jemalloc.git - jemalloc is a general purpose malloc(3) implementation that emphasizes fragmentation avoidance and scalable concurrency support.

	Commit message (Collapse)	Author	Age	Files	Lines
*	Silence miscellaneous 64-to-32-bit data loss warnings.	Jason Evans	2016-02-24	2	-2/+6
\|
*	Compile with -Wshorten-64-to-32.	Jason Evans	2016-02-24	1	-0/+1
\| \| \| \| \|	This will prevent accidental creation of potential integer truncation bugs when developing on LP64 systems.
*	Silence miscellaneous 64-to-32-bit data loss warnings.	Jason Evans	2016-02-24	13	-41/+49
\|
*	Change lg_floor() return type from size_t to unsigned.	Jason Evans	2016-02-24	2	-17/+18
\|
*	Use ssize_t for readlink() rather than int.	Jason Evans	2016-02-24	1	-1/+1
\|
*	Make opt_narenas unsigned rather than size_t.	Jason Evans	2016-02-24	6	-14/+24
\|
*	Make nhbins unsigned rather than size_t.	Jason Evans	2016-02-24	2	-2/+2
\|
*	Explicitly cast mib[] elements to unsigned where appropriate.	Jason Evans	2016-02-24	1	-9/+9
\|
*	Refactor jemalloc_ffs() into ffs_().	Jason Evans	2016-02-24	8	-40/+70
\| \| \| \|	Use appropriate versions to resolve 64-to-32-bit data loss warnings.
*	Fix Windows build issues	Dmitri Smirnov	2016-02-24	4	-6/+31
\| \| \| \|	This resolves #333.
*	Collapse arena_avail_tree_* into arena_run_tree_*.	Jason Evans	2016-02-24	2	-13/+8
\| \| \| \| \|	These tree types converged to become identical, yet they still had independently generated red-black tree implementations.
*	Separate arena_avail trees	Dave Watson	2016-02-24	2	-94/+56
\| \| \| \| \| \| \| \| \| \| \|	Separate run trees by index, replacing the previous quantize logic. Quantization by index is now performed only on insertion / removal from the tree, and not on node comparison, saving some cpu. This also means we don't have to dereference the miscelm* pointers, saving half of the memory loads from miscelms/mapbits that have fallen out of cache. A linear scan of the indicies appears to be fast enough. The only cost of this is an extra tree array in each arena.
*	Remove rbt_nil	Dave Watson	2016-02-24	2	-109/+86
\| \| \| \| \|	Since this is an intrusive tree, rbt_nil is the whole size of the node and can be quite large. For example, miscelm is ~100 bytes.
*	Use table lookup for run_quantize_{floor,ceil}().	Jason Evans	2016-02-23	4	-32/+90
\| \| \| \| \|	Reduce run quantization overhead by generating lookup tables during bootstrapping, and using the tables for all subsequent run quantization.
*	Fix run_quantize_ceil().	Jason Evans	2016-02-23	1	-1/+1
\| \| \| \| \| \| \| \| \| \| \|	In practice this bug had limited impact (and then only by increasing chunk fragmentation) because run_quantize_ceil() returned correct results except for inputs that could only arise from aligned allocation requests that required more than page alignment. This bug existed in the original run quantization implementation, which was introduced by 8a03cf039cd06f9fa6972711195055d865673966 (Implement cache index randomization for large allocations.).
*	Test run quantization.	Jason Evans	2016-02-22	5	-10/+194
\| \| \| \| \|	Also rename run_quantize_*() to improve clarity. These tests demonstrate that run_quantize_ceil() is flawed.
*	Indentation style cleanup.	Jason Evans	2016-02-22	1	-13/+13
\|
*	Refactor time_* into nstime_*.	Jason Evans	2016-02-22	17	-557/+526
\| \| \| \| \| \| \|	Use a single uint64_t in nstime_t to store nanoseconds rather than using struct timespec. This reduces fragility around conversions between long and uint64_t, especially missing casts that only cause problems on 32-bit platforms.
*	Fix Windows-specific prof-related compilation portability issues.	Jason Evans	2016-02-21	2	-5/+16
\|
*	Fix time_update() to compile and work on MinGW.	Jason Evans	2016-02-21	1	-6/+9
\|
*	Remove _WIN32-specific struct timespec declaration.	Jason Evans	2016-02-21	1	-6/+0
\| \| \| \|	struct timespec is already defined by the system (at least on MinGW).
*	Fix overflow in prng_range().	Jason Evans	2016-02-21	5	-6/+40
\| \| \| \| \|	Add jemalloc_ffs64() and use it instead of jemalloc_ffsl() in prng_range(), since long is not guaranteed to be a 64-bit type.
