Unify symbols sharing across native scripts and runtime.

Unify symbols sharing across native scripts and runtime.

We currently have several ways to share symbols that are used in
both native scripts and the runtime. This change unifies this.
We do not use the symbols registry since we don't need the
registry any longer after bootstrapping, but the registry stays
alive afterwards.

R=mlippautz@chromium.org, rossberg@chromium.org

Committed: https://crrev.com/eaba98d97aa9c2847fffd8baa22d10d135ed15e0
Cr-Commit-Position: refs/heads/master@{#30280}

[Michael Lippautz, 2015-08-19 13:41:13]

heap lgtm

https://codereview.chromium.org/1293493004/diff/1/src/heap/heap.h
File src/heap/heap.h (right):

https://codereview.chromium.org/1293493004/diff/1/src/heap/heap.h#newcode297
src/heap/heap.h:297: #define PRIVATE_SYMBOL_LIST(V)              \
I wonder whether it would (at some point) make sense to move this somewhere
else? We create strong roots for those symbols but that's all the heap cares
about ;)

[Yang, 2015-08-19 13:51:49]

On 2015/08/19 13:41:13, Michael Lippautz wrote:
> heap lgtm
> 
> https://codereview.chromium.org/1293493004/diff/1/src/heap/heap.h
> File src/heap/heap.h (right):
> 
> https://codereview.chromium.org/1293493004/diff/1/src/heap/heap.h#newcode297
> src/heap/heap.h:297: #define PRIVATE_SYMBOL_LIST(V)              \
> I wonder whether it would (at some point) make sense to move this somewhere
> else? We create strong roots for those symbols but that's all the heap cares
> about ;)

Not sure. We also have a list of strings right above.

[Yang, 2015-08-20 11:47:03]

On 2015/08/19 13:51:49, Yang wrote:
> On 2015/08/19 13:41:13, Michael Lippautz wrote:
> > heap lgtm
> > 
> > https://codereview.chromium.org/1293493004/diff/1/src/heap/heap.h
> > File src/heap/heap.h (right):
> > 
> > https://codereview.chromium.org/1293493004/diff/1/src/heap/heap.h#newcode297
> > src/heap/heap.h:297: #define PRIVATE_SYMBOL_LIST(V)              \
> > I wonder whether it would (at some point) make sense to move this somewhere
> > else? We create strong roots for those symbols but that's all the heap cares
> > about ;)
> 
> Not sure. We also have a list of strings right above.

ping Andreas :)

[rossberg, 2015-08-20 12:50:12]

LGTM

It's sad that this makes the use of private symbols unmodular. But I guess
uniformity wins...

https://codereview.chromium.org/1293493004/diff/1/src/array-iterator.js
File src/array-iterator.js (right):

https://codereview.chromium.org/1293493004/diff/1/src/array-iterator.js#newco...
src/array-iterator.js:16: var ArrayIterationKindSymbol =
I don't feel too strongly, but don't we typically use capitalised names for
functions only?

[Yang, 2015-08-20 13:13:09]

On 2015/08/20 12:50:12, rossberg wrote:
> LGTM
> 
> It's sad that this makes the use of private symbols unmodular. But I guess
> uniformity wins...
> 
> https://codereview.chromium.org/1293493004/diff/1/src/array-iterator.js
> File src/array-iterator.js (right):
> 
>
https://codereview.chromium.org/1293493004/diff/1/src/array-iterator.js#newco...
> src/array-iterator.js:16: var ArrayIterationKindSymbol =
> I don't feel too strongly, but don't we typically use capitalised names for
> functions only?

Changed all symbols to lower case.

[Yang, 2015-08-20 13:13:13]

The CQ bit was checked by yangguo@chromium.org

[Yang, 2015-08-20 13:13:13]

The patchset sent to the CQ was uploaded after l-g-t-m from
mlippautz@chromium.org, rossberg@chromium.org
Link to the patchset: https://codereview.chromium.org/1293493004/#ps20001
(title: "rebase and use lower case for symbol names")

[commit-bot: I haz the power, 2015-08-20 13:13:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1293493004/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1293493004/20001

[Yang, 2015-08-20 13:14:39]

On 2015/08/20 12:50:12, rossberg wrote:
> LGTM
> 
> It's sad that this makes the use of private symbols unmodular. But I guess
> uniformity wins...

The upside is that we would share private symbols across different native
contexts instead of allocating new ones. This saves us from ICs going
polymorphic.


> 
> https://codereview.chromium.org/1293493004/diff/1/src/array-iterator.js
> File src/array-iterator.js (right):
> 
>
https://codereview.chromium.org/1293493004/diff/1/src/array-iterator.js#newco...
> src/array-iterator.js:16: var ArrayIterationKindSymbol =
> I don't feel too strongly, but don't we typically use capitalised names for
> functions only?

