Lines Matching refs:is

13     allocations when --enable-cache-oblivious configure option is enabled.
15     when resizing from/to a size class that is not a multiple of the chunk size.
27     allocations.  xallocx() is in general incapable of moving small allocations,
30     alignment is non-zero.
36   This is a bugfix release that is somewhat high risk due to the amount of
44   - Avoid a chunk metadata read in arena_prof_tctx_set(), since it is in the
45     fast path when heap profiling is enabled.  Additionally, split a special
53   - Fix TLS configuration such that it is enabled by default for platforms on
57     cleanup functions even after arenas_cache is torn down.
86     + Don't assume that /bin/sh is appropriate when running size_classes.sh
96   cumulative effect is substantial.
101     is now configurable via the --with-lg-size-class-group option.  Also add the
106       is improved because there are far fewer size classes, and therefore
108     + Internal fragmentation is limited to 20% for all but the smallest size
116     control which tcache is used for each operation.
125     function as general purpose independent allocators.  This is important in
141     mallctl), so that it is possible to separately enable junk filling for
156     optimization over dallocx() is the removal of a metadata read, which often
187     that only a single TLS read is necessary per call into the public API.
192     allocation, because a global data structure is critical for determining
214   - Remove the *allocm() API, which is superseded by the *allocx() API.
220     detects whether it is running inside Valgrind.
236     calloc(), and realloc() when profiling is enabled.
238     "secondary" precedence is specified, but sbrk(2) is not supported.
240   - Ensure the default purgeable zone is after the default zone on OS X.
278   - Fix junk filling for mremap(2)-based huge reallocation.  This is only
280   - Avoid compilation failure if 'restrict' C99 keyword is not supported by the
282   - Add a configure test for SSE2 rather than assuming it is usable on i686
297   - Add the *allocx() API, which is a successor to the experimental *allocm()
301     allocm()/rallocm() share with posix_memalign().  Note that *allocm() is
306   - Unless heap profiling is enabled, disable floating point code and don't link
328   - Fix huge deallocation to junk fill when munmap is disabled.
334     --with-private-namespace is specified).
338     this error path is never executed.
361   This version is essentially a small bugfix release, but the addition of
414     is undefined.
451   - Fix heap profiling crash if sampled object is freed via realloc(p, 0).
460   Although this version adds some major new features, the primary focus is on
462   of which is not reflected in the ChangeLog.  This is the first release to
464   result is a more broadly useful allocator (see the git revision history for
526   - Fix chunk_alloc_dss() to stop claiming memory is zeroed.  This bug could
537     is 0.
545   - Don't mangle pthread_create() when --with-private-namespace is specified.
550   - Fix huge_ralloc() race when using mremap(2).  This is a serious bug that
569   - Fix off-by-one backtracing issues (make sure that prof_alloc_prep() is
581   - Add the --with-private-namespace option.  This is a workaround for symbol
605   - Fix symbol visibility so that it is also set on OS X.
609   - Don't use libgcc-based backtracing except on x64, where it is known to work.
680   - Replace JEMALLOC_OPTIONS/JEMALLOC_PROF_PREFIX with MALLOC_CONF, which is
683     is now:
689     This is mainly of interest for OS X, which does not support TLS, but has a
694   - Make cumulative heap profiling data optional, so that it is possible to
740   - Fix compilation when --enable-fill is specified.
776     (ASLR) is in use.