Commit Graph

92 Commits

Author SHA1 Message Date
Jason Evans
45186f0c07 Refactor arena_mapbits unzeroed flag management.
Only set the unzeroed flag when initializing the entire mapbits entry,
rather than mutating just the unzeroed bit.  This simplifies the
possible mapbits state transitions.
2015-08-10 23:03:34 -07:00
Jason Evans
de249c8679 Arena chunk decommit cleanups and fixes.
Decommit arena chunk header during chunk deallocation if the rest of the
chunk is decommitted.
2015-08-10 17:13:59 -07:00
Jason Evans
8fadb1a8c2 Implement chunk hook support for page run commit/decommit.
Cascade from decommit to purge when purging unused dirty pages, so that
it is possible to decommit cleaned memory rather than just purging.  For
non-Windows debug builds, decommit runs rather than purging them, since
this causes access of deallocated runs to segfault.

This resolves #251.
2015-08-07 00:50:58 -07:00
Jason Evans
b49a334a64 Generalize chunk management hooks.
Add the "arena.<i>.chunk_hooks" mallctl, which replaces and expands on
the "arena.<i>.chunk.{alloc,dalloc,purge}" mallctls.  The chunk hooks
allow control over chunk allocation/deallocation, decommit/commit,
purging, and splitting/merging, such that the application can rely on
jemalloc's internal chunk caching and retaining functionality, yet
implement a variety of chunk management mechanisms and policies.

Merge the chunks_[sz]ad_{mmap,dss} red-black trees into
chunks_[sz]ad_retained.  This slightly reduces how hard jemalloc tries
to honor the dss precedence setting; prior to this change the precedence
setting was also consulted when recycling chunks.

Fix chunk purging.  Don't purge chunks in arena_purge_stashed(); instead
deallocate them in arena_unstash_purged(), so that the dirty memory
linkage remains valid until after the last time it is used.

This resolves #176 and #201.
2015-08-03 21:49:02 -07:00
Jason Evans
aa2826621e Revert to first-best-fit run/chunk allocation.
This effectively reverts 97c04a9383 (Use
first-fit rather than first-best-fit run/chunk allocation.).  In some
pathological cases, first-fit search dominates allocation time, and it
also tends not to converge as readily on a steady state of memory
layout, since precise allocation order has a bigger effect than for
first-best-fit.
2015-07-15 17:15:19 -07:00
Jason Evans
241abc601b Fix size class overflow handling when profiling is enabled.
Fix size class overflow handling for malloc(), posix_memalign(),
memalign(), calloc(), and realloc() when profiling is enabled.

Remove an assertion that erroneously caused arena_sdalloc() to fail when
profiling was enabled.

This resolves #232.
2015-06-23 18:56:14 -07:00
Jason Evans
0a9f9a4d51 Convert arena_maybe_purge() recursion to iteration.
This resolves #235.
2015-06-22 18:50:58 -07:00
Jason Evans
6591ff09d8 Fix arena_dalloc() performance regression.
Take into account large_pad when computing whether to pass the
deallocation request to tcache_dalloc_large(), so that the largest
cacheable size makes it back to tcache.  This regression was introduced
by 8a03cf039c (Implement cache index
randomization for large allocations.).
2015-05-19 17:44:45 -07:00
Jason Evans
c451831264 Fix type punning in calls to atomic operation functions. 2015-05-07 22:35:40 -07:00
Jason Evans
8a03cf039c Implement cache index randomization for large allocations.
Extract szad size quantization into {extent,run}_quantize(), and .
quantize szad run sizes to the union of valid small region run sizes and
large run sizes.

Refactor iteration in arena_run_first_fit() to use
run_quantize{,_first,_next(), and add support for padded large runs.

For large allocations that have no specified alignment constraints,
compute a pseudo-random offset from the beginning of the first backing
page that is a multiple of the cache line size.  Under typical
configurations with 4-KiB pages and 64-byte cache lines this results in
a uniform distribution among 64 page boundary offsets.

Add the --disable-cache-oblivious option, primarily intended for
performance testing.

