHT-1)
#undef RTREE_GET_CHILD
#undef RTREE_GET_LEAF
not_reached();
}
void
rtree_delete(rtree_t *rtree)
{
rtree_delete_subtree(rtree, rtree->root, 0);
rtree->dalloc(rtree);
}
void
rtree_prefork(rtree_t *rtree)
{
malloc_mutex_prefork(&rtree->mutex);
}
void
rtree_postfork_parent(rtree_t *rtree)
{
malloc_mutex_postfork_parent(&rtree->mutex);
}
void
rtree_postfork_child(rtree_t *rtree)
{
malloc_mutex_postfork_child(&rtree->mutex);
rtree_ctx_data_init(rtree_ctx_t *ctx) {
for (unsigned i = 0; i < RTREE_CTX_NCACHE; i++) {
rtree_ctx_cache_elm_t *cache = &ctx->cache[i];
cache->leafkey = RTREE_LEAFKEY_INVALID;
cache->leaf = NULL;
}
for (unsigned i = 0; i < RTREE_CTX_NCACHE_L2; i++) {
rtree_ctx_cache_elm_t *cache = &ctx->l2_cache[i];
cache->leafkey = RTREE_LEAFKEY_INVALID;
cache->leaf = NULL;
}
}

1622
deps/jemalloc/src/stats.c vendored
View File

File diff suppressed because it is too large Load Diff

785
deps/jemalloc/src/tcache.c vendored
View File

@@ -1,131 +1,153 @@
#define JEMALLOC_TCACHE_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_TCACHE_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
#include "jemalloc/internal/assert.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/size_classes.h"
/******************************************************************************/
/* Data. */
malloc_tsd_data(, tcache, tcache_t *, NULL)
malloc_tsd_data(, tcache_enabled, tcache_enabled_t, tcache_enabled_default)
bool opt_tcache = true;
ssize_t opt_lg_tcache_max = LG_TCACHE_MAXCLASS_DEFAULT;
tcache_bin_info_t *tcache_bin_info;
static unsigned stack_nelms; /* Total stack elms per tcache. */
size_t nhbins;
unsigned nhbins;
size_t tcache_maxclass;
tcaches_t *tcaches;
/* Index of first element within tcaches that has never been used. */
static unsigned tcaches_past;
/* Head of singly linked list tracking available tcaches elements. */
static tcaches_t *tcaches_avail;
/* Protects tcaches{,_past,_avail}. */
static malloc_mutex_t tcaches_mtx;
/******************************************************************************/
size_t tcache_salloc(const void *ptr)
{
return (arena_salloc(ptr, false));
size_t
tcache_salloc(tsdn_t *tsdn, const void *ptr) {
return arena_salloc(tsdn, ptr);
}
void
tcache_event_hard(tcache_t *tcache)
{
size_t binind = tcache->next_gc_bin;
tcache_bin_t *tbin = &tcache->tbins[binind];
tcache_bin_info_t *tbin_info = &tcache_bin_info[binind];
tcache_event_hard(tsd_t *tsd, tcache_t *tcache) {
szind_t binind = tcache->next_gc_bin;
tcache_bin_t *tbin;
if (binind < NBINS) {
tbin = tcache_small_bin_get(tcache, binind);
} else {
tbin = tcache_large_bin_get(tcache, binind);
}
if (tbin->low_water > 0) {
/*
* Flush (ceiling) 3/4 of the objects below the low water mark.
*/
if (binind < NBINS) {
tcache_bin_flush_small(tbin, binind, tbin->ncached -
tbin->low_water + (tbin->low_water >> 2), tcache);
tcache_bin_flush_small(tsd, tcache, tbin, binind,
tbin->ncached - tbin->low_water + (tbin->low_water
>> 2));
/*
* Reduce fill count by 2X. Limit lg_fill_div such that
* the fill count is always at least 1.
*/
tcache_bin_info_t *tbin_info = &tcache_bin_info[binind];
if ((tbin_info->ncached_max >>
(tcache->lg_fill_div[binind] + 1)) >= 1) {
tcache->lg_fill_div[binind]++;
}
} else {
tcache_bin_flush_large(tbin, binind, tbin->ncached -
tbin->low_water + (tbin->low_water >> 2), tcache);
tcache_bin_flush_large(tsd, tbin, binind, tbin->ncached
- tbin->low_water + (tbin->low_water >> 2), tcache);
}
/*
* Reduce fill count by 2X. Limit lg_fill_div such that the
* fill count is always at least 1.
*/
if ((tbin_info->ncached_max >> (tbin->lg_fill_div+1)) >= 1)
tbin->lg_fill_div++;
} else if (tbin->low_water < 0) {
/*
* Increase fill count by 2X. Make sure lg_fill_div stays
* greater than 0.
* Increase fill count by 2X for small bins. Make sure
* lg_fill_div stays greater than 0.
*/
if (tbin->lg_fill_div > 1)
tbin->lg_fill_div--;
if (binind < NBINS && tcache->lg_fill_div[binind] > 1) {
tcache->lg_fill_div[binind]--;
}
}
tbin->low_water = tbin->ncached;
tcache->next_gc_bin++;
if (tcache->next_gc_bin == nhbins)
if (tcache->next_gc_bin == nhbins) {
tcache->next_gc_bin = 0;
tcache->ev_cnt = 0;
}
}
void *
tcache_alloc_small_hard(tcache_t *tcache, tcache_bin_t *tbin, size_t binind)
{
tcache_alloc_small_hard(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache,
tcache_bin_t *tbin, szind_t binind, bool *tcache_success) {
void *ret;
arena_tcache_fill_small(tcache->arena, tbin, binind,
assert(tcache->arena != NULL);
arena_tcache_fill_small(tsdn, arena, tcache, tbin, binind,
config_prof ? tcache->prof_accumbytes : 0);
if (config_prof)
if (config_prof) {
tcache->prof_accumbytes = 0;
ret = tcache_alloc_easy(tbin);
}
ret = tcache_alloc_easy(tbin, tcache_success);
return (ret);
return ret;
}
void
tcache_bin_flush_small(tcache_bin_t *tbin, size_t binind, unsigned rem,
tcache_t *tcache)
{
void *ptr;
unsigned i, nflush, ndeferred;
tcache_bin_flush_small(tsd_t *tsd, tcache_t *tcache, tcache_bin_t *tbin,
szind_t binind, unsigned rem) {
bool merged_stats = false;
assert(binind < NBINS);
assert(rem <= tbin->ncached);
for (nflush = tbin->ncached - rem; nflush > 0; nflush = ndeferred) {
/* Lock the arena bin associated with the first object. */
arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(
tbin->avail[0]);
arena_t *arena = chunk->arena;
arena_bin_t *bin = &arena->bins[binind];
arena_t *arena = tcache->arena;
assert(arena != NULL);
unsigned nflush = tbin->ncached - rem;
VARIABLE_ARRAY(extent_t *, item_extent, nflush);
/* Look up extent once per item. */
for (unsigned i = 0 ; i < nflush; i++) {
item_extent[i] = iealloc(tsd_tsdn(tsd), *(tbin->avail - 1 - i));
}
if (config_prof && arena == tcache->arena) {
if (arena_prof_accum(arena, tcache->prof_accumbytes))
prof_idump();
while (nflush > 0) {
/* Lock the arena bin associated with the first object. */
extent_t *extent = item_extent[0];
arena_t *bin_arena = extent_arena_get(extent);
arena_bin_t *bin = &bin_arena->bins[binind];
if (config_prof && bin_arena == arena) {
if (arena_prof_accum(tsd_tsdn(tsd), arena,
tcache->prof_accumbytes)) {
prof_idump(tsd_tsdn(tsd));
}
tcache->prof_accumbytes = 0;
}
malloc_mutex_lock(&bin->lock);
if (config_stats && arena == tcache->arena) {
assert(merged_stats == false);
malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock);
if (config_stats && bin_arena == arena) {
assert(!merged_stats);
merged_stats = true;
bin->stats.nflushes++;
bin->stats.nrequests += tbin->tstats.nrequests;
tbin->tstats.nrequests = 0;
}
ndeferred = 0;
for (i = 0; i < nflush; i++) {
ptr = tbin->avail[i];
assert(ptr != NULL);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
if (chunk->arena == arena) {
size_t pageind = ((uintptr_t)ptr -
(uintptr_t)chunk) >> LG_PAGE;
arena_chunk_map_t *mapelm =
arena_mapp_get(chunk, pageind);
if (config_fill && opt_junk) {
arena_alloc_junk_small(ptr,
&arena_bin_info[binind], true);
}
arena_dalloc_bin_locked(arena, chunk, ptr,
mapelm);
unsigned ndeferred = 0;
for (unsigned i = 0; i < nflush; i++) {
void *ptr = *(tbin->avail - 1 - i);
extent = item_extent[i];
assert(ptr != NULL && extent != NULL);
if (extent_arena_get(extent) == bin_arena) {
arena_dalloc_bin_junked_locked(tsd_tsdn(tsd),
bin_arena, extent, ptr);
} else {
/*
* This object was allocated via a different
@@ -133,276 +155,369 @@ tcache_bin_flush_small(tcache_bin_t *tbin, size_t binind, unsigned rem,
* locked. Stash the object, so that it can be
* handled in a future pass.
