diff options
author | Jason Evans <jasone@canonware.com> | 2012-04-21 19:17:21 -0700 |
---|---|---|
committer | Jason Evans <jasone@canonware.com> | 2012-04-21 19:17:21 -0700 |
commit | a8f8d7540d66ddee7337db80c92890916e1063ca (patch) | |
tree | dfd205686200f0b8fb6e2d78aaba0b39ff7aca02 /include/jemalloc/internal/chunk.h | |
parent | 7ad54c1c30e0805e0758690115875f982de46cf2 (diff) | |
download | jemalloc-a8f8d7540d66ddee7337db80c92890916e1063ca.tar.gz |
Remove mmap_unaligned.
Remove mmap_unaligned, which was used to heuristically decide whether to
optimistically call mmap() in such a way that could reduce the total
number of system calls. If I remember correctly, the intention of
mmap_unaligned was to avoid always executing the slow path in the
presence of ASLR. However, that reasoning seems to have been based on a
flawed understanding of how ASLR actually works. Although ASLR
apparently causes mmap() to ignore address requests, it does not cause
total placement randomness, so there is a reasonable expectation that
iterative mmap() calls will start returning chunk-aligned mappings once
the first chunk has been properly aligned.
Diffstat (limited to 'include/jemalloc/internal/chunk.h')
-rw-r--r-- | include/jemalloc/internal/chunk.h | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/include/jemalloc/internal/chunk.h b/include/jemalloc/internal/chunk.h index e047c2b..8fb1fe6 100644 --- a/include/jemalloc/internal/chunk.h +++ b/include/jemalloc/internal/chunk.h @@ -44,8 +44,7 @@ extern size_t arena_maxclass; /* Max size class for arenas. */ void *chunk_alloc(size_t size, size_t alignment, bool base, bool *zero); void chunk_dealloc(void *chunk, size_t size, bool unmap); -bool chunk_boot0(void); -bool chunk_boot1(void); +bool chunk_boot(void); #endif /* JEMALLOC_H_EXTERNS */ /******************************************************************************/ |