This resolves #13.
2015-05-06 13:27:39 -07:00
Jason Evans
562d266511 Add the "stats.arenas.<i>.lg_dirty_mult" mallctl. 2015-03-24 16:41:38 -07:00
Jason Evans
8d6a3e8321 Implement dynamic per arena control over dirty page purging.
Add mallctls:
- arenas.lg_dirty_mult is initialized via opt.lg_dirty_mult, and can be
  modified to change the initial lg_dirty_mult setting for newly created
  arenas.
- arena.<i>.lg_dirty_mult controls an individual arena's dirty page
  purging threshold, and synchronously triggers any purging that may be
  necessary to maintain the constraint.
- arena.<i>.chunk.purge allows the per arena dirty page purging function
  to be replaced.

This resolves #93.
2015-03-18 18:55:33 -07:00
Jason Evans
f5c8f37259 Normalize rdelm/rd structure field naming. 2015-03-10 18:29:49 -07:00
Jason Evans
38e42d311c Refactor dirty run linkage to reduce sizeof(extent_node_t). 2015-03-10 18:15:40 -07:00
Jason Evans
97c04a9383 Use first-fit rather than first-best-fit run/chunk allocation.
This tends to more effectively pack active memory toward low addresses.
However, additional tree searches are required in many cases, so whether
this change stands the test of time will depend on real-world
benchmarks.
2015-03-06 20:21:41 -08:00
Jason Evans
99bd94fb65 Fix chunk cache races.
These regressions were introduced by
ee41ad409a (Integrate whole chunks into
unused dirty page purging machinery.).
2015-02-18 16:40:53 -08:00
Jason Evans
738e089a2e Rename "dirty chunks" to "cached chunks".
Rename "dirty chunks" to "cached chunks", in order to avoid overloading
the term "dirty".

Fix the regression caused by 339c2b23b2
(Fix chunk_unmap() to propagate dirty state.), and actually address what
that change attempted, which is to only purge chunks once, and propagate
whether zeroed pages resulted into chunk_record().
2015-02-18 01:15:50 -08:00
Jason Evans
eafebfdfbe Remove obsolete type arena_chunk_miscelms_t. 2015-02-17 16:12:31 -08:00
Jason Evans
ee41ad409a Integrate whole chunks into unused dirty page purging machinery.
Extend per arena unused dirty page purging to manage unused dirty chunks
in aaddtion to unused dirty runs.  Rather than immediately unmapping
deallocated chunks (or purging them in the --disable-munmap case), store
them in a separate set of trees, chunks_[sz]ad_dirty.  Preferrentially
allocate dirty chunks.  When excessive unused dirty pages accumulate,
purge runs and chunks in ingegrated LRU order (and unmap chunks in the
--enable-munmap case).

Refactor extent_node_t to provide accessor functions.
2015-02-16 21:02:17 -08:00
Jason Evans
40ab8f98e4 Remove more obsolete (incorrect) assertions.
This regression was introduced by
88fef7ceda (Refactor huge_*() calls into
arena internals.), and went undetected because of the --enable-debug
regression.
2015-02-15 20:26:45 -08:00
Jason Evans
cb9b44914e Remove obsolete (incorrect) assertions.
This regression was introduced by
88fef7ceda (Refactor huge_*() calls into
arena internals.), and went undetected because of the --enable-debug
regression.
2015-02-15 20:13:28 -08:00
Jason Evans
41cfe03f39 If MALLOCX_ARENA(a) is specified, use it during tcache fill. 2015-02-13 15:28:56 -08:00
Jason Evans
5f7140b045 Make prof_tctx accesses atomic.
Although exceedingly unlikely, it appears that writes to the prof_tctx
field of arena_chunk_map_misc_t could be reordered such that a stale
value could be read during deallocation, with profiler metadata
corruption and invalid pointer dereferences being the most likely
effects.
2015-02-12 15:54:53 -08:00
Jason Evans
88fef7ceda Refactor huge_*() calls into arena internals.
Make redirects to the huge_*() API the arena code's responsibility,
since arenas now take responsibility for all allocation sizes.
2015-02-12 14:06:37 -08:00
Jason Evans
cbf3a6d703 Move centralized chunk management into arenas.
Migrate all centralized data structures related to huge allocations and
recyclable chunks into arena_t, so that each arena can manage huge
allocations and recyclable virtual memory completely independently of
other arenas.

Add chunk node caching to arenas, in order to avoid contention on the
base allocator.

Use chunks_rtree to look up huge allocations rather than a red-black
tree.  Maintain a per arena unsorted list of huge allocations (which
will be needed to enumerate huge allocations during arena reset).