*/
tbin->avail[ndeferred] = ptr;
*(tbin->avail - 1 - ndeferred) = ptr;
item_extent[ndeferred] = extent;
ndeferred++;
}
}
malloc_mutex_unlock(&bin->lock);
malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock);
arena_decay_ticks(tsd_tsdn(tsd), bin_arena, nflush - ndeferred);
nflush = ndeferred;
}
if (config_stats && merged_stats == false) {
if (config_stats && !merged_stats) {
/*
* The flush loop didn't happen to flush to this thread's
* arena, so the stats didn't get merged. Manually do so now.
*/
arena_bin_t *bin = &tcache->arena->bins[binind];
malloc_mutex_lock(&bin->lock);
arena_bin_t *bin = &arena->bins[binind];
malloc_mutex_lock(tsd_tsdn(tsd), &bin->lock);
bin->stats.nflushes++;
bin->stats.nrequests += tbin->tstats.nrequests;
tbin->tstats.nrequests = 0;
malloc_mutex_unlock(&bin->lock);
malloc_mutex_unlock(tsd_tsdn(tsd), &bin->lock);
}
memmove(tbin->avail, &tbin->avail[tbin->ncached - rem],
rem * sizeof(void *));
memmove(tbin->avail - rem, tbin->avail - tbin->ncached, rem *
sizeof(void *));
tbin->ncached = rem;
if ((int)tbin->ncached < tbin->low_water)
if ((low_water_t)tbin->ncached < tbin->low_water) {
tbin->low_water = tbin->ncached;
}
}
void
tcache_bin_flush_large(tcache_bin_t *tbin, size_t binind, unsigned rem,
tcache_t *tcache)
{
void *ptr;
unsigned i, nflush, ndeferred;
tcache_bin_flush_large(tsd_t *tsd, tcache_bin_t *tbin, szind_t binind,
unsigned rem, tcache_t *tcache) {
bool merged_stats = false;
assert(binind < nhbins);
assert(rem <= tbin->ncached);
for (nflush = tbin->ncached - rem; nflush > 0; nflush = ndeferred) {
arena_t *arena = tcache->arena;
assert(arena != NULL);
unsigned nflush = tbin->ncached - rem;
VARIABLE_ARRAY(extent_t *, item_extent, nflush);
/* Look up extent once per item. */
for (unsigned i = 0 ; i < nflush; i++) {
item_extent[i] = iealloc(tsd_tsdn(tsd), *(tbin->avail - 1 - i));
}
while (nflush > 0) {
/* Lock the arena associated with the first object. */
arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(
tbin->avail[0]);
arena_t *arena = chunk->arena;
extent_t *extent = item_extent[0];
arena_t *locked_arena = extent_arena_get(extent);
UNUSED bool idump;
if (config_prof)
if (config_prof) {
idump = false;
malloc_mutex_lock(&arena->lock);
if ((config_prof || config_stats) && arena == tcache->arena) {
}
malloc_mutex_lock(tsd_tsdn(tsd), &locked_arena->large_mtx);
for (unsigned i = 0; i < nflush; i++) {
void *ptr = *(tbin->avail - 1 - i);
assert(ptr != NULL);
extent = item_extent[i];
if (extent_arena_get(extent) == locked_arena) {
large_dalloc_prep_junked_locked(tsd_tsdn(tsd),
extent);
}
}
if ((config_prof || config_stats) && locked_arena == arena) {
if (config_prof) {
idump = arena_prof_accum_locked(arena,
idump = arena_prof_accum(tsd_tsdn(tsd), arena,
tcache->prof_accumbytes);
tcache->prof_accumbytes = 0;
}
if (config_stats) {
merged_stats = true;
arena->stats.nrequests_large +=
tbin->tstats.nrequests;
arena->stats.lstats[binind - NBINS].nrequests +=
tbin->tstats.nrequests;
arena_stats_large_nrequests_add(tsd_tsdn(tsd),
&arena->stats, binind,
tbin->tstats.nrequests);
tbin->tstats.nrequests = 0;
}
}
ndeferred = 0;
for (i = 0; i < nflush; i++) {
ptr = tbin->avail[i];
assert(ptr != NULL);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
if (chunk->arena == arena)
arena_dalloc_large_locked(arena, chunk, ptr);
else {
malloc_mutex_unlock(tsd_tsdn(tsd), &locked_arena->large_mtx);
unsigned ndeferred = 0;
for (unsigned i = 0; i < nflush; i++) {
void *ptr = *(tbin->avail - 1 - i);
extent = item_extent[i];
assert(ptr != NULL && extent != NULL);
if (extent_arena_get(extent) == locked_arena) {
large_dalloc_finish(tsd_tsdn(tsd), extent);
} else {
/*
* This object was allocated via a different
* arena than the one that is currently locked.
* Stash the object, so that it can be handled
* in a future pass.
*/
tbin->avail[ndeferred] = ptr;
*(tbin->avail - 1 - ndeferred) = ptr;
item_extent[ndeferred] = extent;
ndeferred++;
}
}
malloc_mutex_unlock(&arena->lock);
if (config_prof && idump)
prof_idump();
if (config_prof && idump) {
prof_idump(tsd_tsdn(tsd));
}
arena_decay_ticks(tsd_tsdn(tsd), locked_arena, nflush -
ndeferred);
nflush = ndeferred;
}
if (config_stats && merged_stats == false) {
if (config_stats && !merged_stats) {
/*
* The flush loop didn't happen to flush to this thread's
* arena, so the stats didn't get merged. Manually do so now.
*/
arena_t *arena = tcache->arena;
malloc_mutex_lock(&arena->lock);
arena->stats.nrequests_large += tbin->tstats.nrequests;
arena->stats.lstats[binind - NBINS].nrequests +=
tbin->tstats.nrequests;
arena_stats_large_nrequests_add(tsd_tsdn(tsd), &arena->stats,
binind, tbin->tstats.nrequests);
tbin->tstats.nrequests = 0;
malloc_mutex_unlock(&arena->lock);
}
memmove(tbin->avail, &tbin->avail[tbin->ncached - rem],
rem * sizeof(void *));
memmove(tbin->avail - rem, tbin->avail - tbin->ncached, rem *
sizeof(void *));
tbin->ncached = rem;
if ((int)tbin->ncached < tbin->low_water)
if ((low_water_t)tbin->ncached < tbin->low_water) {
tbin->low_water = tbin->ncached;
}
}
void
tcache_arena_associate(tcache_t *tcache, arena_t *arena)
{
tcache_arena_associate(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena) {
assert(tcache->arena == NULL);
tcache->arena = arena;
if (config_stats) {
/* Link into list of extant tcaches. */
malloc_mutex_lock(&arena->lock);
malloc_mutex_lock(tsdn, &arena->tcache_ql_mtx);
ql_elm_new(tcache, link);
ql_tail_insert(&arena->tcache_ql, tcache, link);
malloc_mutex_unlock(&arena->lock);
malloc_mutex_unlock(tsdn, &arena->tcache_ql_mtx);
}
tcache->arena = arena;
}
static void
tcache_arena_dissociate(tsdn_t *tsdn, tcache_t *tcache) {
arena_t *arena = tcache->arena;
assert(arena != NULL);
if (config_stats) {
/* Unlink from list of extant tcaches. */
malloc_mutex_lock(tsdn, &arena->tcache_ql_mtx);
if (config_debug) {
bool in_ql = false;
tcache_t *iter;
ql_foreach(iter, &arena->tcache_ql, link) {
if (iter == tcache) {
in_ql = true;
break;
}
}
assert(in_ql);
}
ql_remove(&arena->tcache_ql, tcache, link);
tcache_stats_merge(tsdn, tcache, arena);
malloc_mutex_unlock(tsdn, &arena->tcache_ql_mtx);
}
tcache->arena = NULL;
}
void
tcache_arena_dissociate(tcache_t *tcache)
{
if (config_stats) {
/* Unlink from list of extant tcaches. */
malloc_mutex_lock(&tcache->arena->lock);
ql_remove(&tcache->arena->tcache_ql, tcache, link);
tcache_stats_merge(tcache, tcache->arena);
malloc_mutex_unlock(&tcache->arena->lock);
}
tcache_arena_reassociate(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena) {
tcache_arena_dissociate(tsdn, tcache);
tcache_arena_associate(tsdn, tcache, arena);
}
bool
tsd_tcache_enabled_data_init(tsd_t *tsd) {
/* Called upon tsd initialization. */
tsd_tcache_enabled_set(tsd, opt_tcache);
tsd_slow_update(tsd);
if (opt_tcache) {
/* Trigger tcache init. */
tsd_tcache_data_init(tsd);
}
return false;
}
/* Initialize auto tcache (embedded in TSD). */
static void
tcache_init(tsd_t *tsd, tcache_t *tcache, void *avail_stack) {
memset(&tcache->link, 0, sizeof(ql_elm(tcache_t)));
tcache->prof_accumbytes = 0;
tcache->next_gc_bin = 0;
tcache->arena = NULL;
ticker_init(&tcache->gc_ticker, TCACHE_GC_INCR);
size_t stack_offset = 0;
assert((TCACHE_NSLOTS_SMALL_MAX & 1U) == 0);
memset(tcache->tbins_small, 0, sizeof(tcache_bin_t) * NBINS);
memset(tcache->tbins_large, 0, sizeof(tcache_bin_t) * (nhbins - NBINS));
unsigned i = 0;
for (; i < NBINS; i++) {
tcache->lg_fill_div[i] = 1;
stack_offset += tcache_bin_info[i].ncached_max * sizeof(void *);
/*
* avail points past the available space. Allocations will
* access the slots toward higher addresses (for the benefit of
* prefetch).
