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.
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 8a03cf039c (Implement
cache index randomization for large allocations.).
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.
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.
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.
Summary:
Currently an HTTP error response will still try to be parsed, resulting in these messages:
substr outside of string at /home/davejwatson/local/jemalloc-github/bin/jeprof line 3635, <PROFILE> line 1.
Use of uninitialized value in string eq at /home/davejwatson/local/jemalloc-github/bin/jeprof line 3635, <PROFILE> line 1.
substr outside of string at /home/davejwatson/local/jemalloc-github/bin/jeprof line 3637, <PROFILE> line 1.
Use of uninitialized value in string eq at /home/davejwatson/local/jemalloc-github/bin/jeprof line 3637, <PROFILE> line 1.
/home/davejwatson/jeprof/server.1452638936.localhost.pprof.heap: header size >= 2**16
After this fix, curl will return an error status code that will be correctly checked at line 3536, resulting in this error message:
Failed to get profile: curl -s --fail 'http://localhost:4010/pprof/heap' > /home/davejwatson/jeprof/.tmp.server.1452639085.localhost.pprof.heap: No such file or directory
Test Plan:
Tested with MALLOC_CONF="prof:false". Also tested fetching symbols. Didn't test redirects, but this should only affect http error codes >= 400
When using LinuxThreads, malloc bootstrapping deadlocks, since
malloc_tsd_boot0() ends up calling pthread_setspecific(), which causes
recursive allocation. Fix it by moving the malloc_tsd_boot0() call to
malloc_init_hard_recursible().
The deadlock was introduced by 8bb3198f72
(Refactor/fix arenas manipulation.), when tsd_boot() was split and the
top half, tsd_boot0(), got an extra tsd_wrapper_set() call.
- Combine multiple runtime branches into a single malloc_slow check.
- Avoid calling arena_choose / size2index / index2size on fast path.
- A few micro optimizations.
ex_destroy iterates over the tree using post-order traversal so nodes
can be removed and processed by the callback function without paying the
cost to rebalance the tree. The destruction process cannot be stopped
once started.