*	Add symbol mangling for prng_[lg_]range().	Jason Evans	2016-02-20	1	-0/+2
\|
*	Add MS Visual Studio 2015 support	rustyx	2016-02-20	10	-0/+1204
\|
*	Fix warning in ipalloc	rustyx	2016-02-20	1	-2/+2
\|
*	Prevent MSVC from optimizing away tls_callback (resolves #318)	rustyx	2016-02-20	1	-1/+3
\|
*	getpid() fix for Win32	rustyx	2016-02-20	2	-0/+4
\|
*	Add CPU "pause" intrinsic for MSVC	rustyx	2016-02-20	1	-6/+16
\|
*	Fix error "+ 2")syntax error: invalid arithmetic operator (error token is " ↵	rustyx	2016-02-20	1	-1/+1
\| \| \| \|	in Cygwin x64
*	Detect LG_SIZEOF_PTR depending on MSVC platform target	rustyx	2016-02-20	2	-6/+19
\|
*	Fix a typo in the ckh_search() prototype.	Christopher Ferris	2016-02-20	1	-1/+1
\|
*	Handle unaligned keys in hash().	Jason Evans	2016-02-20	2	-4/+33
\| \| \| \|	Reported by Christopher Ferris <cferris@google.com>.
*	Increase test coverage in test_decay_ticks.	Jason Evans	2016-02-20	1	-123/+98
\|
*	Implement decay-based unused dirty page purging.	Jason Evans	2016-02-20	18	-112/+1268
\| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \|	This is an alternative to the existing ratio-based unused dirty page purging, and is intended to eventually become the sole purging mechanism. Add mallctls: - opt.purge - opt.decay_time - arena.<i>.decay - arena.<i>.decay_time - arenas.decay_time - stats.arenas.<i>.decay_time This resolves #325.
*	Implement smoothstep table generation.	Jason Evans	2016-02-20	5	-0/+472
\| \| \| \| \| \|	Check in a generated smootherstep table as smoothstep.h rather than generating it at configure time, since not all systems (e.g. Windows) have dc.
*	Refactor out arena_compute_npurge().	Jason Evans	2016-02-20	1	-43/+37
\| \| \| \| \|	Refactor out arena_compute_npurge() by integrating its logic into arena_stash_dirty() as an incremental computation.
*	Refactor arenas_cache tsd.	Jason Evans	2016-02-20	6	-92/+142
\| \| \| \| \|	Refactor arenas_cache tsd into arenas_tdata, which is a structure of type arena_tdata_t.
*	Refactor arena_ralloc_no_move().	Jason Evans	2016-02-20	1	-11/+10
\| \| \| \| \|	Refactor early return logic in arena_ralloc_no_move() to return early on failure rather than on success.
*	Refactor arena_malloc_hard() out of arena_malloc().	Jason Evans	2016-02-20	3	-17/+25
\|
*	Refactor prng* from cpp macros into inline functions.	Jason Evans	2016-02-20	12	-69/+205
\| \| \| \| \|	Remove 32-bit variant, convert prng64() to prng_lg_range(), and add prng_range().
*	Use ticker for incremental tcache GC.	Jason Evans	2016-02-20	2	-5/+4
\|
*	Implement ticker.	Jason Evans	2016-02-20	6	-2/+163
\| \| \| \| \|	Implement ticker, which provides a simple API for ticking off some number of events before indicating that the ticker has hit its limit.
*	Flesh out time_*() API.	Jason Evans	2016-02-20	8	-56/+397
\|
*	Add time_update().	Cameron Evans	2016-02-20	7	-2/+97
\|
*	Add --with-malloc-conf.	Jason Evans	2016-02-20	9	-44/+87
\| \| \| \| \|	Add --with-malloc-conf, which makes it possible to embed a default options string during configuration.
*	Fix arena_sdalloc() line wrapping.	Jason Evans	2016-02-20	1	-5/+8
\|
*	Fix test_stats_arenas_summary fragility.	Jason Evans	2016-02-20	1	-4/+4
\| \| \| \| \|	Fix test_stats_arenas_summary to deallocate before asserting that purging must have happened.
*	Don't rely on unpurged chunks in xallocx() test.	Jason Evans	2016-02-20	1	-20/+20
\|
*	Fix a documentation editing error.	Jason Evans	2016-02-20	1	-1/+1
\|
*	Pass retain and exclude parameters to /pprof/symbol.	Jason Evans	2016-01-30	1	-0/+23
\| \| \| \| \| \|	Pass the retain and exclude parameters to the /pprof/symbol pprof server endpoint so that the server has the opportunity to optimize which symbols it looks up and/or returns mappings for.