*/
tcache_small_bin_get(tcache, i)->avail =
(void **)((uintptr_t)avail_stack + (uintptr_t)stack_offset);
}
for (; i < nhbins; i++) {
stack_offset += tcache_bin_info[i].ncached_max * sizeof(void *);
tcache_large_bin_get(tcache, i)->avail =
(void **)((uintptr_t)avail_stack + (uintptr_t)stack_offset);
}
assert(stack_offset == stack_nelms * sizeof(void *));
}
/* Initialize auto tcache (embedded in TSD). */
bool
tsd_tcache_data_init(tsd_t *tsd) {
tcache_t *tcache = tsd_tcachep_get_unsafe(tsd);
assert(tcache_small_bin_get(tcache, 0)->avail == NULL);
size_t size = stack_nelms * sizeof(void *);
/* Avoid false cacheline sharing. */
size = sz_sa2u(size, CACHELINE);
void *avail_array = ipallocztm(tsd_tsdn(tsd), size, CACHELINE, true,
NULL, true, arena_get(TSDN_NULL, 0, true));
if (avail_array == NULL) {
return true;
}
tcache_init(tsd, tcache, avail_array);
/*
* Initialization is a bit tricky here. After malloc init is done, all
* threads can rely on arena_choose and associate tcache accordingly.
* However, the thread that does actual malloc bootstrapping relies on
* functional tsd, and it can only rely on a0. In that case, we
* associate its tcache to a0 temporarily, and later on
* arena_choose_hard() will re-associate properly.
*/
tcache->arena = NULL;
arena_t *arena;
if (!malloc_initialized()) {
/* If in initialization, assign to a0. */
arena = arena_get(tsd_tsdn(tsd), 0, false);
tcache_arena_associate(tsd_tsdn(tsd), tcache, arena);
} else {
arena = arena_choose(tsd, NULL);
/* This may happen if thread.tcache.enabled is used. */
if (tcache->arena == NULL) {
tcache_arena_associate(tsd_tsdn(tsd), tcache, arena);
}
}
assert(arena == tcache->arena);
return false;
}
/* Created manual tcache for tcache.create mallctl. */
tcache_t *
tcache_create(arena_t *arena)
{
tcache_create_explicit(tsd_t *tsd) {
tcache_t *tcache;
size_t size, stack_offset;
unsigned i;
size = offsetof(tcache_t, tbins) + (sizeof(tcache_bin_t) * nhbins);
size = sizeof(tcache_t);
/* Naturally align the pointer stacks. */
size = PTR_CEILING(size);
stack_offset = size;
size += stack_nelms * sizeof(void *);
/*
* Round up to the nearest multiple of the cacheline size, in order to
* avoid the possibility of false cacheline sharing.
*
* That this works relies on the same logic as in ipalloc(), but we
* cannot directly call ipalloc() here due to tcache bootstrapping
* issues.
*/
size = (size + CACHELINE_MASK) & (-CACHELINE);
/* Avoid false cacheline sharing. */
size = sz_sa2u(size, CACHELINE);
if (size <= SMALL_MAXCLASS)
tcache = (tcache_t *)arena_malloc_small(arena, size, true);
else if (size <= tcache_maxclass)
tcache = (tcache_t *)arena_malloc_large(arena, size, true);
else
tcache = (tcache_t *)icalloct(size, false, arena);
if (tcache == NULL)
return (NULL);
tcache_arena_associate(tcache, arena);
assert((TCACHE_NSLOTS_SMALL_MAX & 1U) == 0);
for (i = 0; i < nhbins; i++) {
tcache->tbins[i].lg_fill_div = 1;
tcache->tbins[i].avail = (void **)((uintptr_t)tcache +
(uintptr_t)stack_offset);
stack_offset += tcache_bin_info[i].ncached_max * sizeof(void *);
tcache = ipallocztm(tsd_tsdn(tsd), size, CACHELINE, true, NULL, true,
arena_get(TSDN_NULL, 0, true));
if (tcache == NULL) {
return NULL;
}
tcache_tsd_set(&tcache);
tcache_init(tsd, tcache,
(void *)((uintptr_t)tcache + (uintptr_t)stack_offset));
tcache_arena_associate(tsd_tsdn(tsd), tcache, arena_ichoose(tsd, NULL));
return (tcache);
return tcache;
}
void
tcache_destroy(tcache_t *tcache)
{
unsigned i;
size_t tcache_size;
static void
tcache_flush_cache(tsd_t *tsd, tcache_t *tcache) {
assert(tcache->arena != NULL);
tcache_arena_dissociate(tcache);
for (unsigned i = 0; i < NBINS; i++) {
tcache_bin_t *tbin = tcache_small_bin_get(tcache, i);
tcache_bin_flush_small(tsd, tcache, tbin, i, 0);
for (i = 0; i < NBINS; i++) {
tcache_bin_t *tbin = &tcache->tbins[i];
tcache_bin_flush_small(tbin, i, 0, tcache);
if (config_stats && tbin->tstats.nrequests != 0) {
arena_t *arena = tcache->arena;
arena_bin_t *bin = &arena->bins[i];
malloc_mutex_lock(&bin->lock);
bin->stats.nrequests += tbin->tstats.nrequests;
malloc_mutex_unlock(&bin->lock);
if (config_stats) {
assert(tbin->tstats.nrequests == 0);
}
}
for (unsigned i = NBINS; i < nhbins; i++) {
tcache_bin_t *tbin = tcache_large_bin_get(tcache, i);
tcache_bin_flush_large(tsd, tbin, i, 0, tcache);
for (; i < nhbins; i++) {
tcache_bin_t *tbin = &tcache->tbins[i];
tcache_bin_flush_large(tbin, i, 0, tcache);
if (config_stats && tbin->tstats.nrequests != 0) {
arena_t *arena = tcache->arena;
malloc_mutex_lock(&arena->lock);
arena->stats.nrequests_large += tbin->tstats.nrequests;
arena->stats.lstats[i - NBINS].nrequests +=
tbin->tstats.nrequests;
malloc_mutex_unlock(&arena->lock);
if (config_stats) {
assert(tbin->tstats.nrequests == 0);
}
}
if (config_prof && tcache->prof_accumbytes > 0 &&
arena_prof_accum(tcache->arena, tcache->prof_accumbytes))
prof_idump();
tcache_size = arena_salloc(tcache, false);
if (tcache_size <= SMALL_MAXCLASS) {
arena_chunk_t *chunk = CHUNK_ADDR2BASE(tcache);
arena_t *arena = chunk->arena;
size_t pageind = ((uintptr_t)tcache - (uintptr_t)chunk) >>
LG_PAGE;
arena_chunk_map_t *mapelm = arena_mapp_get(chunk, pageind);
arena_dalloc_bin(arena, chunk, tcache, pageind, mapelm);
} else if (tcache_size <= tcache_maxclass) {
arena_chunk_t *chunk = CHUNK_ADDR2BASE(tcache);
arena_t *arena = chunk->arena;
arena_dalloc_large(arena, chunk, tcache);
} else
idalloct(tcache, false);
}
void
tcache_thread_cleanup(void *arg)
{
tcache_t *tcache = *(tcache_t **)arg;
if (tcache == TCACHE_STATE_DISABLED) {
/* Do nothing. */
} else if (tcache == TCACHE_STATE_REINCARNATED) {
/*
* Another destructor called an allocator function after this
* destructor was called. Reset tcache to
* TCACHE_STATE_PURGATORY in order to receive another callback.
