project-base/server-skynet-source-3rd-jemalloc
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update ChangeLog for 4.0.1.

Browse Source
This commit is contained in:
Jason Evans 2015-09-15 15:26:23 -07:00
parent 65b940a3c5
commit 1d7540c9d7
1 changed files with 47 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

71
ChangeLog
View File

@ -4,39 +4,62 @@ brevity. Much more detail can be found in the git revision history:
https://github.com/jemalloc/jemalloc https://github.com/jemalloc/jemalloc
* 4.0.1 (XXX) * 4.0.1 (September 15, 2015)
This is a bugfix release that is somewhat high risk due to the amount of
refactoring required to address deep xallocx() problems. As a side effect of
these fixes, xallocx() now tries harder to partially fulfill requests for
optional extra space. Note that a couple of minor heap profiling
optimizations are included, but these are better thought of as performance
fixes that were integral to disovering most of the other bugs.
Optimizations:
- Avoid a chunk metadata read in arena_prof_tctx_set(), since it is in the
fast path when heap profiling is enabled. Additionally, split a special
case out into arena_prof_tctx_reset(), which also avoids chunk metadata
reads.
- Optimize irallocx_prof() to optimistically update the sampler state. The
prior implementation appears to have been a holdover from when
rallocx()/xallocx() functionality was combined as rallocm().
Bug fixes: Bug fixes:
- Fix TLS configuration such that it is enabled by default for platforms on
which it works correctly.
- Fix arenas_cache_cleanup() and arena_get_hard() to handle - Fix arenas_cache_cleanup() and arena_get_hard() to handle
allocation/deallocation within the application's thread-specific data allocation/deallocation within the application's thread-specific data
cleanup functions even after arenas_cache is torn down. cleanup functions even after arenas_cache is torn down.
- Don't bitshift by negative amounts when encoding/decoding run sizes in chunk - Fix xallocx() bugs related to size+extra exceeding HUGE_MAXCLASS.
header maps. This affected systems with page sizes greater than 8 KiB.
- Rename index_t to szind_t to avoid an existing type on Solaris.
- Add JEMALLOC_CXX_THROW to the memalign() function prototype, in order to
match glibc and avoid compilation errors when including both
jemalloc/jemalloc.h and malloc.h in C++ code.
- Fix chunk purge hook calls for in-place huge shrinking reallocation to - Fix chunk purge hook calls for in-place huge shrinking reallocation to
specify the old chunk size rather than the new chunk size. This bug caused specify the old chunk size rather than the new chunk size. This bug caused
no correctness issues for the default chunk purge function, but was no correctness issues for the default chunk purge function, but was
visible to custom functions set via the "arena.<i>.chunk_hooks" mallctl. visible to custom functions set via the "arena.<i>.chunk_hooks" mallctl.
- Fix TLS configuration such that it is enabled by default for platforms on - Fix heap profiling bugs:
which it works correctly. + Fix heap profiling to distinguish among otherwise identical sample sites
- Fix heap profiling to distinguish among otherwise identical sample sites with interposed resets (triggered via the "prof.reset" mallctl). This bug
with interposed resets (triggered via the "prof.reset" mallctl). This bug could cause data structure corruption that would most likely result in a
could cause data structure corruption that would most likely result in a segfault.
segfault. + Fix irealloc_prof() to prof_alloc_rollback() on OOM.
- Fix xallocx() bugs related to size+extra exceeding HUGE_MAXCLASS. + Make one call to prof_active_get_unlocked() per allocation event, and use
- Fix irealloc_prof() to prof_alloc_rollback() on OOM. the result throughout the relevant functions that handle an allocation
- Make one call to prof_active_get_unlocked() per allocation event, and use event. Also add a missing check in prof_realloc(). These fixes protect
the result throughout the relevant functions that handle an allocation allocation events against concurrent prof_active changes.
event. Also add a missing check in prof_realloc(). These fixes protect + Fix ixallocx_prof() to pass usize_max and zero to ixallocx_prof_sample()
allocation events against concurrent prof_active changes. in the correct order.
- Fix ixallocx_prof() to pass usize_max and zero to ixallocx_prof_sample() in + Fix prof_realloc() to call prof_free_sampled_object() after calling
the correct order. prof_malloc_sample_object(). Prior to this fix, if tctx and old_tctx were
- Fix prof_realloc() to call prof_free_sampled_object() after calling the same, the tctx could have been prematurely destroyed.
prof_malloc_sample_object(). Prior to this fix, if tctx and old_tctx were - Fix portability bugs:
the same, the tctx could have been prematurely destroyed. + Don't bitshift by negative amounts when encoding/decoding run sizes in
chunk header maps. This affected systems with page sizes greater than 8
KiB.
+ Rename index_t to szind_t to avoid an existing type on Solaris.
+ Add JEMALLOC_CXX_THROW to the memalign() function prototype, in order to
match glibc and avoid compilation errors when including both
jemalloc/jemalloc.h and malloc.h in C++ code.
+ Don't assume that /bin/sh is appropriate when running size_classes.sh
during configuration.
+ Consider __sparcv9 a synonym for __sparc64__ when defining LG_QUANTUM.
+ Link tests to librt if it contains clock_gettime(2).
* 4.0.0 (August 17, 2015) * 4.0.0 (August 17, 2015)