Add the --with-lg-hugepage configure option, but automatically configure
LG_HUGEPAGE even if it isn't specified.
Add the pages_[no]huge() functions, which toggle huge page state via
madvise(..., MADV_[NO]HUGEPAGE) calls.
These bugs were introduced by 0ba5b9b618
(Add "J" (JSON) support to malloc_stats_print().), which was backported
as b599b32280 (with the same bugs except
the inapplicable "metatata" misspelling) and first released in 4.3.0.
Rewrite arena_slab_regind() to provide sufficient constant data for
the compiler to perform division strength reduction. This replaces
more general manual strength reduction that was implemented before
arena_bin_info was compile-time-constant. It would be possible to
slightly improve on the compiler-generated division code by taking
advantage of range limits that the compiler doesn't know about.
Adds cpp bindings for jemalloc, along with necessary autoconf settings.
This is mostly to add sized deallocation support, which can't be added
from C directly. Sized deallocation is ~10% microbench improvement.
* Import ax_cxx_compile_stdcxx.m4 from the autoconf repo, seems like the
easiest way to get c++14 detection.
* Adds various other changes, like CXXFLAGS, to configure.ac.
* Adds new rules to Makefile.in for src/jemalloc-cpp.cpp, and a basic
unittest.
* Both new and delete are overridden, to ensure jemalloc is used for
both.
* TODO future enhancement of avoiding extra PLT thunks for new and
delete - sdallocx and malloc are publicly exported jemalloc symbols,
using an alias would link them directly. Unfortunately, was having
trouble getting it to play nice with jemalloc's namespace support.
Testing:
Tested gcc 4.8, gcc 5, gcc 5.2, clang 4.0. Only gcc >= 5 has sized
deallocation support, verified that the rest build correctly.
Tested mac osx and Centos.
Tested --with-jemalloc-prefix and --without-export.
This resolves#202.
Some versions of Android provide a pthreads library without providing
pthread_atfork(), so in practice a separate feature test is necessary
for the latter.
Add feature tests for the MADV_FREE and MADV_DONTNEED flags to
madvise(2), so that MADV_FREE is detected and used for Linux kernel
versions 4.5 and newer. Refactor pages_purge() so that on systems which
support both flags, MADV_FREE is preferred over MADV_DONTNEED.
This resolves#387.
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 resolves#147.
Do not call s2u() during alloc_size computation, since any necessary
ceiling increase is taken care of later by extent_first_best_fit() -->
extent_size_quantize_ceil(), and the s2u() call may erroneously cause a
higher quantization result.
Remove an overly strict overflow check that was added in
4a7852137d (Fix extent_recycle()'s
cache-oblivious padding support.).
Add padding *after* computing the size class, so that the optimal size
class isn't skipped during search for a usable extent. This regression
was caused by b46261d58b (Implement
cache-oblivious support for huge size classes.).
Add an "over-size" extent heap in which to store extents which exceed
the maximum size class (plus cache-oblivious padding, if enabled).
Remove psz2ind_clamp() and use psz2ind() instead so that trying to
allocate the maximum size class can in principle succeed. In practice,
this allows assertions to hold so that OOM errors can be successfully
generated.
Fix extent_alloc_cache[_locked]() to support decommitted allocation, and
use this ability in arena_stash_dirty(), so that decommitted extents are
not needlessly committed during purging. In practice this does not
happen on any currently supported systems, because both extent merging
and decommit must be implemented; all supported systems implement one
xor the other.
rtree_node_init spinlocks the node, allocates, and then sets the node.
This is under heavy contention at the top of the tree if many threads
start to allocate at the same time.
Instead, take a per-rtree sleeping mutex to reduce spinning. Tested
both pthreads and osx OSSpinLock, and both reduce spinning adequately
Previous benchmark time:
./ttest1 500 100
~15s
New benchmark time:
./ttest1 500 100
.57s
Fix zone_force_unlock() to reinitialize, rather than unlocking mutexes,
since OS X 10.12 cannot tolerate a child unlocking mutexes that were
locked by its parent.
Refactor; this was a side effect of experimenting with zone
{de,re}registration during fork(2).
The raw clock variant is slow (even relative to plain CLOCK_MONOTONIC),
whereas the coarse clock variant is faster than CLOCK_MONOTONIC, but
still has resolution (~1ms) that is adequate for our purposes.
This resolves#479.
