Reviewers: Erik Corry,
Message:
Is this comment out of sync? I am trying to figure out what is the rule to
use
roots_array_start in the IA32 code stub generation. Sometimes
roots_array_start
is used and sometimes factory->the_hole_value or others are used.
From this comment, I got an impression that using the entry above 32 might
be
unsafe but I see "factory->single_character_string_cache (above 32)" used in
code-stubs-ia32.cc.
It would be great if the reviewer could explain how V8 ensures the first 32
roots are never on a page that is being compacted which would help me
understand
V8 better :)
Description:
Move a comment to its right position
Please review this at https://codereview.chromium.org/11593004/
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affected files:
M src/heap.h
Index: src/heap.h
===================================================================
--- src/heap.h (revision 13225)
+++ src/heap.h (working copy)
@@ -78,11 +78,11 @@
V(DescriptorArray, empty_descriptor_array,
EmptyDescriptorArray) \
V(Smi, stack_limit,
StackLimit) \
V(Oddball, arguments_marker,
ArgumentsMarker) \
+ V(FixedArray, number_string_cache,
NumberStringCache) \
+ V(Object, instanceof_cache_function,
InstanceofCacheFunction) \
/* The first 32 roots above this line should be boring from a GC point
of */ \
/* view. This means they are never in new space and never on a page
that */ \
/* is being
compacted. */ \
- V(FixedArray, number_string_cache,
NumberStringCache) \
- V(Object, instanceof_cache_function,
InstanceofCacheFunction) \
V(Object, instanceof_cache_map,
InstanceofCacheMap) \
V(Object, instanceof_cache_answer,
InstanceofCacheAnswer) \
V(FixedArray, single_character_string_cache,
SingleCharacterStringCache) \
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
