Commit Graph

49 Commits

Author SHA1 Message Date
Jason Evans
cfc5706f69 Miscellaneous cleanups. 2014-10-30 23:18:45 -07:00
Daniel Micay
809b0ac391 mark huge allocations as unlikely
This cleans up the fast path a bit more by moving away more code.
2014-10-30 17:06:38 -07:00
Jason Evans
44c97b712e Fix a prof_tctx_t/prof_tdata_t cleanup race.
Fix a prof_tctx_t/prof_tdata_t cleanup race by storing a copy of thr_uid
in prof_tctx_t, so that the associated tdata need not be present during
tctx teardown.
2014-10-12 13:03:20 -07:00
Jason Evans
34e85b4182 Make prof-related inline functions always-inline. 2014-10-04 11:26:05 -07:00
Jason Evans
029d44cf8b Fix tsd cleanup regressions.
Fix tsd cleanup regressions that were introduced in
5460aa6f66 (Convert all tsd variables to
reside in a single tsd structure.).  These regressions were twofold:

1) tsd_tryget() should never (and need never) return NULL.  Rename it to
   tsd_fetch() and simplify all callers.
2) tsd_*_set() must only be called when tsd is in the nominal state,
   because cleanup happens during the nominal-->purgatory transition,
   and re-initialization must not happen while in the purgatory state.
   Add tsd_nominal() and use it as needed.  Note that tsd_*{p,}_get()
   can still be used as long as no re-initialization that would require
   cleanup occurs.  This means that e.g. the thread_allocated counter
   can be updated unconditionally.
2014-10-04 11:22:55 -07:00
Jason Evans
fc12c0b8bc Implement/test/fix prof-related mallctl's.
Implement/test/fix the opt.prof_thread_active_init,
prof.thread_active_init, and thread.prof.active mallctl's.

Test/fix the thread.prof.name mallctl.

Refactor opt_prof_active to be read-only and move mutable state into the
prof_active variable.  Stop leaning on ctl-related locking for
protection.
2014-10-03 23:25:30 -07:00
Jason Evans
551ebc4364 Convert to uniform style: cond == false --> !cond 2014-10-03 10:16:09 -07:00
Jason Evans
20c31deaae Test prof.reset mallctl and fix numerous discovered bugs. 2014-10-02 23:01:10 -07:00
Jason Evans
6ef80d68f0 Fix profile dumping race.
Fix a race that caused a non-critical assertion failure.  To trigger the
race, a thread had to be part way through initializing a new sample,
such that it was discoverable by the dumping thread, but not yet linked
into its gctx by the time a later dump phase would normally have reset
its state to 'nominal'.

Additionally, lock access to the state field during modification to
transition to the dumping state.  It's not apparent that this oversight
could have caused an actual problem due to outer locking that protects
the dumping machinery, but the added locking pedantically follows the
stated locking protocol for the state field.
2014-09-24 22:23:43 -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
Jason Evans
6e73dc194e Fix a profile sampling race.
Fix a profile sampling race that was due to preparing to sample, yet
doing nothing to assure that the context remains valid until the stats
are updated.

These regressions were caused by
602c8e0971 (Implement per thread heap
profiling.), which did not make it into any releases prior to these
fixes.
2014-09-09 19:47:09 -07:00
Jason Evans
6fd53da030 Fix prof_tdata_get()-related regressions.
Fix prof_tdata_get() to avoid dereferencing an invalid tdata pointer
(when it's PROF_TDATA_STATE_{REINCARNATED,PURGATORY}).

Fix prof_tdata_get() callers to check for invalid results besides NULL
(PROF_TDATA_STATE_{REINCARNATED,PURGATORY}).