Some applications wrap various system calls, and if they call the
allocator in their wrappers, unexpected reentry can result. This is not
a general solution (many other syscalls are spread throughout the code),
but this resolves a bootstrapping issue that is apparently common.
This resolves#443.
This works around malloc_conf not being properly initialized by at least
the cygwin toolchain. Prior build system changes to use
-Wl,--[no-]whole-archive may be necessary for malloc_conf resolution to
work properly as a non-weak symbol (not tested).
This is generally correct (no need for weak symbols since no jemalloc
library is involved in the link phase), and avoids linking problems
(apparently unininitialized non-NULL malloc_conf) when using cygwin with
gcc.
glibc defines its malloc implementation with several weak and strong
symbols:
strong_alias (__libc_calloc, __calloc) weak_alias (__libc_calloc, calloc)
strong_alias (__libc_free, __cfree) weak_alias (__libc_free, cfree)
strong_alias (__libc_free, __free) strong_alias (__libc_free, free)
strong_alias (__libc_malloc, __malloc) strong_alias (__libc_malloc, malloc)
The issue is not with the weak symbols, but that other parts of glibc
depend on __libc_malloc explicitly. Defining them in terms of jemalloc
API's allows the linker to drop glibc's malloc.o completely from the link,
and static linking no longer results in symbol collisions.
Another wrinkle: jemalloc during initialization calls sysconf to
get the number of CPU's. GLIBC allocates for the first time before
setting up isspace (and other related) tables, which are used by
sysconf. Instead, use the pthread API to get the number of
CPUs with GLIBC, which seems to work.
This resolves#442.
Rather than protecting dss operations with a mutex, use atomic
operations. This has negligible impact on synchronization overhead
during typical dss allocation, but is a substantial improvement for
extent_in_dss() and the newly added extent_dss_mergeable(), which can be
called multiple times during extent deallocations.
This change also has the advantage of avoiding tsd in deallocation paths
associated with purging, which resolves potential deadlocks during
thread exit due to attempted tsd resurrection.
This resolves#425.
Add spin_t and spin_{init,adaptive}(), which provide a simple
abstraction for adaptive spinning.
Adaptively spin during busy waits in bootstrapping and rtree node
initialization.
Simplify decay-based purging attempts to only be triggered when the
epoch is advanced, rather than every time purgeable memory increases.
In a correctly functioning system (not previously the case; see below),
this only causes a behavior difference if during subsequent purge
attempts the least recently used (LRU) purgeable memory extent is
initially too large to be purged, but that memory is reused between
attempts and one or more of the next LRU purgeable memory extents are
small enough to be purged. In practice this is an arbitrary behavior
change that is within the set of acceptable behaviors.
As for the purging fix, assure that arena->decay.ndirty is recorded
*after* the epoch advance and associated purging occurs. Prior to this
fix, it was possible for purging during epoch advance to cause a
substantially underrepresentative (arena->ndirty - arena->decay.ndirty),
i.e. the number of dirty pages attributed to the current epoch was too
low, and a series of unintended purges could result. This fix is also
relevant in the context of the simplification described above, but the
bug's impact would be limited to over-purging at epoch advances.
Instead, move the epoch backward in time. Additionally, add
nstime_monotonic() and use it in debug builds to assert that time only
goes backward if nstime_update() is using a non-monotonic time source.
Add missing #include <time.h>. The critical time facilities appear to
have been transitively included via unistd.h and sys/time.h, but in
principle this omission was capable of having caused
clock_gettime(CLOCK_MONOTONIC, ...) to have been overlooked in favor of
gettimeofday(), which in turn could cause spurious non-monotonic time
updates.
Refactor nstime_get() out of nstime_update() and add configure tests for
all variants.
Add CLOCK_MONOTONIC_RAW support (Linux-specific) and
mach_absolute_time() support (OS X-specific).
Do not fall back to clock_gettime(CLOCK_REALTIME, ...). This was a
fragile Linux-specific workaround, which we're unlikely to use at all
now that clock_gettime(CLOCK_MONOTONIC_RAW, ...) is supported, and if we
have no choice besides non-monotonic clocks, gettimeofday() is only
incrementally worse.
Avoid calling s2u() on raw extent sizes in extent_recycle().
Clamp psz2ind() (implemented as psz2ind_clamp()) when inserting/removing
into/from size-segregated extent heaps.
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.
847ff22 added a call to malloc_default_zone() before the main loop in
register_zone, effectively making malloc_default_zone() called twice
without any different outcome expected in the returned result.
