Merge branch.

Author: jasone

ChangeLog

@@ -4,6 +4,27 @@ brevity.  Much more detail can be found in the git revision history:
 
     https://github.com/jemalloc/jemalloc
 
+* 4.1.1 (May 3, 2016)
+
+  This bugfix release resolves a variety of mostly minor issues, though the
+  bitmap fix is critical for 64-bit Windows.
+
+  Bug fixes:
+  - Fix the linear scan version of bitmap_sfu() to shift by the proper amount
+    even when sizeof(long) is not the same as sizeof(void *), as on 64-bit
+    Windows.  (@jasone)
+  - Fix hashing functions to avoid unaligned memory accesses (and resulting
+    crashes).  This is relevant at least to some ARM-based platforms.
+    (@rkmisra)
+  - Fix fork()-related lock rank ordering reversals.  These reversals were
+    unlikely to cause deadlocks in practice except when heap profiling was
+    enabled and active.  (@jasone)
+  - Fix various chunk leaks in OOM code paths.  (@jasone)
+  - Fix malloc_stats_print() to print opt.narenas correctly.  (@jasone)
+  - Fix MSVC-specific build/test issues.  (@rustyx, yuslepukhin)
+  - Fix a variety of test failures that were due to test fragility rather than
+    core bugs.  (@jasone)
+
 * 4.1.0 (February 28, 2016)
 
   This release is primarily about optimizations, but it also incorporates a lot

Makefile.in

@@ -138,6 +138,7 @@ TESTS_UNIT := $(srcroot)test/unit/atomic.c \
 	$(srcroot)test/unit/bitmap.c \
 	$(srcroot)test/unit/ckh.c \
 	$(srcroot)test/unit/decay.c \
+	$(srcroot)test/unit/fork.c \
 	$(srcroot)test/unit/hash.c \
 	$(srcroot)test/unit/junk.c \
 	$(srcroot)test/unit/junk_alloc.c \

doc/jemalloc.xml.in

@@ -1016,7 +1016,7 @@ for (i = 0; i < nbins; i++) {
         allocate memory during application initialization and then deadlock
         internally when jemalloc in turn calls
         <function>atexit<parameter/></function>, so this option is not
-        univerally usable (though the application can register its own
+        universally usable (though the application can register its own
         <function>atexit<parameter/></function> function with equivalent
         functionality).  Therefore, this option should only be used with care;
         it is primarily intended as a performance tuning aid during application
@@ -1320,7 +1320,7 @@ malloc_conf = "xmalloc:true";]]></programlisting>
         option.  Note that <function>atexit<parameter/></function> may allocate
         memory during application initialization and then deadlock internally
         when jemalloc in turn calls <function>atexit<parameter/></function>, so
-        this option is not univerally usable (though the application can
+        this option is not universally usable (though the application can
         register its own <function>atexit<parameter/></function> function with
         equivalent functionality).  This option is disabled by
         default.</para></listitem>
@@ -2062,7 +2062,7 @@ typedef struct {
           [<option>--enable-prof</option>]
         </term>
         <listitem><para>Average number of bytes allocated between
-        inverval-based profile dumps.  See the
+        interval-based profile dumps.  See the
         <link
         linkend="opt.lg_prof_interval"><mallctl>opt.lg_prof_interval</mallctl></link>
         option for additional information.</para></listitem>

include/jemalloc/internal/arena.h

@@ -584,7 +584,10 @@ void	arena_nthreads_inc(arena_t *arena);
 void	arena_nthreads_dec(arena_t *arena);
 arena_t	*arena_new(unsigned ind);
 bool	arena_boot(void);
-void	arena_prefork(arena_t *arena);
+void	arena_prefork0(arena_t *arena);
+void	arena_prefork1(arena_t *arena);
+void	arena_prefork2(arena_t *arena);
+void	arena_prefork3(arena_t *arena);
 void	arena_postfork_parent(arena_t *arena);
 void	arena_postfork_child(arena_t *arena);
 

include/jemalloc/internal/bitmap.h

@@ -223,7 +223,7 @@ bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo)
 		i++;
 		g = bitmap[i];
 	}
-	bit = (bit - 1) + (i << 6);
+	bit = (bit - 1) + (i << LG_BITMAP_GROUP_NBITS);
 #endif
 	bitmap_set(bitmap, binfo, bit);
 	return (bit);