[commit-bot: I haz the power, 2015-08-20 14:09:49]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-20 14:09:50]

Failed to apply patch for src/heap/heap.h:
While running git apply --index -3 -p1;
  error: patch failed: src/heap/heap.h:1812
  Falling back to three-way merge...
  Applied patch to 'src/heap/heap.h' with conflicts.
  U src/heap/heap.h

Patch:       src/heap/heap.h
Index: src/heap/heap.h
diff --git a/src/heap/heap.h b/src/heap/heap.h
index
040144d4d61c3b9eec755ce77d1fe3b29f354daa..dd75b1e4fda95297940482f34f519699207b0165
100644
--- a/src/heap/heap.h
+++ b/src/heap/heap.h
@@ -267,9 +267,6 @@ namespace internal {
   V(WeakSet_string, "WeakSet")                                 \
   V(for_string, "for")                                         \
   V(for_api_string, "for_api")                                 \
-  V(for_intern_string, "for_intern")                           \
-  V(private_api_string, "private_api")                         \
-  V(private_intern_string, "private_intern")                   \
   V(Date_string, "Date")                                       \
   V(char_at_string, "CharAt")                                  \
   V(undefined_string, "undefined")                             \
@@ -300,31 +297,47 @@ namespace internal {
   V(Error_string, "Error")                                     \
   V(RegExp_string, "RegExp")
 
-#define PRIVATE_SYMBOL_LIST(V)      \
-  V(nonextensible_symbol)           \
-  V(sealed_symbol)                  \
-  V(hash_code_symbol)               \
-  V(frozen_symbol)                  \
-  V(nonexistent_symbol)             \
-  V(elements_transition_symbol)     \
-  V(observed_symbol)                \
-  V(uninitialized_symbol)           \
-  V(megamorphic_symbol)             \
-  V(premonomorphic_symbol)          \
-  V(stack_trace_symbol)             \
-  V(detailed_stack_trace_symbol)    \
-  V(normal_ic_symbol)               \
-  V(home_object_symbol)             \
-  V(intl_initialized_marker_symbol) \
-  V(intl_impl_object_symbol)        \
-  V(promise_debug_marker_symbol)    \
-  V(promise_has_handler_symbol)     \
-  V(class_start_position_symbol)    \
-  V(class_end_position_symbol)      \
-  V(error_start_pos_symbol)         \
-  V(error_end_pos_symbol)           \
-  V(error_script_symbol)            \
-  V(internal_error_symbol)
+#define PRIVATE_SYMBOL_LIST(V)              \
+  V(array_iteration_kind_symbol)            \
+  V(array_iterator_next_symbol)             \
+  V(array_iterator_object_symbol)           \
+  V(call_site_function_symbol)              \
+  V(call_site_position_symbol)              \
+  V(call_site_receiver_symbol)              \
+  V(call_site_strict_symbol)                \
+  V(class_end_position_symbol)              \
+  V(class_start_position_symbol)            \
+  V(detailed_stack_trace_symbol)            \
+  V(elements_transition_symbol)             \
+  V(error_end_pos_symbol)                   \
+  V(error_script_symbol)                    \
+  V(error_start_pos_symbol)                 \
+  V(formatted_stack_trace_symbol)           \
+  V(frozen_symbol)                          \
+  V(hash_code_symbol)                       \
+  V(home_object_symbol)                     \
+  V(internal_error_symbol)                  \
+  V(intl_impl_object_symbol)                \
+  V(intl_initialized_marker_symbol)         \
+  V(megamorphic_symbol)                     \
+  V(nonexistent_symbol)                     \
+  V(nonextensible_symbol)                   \
+  V(normal_ic_symbol)                       \
+  V(observed_symbol)                        \
+  V(premonomorphic_symbol)                  \
+  V(promise_debug_marker_symbol)            \
+  V(promise_has_handler_symbol)             \
+  V(promise_on_resolve_symbol)              \
+  V(promise_on_reject_symbol)               \
+  V(promise_raw_symbol)                     \
+  V(promise_status_symbol)                  \
+  V(promise_value_symbol)                   \
+  V(sealed_symbol)                          \
+  V(stack_trace_symbol)                     \
+  V(string_iterator_iterated_string_symbol) \
+  V(string_iterator_next_index_symbol)      \
+  V(uninitialized_symbol)
+
 
 #define PUBLIC_SYMBOL_LIST(V)                                    \
   V(has_instance_symbol, symbolHasInstance, Symbol.hasInstance)  \
@@ -1812,8 +1825,6 @@ class Heap {
   // any string when looked up in properties.
   String* hidden_string_;
 
-  void AddPrivateGlobalSymbols(Handle<Object> private_intern_table);
-
   struct GCCallbackPair {
     GCCallbackPair(v8::Isolate::GCCallback callback, GCType gc_type,
                    bool pass_isolate)

[commit-bot: I haz the power, 2015-08-20 14:09:55]

The CQ bit was unchecked by commit-bot@chromium.org

[Yang, 2015-08-21 06:18:47]

The CQ bit was checked by yangguo@chromium.org

[Yang, 2015-08-21 06:18:48]

The patchset sent to the CQ was uploaded after l-g-t-m from
mlippautz@chromium.org, rossberg@chromium.org
Link to the patchset: https://codereview.chromium.org/1293493004/#ps40001
(title: "rebase")

[commit-bot: I haz the power, 2015-08-21 06:19:01]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1293493004/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1293493004/40001

[commit-bot: I haz the power, 2015-08-21 06:44:24]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-08-21 06:44:43]

Patchset 3 (id:??) landed as
https://crrev.com/eaba98d97aa9c2847fffd8baa22d10d135ed15e0
Cr-Commit-Position: refs/heads/master@{#30280}