Remove the --enable-ivsalloc option, make ivsalloc() always available,
and use it for size queries if --enable-debug is enabled.  The only
practical implications to this removal are that 1) ivsalloc() is now
always available during live debugging (and the underlying radix tree is
available during core-based debugging), and 2) size query validation can
no longer be enabled independent of --enable-debug.

Remove the stats.chunks.{current,total,high} mallctls, and replace their
underlying statistics with simpler atomically updated counters used
exclusively for gdump triggering.  These statistics are no longer very
useful because each arena manages chunks independently, and per arena
statistics provide similar information.

Simplify chunk synchronization code, now that base chunk allocation
cannot cause recursive lock acquisition.
2015-02-12 00:15:56 -08:00
Jason Evans
1cb181ed63 Implement explicit tcache support.
Add the MALLOCX_TCACHE() and MALLOCX_TCACHE_NONE macros, which can be
used in conjunction with the *allocx() API.

Add the tcache.create, tcache.flush, and tcache.destroy mallctls.

This resolves #145.
2015-02-09 17:44:48 -08:00
Jason Evans
4581b97809 Implement metadata statistics.
There are three categories of metadata:

- Base allocations are used for bootstrap-sensitive internal allocator
  data structures.
- Arena chunk headers comprise pages which track the states of the
  non-metadata pages.
- Internal allocations differ from application-originated allocations
  in that they are for internal use, and that they are omitted from heap
  profiles.

The metadata statistics comprise the metadata categories as follows:

- stats.metadata: All metadata -- base + arena chunk headers + internal
  allocations.
- stats.arenas.<i>.metadata.mapped: Arena chunk headers.
- stats.arenas.<i>.metadata.allocated: Internal allocations.  This is
  reported separately from the other metadata statistics because it
  overlaps with the allocated and active statistics, whereas the other
  metadata statistics do not.

Base allocations are not reported separately, though their magnitude can
be computed by subtracting the arena-specific metadata.

This resolves #163.
2015-01-23 23:34:43 -08:00
Jason Evans
e12eaf93dc Style and spelling fixes. 2014-12-08 16:34:04 -08:00
Jason Evans
d7a9bab92d Fix arena_sdalloc() to use promoted size (second attempt).
Unlike the preceeding attempted fix, this version avoids the potential
for converting an invalid bin index to a size class.
2014-10-31 22:26:24 -07:00
Jason Evans
6da2e9d4f6 Fix arena_sdalloc() to use promoted size. 2014-10-31 17:08:13 -07:00
Jason Evans
9b41ac909f Fix huge allocation statistics. 2014-10-14 22:20:00 -07:00
Jason Evans
3c4d92e82a Add per size class huge allocation statistics.
Add per size class huge allocation statistics, and normalize various
stats:
- Change the arenas.nlruns type from size_t to unsigned.
- Add the arenas.nhchunks and arenas.hchunks.<i>.size mallctl's.
- Replace the stats.arenas.<i>.bins.<j>.allocated mallctl with
  stats.arenas.<i>.bins.<j>.curregs .
- Add the 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 mallctl's.
2014-10-12 23:02:10 -07:00
Jason Evans
381c23dd9d Remove arena_dalloc_bin_run() clean page preservation.
Remove code in arena_dalloc_bin_run() that preserved the "clean" state
of trailing clean pages by splitting them into a separate run during
deallocation.  This was a useful mechanism for reducing dirty page
churn when bin runs comprised many pages, but bin runs are now quite
small.

Remove the nextind field from arena_run_t now that it is no longer
needed, and change arena_run_t's bin field (arena_bin_t *) to binind
(index_t).  These two changes remove 8 bytes of chunk header overhead
per page, which saves 1/512 of all arena chunk memory.
2014-10-10 23:01:03 -07:00
Jason Evans
fc0b3b7383 Add configure options.
Add:
  --with-lg-page
  --with-lg-page-sizes
  --with-lg-size-class-group
  --with-lg-quantum

Get rid of STATIC_PAGE_SHIFT, in favor of directly setting LG_PAGE.

Fix various edge conditions exposed by the configure options.
2014-10-09 22:44:37 -07:00
Jason Evans
8bb3198f72 Refactor/fix arenas manipulation.
Abstract arenas access to use arena_get() (or a0get() where appropriate)
rather than directly reading e.g. arenas[ind].  Prior to the addition of
the arenas.extend mallctl, the worst possible outcome of directly
accessing arenas was a stale read, but arenas.extend may allocate and
assign a new array to arenas.