*/
tcache = TCACHE_STATE_PURGATORY;
tcache_tsd_set(&tcache);
} else if (tcache == TCACHE_STATE_PURGATORY) {
/*
* The previous time this destructor was called, we set the key
* to TCACHE_STATE_PURGATORY so that other destructors wouldn't
* cause re-creation of the tcache. This time, do nothing, so
* that the destructor will not be called again.
*/
} else if (tcache != NULL) {
assert(tcache != TCACHE_STATE_PURGATORY);
tcache_destroy(tcache);
tcache = TCACHE_STATE_PURGATORY;
tcache_tsd_set(&tcache);
arena_prof_accum(tsd_tsdn(tsd), tcache->arena,
tcache->prof_accumbytes)) {
prof_idump(tsd_tsdn(tsd));
}
}
/* Caller must own arena->lock. */
void
tcache_stats_merge(tcache_t *tcache, arena_t *arena)
{
tcache_flush(tsd_t *tsd) {
assert(tcache_available(tsd));
tcache_flush_cache(tsd, tsd_tcachep_get(tsd));
}
static void
tcache_destroy(tsd_t *tsd, tcache_t *tcache, bool tsd_tcache) {
tcache_flush_cache(tsd, tcache);
tcache_arena_dissociate(tsd_tsdn(tsd), tcache);
if (tsd_tcache) {
/* Release the avail array for the TSD embedded auto tcache. */
void *avail_array =
(void *)((uintptr_t)tcache_small_bin_get(tcache, 0)->avail -
(uintptr_t)tcache_bin_info[0].ncached_max * sizeof(void *));
idalloctm(tsd_tsdn(tsd), avail_array, NULL, NULL, true, true);
} else {
/* Release both the tcache struct and avail array. */
idalloctm(tsd_tsdn(tsd), tcache, NULL, NULL, true, true);
}
}
/* For auto tcache (embedded in TSD) only. */
void
tcache_cleanup(tsd_t *tsd) {
tcache_t *tcache = tsd_tcachep_get(tsd);
if (!tcache_available(tsd)) {
assert(tsd_tcache_enabled_get(tsd) == false);
if (config_debug) {
assert(tcache_small_bin_get(tcache, 0)->avail == NULL);
}
return;
}
assert(tsd_tcache_enabled_get(tsd));
assert(tcache_small_bin_get(tcache, 0)->avail != NULL);
tcache_destroy(tsd, tcache, true);
if (config_debug) {
tcache_small_bin_get(tcache, 0)->avail = NULL;
}
}
void
tcache_stats_merge(tsdn_t *tsdn, tcache_t *tcache, arena_t *arena) {
unsigned i;
cassert(config_stats);
@@ -410,48 +525,151 @@ tcache_stats_merge(tcache_t *tcache, arena_t *arena)
/* Merge and reset tcache stats. */
for (i = 0; i < NBINS; i++) {
arena_bin_t *bin = &arena->bins[i];
tcache_bin_t *tbin = &tcache->tbins[i];
malloc_mutex_lock(&bin->lock);
tcache_bin_t *tbin = tcache_small_bin_get(tcache, i);
malloc_mutex_lock(tsdn, &bin->lock);
bin->stats.nrequests += tbin->tstats.nrequests;
malloc_mutex_unlock(&bin->lock);
malloc_mutex_unlock(tsdn, &bin->lock);
tbin->tstats.nrequests = 0;
}
for (; i < nhbins; i++) {
malloc_large_stats_t *lstats = &arena->stats.lstats[i - NBINS];
tcache_bin_t *tbin = &tcache->tbins[i];
arena->stats.nrequests_large += tbin->tstats.nrequests;
lstats->nrequests += tbin->tstats.nrequests;
tcache_bin_t *tbin = tcache_large_bin_get(tcache, i);
arena_stats_large_nrequests_add(tsdn, &arena->stats, i,
tbin->tstats.nrequests);
tbin->tstats.nrequests = 0;
}
}
static bool
tcaches_create_prep(tsd_t *tsd) {
bool err;
malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx);
if (tcaches == NULL) {
tcaches = base_alloc(tsd_tsdn(tsd), b0get(), sizeof(tcache_t *)
* (MALLOCX_TCACHE_MAX+1), CACHELINE);
if (tcaches == NULL) {
err = true;
goto label_return;
}
}
if (tcaches_avail == NULL && tcaches_past > MALLOCX_TCACHE_MAX) {
err = true;
goto label_return;
}
err = false;
label_return:
malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx);
return err;
}
bool
tcache_boot0(void)
{
unsigned i;
tcaches_create(tsd_t *tsd, unsigned *r_ind) {
witness_assert_depth(tsdn_witness_tsdp_get(tsd_tsdn(tsd)), 0);
/*
* If necessary, clamp opt_lg_tcache_max, now that arena_maxclass is
* known.
*/
if (opt_lg_tcache_max < 0 || (1U << opt_lg_tcache_max) < SMALL_MAXCLASS)
bool err;
if (tcaches_create_prep(tsd)) {
err = true;
goto label_return;
}
tcache_t *tcache = tcache_create_explicit(tsd);
if (tcache == NULL) {
err = true;
goto label_return;
}
tcaches_t *elm;
malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx);
if (tcaches_avail != NULL) {
elm = tcaches_avail;
tcaches_avail = tcaches_avail->next;
elm->tcache = tcache;
*r_ind = (unsigned)(elm - tcaches);
} else {
elm = &tcaches[tcaches_past];
elm->tcache = tcache;
*r_ind = tcaches_past;
tcaches_past++;
}
malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx);
err = false;
label_return:
witness_assert_depth(tsdn_witness_tsdp_get(tsd_tsdn(tsd)), 0);
return err;
}
static tcache_t *
tcaches_elm_remove(tsd_t *tsd, tcaches_t *elm) {
malloc_mutex_assert_owner(tsd_tsdn(tsd), &tcaches_mtx);
if (elm->tcache == NULL) {
return NULL;
}
tcache_t *tcache = elm->tcache;
elm->tcache = NULL;
return tcache;
}
void
tcaches_flush(tsd_t *tsd, unsigned ind) {
malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx);
tcache_t *tcache = tcaches_elm_remove(tsd, &tcaches[ind]);
malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx);
if (tcache != NULL) {
tcache_destroy(tsd, tcache, false);
}
}
void
tcaches_destroy(tsd_t *tsd, unsigned ind) {
malloc_mutex_lock(tsd_tsdn(tsd), &tcaches_mtx);
tcaches_t *elm = &tcaches[ind];
tcache_t *tcache = tcaches_elm_remove(tsd, elm);
elm->next = tcaches_avail;
tcaches_avail = elm;
malloc_mutex_unlock(tsd_tsdn(tsd), &tcaches_mtx);
if (tcache != NULL) {
tcache_destroy(tsd, tcache, false);
}
}
bool
tcache_boot(tsdn_t *tsdn) {
/* If necessary, clamp opt_lg_tcache_max. */
if (opt_lg_tcache_max < 0 || (ZU(1) << opt_lg_tcache_max) <
SMALL_MAXCLASS) {
tcache_maxclass = SMALL_MAXCLASS;
else if ((1U << opt_lg_tcache_max) > arena_maxclass)
tcache_maxclass = arena_maxclass;
else
tcache_maxclass = (1U << opt_lg_tcache_max);
} else {
tcache_maxclass = (ZU(1) << opt_lg_tcache_max);
}
nhbins = NBINS + (tcache_maxclass >> LG_PAGE);
if (malloc_mutex_init(&tcaches_mtx, "tcaches", WITNESS_RANK_TCACHES,
malloc_mutex_rank_exclusive)) {
return true;
}
nhbins = sz_size2index(tcache_maxclass) + 1;
/* Initialize tcache_bin_info. */
tcache_bin_info = (tcache_bin_info_t *)base_alloc(nhbins *
sizeof(tcache_bin_info_t));
if (tcache_bin_info == NULL)
return (true);
tcache_bin_info = (tcache_bin_info_t *)base_alloc(tsdn, b0get(), nhbins
* sizeof(tcache_bin_info_t), CACHELINE);
if (tcache_bin_info == NULL) {
return true;
}
stack_nelms = 0;
unsigned i;
for (i = 0; i < NBINS; i++) {
if ((arena_bin_info[i].nregs << 1) <= TCACHE_NSLOTS_SMALL_MAX) {
if ((arena_bin_info[i].nregs << 1) <= TCACHE_NSLOTS_SMALL_MIN) {
tcache_bin_info[i].ncached_max =
TCACHE_NSLOTS_SMALL_MIN;
} else if ((arena_bin_info[i].nregs << 1) <=
TCACHE_NSLOTS_SMALL_MAX) {
tcache_bin_info[i].ncached_max =
(arena_bin_info[i].nregs << 1);
} else {
@@ -465,15 +683,26 @@ tcache_boot0(void)