include/jemalloc/internal/chunk.h

@@ -62,12 +62,8 @@ void	*chunk_alloc_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks,
     void *new_addr, size_t size, size_t alignment, bool *zero, bool *commit);
 void	chunk_dalloc_cache(arena_t *arena, chunk_hooks_t *chunk_hooks,
     void *chunk, size_t size, bool committed);
-void	chunk_dalloc_arena(arena_t *arena, chunk_hooks_t *chunk_hooks,
-    void *chunk, size_t size, bool zeroed, bool committed);
 void	chunk_dalloc_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks,
-    void *chunk, size_t size, bool committed);
-bool	chunk_purge_arena(arena_t *arena, void *chunk, size_t offset,
-    size_t length);
+    void *chunk, size_t size, bool zeroed, bool committed);
 bool	chunk_purge_wrapper(arena_t *arena, chunk_hooks_t *chunk_hooks,
     void *chunk, size_t size, size_t offset, size_t length);
 bool	chunk_boot(void);

include/jemalloc/internal/hash.h

@@ -53,7 +53,7 @@ hash_get_block_32(const uint32_t *p, int i)
 	if (unlikely((uintptr_t)p & (sizeof(uint32_t)-1)) != 0) {
 		uint32_t ret;
 
-		memcpy(&ret, &p[i], sizeof(uint32_t));
+		memcpy(&ret, (uint8_t *)(p + i), sizeof(uint32_t));
 		return (ret);
 	}
 
@@ -68,7 +68,7 @@ hash_get_block_64(const uint64_t *p, int i)
 	if (unlikely((uintptr_t)p & (sizeof(uint64_t)-1)) != 0) {
 		uint64_t ret;
 
-		memcpy(&ret, &p[i], sizeof(uint64_t));
+		memcpy(&ret, (uint8_t *)(p + i), sizeof(uint64_t));
 		return (ret);
 	}
 

