This introduces a backport of C11 atomics. It has four implementations; ranked
in order of preference, they are:
- GCC/Clang __atomic builtins
- GCC/Clang __sync builtins
- MSVC _Interlocked builtins
- C11 atomics, from <stdatomic.h>
The primary advantages are:
- Close adherence to the standard API gives us a defined memory model.
- Type safety: atomic objects are now separate types from non-atomic ones, so
that it's impossible to mix up atomic and non-atomic updates (which is
undefined behavior that compilers are starting to take advantage of).
- Efficiency: we can specify ordering for operations, avoiding fences and
atomic operations on strongly ordered architectures (example:
`atomic_write_u32(ptr, val);` involves a CAS loop, whereas
`atomic_store(ptr, val, ATOMIC_RELEASE);` is a plain store.
This diff leaves in the current atomics API (implementing them in terms of the
backport). This lets us transition uses over piecemeal.
Testing:
This is by nature hard to test. I've manually tested the first three options on
Linux on gcc by futzing with the #defines manually, on freebsd with gcc and
clang, on MSVC, and on OS X with clang. All of these were x86 machines though,
and we don't have any test infrastructure set up for non-x86 platforms.
In the long term, we'll transition to C99-style inline semantics. In the
short-term, this will allow both styles to coexist without breaking one another.
This avoids signed/unsigned comparison warnings when specifying integer
constants as inputs.
Clean up whitespace and add clarifying parentheses for
CONF_HANDLE_SIZE_T(opt_lg_chunk, ...).
Detect whether chunks start off as THP-capable by default (according to
the state of /sys/kernel/mm/transparent_hugepage/enabled), and use this
as the basis for whether to call pages_nohuge() once per chunk during
first purge of any of the chunk's page runs.
Add the --disable-thp configure option, as well as the the opt.thp
mallctl.
This resolves#541.
Convert CFLAGS to be a concatenation:
CFLAGS := CONFIGURE_CFLAGS SPECIFIED_CFLAGS EXTRA_CFLAGS
This ordering makes it possible to override the flags set by the
configure script both during and after configuration, with CFLAGS and
EXTRA_CFLAGS, respectively.
This resolves#619.
When multiple threads calling stats_print, race could happen as we read the
counters in separate mallctl calls; and the removed assertion could fail when
other operations happened in between the mallctl calls. For simplicity, output
"race" in the utilization field in this case.
This resolves#616.
Fix lg_chunk clamping to take into account cache-oblivious large
allocation. This regression only resulted in incorrect behavior if
!config_fill (false unless --disable-fill specified) and
config_cache_oblivious (true unless --disable-cache-oblivious
specified).
This regression was introduced by
8a03cf039c (Implement cache index
randomization for large allocations.), which was first released in
4.0.0.
This resolves#555.
Remove obsolete unit test scaffolding for extent quantization. Remove
redundant assertions. Add an assertion to
extents_first_best_fit_locked() that should help prevent aligned
allocation regressions.
Implement and test a JSON validation parser. Use the parser to validate
JSON output from malloc_stats_print(), with a significant subset of
supported output options.
This resolves#583.
Fix chunk_alloc_dss() to account for bytes that are not a multiple of
the chunk size. This regression was introduced by
e2bcf037d4 (Make dss operations
lockless.), which was first released in 4.3.0.
We don't touch witness at all when config_debug == false. Let's only pay the
memory cost in malloc_mutex_s when needed. Note that when !config_debug, we keep
the field in a union so that we don't have to do #ifdefs in multiple places.
In some cases the prof machinery allocates (in order to modify the
bt2gctx hash table), and such operations are synchronized via
bt2gctx_mtx. Rather than asserting that no locks are held on entry
into functions that may call prof_gdump(), make the weaker assertion
that no "core" locks are held. The prof machinery enqueues dumps
triggered by prof_gdump() calls when bt2gctx_mtx is held, so this
weakened assertion avoids false failures in such cases.
This fixes interactions with witness_assert_depth[_to_rank](), which was
added in dad74bd3c8 (Convert
witness_assert_lockless() to witness_assert_lock_depth().).
malloc_conf does not reliably work with MSVC, which complains of
"inconsistent dll linkage", i.e. its inability to support the
application overriding malloc_conf when dynamically linking/loading.
Work around this limitation by adding test harness support for per test
shell script sourcing, and converting all tests to use MALLOC_CONF
instead of malloc_conf.
Synchronize tcaches with tcaches_mtx rather than ctl_mtx. Add missing
synchronization for tcache flushing. This bug was introduced by
1cb181ed63 (Implement explicit tcache
support.), which was first released in 4.0.0.
malloc_conf does not reliably work with MSVC, which complains of
"inconsistent dll linkage", i.e. its inability to support the
application overriding malloc_conf when dynamically linking/loading.
Work around this limitation by adding test harness support for per test
shell script sourcing, and converting all tests to use MALLOC_CONF
instead of malloc_conf.
Introduces gen_travis.py, which generates .travis.yml, and updates .travis.yml
to be the generated version.
The travis build matrix approach doesn't play well with mixing and matching
various different environment settings, so we generate every build explicitly,
rather than letting them do it for us.
To avoid abusing travis resources (and save us time waiting for CI results), we
don't test every possible combination of options; we only check up to 2 unusual
settings at a time.
Extent splitting and coalescing is a major component of large allocation
overhead, and disabling coalescing of cached extents provides a simple
and effective hysteresis mechanism. Once two-phase purging is
implemented, it will probably make sense to leave coalescing disabled
for the first phase, but coalesce during the second phase.