stack_nelms += tcache_bin_info[i].ncached_max;
}
return (false);
return false;
}
bool
tcache_boot1(void)
{
if (tcache_tsd_boot() || tcache_enabled_tsd_boot())
return (true);
return (false);
void
tcache_prefork(tsdn_t *tsdn) {
if (!config_prof && opt_tcache) {
malloc_mutex_prefork(tsdn, &tcaches_mtx);
}
}
void
tcache_postfork_parent(tsdn_t *tsdn) {
if (!config_prof && opt_tcache) {
malloc_mutex_postfork_parent(tsdn, &tcaches_mtx);
}
}
void
tcache_postfork_child(tsdn_t *tsdn) {
if (!config_prof && opt_tcache) {
malloc_mutex_postfork_child(tsdn, &tcaches_mtx);
}
}

288
deps/jemalloc/src/tsd.c vendored
View File

@@ -1,5 +1,10 @@
#define JEMALLOC_TSD_C_
#include "jemalloc/internal/jemalloc_internal.h"
#define JEMALLOC_TSD_C_
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
#include "jemalloc/internal/assert.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/rtree.h"
/******************************************************************************/
/* Data. */
@@ -7,28 +12,148 @@
static unsigned ncleanups;
static malloc_tsd_cleanup_t cleanups[MALLOC_TSD_CLEANUPS_MAX];
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
__thread tsd_t JEMALLOC_TLS_MODEL tsd_tls = TSD_INITIALIZER;
__thread bool JEMALLOC_TLS_MODEL tsd_initialized = false;
bool tsd_booted = false;
#elif (defined(JEMALLOC_TLS))
__thread tsd_t JEMALLOC_TLS_MODEL tsd_tls = TSD_INITIALIZER;
pthread_key_t tsd_tsd;
bool tsd_booted = false;
#elif (defined(_WIN32))
DWORD tsd_tsd;
tsd_wrapper_t tsd_boot_wrapper = {false, TSD_INITIALIZER};
bool tsd_booted = false;
#else
/*
* This contains a mutex, but it's pretty convenient to allow the mutex code to
* have a dependency on tsd. So we define the struct here, and only refer to it
* by pointer in the header.
*/
struct tsd_init_head_s {
ql_head(tsd_init_block_t) blocks;
malloc_mutex_t lock;
};
pthread_key_t tsd_tsd;
tsd_init_head_t tsd_init_head = {
ql_head_initializer(blocks),
MALLOC_MUTEX_INITIALIZER
};
tsd_wrapper_t tsd_boot_wrapper = {
false,
TSD_INITIALIZER
};
bool tsd_booted = false;
#endif
/******************************************************************************/
void
tsd_slow_update(tsd_t *tsd) {
if (tsd_nominal(tsd)) {
if (malloc_slow || !tsd_tcache_enabled_get(tsd) ||
tsd_reentrancy_level_get(tsd) > 0) {
tsd->state = tsd_state_nominal_slow;
} else {
tsd->state = tsd_state_nominal;
}
}
}
static bool
tsd_data_init(tsd_t *tsd) {
/*
* We initialize the rtree context first (before the tcache), since the
* tcache initialization depends on it.
*/
rtree_ctx_data_init(tsd_rtree_ctxp_get_unsafe(tsd));
return tsd_tcache_enabled_data_init(tsd);
}
static void
assert_tsd_data_cleanup_done(tsd_t *tsd) {
assert(!tsd_nominal(tsd));
assert(*tsd_arenap_get_unsafe(tsd) == NULL);
assert(*tsd_iarenap_get_unsafe(tsd) == NULL);
assert(*tsd_arenas_tdata_bypassp_get_unsafe(tsd) == true);
assert(*tsd_arenas_tdatap_get_unsafe(tsd) == NULL);
assert(*tsd_tcache_enabledp_get_unsafe(tsd) == false);
assert(*tsd_prof_tdatap_get_unsafe(tsd) == NULL);
}
static bool
tsd_data_init_nocleanup(tsd_t *tsd) {
assert(tsd->state == tsd_state_reincarnated ||
tsd->state == tsd_state_minimal_initialized);
/*
* During reincarnation, there is no guarantee that the cleanup function
* will be called (deallocation may happen after all tsd destructors).
* We set up tsd in a way that no cleanup is needed.
*/
rtree_ctx_data_init(tsd_rtree_ctxp_get_unsafe(tsd));
*tsd_arenas_tdata_bypassp_get(tsd) = true;
*tsd_tcache_enabledp_get_unsafe(tsd) = false;
*tsd_reentrancy_levelp_get(tsd) = 1;
assert_tsd_data_cleanup_done(tsd);
return false;
}
tsd_t *
tsd_fetch_slow(tsd_t *tsd, bool minimal) {
assert(!tsd_fast(tsd));
if (tsd->state == tsd_state_nominal_slow) {
/* On slow path but no work needed. */
assert(malloc_slow || !tsd_tcache_enabled_get(tsd) ||
tsd_reentrancy_level_get(tsd) > 0 ||
*tsd_arenas_tdata_bypassp_get(tsd));
} else if (tsd->state == tsd_state_uninitialized) {
if (!minimal) {
tsd->state = tsd_state_nominal;
tsd_slow_update(tsd);
/* Trigger cleanup handler registration. */
tsd_set(tsd);
tsd_data_init(tsd);
} else {
tsd->state = tsd_state_minimal_initialized;
tsd_set(tsd);
tsd_data_init_nocleanup(tsd);
}
} else if (tsd->state == tsd_state_minimal_initialized) {
if (!minimal) {
/* Switch to fully initialized. */
tsd->state = tsd_state_nominal;
assert(*tsd_reentrancy_levelp_get(tsd) >= 1);
(*tsd_reentrancy_levelp_get(tsd))--;
tsd_slow_update(tsd);
tsd_data_init(tsd);
} else {
assert_tsd_data_cleanup_done(tsd);
}
} else if (tsd->state == tsd_state_purgatory) {
tsd->state = tsd_state_reincarnated;
tsd_set(tsd);
tsd_data_init_nocleanup(tsd);
} else {
assert(tsd->state == tsd_state_reincarnated);
}
return tsd;
}
void *
malloc_tsd_malloc(size_t size)
{
/* Avoid choose_arena() in order to dodge bootstrapping issues. */
return (arena_malloc(arenas[0], size, false, false));
malloc_tsd_malloc(size_t size) {
return a0malloc(CACHELINE_CEILING(size));
}
void
malloc_tsd_dalloc(void *wrapper)
{
idalloct(wrapper, false);
}
void
malloc_tsd_no_cleanup(void *arg)
{
not_reached();
malloc_tsd_dalloc(void *wrapper) {
a0dalloc(wrapper);
}
#if defined(JEMALLOC_MALLOC_THREAD_CLEANUP) || defined(_WIN32)
@@ -36,21 +161,22 @@ malloc_tsd_no_cleanup(void *arg)
JEMALLOC_EXPORT
#endif
void
_malloc_thread_cleanup(void)
{
_malloc_thread_cleanup(void) {
bool pending[MALLOC_TSD_CLEANUPS_MAX], again;
unsigned i;
for (i = 0; i < ncleanups; i++)
for (i = 0; i < ncleanups; i++) {
pending[i] = true;
}
do {
again = false;
for (i = 0; i < ncleanups; i++) {
if (pending[i]) {
pending[i] = cleanups[i]();
if (pending[i])
if (pending[i]) {
again = true;
}
}
}
} while (again);
@@ -58,26 +184,92 @@ _malloc_thread_cleanup(void)
#endif
void
malloc_tsd_cleanup_register(bool (*f)(void))
{
malloc_tsd_cleanup_register(bool (*f)(void)) {
assert(ncleanups < MALLOC_TSD_CLEANUPS_MAX);
cleanups[ncleanups] = f;
ncleanups++;
}
static void
tsd_do_data_cleanup(tsd_t *tsd) {
prof_tdata_cleanup(tsd);
iarena_cleanup(tsd);
arena_cleanup(tsd);
arenas_tdata_cleanup(tsd);
tcache_cleanup(tsd);
witnesses_cleanup(tsd_witness_tsdp_get_unsafe(tsd));
}
void
malloc_tsd_boot(void)
{
tsd_cleanup(void *arg) {
tsd_t *tsd = (tsd_t *)arg;
switch (tsd->state) {
case tsd_state_uninitialized:
/* Do nothing. */
break;
case tsd_state_minimal_initialized:
/* This implies the thread only did free() in its life time. */
/* Fall through. */
case tsd_state_reincarnated:
/*
* Reincarnated means another destructor deallocated memory
* after the destructor was called. Cleanup isn't required but
* is still called for testing and completeness.
