Tune rosalloc buffer sizes.
We now use one-page buffers for size brackets 4-7, instead of two-page
buffers, and the first 8 size brackets for thread-local allocations,
instead of 11.
No slowdown observed with MemAllocTest, EvaluateAndApplyChanges, and
BinaryTrees.
(cherrypick commit c4cd95fa37b7138a0fa26d07c235aa409542aecd)
Bug: 18377775
Change-Id: I311f3adf9cab660d258833b17df7e6d905f73c72
diff --git a/runtime/gc/allocator/rosalloc.h b/runtime/gc/allocator/rosalloc.h
index 431686a..3269e10 100644
--- a/runtime/gc/allocator/rosalloc.h
+++ b/runtime/gc/allocator/rosalloc.h
@@ -414,8 +414,7 @@
// We use thread-local runs for the size Brackets whose indexes
// are less than this index. We use shared (current) runs for the rest.
-
- static const size_t kNumThreadLocalSizeBrackets = 11;
+ static const size_t kNumThreadLocalSizeBrackets = 8;
private:
// The base address of the memory region that's managed by this allocator.