Add a tsd-based arenas_cache, which amortizes arenas reads.  This
introduces some subtle bootstrapping issues, with tsd_boot() now being
split into tsd_boot[01]() to support tsd wrapper allocation
bootstrapping, as well as an arenas_cache_bypass tsd variable which
dynamically terminates allocation of arenas_cache itself.

Promote a0malloc(), a0calloc(), and a0free() to be generally useful for
internal allocation, and use them in several places (more may be
appropriate).

Abstract arena->nthreads management and fix a missing decrement during
thread destruction (recent tsd refactoring left arenas_cleanup()
unused).

Change arena_choose() to propagate OOM, and handle OOM in all callers.
This is important for providing consistent allocation behavior when the
MALLOCX_ARENA() flag is being used.  Prior to this fix, it was possible
for an OOM to result in allocation silently allocating from a different
arena than the one specified.
2014-10-07 23:14:57 -07:00
Jason Evans
155bfa7da1 Normalize size classes.
Normalize size classes to use the same number of size classes per size
doubling (currently hard coded to 4), across the intire range of size
classes.  Small size classes already used this spacing, but in order to
support this change, additional small size classes now fill [4 KiB .. 16
KiB).  Large size classes range from [16 KiB .. 4 MiB).  Huge size
classes now support non-multiples of the chunk size in order to fill (4
MiB .. 16 MiB).
2014-10-06 01:45:13 -07:00
Daniel Micay
a95018ee81 Attempt to expand huge allocations in-place.
This adds support for expanding huge allocations in-place by requesting
memory at a specific address from the chunk allocator.

It's currently only implemented for the chunk recycling path, although
in theory it could also be done by optimistically allocating new chunks.
On Linux, it could attempt an in-place mremap. However, that won't work
in practice since the heap is grown downwards and memory is not unmapped
(in a normal build, at least).

Repeated vector reallocation micro-benchmark:

    #include <string.h>
    #include <stdlib.h>

    int main(void) {
        for (size_t i = 0; i < 100; i++) {
            void *ptr = NULL;
            size_t old_size = 0;
            for (size_t size = 4; size < (1 << 30); size *= 2) {
                ptr = realloc(ptr, size);
                if (!ptr) return 1;
                memset(ptr + old_size, 0xff, size - old_size);
                old_size = size;
            }
            free(ptr);
        }
    }

The glibc allocator fails to do any in-place reallocations on this
benchmark once it passes the M_MMAP_THRESHOLD (default 128k) but it
elides the cost of copies via mremap, which is currently not something
that jemalloc can use.

With this improvement, jemalloc still fails to do any in-place huge
reallocations for the first outer loop, but then succeeds 100% of the
time for the remaining 99 iterations. The time spent doing allocations
and copies drops down to under 5%, with nearly all of it spent doing
purging + faulting (when huge pages are disabled) and the array memset.

An improved mremap API (MREMAP_RETAIN - #138) would be far more general
but this is a portable optimization and would still be useful on Linux
for xallocx.

Numbers with transparent huge pages enabled:

glibc (copies elided via MREMAP_MAYMOVE): 8.471s

jemalloc: 17.816s
jemalloc + no-op madvise: 13.236s

jemalloc + this commit: 6.787s
jemalloc + this commit + no-op madvise: 6.144s

Numbers with transparent huge pages disabled:

glibc (copies elided via MREMAP_MAYMOVE): 15.403s

jemalloc: 39.456s
jemalloc + no-op madvise: 12.768s

jemalloc + this commit: 15.534s
jemalloc + this commit + no-op madvise: 6.354s

Closes #137
2014-10-05 14:47:01 -07:00
Jason Evans
551ebc4364 Convert to uniform style: cond == false --> !cond 2014-10-03 10:16:09 -07:00
Jason Evans
0c5dd03e88 Move small run metadata into the arena chunk header.
Move small run metadata into the arena chunk header, with multiple
expected benefits:
- Lower run fragmentation due to reduced run sizes; runs are more likely
  to completely drain when there are fewer total regions.
- Improved cache behavior.  Prior to this change, run headers were
  always page-aligned, which put extra pressure on some CPU cache sets.
  The degree to which this was a problem was hardware dependent, but it
  likely hurt some even for the most advanced modern hardware.
- Buffer overruns/underruns are less likely to corrupt allocator
  metadata.
- Size classes between 4 KiB and 16 KiB become reasonable to support
  without any special handling, and the runs are small enough that dirty
  unused pages aren't a significant concern.
2014-09-29 01:31:39 -07:00
Jason Evans
5460aa6f66 Convert all tsd variables to reside in a single tsd structure. 2014-09-23 02:36:08 -07:00
Jason Evans
9c640bfdd4 Apply likely()/unlikely() to allocation/deallocation fast paths. 2014-09-11 17:01:58 -07:00
Daniel Micay
4cfe55166e Add support for sized deallocation.
This adds a new `sdallocx` function to the external API, allowing the
size to be passed by the caller.  It avoids some extra reads in the
thread cache fast path.  In the case where stats are enabled, this
avoids the work of calculating the size from the pointer.

An assertion validates the size that's passed in, so enabling debugging
will allow users of the API to debug cases where an incorrect size is
passed in.

The performance win for a contrived microbenchmark doing an allocation
and immediately freeing it is ~10%.  It may have a different impact on a
real workload.

Closes #28
2014-09-08 17:34:24 -07:00
Jason Evans
b718cf77e9 Optimize [nmd]alloc() fast paths.
Optimize [nmd]alloc() fast paths such that the (flags == 0) case is
streamlined, flags decoding only happens to the minimum degree
necessary, and no conditionals are repeated.
2014-09-07 14:40:19 -07:00
Qinfan Wu
ff6a31d3b9 Refactor chunk map.
Break the chunk map into two separate arrays, in order to improve cache
locality. This is related to issue #23.
2014-09-04 22:22:52 -07:00
Jason Evans
602c8e0971 Implement per thread heap profiling.
Rename data structures (prof_thr_cnt_t-->prof_tctx_t,
prof_ctx_t-->prof_gctx_t), and convert to storing a prof_tctx_t for
sampled objects.

Convert PROF_ALLOC_PREP() to prof_alloc_prep(), since precise backtrace
depth within jemalloc functions is no longer an issue (pprof prunes
irrelevant frames).

Implement mallctl's:
- prof.reset implements full sample data reset, and optional change of
  sample interval.
- prof.lg_sample reads the current sample interval (opt.lg_prof_sample
  was the permanent source of truth prior to prof.reset).
- thread.prof.name provides naming capability for threads within heap
  profile dumps.
- thread.prof.active makes it possible to activate/deactivate heap
  profiling for individual threads.

Modify the heap dump files to contain per thread heap profile data.
This change is incompatible with the existing pprof, which will require
enhancements to read and process the enriched data.
2014-08-19 21:31:16 -07:00
Jason Evans
070b3c3fbd Fix and refactor runs_dirty-based purging.
Fix runs_dirty-based purging to also purge dirty pages in the spare
chunk.

Refactor runs_dirty manipulation into arena_dirty_{insert,remove}(), and
move the arena->ndirty accounting into those functions.

Remove the u.ql_link field from arena_chunk_map_t, and get rid of the
enclosing union for u.rb_link, since only rb_link remains.

Remove the ndirty field from arena_chunk_t.
2014-08-14 14:45:58 -07:00
Qinfan Wu
e8a2fd83a2 arena->npurgatory is no longer needed since we drop arena's lock
after stashing all the purgeable runs.
2014-08-12 09:50:01 -07:00
Qinfan Wu
90737fcda1 Remove chunks_dirty tree, nruns_avail and nruns_adjac since we no
longer need to maintain the tree for dirty page purging.
2014-08-12 09:50:00 -07:00
Qinfan Wu
04d60a132b Maintain all the dirty runs in a linked list for each arena 2014-08-12 09:50:00 -07:00
Richard Diamond
9c3a10fdf6 Try to use __builtin_ffsl if ffsl is unavailable.
Some platforms (like those using Newlib) don't have ffs/ffsl.  This
commit adds a check to configure.ac for __builtin_ffsl if ffsl isn't
found.  __builtin_ffsl performs the same function as ffsl, and has the
added benefit of being available on any platform utilizing
Gcc-compatible compiler.

This change does not address the used of ffs in the MALLOCX_ARENA()
macro.
2014-06-02 07:44:50 -07:00