It is also called once at the beginning, and a second time at the end
of the loop block.
Instead, call it only once per iteration.
Revert 245ae6036c (Support --with-lg-page
values larger than actual page size.), because it could cause VM map
fragmentation if the kernel grows mmap()ed memory downward.
This resolves#391.
Fix a fundamental extent_split_wrapper() bug in an error path.
Fix extent_recycle() to deregister unsplittable extents before leaking
them.
Relax xallocx() test assertions so that unsplittable extents don't cause
test failures.
rtree-based extent lookups remain more expensive than chunk-based run
lookups, but with this optimization the fast path slowdown is ~3 CPU
cycles per metadata lookup (on Intel Core i7-4980HQ), versus ~11 cycles
prior. The path caching speedup tends to degrade gracefully unless
allocated memory is spread far apart (as is the case when using a
mixture of sbrk() and mmap()).
rallocx() for an alignment-constrained request may end up with a
smaller-than-worst-case size if in-place reallocation succeeds due to
serendipitous alignment. In such cases, sampling may not happen.
When an allocation is large enough to trigger multiple dumps, use
modular math rather than subtraction to reset the interval counter.
Prior to this change, it was possible for a single allocation to cause
many subsequent allocations to all trigger profile dumps.
When updating usable size for a sampled object, try to cancel out
the difference between LARGE_MINCLASS and usable size from the interval
counter.
Fix huge_ralloc_no_move_expand() to update the extent's zeroed attribute
based on the intersection of the previous value and that of the newly
merged trailing extent.
Look up chunk metadata via the radix tree, rather than using
CHUNK_ADDR2BASE().
Propagate pointer's containing extent.
Minimize extent lookups by doing a single lookup (e.g. in free()) and
propagating the pointer's extent into nearly all the functions that may
need it.
This makes it possible to acquire short-term "ownership" of rtree
elements so that it is possible to read an extent pointer *and* read the
extent's contents with a guarantee that the element will not be modified
until the ownership is released. This is intended as a mechanism for
resolving rtree read/write races rather than as a way to lock extents.
Use pszind_t size classes rather than szind_t size classes, and always
reserve space for NPSIZES elements. This removes unused heaps that are
not multiples of the page size, and adds (currently) unused heaps for
all huge size classes, with the immediate benefit that the size of
arena_t allocations is constant (no longer dependent on chunk size).
These compute size classes and indices similarly to size2index(),
index2size() and s2u(), respectively, but using the subset of size
classes that are multiples of the page size. Note that pszind_t and
szind_t are not interchangeable.
Short-circuit commonly called witness functions so that they only
execute in debug builds, and remove equivalent guards from mutex
functions. This avoids pointless code execution in
witness_assert_lockless(), which is typically called twice per
allocation/deallocation function invocation.
Inline commonly called witness functions so that optimized builds can
completely remove calls as dead code.
b2c0d6322d (Add witness, a simple online
locking validator.) caused a broad propagation of tsd throughout the
internal API, but tsd_fetch() was designed to fail prior to tsd
bootstrapping. Fix this by splitting tsd_t into non-nullable tsd_t and
nullable tsdn_t, and modifying all internal APIs that do not critically
rely on tsd to take nullable pointers. Furthermore, add the
tsd_booted_get() function so that tsdn_fetch() can probe whether tsd
bootstrapping is complete and return NULL if not. All dangerous
conversions of nullable pointers are tsdn_tsd() calls that assert-fail
on invalid conversion.
This is a broader application of optimizations to malloc() and free() in
f4a0f32d34 (Fast-path improvement:
reduce # of branches and unnecessary operations.).
This resolves#321.
If the OS overcommits:
- Commit all mappings in pages_map() regardless of whether the caller
requested committed memory.
- Linux-specific: Specify MAP_NORESERVE to avoid
unfortunate interactions with heuristic overcommit mode during
fork(2).
This resolves#193.
Split arena_choose() into arena_[i]choose() and use arena_ichoose() for
arena lookup during internal allocation. This fixes huge_palloc() so
that it always succeeds during extent node allocation.
This regression was introduced by
66cd953514 (Do not allocate metadata via
non-auto arenas, nor tcaches.).
Change test-related mangling to simplify symbol filtering.