*/
assert_tsd_data_cleanup_done(tsd);
/* Fall through. */
case tsd_state_nominal:
case tsd_state_nominal_slow:
tsd_do_data_cleanup(tsd);
tsd->state = tsd_state_purgatory;
tsd_set(tsd);
break;
case tsd_state_purgatory:
/*
* The previous time this destructor was called, we set the
* state to tsd_state_purgatory so that other destructors
* wouldn't cause re-creation of the tsd. This time, do
* nothing, and do not request another callback.
*/
break;
default:
not_reached();
}
#ifdef JEMALLOC_JET
test_callback_t test_callback = *tsd_test_callbackp_get_unsafe(tsd);
int *data = tsd_test_datap_get_unsafe(tsd);
if (test_callback != NULL) {
test_callback(data);
}
#endif
}
tsd_t *
malloc_tsd_boot0(void) {
tsd_t *tsd;
ncleanups = 0;
if (tsd_boot0()) {
return NULL;
}
tsd = tsd_fetch();
*tsd_arenas_tdata_bypassp_get(tsd) = true;
return tsd;
}
void
malloc_tsd_boot1(void) {
tsd_boot1();
tsd_t *tsd = tsd_fetch();
/* malloc_slow has been set properly. Update tsd_slow. */
tsd_slow_update(tsd);
*tsd_arenas_tdata_bypassp_get(tsd) = false;
}
#ifdef _WIN32
static BOOL WINAPI
_tls_callback(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved)
{
_tls_callback(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved) {
switch (fdwReason) {
#ifdef JEMALLOC_LAZY_LOCK
case DLL_THREAD_ATTACH:
@@ -90,52 +282,60 @@ _tls_callback(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved)
default:
break;
}
return (true);
return true;
}
/*
* We need to be able to say "read" here (in the "pragma section"), but have
* hooked "read". We won't read for the rest of the file, so we can get away
* with unhooking.
*/
#ifdef read
# undef read
#endif
#ifdef _MSC_VER
# ifdef _M_IX86
# pragma comment(linker, "/INCLUDE:__tls_used")
# pragma comment(linker, "/INCLUDE:_tls_callback")
# else
# pragma comment(linker, "/INCLUDE:_tls_used")
# pragma comment(linker, "/INCLUDE:tls_callback")
# endif
# pragma section(".CRT$XLY",long,read)
#endif
JEMALLOC_SECTION(".CRT$XLY") JEMALLOC_ATTR(used)
static const BOOL (WINAPI *tls_callback)(HINSTANCE hinstDLL,
BOOL (WINAPI *const tls_callback)(HINSTANCE hinstDLL,
DWORD fdwReason, LPVOID lpvReserved) = _tls_callback;
#endif
#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
!defined(_WIN32))
void *
tsd_init_check_recursion(tsd_init_head_t *head, tsd_init_block_t *block)
{
tsd_init_check_recursion(tsd_init_head_t *head, tsd_init_block_t *block) {
pthread_t self = pthread_self();
tsd_init_block_t *iter;
/* Check whether this thread has already inserted into the list. */
malloc_mutex_lock(&head->lock);
malloc_mutex_lock(TSDN_NULL, &head->lock);
ql_foreach(iter, &head->blocks, link) {
if (iter->thread == self) {
malloc_mutex_unlock(&head->lock);
return (iter->data);
malloc_mutex_unlock(TSDN_NULL, &head->lock);
return iter->data;
}
}
/* Insert block into list. */
ql_elm_new(block, link);
block->thread = self;
ql_tail_insert(&head->blocks, block, link);
malloc_mutex_unlock(&head->lock);
return (NULL);
malloc_mutex_unlock(TSDN_NULL, &head->lock);
return NULL;
}
void
tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block)
{
malloc_mutex_lock(&head->lock);
tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block) {
malloc_mutex_lock(TSDN_NULL, &head->lock);
ql_remove(&head->blocks, block, link);
malloc_mutex_unlock(&head->lock);
malloc_mutex_unlock(TSDN_NULL, &head->lock);
}
#endif

453
deps/jemalloc/src/zone.c vendored
View File

@@ -1,10 +1,83 @@
#include "jemalloc/internal/jemalloc_internal.h"
#include "jemalloc/internal/jemalloc_preamble.h"
#include "jemalloc/internal/jemalloc_internal_includes.h"
#include "jemalloc/internal/assert.h"
#ifndef JEMALLOC_ZONE
# error "This source file is for zones on Darwin (OS X)."
#endif
/* Definitions of the following structs in malloc/malloc.h might be too old
* for the built binary to run on newer versions of OSX. So use the newest
* possible version of those structs.
*/
typedef struct _malloc_zone_t {
void *reserved1;
void *reserved2;
size_t (*size)(struct _malloc_zone_t *, const void *);
void *(*malloc)(struct _malloc_zone_t *, size_t);
void *(*calloc)(struct _malloc_zone_t *, size_t, size_t);
void *(*valloc)(struct _malloc_zone_t *, size_t);
void (*free)(struct _malloc_zone_t *, void *);
void *(*realloc)(struct _malloc_zone_t *, void *, size_t);
void (*destroy)(struct _malloc_zone_t *);
const char *zone_name;
unsigned (*batch_malloc)(struct _malloc_zone_t *, size_t, void **, unsigned);
void (*batch_free)(struct _malloc_zone_t *, void **, unsigned);
struct malloc_introspection_t *introspect;
unsigned version;
void *(*memalign)(struct _malloc_zone_t *, size_t, size_t);
void (*free_definite_size)(struct _malloc_zone_t *, void *, size_t);
size_t (*pressure_relief)(struct _malloc_zone_t *, size_t);
} malloc_zone_t;
typedef struct {
vm_address_t address;
vm_size_t size;
} vm_range_t;
typedef struct malloc_statistics_t {
unsigned blocks_in_use;
size_t size_in_use;
size_t max_size_in_use;
size_t size_allocated;
} malloc_statistics_t;
typedef kern_return_t memory_reader_t(task_t, vm_address_t, vm_size_t, void **);
typedef void vm_range_recorder_t(task_t, void *, unsigned type, vm_range_t *, unsigned);
typedef struct malloc_introspection_t {
kern_return_t (*enumerator)(task_t, void *, unsigned, vm_address_t, memory_reader_t, vm_range_recorder_t);
size_t (*good_size)(malloc_zone_t *, size_t);
boolean_t (*check)(malloc_zone_t *);
void (*print)(malloc_zone_t *, boolean_t);
void (*log)(malloc_zone_t *, void *);
void (*force_lock)(malloc_zone_t *);
void (*force_unlock)(malloc_zone_t *);
void (*statistics)(malloc_zone_t *, malloc_statistics_t *);
boolean_t (*zone_locked)(malloc_zone_t *);
boolean_t (*enable_discharge_checking)(malloc_zone_t *);
boolean_t (*disable_discharge_checking)(malloc_zone_t *);
void (*discharge)(malloc_zone_t *, void *);
#ifdef __BLOCKS__
void (*enumerate_discharged_pointers)(malloc_zone_t *, void (^)(void *, void *));
#else
void *enumerate_unavailable_without_blocks;
#endif
void (*reinit_lock)(malloc_zone_t *);
} malloc_introspection_t;
extern kern_return_t malloc_get_all_zones(task_t, memory_reader_t, vm_address_t **, unsigned *);
extern malloc_zone_t *malloc_default_zone(void);
extern void malloc_zone_register(malloc_zone_t *zone);
extern void malloc_zone_unregister(malloc_zone_t *zone);
/*
* The malloc_default_purgeable_zone function is only available on >= 10.6.