These regressions were caused by
602c8e0971 (Implement per thread heap
profiling.), which did not make it into any releases prior to these
fixes.
2014-09-09 15:29:34 -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
3a81cbd2d4 Dump heap profile backtraces in a stable order.
Also iterate over per thread stats in a stable order, which prepares the
way for stable ordering of per thread heap profile dumps.
2014-08-19 21:05:54 -07:00
Jason Evans
ab532e9799 Directly embed prof_ctx_t's bt. 2014-08-19 21:05:54 -07:00
Jason Evans
b41ccdb125 Convert prof_tdata_t's bt2cnt to a comprehensive map.
Treat prof_tdata_t's bt2cnt as a comprehensive map of the thread's
extant allocation samples (do not limit the total number of entries).
This helps prepare the way for per thread heap profiling.
2014-08-19 21:05:54 -07:00
Jason Evans
6f001059aa Simplify backtracing.
Simplify backtracing to not ignore any frames, and compensate for this
in pprof in order to increase flexibility with respect to function-based
refactoring even in the presence of non-deterministic inlining.  Modify
pprof to blacklist all jemalloc allocation entry points including
non-standard ones like mallocx(), and ignore all allocator-internal
frames.  Prior to this change, pprof excluded the specifically
blacklisted functions from backtraces, but it left allocator-internal
frames intact.
2014-04-22 20:55:09 -07:00
Jason Evans
0b49403958 Fix debug-only compilation failures.
Fix debug-only compilation failures introduced by changes to
prof_sample_accum_update() in:

    6c39f9e059
    refactor profiling. only use a bytes till next sample variable.
2014-04-16 16:38:22 -07:00
Ben Maurer
6c39f9e059 refactor profiling. only use a bytes till next sample variable. 2014-04-16 13:43:30 -07:00
Jason Evans
9b0cbf0850 Remove support for non-prof-promote heap profiling metadata.
Make promotion of sampled small objects to large objects mandatory, so
that profiling metadata can always be stored in the chunk map, rather
than requiring one pointer per small region in each small-region page
run.  In practice the non-prof-promote code was only useful when using
jemalloc to track all objects and report them as leaks at program exit.
However, Valgrind is at least as good a tool for this particular use
case.

Furthermore, the non-prof-promote code is getting in the way of
some optimizations that will make heap profiling much cheaper for the
predominant use case (sampling a small representative proportion of all
allocations).
2014-04-11 14:24:51 -07:00
Jason Evans
5f60afa01e Avoid a compiler warning.
Avoid copying "jeprof" to a 1-byte buffer within prof_boot0() when heap
profiling is disabled.  Although this is dead code under such
conditions, the compiler doesn't figure that part out.

Reported by Eduardo Silva.
2014-01-28 23:04:02 -08:00
Jason Evans
772163b4f3 Add heap profiling tests.
Fix a regression in prof_dump_ctx() due to an uninitized variable.  This
was caused by revision 4f37ef693e, so no
releases are affected.
2014-01-17 15:40:52 -08:00
Jason Evans
eefdd02e70 Fix a variable prototype/definition mismatch. 2014-01-16 18:04:30 -08:00
Jason Evans
4f37ef693e Refactor prof_dump() to reduce contention.
Refactor prof_dump() to use a two pass algorithm, and prof_leave() prior
to the second pass.  This avoids write(2) system calls while holding
critical prof resources.

Fix prof_dump() to close the dump file descriptor for all relevant error
paths.

Minimize the size of prof-related static buffers when prof is disabled.
This saves roughly 65 KiB of application memory for non-prof builds.

Refactor prof_ctx_init() out of prof_lookup_global().
2014-01-16 13:36:38 -08:00
Jason Evans
665769357c Optimize arena_prof_ctx_set().
Refactor such that arena_prof_ctx_set() receives usize as an argument,
and use it to determine whether to handle ptr as a small region, rather
than reading the chunk page map.
2013-12-15 21:57:02 -08:00
Jason Evans
b1941c6150 Add probabability distribution utility code.
Add probabability distribution utility code that enables generation of
random deviates drawn from normal, Chi-square, and Gamma distributions.

Fix format strings in several of the assert_* macros (remove a %s).