include/jemalloc/internal/private_symbols.txt

@@ -21,7 +21,9 @@ arena_dalloc
 arena_dalloc_bin
 arena_dalloc_bin_junked_locked
 arena_dalloc_junk_large
+arena_dalloc_junk_large_impl
 arena_dalloc_junk_small
+arena_dalloc_junk_small_impl
 arena_dalloc_large
 arena_dalloc_large_junked_locked
 arena_dalloc_small
@@ -81,7 +83,10 @@ arena_nthreads_inc
 arena_palloc
 arena_postfork_child
 arena_postfork_parent
-arena_prefork
+arena_prefork0
+arena_prefork1
+arena_prefork2
+arena_prefork3
 arena_prof_accum
 arena_prof_accum_impl
 arena_prof_accum_locked
@@ -123,6 +128,11 @@ atomic_sub_u
 atomic_sub_uint32
 atomic_sub_uint64
 atomic_sub_z
+atomic_write_p
+atomic_write_u
+atomic_write_uint32
+atomic_write_uint64
+atomic_write_z
 base_alloc
 base_boot
 base_postfork_child
@@ -148,7 +158,6 @@ chunk_alloc_dss
 chunk_alloc_mmap
 chunk_alloc_wrapper
 chunk_boot
-chunk_dalloc_arena
 chunk_dalloc_cache
 chunk_dalloc_mmap
 chunk_dalloc_wrapper
@@ -168,7 +177,6 @@ chunk_npages
 chunk_postfork_child
 chunk_postfork_parent
 chunk_prefork
-chunk_purge_arena
 chunk_purge_wrapper
 chunk_register
 chunks_rtree
@@ -200,6 +208,8 @@ extent_node_addr_get
 extent_node_addr_set
 extent_node_arena_get
 extent_node_arena_set
+extent_node_committed_get
+extent_node_committed_set
 extent_node_dirty_insert
 extent_node_dirty_linkage_init
 extent_node_dirty_remove
@@ -210,6 +220,8 @@ extent_node_size_get
 extent_node_size_set
 extent_node_zeroed_get
 extent_node_zeroed_set
+extent_tree_ad_destroy
+extent_tree_ad_destroy_recurse
 extent_tree_ad_empty
 extent_tree_ad_first
 extent_tree_ad_insert
@@ -227,6 +239,8 @@ extent_tree_ad_reverse_iter
 extent_tree_ad_reverse_iter_recurse
 extent_tree_ad_reverse_iter_start
 extent_tree_ad_search
+extent_tree_szad_destroy
+extent_tree_szad_destroy_recurse
 extent_tree_szad_empty
 extent_tree_szad_first
 extent_tree_szad_insert
@@ -304,6 +318,7 @@ jemalloc_postfork_parent
 jemalloc_prefork
 large_maxclass
 lg_floor
+lg_prof_sample
 malloc_cprintf
 malloc_mutex_init
 malloc_mutex_lock
@@ -331,6 +346,8 @@ narenas_tdata_cleanup
 narenas_total_get
 ncpus
 nhbins
+nhclasses
+nlclasses
 nstime_add
 nstime_compare
 nstime_copy
@@ -344,6 +361,7 @@ nstime_nsec
 nstime_sec
 nstime_subtract
 nstime_update
+nstime_update_impl
 opt_abort
 opt_decay_time
 opt_dss
@@ -384,6 +402,7 @@ pow2_ceil_u64
 pow2_ceil_zu
 prng_lg_range
 prng_range
+prof_active
 prof_active_get
 prof_active_get_unlocked
 prof_active_set
@@ -393,6 +412,7 @@ prof_backtrace
 prof_boot0
 prof_boot1
 prof_boot2
+prof_bt_count
 prof_dump_header
 prof_dump_open
 prof_free
@@ -410,7 +430,8 @@ prof_malloc_sample_object
 prof_mdump
 prof_postfork_child
 prof_postfork_parent
-prof_prefork
+prof_prefork0
+prof_prefork1
 prof_realloc
 prof_reset
 prof_sample_accum_update
@@ -419,6 +440,7 @@ prof_tctx_get
 prof_tctx_reset
 prof_tctx_set
 prof_tdata_cleanup
+prof_tdata_count
 prof_tdata_get
 prof_tdata_init
 prof_tdata_reinit
@@ -506,6 +528,13 @@ ticker_tick
 ticker_ticks
 tsd_arena_get
 tsd_arena_set
+tsd_arenap_get
+tsd_arenas_tdata_bypass_get
+tsd_arenas_tdata_bypass_set
+tsd_arenas_tdata_bypassp_get
+tsd_arenas_tdata_get
+tsd_arenas_tdata_set
+tsd_arenas_tdatap_get
 tsd_boot
 tsd_boot0
 tsd_boot1
@@ -514,6 +543,9 @@ tsd_cleanup
 tsd_cleanup_wrapper
 tsd_fetch
 tsd_get
+tsd_narenas_tdata_get
+tsd_narenas_tdata_set
+tsd_narenas_tdatap_get
 tsd_wrapper_get
 tsd_wrapper_set
 tsd_initialized
@@ -523,17 +555,23 @@ tsd_init_head
 tsd_nominal
 tsd_prof_tdata_get
 tsd_prof_tdata_set
+tsd_prof_tdatap_get
 tsd_quarantine_get
 tsd_quarantine_set
+tsd_quarantinep_get
 tsd_set
 tsd_tcache_enabled_get
 tsd_tcache_enabled_set
+tsd_tcache_enabledp_get
 tsd_tcache_get
 tsd_tcache_set
+tsd_tcachep_get
 tsd_thread_allocated_get
 tsd_thread_allocated_set
+tsd_thread_allocatedp_get
 tsd_thread_deallocated_get
 tsd_thread_deallocated_set
+tsd_thread_deallocatedp_get
 tsd_tls
 tsd_tsd
 u2rz

include/jemalloc/internal/prof.h

@@ -316,7 +316,8 @@ bool	prof_gdump_set(bool active);
 void	prof_boot0(void);
 void	prof_boot1(void);
 bool	prof_boot2(void);
-void	prof_prefork(void);
+void	prof_prefork0(void);
+void	prof_prefork1(void);
 void	prof_postfork_parent(void);
 void	prof_postfork_child(void);
 void	prof_sample_threshold_update(prof_tdata_t *tdata);

include/msvc_compat/windows_extra.h

@@ -1,26 +1,6 @@
 #ifndef MSVC_COMPAT_WINDOWS_EXTRA_H
 #define	MSVC_COMPAT_WINDOWS_EXTRA_H
 
-#ifndef ENOENT
-#  define ENOENT ERROR_PATH_NOT_FOUND
-#endif
-#ifndef EINVAL
-#  define EINVAL ERROR_BAD_ARGUMENTS
-#endif
-#ifndef EAGAIN
-#  define EAGAIN ERROR_OUTOFMEMORY
-#endif
-#ifndef EPERM
-#  define EPERM  ERROR_WRITE_FAULT
-#endif
-#ifndef EFAULT
-#  define EFAULT ERROR_INVALID_ADDRESS
-#endif
-#ifndef ENOMEM
-#  define ENOMEM ERROR_NOT_ENOUGH_MEMORY
-#endif
-#ifndef ERANGE
-#  define ERANGE ERROR_INVALID_DATA
-#endif
+#include <errno.h>
 
 #endif /* MSVC_COMPAT_WINDOWS_EXTRA_H */