* The malloc_default_purgeable_zone() function is only available on >= 10.6.
* We need to check whether it is present at runtime, thus the weak_import.
*/
extern malloc_zone_t *malloc_default_purgeable_zone(void)
@@ -13,30 +86,42 @@ JEMALLOC_ATTR(weak_import);
/******************************************************************************/
/* Data. */
static malloc_zone_t zone;
static struct malloc_introspection_t zone_introspect;
static malloc_zone_t *default_zone, *purgeable_zone;
static malloc_zone_t jemalloc_zone;
static struct malloc_introspection_t jemalloc_zone_introspect;
/******************************************************************************/
/* Function prototypes for non-inline static functions. */
static size_t zone_size(malloc_zone_t *zone, void *ptr);
static size_t zone_size(malloc_zone_t *zone, const void *ptr);
static void *zone_malloc(malloc_zone_t *zone, size_t size);
static void *zone_calloc(malloc_zone_t *zone, size_t num, size_t size);
static void *zone_valloc(malloc_zone_t *zone, size_t size);
static void zone_free(malloc_zone_t *zone, void *ptr);
static void *zone_realloc(malloc_zone_t *zone, void *ptr, size_t size);
#if (JEMALLOC_ZONE_VERSION >= 5)
static void *zone_memalign(malloc_zone_t *zone, size_t alignment,
#endif
#if (JEMALLOC_ZONE_VERSION >= 6)
size_t size);
static void zone_free_definite_size(malloc_zone_t *zone, void *ptr,
size_t size);
#endif
static void *zone_destroy(malloc_zone_t *zone);
static void zone_destroy(malloc_zone_t *zone);
static unsigned zone_batch_malloc(struct _malloc_zone_t *zone, size_t size,
void **results, unsigned num_requested);
static void zone_batch_free(struct _malloc_zone_t *zone,
void **to_be_freed, unsigned num_to_be_freed);
static size_t zone_pressure_relief(struct _malloc_zone_t *zone, size_t goal);
static size_t zone_good_size(malloc_zone_t *zone, size_t size);
static kern_return_t zone_enumerator(task_t task, void *data, unsigned type_mask,
vm_address_t zone_address, memory_reader_t reader,
vm_range_recorder_t recorder);
static boolean_t zone_check(malloc_zone_t *zone);
static void zone_print(malloc_zone_t *zone, boolean_t verbose);
static void zone_log(malloc_zone_t *zone, void *address);
static void zone_force_lock(malloc_zone_t *zone);
static void zone_force_unlock(malloc_zone_t *zone);
static void zone_statistics(malloc_zone_t *zone,
malloc_statistics_t *stats);
static boolean_t zone_locked(malloc_zone_t *zone);
static void zone_reinit_lock(malloc_zone_t *zone);
/******************************************************************************/
/*
@@ -44,9 +129,7 @@ static void zone_force_unlock(malloc_zone_t *zone);
*/
static size_t
zone_size(malloc_zone_t *zone, void *ptr)
{
zone_size(malloc_zone_t *zone, const void *ptr) {
/*
* There appear to be places within Darwin (such as setenv(3)) that
* cause calls to this function with pointers that *no* zone owns. If
@@ -54,40 +137,33 @@ zone_size(malloc_zone_t *zone, void *ptr)
* our zone into two parts, and use one as the default allocator and
* the other as the default deallocator/reallocator. Since that will
* not work in practice, we must check all pointers to assure that they
* reside within a mapped chunk before determining size.
* reside within a mapped extent before determining size.
*/
return (ivsalloc(ptr, config_prof));
return ivsalloc(tsdn_fetch(), ptr);
}
static void *
zone_malloc(malloc_zone_t *zone, size_t size)
{
return (je_malloc(size));
zone_malloc(malloc_zone_t *zone, size_t size) {
return je_malloc(size);
}
static void *
zone_calloc(malloc_zone_t *zone, size_t num, size_t size)
{
return (je_calloc(num, size));
zone_calloc(malloc_zone_t *zone, size_t num, size_t size) {
return je_calloc(num, size);
}
static void *
zone_valloc(malloc_zone_t *zone, size_t size)
{
zone_valloc(malloc_zone_t *zone, size_t size) {
void *ret = NULL; /* Assignment avoids useless compiler warning. */
je_posix_memalign(&ret, PAGE, size);
return (ret);
return ret;
}
static void
zone_free(malloc_zone_t *zone, void *ptr)
{
if (ivsalloc(ptr, config_prof) != 0) {
zone_free(malloc_zone_t *zone, void *ptr) {
if (ivsalloc(tsdn_fetch(), ptr) != 0) {
je_free(ptr);
return;
}
@@ -96,163 +172,280 @@ zone_free(malloc_zone_t *zone, void *ptr)
}
static void *
zone_realloc(malloc_zone_t *zone, void *ptr, size_t size)
{
zone_realloc(malloc_zone_t *zone, void *ptr, size_t size) {
if (ivsalloc(tsdn_fetch(), ptr) != 0) {
return je_realloc(ptr, size);
}
if (ivsalloc(ptr, config_prof) != 0)
return (je_realloc(ptr, size));
return (realloc(ptr, size));
return realloc(ptr, size);
}
#if (JEMALLOC_ZONE_VERSION >= 5)
static void *
zone_memalign(malloc_zone_t *zone, size_t alignment, size_t size)
{
zone_memalign(malloc_zone_t *zone, size_t alignment, size_t size) {
void *ret = NULL; /* Assignment avoids useless compiler warning. */
je_posix_memalign(&ret, alignment, size);
return (ret);
return ret;
}
#endif
#if (JEMALLOC_ZONE_VERSION >= 6)
static void
zone_free_definite_size(malloc_zone_t *zone, void *ptr, size_t size)
{
zone_free_definite_size(malloc_zone_t *zone, void *ptr, size_t size) {
size_t alloc_size;
if (ivsalloc(ptr, config_prof) != 0) {
assert(ivsalloc(ptr, config_prof) == size);
alloc_size = ivsalloc(tsdn_fetch(), ptr);
if (alloc_size != 0) {
assert(alloc_size == size);
je_free(ptr);
return;
}
free(ptr);
}
#endif
static void *
zone_destroy(malloc_zone_t *zone)
{
static void
zone_destroy(malloc_zone_t *zone) {
/* This function should never be called. */
not_reached();
return (NULL);
}
static unsigned
zone_batch_malloc(struct _malloc_zone_t *zone, size_t size, void **results,
unsigned num_requested) {
unsigned i;
for (i = 0; i < num_requested; i++) {
results[i] = je_malloc(size);
if (!results[i])
break;
}
return i;
}
static void
zone_batch_free(struct _malloc_zone_t *zone, void **to_be_freed,
unsigned num_to_be_freed) {
unsigned i;
for (i = 0; i < num_to_be_freed; i++) {
zone_free(zone, to_be_freed[i]);
to_be_freed[i] = NULL;
}
}
static size_t
zone_good_size(malloc_zone_t *zone, size_t size)
{
zone_pressure_relief(struct _malloc_zone_t *zone, size_t goal) {
return 0;
}
if (size == 0)
static size_t
zone_good_size(malloc_zone_t *zone, size_t size) {
if (size == 0) {
size = 1;
return (s2u(size));
}
return sz_s2u(size);
}
static kern_return_t
zone_enumerator(task_t task, void *data, unsigned type_mask,
vm_address_t zone_address, memory_reader_t reader,
vm_range_recorder_t recorder) {
return KERN_SUCCESS;
}
static boolean_t
zone_check(malloc_zone_t *zone) {
return true;
}
static void
zone_force_lock(malloc_zone_t *zone)
{
zone_print(malloc_zone_t *zone, boolean_t verbose) {
}
if (isthreaded)
static void
zone_log(malloc_zone_t *zone, void *address) {
}
static void
zone_force_lock(malloc_zone_t *zone) {
if (isthreaded) {
jemalloc_prefork();
}
}
static void
zone_force_unlock(malloc_zone_t *zone)
{
zone_force_unlock(malloc_zone_t *zone) {
/*
* Call jemalloc_postfork_child() rather than
* jemalloc_postfork_parent(), because this function is executed by both
* parent and child. The parent can tolerate having state
* reinitialized, but the child cannot unlock mutexes that were locked
* by the parent.