Clean up header issues; it's critical that system headers are not
included after internal definitions potentially do things like:

  #define inline

Fix the build system to incorporate header dependencies for the test
library C files.
2013-12-09 23:42:08 -08:00
Jason Evans
d37d5adee4 Disable floating point code/linking when possible.
Unless heap profiling is enabled, disable floating point code and don't
link with libm.  This, in combination with e.g. EXTRA_CFLAGS=-mno-sse on
x64 systems, makes it possible to completely disable floating point
register use.  Some versions of glibc neglect to save/restore
caller-saved floating point registers during dynamic lazy symbol
loading, and the symbol loading code uses whatever malloc the
application happens to have linked/loaded with, the result being
potential floating point register corruption.
2013-12-05 23:01:50 -08:00
Jason Evans
bbe29d374d Fix potential TLS-related memory corruption.
Avoid writing to uninitialized TLS as a side effect of deallocation.
Initializing TLS during deallocation is unsafe because it is possible
that a thread never did any allocation, and that TLS has already been
deallocated by the threads library, resulting in write-after-free
corruption.  These fixes affect prof_tdata and quarantine; all other
uses of TLS are already safe, whether intentionally (as for tcache) or
unintentionally (as for arenas).
2013-01-31 14:23:48 -08:00
Jason Evans
20f1fc95ad Fix fork(2)-related deadlocks.
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.
2012-10-09 15:21:46 -07:00
Jason Evans
3860eac170 Fix heap profiling crash for realloc(p, 0) case.
Fix prof_realloc() to not call prof_ctx_set() if a sampled object is
being freed via realloc(p, 0).
2012-05-15 13:56:28 -07:00
Mike Hommey
8b49971d0c Avoid variable length arrays and remove declarations within code
MSVC doesn't support C99, and building as C++ to be able to use them is
dangerous, as C++ and C99 are incompatible.

Introduce a VARIABLE_ARRAY macro that either uses VLA when supported,
or alloca() otherwise. Note that using alloca() inside loops doesn't
quite work like VLAs, thus the use of VARIABLE_ARRAY there is discouraged.
It might be worth investigating ways to check whether VARIABLE_ARRAY is
used in such context at runtime in debug builds and bail out if that
happens.
2012-04-29 00:25:34 -07:00
Jason Evans
f278994029 Fix more prof_tdata resurrection corner cases. 2012-04-28 23:27:13 -07:00
Jason Evans
0050a0f7e6 Handle prof_tdata resurrection.
Handle prof_tdata resurrection during thread shutdown, similarly to how
tcache and quarantine handle resurrection.
2012-04-28 18:14:24 -07:00
Jason Evans
3fb50b0407 Fix a PROF_ALLOC_PREP() error path.
Fix a PROF_ALLOC_PREP() error path to initialize the return value to
NULL.
2012-04-25 13:13:44 -07:00
Jason Evans
52386b2dc6 Fix heap profiling bugs.
Fix a potential deadlock that could occur during interval- and
growth-triggered heap profile dumps.

Fix an off-by-one heap profile statistics bug that could be observed in
interval- and growth-triggered heap profiles.

Fix heap profile dump filename sequence numbers (regression during
conversion to malloc_snprintf()).
2012-04-22 16:00:11 -07:00
Jason Evans
0b25fe79aa Update prof defaults to match common usage.
Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB).

Change the "opt.prof_accum" default from true to false.

Add the "opt.prof_final" mallctl, so that "opt.prof_prefix" need not be
abused to disable final profile dumping.
2012-04-17 16:39:33 -07:00
Jason Evans
122449b073 Implement Valgrind support, redzones, and quarantine.
Implement Valgrind support, as well as the redzone and quarantine
features, which help Valgrind detect memory errors.  Redzones are only
implemented for small objects because the changes necessary to support
redzones around large and huge objects are complicated by in-place
reallocation, to the point that it isn't clear that the maintenance
burden is worth the incremental improvement to Valgrind support.

