Tweak chunk purge order to purge unfragmented chunks from high to low
memory.  This facilitates dirty run reuse.

Based on http://www.canonware.com/pipermail/jemalloc-discuss/2012-March/000164.html

Purge unused dirty pages in an order that first performs clean/dirty run
defragmentation, in order to mitigate available run fragmentation.

Remove the limitation that prevented purging unless at least one chunk
worth of dirty pages had accumulated in an arena.  This limitation was
intended to avoid excessive purging for small applications, but the
threshold was arbitrary, and the effect of questionable utility.

Relax opt_lg_dirty_mult from 5 to 3.  This compensates for increased
likelihood of allocating clean runs, given the same ratio of clean:dirty
runs, and reduces the potential for repeated purging in pathological
large malloc/free loops that push the active:dirty page ratio just over
the purge threshold.

Fix deadlock in the arenas.purge mallctl due to recursive mutex
acquisition.

Fix dss/mmap allocation precedence code to use recyclable mmap memory
only after primary dss allocation fails.

Add ctl_mutex proection to arena_i_dss_ctl(), since ctl_stats.narenas is
accessed.

Add the "arenas.extend" mallctl, so that it is possible to create new
arenas that are outside the set that jemalloc automatically multiplexes
threads onto.

Add the ALLOCM_ARENA() flag for {,r,d}allocm(), so that it is possible
to explicitly allocate from a particular arena.

Add the "opt.dss" mallctl, which controls the default precedence of dss
allocation relative to mmap allocation.

Add the "arena.<i>.dss" mallctl, which makes it possible to set the
default dss precedence on a per arena or global basis.

Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge".

Add the "stats.arenas.<i>.dss" mallctl.

Mozilla build hides everything by default using visibility pragma and
unhides only explicitly listed headers. But this doesn't work on FreeBSD
because _pthread_mutex_init_calloc_cb is neither documented nor exposed
via any header.

Use JEMALLOC_USABLE_SIZE_CONST for the malloc_usable_size()
implementation as well as the prototype, for consistency's sake.

Drop const from malloc_usable_size() argument on Linux, in order to
match the prototype in Linux's malloc.h.

Fix mutex acquisition order inversion for the chunks rtree and the base
mutex.  Chunks rtree acquisition was introduced by the previous commit,
so this bug was short-lived.

Add a library constructor for jemalloc that initializes the allocator.
This fixes a race that could occur if threads were created by the main
thread prior to any memory allocation, followed by fork(2), and then
memory allocation in the child process.

Fix the prefork/postfork functions to acquire/release the ctl, prof, and
rtree mutexes.  This fixes various fork() child process deadlocks, but
one possible deadlock remains (intentionally) unaddressed: prof
backtracing can acquire runtime library mutexes, so deadlock is still
possible if heap profiling is enabled during fork().  This deadlock is
known to be a real issue in at least the case of libgcc-based
backtracing.

Reported by tfengjun.

mlockall(2) can cause purging via madvise(2) to fail.  Fix purging code
to check whether madvise() succeeded, and base zeroed page metadata on
the result.

Reported by Olivier Lecomte.

Reported by Corey Richardson.

Submitted by Jory Pratt.

Remove const from __*_hook variable declarations, so that glibc can
modify them during process forking.

Disable tcache by default if running inside Valgrind, in order to avoid
making unallocated objects appear reachable to Valgrind.

Auto-detect whether running inside Valgrind, thus removing the need to
manually specify MALLOC_CONF=valgrind:true.

Fix prof_realloc() to not call prof_ctx_set() if a sampled object is
being freed via realloc(p, 0).

Avoid mutex operations in _malloc_{pre,post}fork() unless jemalloc has
been initialized.

Reported by David Xu.

Refactor code such that arena_mapbits_{large,small}_set() always
preserves the unzeroed flag, and manually manipulate the unzeroed flag
in the one case where it actually gets reset (in arena_chunk_purge()).
This fixes unzeroed preservation bugs in arena_run_split() and
arena_ralloc_large_grow().  These bugs caused large calloc() to return
non-zeroed memory under some circumstances.

Refactor duplicated arena_run_alloc() code into
arena_run_alloc_helper().

Generalize "stats.mapped" documentation to state that all inactive
chunks are omitted, now that it is possible for mmap'ed chunks to be
omitted in addition to DSS chunks.

da99e311 removed attributes on je_memalign and je_valloc, while they didn't
have a definition in the jemalloc.h header, thus making them non-exported.
Export them again, by defining them in the jemalloc.h header.

It doesn't allow the je_malloc_message and je_malloc_conf symbols to be
overridden when linking statically.

Add the --enable-mremap option, and disable the use of mremap(2) by
default, for the same reason that freeing chunks via munmap(2) is
disabled by default on Linux: semi-permanent VM map fragmentation.

Fix chunk_recycle() to correctly compute trailsize and re-insert
trailing chunks.  This fixes a major virtual memory leak.

Simplify chunk_record() to avoid dropping/re-acquiring chunks_mtx.

Simplify chunk_alloc_mmap() to no longer attempt map extension.  The
extra complexity isn't warranted, because although in the success case
it saves one system call as compared to immediately falling back to
chunk_alloc_mmap_slow(), it also makes the failure case even more
expensive.  This simplification removes two bugs:

- For Windows platforms, pages_unmap() wasn't being called for unaligned
  mappings prior to falling back to chunk_alloc_mmap_slow().  This
  caused permanent virtual memory leaks.
- For non-Windows platforms, alignment greater than chunksize caused
  pages_map() to be called with size 0 when attempting map extension.
  This always resulted in an mmap() error, and subsequent fallback to
  chunk_alloc_mmap_slow().

Fix a base allocator deadlock due to chunk_recycle() calling back into
the base allocator.

In the alloca() case, this fails to be the right size.

We don't use MSVC's printf, but ours, and it doesn't support the I32 and I64
prefixes.