*/
if (isthreaded) {
jemalloc_postfork_child();
}
}
if (isthreaded)
jemalloc_postfork_parent();
static void
zone_statistics(malloc_zone_t *zone, malloc_statistics_t *stats) {
/* We make no effort to actually fill the values */
stats->blocks_in_use = 0;
stats->size_in_use = 0;
stats->max_size_in_use = 0;
stats->size_allocated = 0;
}
static boolean_t
zone_locked(malloc_zone_t *zone) {
/* Pretend no lock is being held */
return false;
}
static void
zone_reinit_lock(malloc_zone_t *zone) {
/* As of OSX 10.12, this function is only used when force_unlock would
* be used if the zone version were < 9. So just use force_unlock. */
zone_force_unlock(zone);
}
static void
zone_init(void) {
jemalloc_zone.size = zone_size;
jemalloc_zone.malloc = zone_malloc;
jemalloc_zone.calloc = zone_calloc;
jemalloc_zone.valloc = zone_valloc;
jemalloc_zone.free = zone_free;
jemalloc_zone.realloc = zone_realloc;
jemalloc_zone.destroy = zone_destroy;
jemalloc_zone.zone_name = "jemalloc_zone";
jemalloc_zone.batch_malloc = zone_batch_malloc;
jemalloc_zone.batch_free = zone_batch_free;
jemalloc_zone.introspect = &jemalloc_zone_introspect;
jemalloc_zone.version = 9;
jemalloc_zone.memalign = zone_memalign;
jemalloc_zone.free_definite_size = zone_free_definite_size;
jemalloc_zone.pressure_relief = zone_pressure_relief;
jemalloc_zone_introspect.enumerator = zone_enumerator;
jemalloc_zone_introspect.good_size = zone_good_size;
jemalloc_zone_introspect.check = zone_check;
jemalloc_zone_introspect.print = zone_print;
jemalloc_zone_introspect.log = zone_log;
jemalloc_zone_introspect.force_lock = zone_force_lock;
jemalloc_zone_introspect.force_unlock = zone_force_unlock;
jemalloc_zone_introspect.statistics = zone_statistics;
jemalloc_zone_introspect.zone_locked = zone_locked;
jemalloc_zone_introspect.enable_discharge_checking = NULL;
jemalloc_zone_introspect.disable_discharge_checking = NULL;
jemalloc_zone_introspect.discharge = NULL;
#ifdef __BLOCKS__
jemalloc_zone_introspect.enumerate_discharged_pointers = NULL;
#else
jemalloc_zone_introspect.enumerate_unavailable_without_blocks = NULL;
#endif
jemalloc_zone_introspect.reinit_lock = zone_reinit_lock;
}
static malloc_zone_t *
zone_default_get(void) {
malloc_zone_t **zones = NULL;
unsigned int num_zones = 0;
/*
* On OSX 10.12, malloc_default_zone returns a special zone that is not
* present in the list of registered zones. That zone uses a "lite zone"
* if one is present (apparently enabled when malloc stack logging is
* enabled), or the first registered zone otherwise. In practice this
* means unless malloc stack logging is enabled, the first registered
* zone is the default. So get the list of zones to get the first one,
* instead of relying on malloc_default_zone.
*/
if (KERN_SUCCESS != malloc_get_all_zones(0, NULL,
(vm_address_t**)&zones, &num_zones)) {
/*
* Reset the value in case the failure happened after it was
* set.
*/
num_zones = 0;
}
if (num_zones) {
return zones[0];
}
return malloc_default_zone();
}
/* As written, this function can only promote jemalloc_zone. */
static void
zone_promote(void) {
malloc_zone_t *zone;
do {
/*
* Unregister and reregister the default zone. On OSX >= 10.6,
* unregistering takes the last registered zone and places it
* at the location of the specified zone. Unregistering the
* default zone thus makes the last registered one the default.
* On OSX < 10.6, unregistering shifts all registered zones.
* The first registered zone then becomes the default.
*/
malloc_zone_unregister(default_zone);
malloc_zone_register(default_zone);
/*
* On OSX 10.6, having the default purgeable zone appear before
* the default zone makes some things crash because it thinks it
* owns the default zone allocated pointers. We thus
* unregister/re-register it in order to ensure it's always
* after the default zone. On OSX < 10.6, there is no purgeable
* zone, so this does nothing. On OSX >= 10.6, unregistering
* replaces the purgeable zone with the last registered zone
* above, i.e. the default zone. Registering it again then puts
* it at the end, obviously after the default zone.
*/
if (purgeable_zone != NULL) {
malloc_zone_unregister(purgeable_zone);
malloc_zone_register(purgeable_zone);
}
zone = zone_default_get();
} while (zone != &jemalloc_zone);
}
JEMALLOC_ATTR(constructor)
void
register_zone(void)
{
zone_register(void) {
/*
* If something else replaced the system default zone allocator, don't
* register jemalloc's.
*/
malloc_zone_t *default_zone = malloc_default_zone();
if (!default_zone->zone_name ||
strcmp(default_zone->zone_name, "DefaultMallocZone") != 0) {
default_zone = zone_default_get();
if (!default_zone->zone_name || strcmp(default_zone->zone_name,
"DefaultMallocZone") != 0) {
return;
}
zone.size = (void *)zone_size;
zone.malloc = (void *)zone_malloc;
zone.calloc = (void *)zone_calloc;
zone.valloc = (void *)zone_valloc;
zone.free = (void *)zone_free;
zone.realloc = (void *)zone_realloc;
zone.destroy = (void *)zone_destroy;
zone.zone_name = "jemalloc_zone";
zone.batch_malloc = NULL;
zone.batch_free = NULL;
zone.introspect = &zone_introspect;
zone.version = JEMALLOC_ZONE_VERSION;
#if (JEMALLOC_ZONE_VERSION >= 5)
zone.memalign = zone_memalign;
#endif
#if (JEMALLOC_ZONE_VERSION >= 6)
zone.free_definite_size = zone_free_definite_size;
#endif
#if (JEMALLOC_ZONE_VERSION >= 8)
zone.pressure_relief = NULL;
#endif
zone_introspect.enumerator = NULL;
zone_introspect.good_size = (void *)zone_good_size;
zone_introspect.check = NULL;
zone_introspect.print = NULL;
zone_introspect.log = NULL;
zone_introspect.force_lock = (void *)zone_force_lock;
zone_introspect.force_unlock = (void *)zone_force_unlock;
zone_introspect.statistics = NULL;
#if (JEMALLOC_ZONE_VERSION >= 6)
zone_introspect.zone_locked = NULL;
#endif
#if (JEMALLOC_ZONE_VERSION >= 7)
zone_introspect.enable_discharge_checking = NULL;
zone_introspect.disable_discharge_checking = NULL;
zone_introspect.discharge = NULL;
#ifdef __BLOCKS__
zone_introspect.enumerate_discharged_pointers = NULL;
#else
zone_introspect.enumerate_unavailable_without_blocks = NULL;
#endif
#endif
/*
* The default purgeable zone is created lazily by OSX's libc. It uses
* the default zone when it is created for "small" allocations
* (< 15 KiB), but assumes the default zone is a scalable_zone. This
* obviously fails when the default zone is the jemalloc zone, so
* malloc_default_purgeable_zone is called beforehand so that the
* malloc_default_purgeable_zone() is called beforehand so that the
* default purgeable zone is created when the default zone is still
* a scalable_zone. As purgeable zones only exist on >= 10.6, we need
* to check for the existence of malloc_default_purgeable_zone() at
* run time.
*/
if (malloc_default_purgeable_zone != NULL)
malloc_default_purgeable_zone();
purgeable_zone = (malloc_default_purgeable_zone == NULL) ? NULL :
malloc_default_purgeable_zone();
/* Register the custom zone. At this point it won't be the default. */
malloc_zone_register(&zone);
zone_init();
malloc_zone_register(&jemalloc_zone);
/*
* Unregister and reregister the default zone. On OSX >= 10.6,
* unregistering takes the last registered zone and places it at the
* location of the specified zone. Unregistering the default zone thus
* makes the last registered one the default. On OSX < 10.6,
* unregistering shifts all registered zones. The first registered zone
* then becomes the default.
*/
do {
default_zone = malloc_default_zone();
malloc_zone_unregister(default_zone);
malloc_zone_register(default_zone);
} while (malloc_default_zone() != &zone);
/* Promote the custom zone to be default. */
zone_promote();
}