The following commands can be used to detect missing/obsolete symbol
mangling, with the caveat that the full set of symbols is based on the
union of symbols generated by all configurations, some of which are
platform-specific:
./autogen.sh --enable-debug --enable-prof --enable-lazy-lock
make all tests
nm -a lib/libjemalloc.a src/*.jet.o \
|grep " [TDBCR] " \
|awk '{print $3}' \
|sed -e 's/^\(je_\|jet_\(n_\)\?\)\([a-zA-Z0-9_]*\)/\3/g' \
|LC_COLLATE=C sort -u \
|grep -v \
-e '^\(malloc\|calloc\|posix_memalign\|aligned_alloc\|realloc\|free\)$' \
-e '^\(m\|r\|x\|s\|d\|sd\|n\)allocx$' \
-e '^mallctl\(\|nametomib\|bymib\)$' \
-e '^malloc_\(stats_print\|usable_size\|message\)$' \
-e '^\(memalign\|valloc\)$' \
-e '^__\(malloc\|memalign\|realloc\|free\)_hook$' \
-e '^pthread_create$' \
> /tmp/private_symbols.txt
During over-allocation in preparation for creating aligned mappings,
allocate one more page than necessary if PAGE is the actual page size,
so that trimming still succeeds even if the system returns a mapping
that has less than PAGE alignment. This allows compiling with e.g. 64
KiB "pages" on systems that actually use 4 KiB pages.
Note that for e.g. --with-lg-page=21, it is also necessary to increase
the chunk size (e.g. --with-malloc-conf=lg_chunk:22) so that there are
at least two "pages" per chunk. In practice this isn't a particularly
compelling configuration because so much (unusable) virtual memory is
dedicated to chunk headers.
Refactor ph to support configurable comparison functions. Use a cpp
macro code generation form equivalent to the rb macros so that pairing
heaps can be used for both run heaps and chunk heaps.
Remove per node parent pointers, and instead use leftmost siblings' prev
pointers to track parents.
Fix multi-pass sibling merging to iterate over intermediate results
using a FIFO, rather than a LIFO. Use this fixed sibling merging
implementation for both merge phases of the auxiliary twopass algorithm
(first merging the aux list, then replacing the root with its merged
children). This fixes both degenerate merge behavior and the potential
for deep recursion.
This regression was introduced by
6bafa6678f (Pairing heap).
This resolves#371.
Move chunk_dalloc_arena()'s implementation into chunk_dalloc_wrapper(),
so that if the dalloc hook fails, proper decommit/purge/retain cascading
occurs. This fixes three potential chunk leaks on OOM paths, one during
dss-based chunk allocation, one during chunk header commit (currently
relevant only on Windows), and one during rtree write (e.g. if rtree
node allocation fails).
Merge chunk_purge_arena() into chunk_purge_default() (refactor, no
change to functionality).
Variables s and slen are declared inside a switch statement, but outside
a case scope. clang reports these variable definitions as "unreachable",
though this is not really meaningful in this case. This is the only
-Wunreachable-code warning in jemalloc.
src/util.c:501:5 [-Wunreachable-code] code will never be executed
This resolves#364.
Use pairing heap instead of red black tree in arena runs_avail. The
extra links are unioned with the bitmap_t, so this change doesn't use
any extra memory.
Canaries show this change to be a 1% cpu win, and 2% latency win. In
particular, large free()s, and small bin frees are now O(1) (barring
coalescing).
I also tested changing bin->runs to be a pairing heap, but saw a much
smaller win, and it would mean increasing the size of arena_run_s by two
pointers, so I left that as an rb-tree for now.
Initial implementation of a twopass pairing heap with aux list.
Research papers linked in comments.
Where search/nsearch/last aren't needed, this gives much faster first(),
delete(), and insert(). Insert is O(1), and first/delete don't have to
walk the whole tree.
Also tested rb_old with parent pointers - it was better than the current
rb.h for memory loads, but still much worse than a pairing heap.
An array-based heap would be much faster if everything fits in memory,
but on a cold cache it has many more memory loads for most operations.
Add a cast to avoid comparing a ssize_t value to a uint64_t value that
is always larger than a 32-bit ssize_t. This silences an innocuous
compiler warning from e.g. gcc 4.2.1 about the comparison always having
the same result.