Merge arena_salloc() and arena_salloc_demote().

Refactor i[v]salloc() to expose the 'demote' option.
2012-04-11 11:46:18 -07:00
Jason Evans
6da5418ded Remove ephemeral mutexes.
Remove ephemeral mutexes from the prof machinery, and remove
malloc_mutex_destroy().  This simplifies mutex management on systems
that call malloc()/free() inside pthread_mutex_{create,destroy}().

Add atomic_*_u() for operation on unsigned values.

Fix prof_printf() to call malloc_vsnprintf() rather than
malloc_snprintf().
2012-03-23 18:05:51 -07:00
Jason Evans
cd9a1346e9 Implement tsd.
Implement tsd, which is a TLS/TSD abstraction that uses one or both
internally.  Modify bootstrapping such that no tsd's are utilized until
allocation is safe.

Remove malloc_[v]tprintf(), and use malloc_snprintf() instead.

Fix %p argument size handling in malloc_vsnprintf().

Fix a long-standing statistics-related bug in the "thread.arena"
mallctl that could cause crashes due to linked list corruption.
2012-03-23 15:14:55 -07:00
Jason Evans
e24c7af35d Invert NO_TLS to JEMALLOC_TLS. 2012-03-19 10:21:17 -07:00
Jason Evans
b8c8be7f8a Use UINT64_C() rather than LLU for 64-bit constants. 2012-03-05 12:26:26 -08:00
Jason Evans
84f7cdb0c5 Rename prn to prng.
Rename prn to prng so that Windows doesn't choke when trying to create
a file named prn.h.
2012-03-02 15:59:45 -08:00
Jason Evans
5389146191 Remove the opt.lg_prof_bt_max option.
Remove opt.lg_prof_bt_max, and hard code it to 7.  The original
intention of this option was to enable faster backtracing by limiting
backtrace depth.  However, this makes graphical pprof output very
difficult to interpret.  In practice, decreasing sampling frequency is a
better mechanism for limiting profiling overhead.
2012-02-13 18:41:36 -08:00
Jason Evans
0b526ff94d Remove the opt.lg_prof_tcmax option.
Remove the opt.lg_prof_tcmax option and hard-code a cache size of 1024.
This setting is something that users just shouldn't have to worry about.
If lock contention actually ends up being a problem, the simple solution
available to the user is to reduce sampling frequency.
2012-02-13 18:04:26 -08:00
Jason Evans
fd56043c53 Remove magic.
Remove structure magic, because 1) it is no longer conditional, and 2)
it stopped being very effective at detecting memory corruption several
years ago.
2012-02-13 10:24:43 -08:00
Jason Evans
7372b15a31 Reduce cpp conditional logic complexity.
Convert configuration-related cpp conditional logic to use static
constant variables, e.g.:

  #ifdef JEMALLOC_DEBUG
    [...]
  #endif

becomes:

  if (config_debug) {
    [...]
  }

The advantage is clearer, more concise code.  The main disadvantage is
that data structures no longer have conditionally defined fields, so
they pay the cost of all fields regardless of whether they are used.  In
practice, this is only a minor concern; config_stats will go away in an
upcoming change, and config_prof is the only other major feature that
depends on more than a few special-purpose fields.
2012-02-10 20:22:09 -08:00
Jason Evans
a507004d29 Fix off-by-one backtracing issues.
Rewrite prof_alloc_prep() as a cpp macro, PROF_ALLOC_PREP(), in order to
remove any doubt as to whether an additional stack frame is created.
Prior to this change, it was assumed that inlining would reduce the
total number of frames in the backtrace, but in practice behavior wasn't
completely predictable.

Create imemalign() and call it from posix_memalign(), memalign(), and
valloc(), so that all entry points require the same number of stack
frames to be ignored during backtracing.
2011-08-12 13:48:27 -07:00
Jason Evans
7427525c28 Move repo contents in jemalloc/ to top level. 2011-03-31 20:36:17 -07:00