romenskiy2012/jemalloc
1
0
Fork 0
mirror of https://github.com/jemalloc/jemalloc.git synced 2026-06-05 19:54:16 +03:00
Code Issues Projects Releases Packages Wiki Activity

Implement cache index randomization for large allocations.

Browse source 
Extract szad size quantization into {extent,run}_quantize(), and .
quantize szad run sizes to the union of valid small region run sizes and
large run sizes.

Refactor iteration in arena_run_first_fit() to use
run_quantize{,_first,_next(), and add support for padded large runs.

For large allocations that have no specified alignment constraints,
compute a pseudo-random offset from the beginning of the first backing
page that is a multiple of the cache line size.  Under typical
configurations with 4-KiB pages and 64-byte cache lines this results in
a uniform distribution among 64 page boundary offsets.

Add the --disable-cache-oblivious option, primarily intended for
performance testing.

This resolves #13.
This commit is contained in:
Jason Evans 2015-05-04 09:58:36 -07:00
parent 6bb54cb9da
commit 8a03cf039c
10 changed files with 280 additions and 74 deletions
Download patch file Download diff file Expand all files Collapse all files

3
src/jemalloc.c
View file

@ -1182,7 +1182,8 @@ malloc_init_hard_a0_locked(void)
return (true);
if (config_prof)
prof_boot1();
arena_boot();
if (arena_boot())
return (true);
if (config_tcache && tcache_boot())
return (true);
if (malloc_mutex_init(&arenas_lock))