Prior to 767d85061a (Refactor arenas array
(fixes deadlock).), it was possible under some circumstances for
arena_get() to trigger recreation of the arenas cache during tsd
cleanup, and the arenas cache would then be leaked. In principle a
similar issue could still occur as a side effect of decay-based purging,
which calls arena_tdata_get(). Fix arenas_tdata_cleanup() by setting
tsd->arenas_tdata_bypass to true, so that arena_tdata_get() will
gracefully fail (an expected behavior) rather than recreating
tsd->arena_tdata.
Reported by Christopher Ferris <cferris@google.com>.
Add missing stats.arenas.<i>.{dss,lg_dirty_mult,decay_time}
initialization.
Fix stats.arenas.<i>.{pactive,pdirty} to read under the protection of
the arena mutex.
Fix stats.cactive accounting to always increase/decrease by multiples of
the chunk size, even for huge size classes that are not multiples of the
chunk size, e.g. {2.5, 3, 3.5, 5, 7} MiB with 2 MiB chunk size. This
regression was introduced by 155bfa7da1
(Normalize size classes.) and first released in 4.0.0.
This resolves#336.
This removes an implicit conversion from size_t to ssize_t. For cactive
decreases, the size_t value was intentionally underflowed to generate
"negative" values (actually positive values above the positive range of
ssize_t), and the conversion to ssize_t was undefined according to C
language semantics.
This regression was perpetuated by
1522937e9c (Fix the cactive statistic.)
and first release in 4.0.0, which in retrospect only fixed one of two
problems introduced by aa5113b1fd
(Refactor overly large/complex functions) and first released in 3.5.0.
For small bitmaps, a linear scan of the bitmap is slightly faster than
a tree search - bitmap_t is more compact, and there are fewer writes
since we don't have to propogate state transitions up the tree.
On x86_64 with the current settings, I'm seeing ~.5%-1% CPU improvement
in production canaries with this change.
The old tree code is left since 32bit sizes are much larger (and ffsl
smaller), and maybe the run sizes will change in the future.
This resolves#339.
Add HUGE_MAXCLASS overflow checks that are specific to heap profiling
code paths. This fixes test failures that were introduced by
0c516a00c4 (Make *allocx() size class
overflow behavior defined.).
Refactor the arenas array, which contains pointers to all extant arenas,
such that it starts out as a sparse array of maximum size, and use
double-checked atomics-based reads as the basis for fast and simple
arena_get(). Additionally, reduce arenas_lock's role such that it only
protects against arena initalization races. These changes remove the
possibility for arena lookups to trigger locking, which resolves at
least one known (fork-related) deadlock.
This resolves#315.
Fix arena_size arena_new() computation to incorporate
runs_avail_nclasses elements for runs_avail, rather than
(runs_avail_nclasses - 1) elements. Since offsetof(arena_t, runs_avail)
is used rather than sizeof(arena_t) for the first term of the
computation, all of the runs_avail elements must be added into the
second term.
This bug was introduced (by Jason Evans) while merging pull request #330
as 3417a304cc (Separate arena_avail
trees).
Merge of 3417a304cc looks like a small
bug: first_best_fit doesn't scan through all the classes, since ind is
offset from runs_avail_nclasses by run_avail_bias.
Attempt mmap-based in-place huge reallocation by plumbing new_addr into
chunk_alloc_mmap(). This can dramatically speed up incremental huge
reallocation.
This resolves#335.
Separate run trees by index, replacing the previous quantize logic.
Quantization by index is now performed only on insertion / removal from
the tree, and not on node comparison, saving some cpu. This also means
we don't have to dereference the miscelm* pointers, saving half of the
memory loads from miscelms/mapbits that have fallen out of cache. A
linear scan of the indicies appears to be fast enough.
The only cost of this is an extra tree array in each arena.
In practice this bug had limited impact (and then only by increasing
chunk fragmentation) because run_quantize_ceil() returned correct
results except for inputs that could only arise from aligned allocation
requests that required more than page alignment.
This bug existed in the original run quantization implementation, which
was introduced by 8a03cf039c (Implement
cache index randomization for large allocations.).
Use a single uint64_t in nstime_t to store nanoseconds rather than using
struct timespec. This reduces fragility around conversions between long
and uint64_t, especially missing casts that only cause problems on
32-bit platforms.
This is an alternative to the existing ratio-based unused dirty page
purging, and is intended to eventually become the sole purging
mechanism.
Add mallctls:
- opt.purge
- opt.decay_time
- arena.<i>.decay
- arena.<i>.decay_time
- arenas.decay_time
- stats.arenas.<i>.decay_time
This resolves#325.
