[counters] Annotate v8 with more runtime call counters.

[counters] Annotate v8 with more runtime call counters.

By fully annotating the API with runtime counters we can properly measure
how much time we spend in total in v8. When --runtime-call-stats is specified
we now disable the fast-paths for callbacks to properly measure them.
As a drive-by-fix this CL unifies the LOG messages in api.cc.
Additionally we added missing timers to gain better resolution in the parser
and callbacks.

BUG=

Committed: https://crrev.com/407d9fcec722d21e886eb581173240887fdd9c4c
Cr-Commit-Position: refs/heads/master@{#36248}

[Camillo Bruni, 2016-04-27 12:11:05]

Description was changed from

==========
[counter] Annotate api.cc with more runtime call counters.

BUG=
==========

to

==========
[counters] Annotate api.cc with more runtime call counters.

BUG=
==========

[Camillo Bruni, 2016-05-04 12:31:49]

cbruni@chromium.org changed reviewers:
+ jochen@chromium.org, mstarzinger@chromium.org

[Camillo Bruni, 2016-05-04 12:31:50]

mstartzinger@ PTAL compiler.cc
jochen@ the rest

[Camillo Bruni, 2016-05-06 08:53:54]

The CQ bit was checked by cbruni@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-05-06 08:54:06]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1923893002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1923893002/120001

[commit-bot: I haz the power, 2016-05-06 08:55:16]

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

[commit-bot: I haz the power, 2016-05-06 08:55:17]

Dry run: Try jobs failed on following builders:
  v8_android_arm_compile_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_android_arm_compile_rel/...)
  v8_linux64_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_rel_ng/builds/5353)
  v8_linux_dbg_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_dbg_ng/builds/5364)
  v8_linux_mipsel_compile_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_mipsel_compile_rel...)
  v8_mac_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel_ng/builds/1239)
  v8_presubmit on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_presubmit/builds/14750)

[Michael Starzinger, 2016-05-06 09:36:00]

LGTM on compiler and full-codegen. Didn't look at the rest.

[jochen (gone - plz use gerrit), 2016-05-06 09:39:54]

https://codereview.chromium.org/1923893002/diff/120001/src/api.cc
File src/api.cc (right):

https://codereview.chromium.org/1923893002/diff/120001/src/api.cc#newcode72
src/api.cc:72: i::Isolate* _log_isolate = isolate;      \
why's that required?

[Camillo Bruni, 2016-05-06 10:47:28]

jochen@: removed isolate tmp var

[jochen (gone - plz use gerrit), 2016-05-06 10:54:58]

https://codereview.chromium.org/1923893002/diff/180001/src/api.cc
File src/api.cc (right):

https://codereview.chromium.org/1923893002/diff/180001/src/api.cc#newcode1154
src/api.cc:1154: LOG_API(i_isolate, FunctionTemplate_New,
"FunctionTemplate::New");
sorry for not adding this comment earlier, I promise it's the last comment

why not LOG_API(i_isolate, FunctionTemplate, New) and

#def LOG_API(isolate, obj, meth) \
  RUNTIME_COUNTER(isolate obj ## _ ## meth); \
  LOG(isolate, #obj "::" #meth)

[Camillo Bruni, 2016-05-06 11:01:17]

https://codereview.chromium.org/1923893002/diff/180001/src/api.cc
File src/api.cc (right):

https://codereview.chromium.org/1923893002/diff/180001/src/api.cc#newcode1154
src/api.cc:1154: LOG_API(i_isolate, FunctionTemplate_New,
"FunctionTemplate::New");
On 2016/05/06 at 10:54:58, jochen wrote:
> sorry for not adding this comment earlier, I promise it's the last comment
> 
> why not LOG_API(i_isolate, FunctionTemplate, New) and
> 
> #def LOG_API(isolate, obj, meth) \
>   RUNTIME_COUNTER(isolate obj ## _ ## meth); \
>   LOG(isolate, #obj "::" #meth)

if the naming were more consistent I would agree to this, but you have a
"random" set of api-functions which use the "v8::" prefix and some don't.
Unless of course we wan't to just change it entirely (though I see now that I
acidentally stripped "()" from some log messages :/)

[jochen (gone - plz use gerrit), 2016-05-09 15:44:58]

On 2016/05/06 at 11:01:17, cbruni wrote:
> https://codereview.chromium.org/1923893002/diff/180001/src/api.cc
> File src/api.cc (right):
> 
> https://codereview.chromium.org/1923893002/diff/180001/src/api.cc#newcode1154
> src/api.cc:1154: LOG_API(i_isolate, FunctionTemplate_New,
"FunctionTemplate::New");
> On 2016/05/06 at 10:54:58, jochen wrote:
> > sorry for not adding this comment earlier, I promise it's the last comment
> > 
> > why not LOG_API(i_isolate, FunctionTemplate, New) and
> > 
> > #def LOG_API(isolate, obj, meth) \
> >   RUNTIME_COUNTER(isolate obj ## _ ## meth); \
> >   LOG(isolate, #obj "::" #meth)
> 
> if the naming were more consistent I would agree to this, but you have a
"random" set of api-functions which use the "v8::" prefix and some don't.
> Unless of course we wan't to just change it entirely (though I see now that I
acidentally stripped "()" from some log messages :/)

I think it's fine for the sake of unifying it

[Camillo Bruni, 2016-05-11 14:27:46]

jochen PTAL again, changed now all messages and names to be predictable

[jochen (gone - plz use gerrit), 2016-05-11 14:30:44]

lgtm

[Camillo Bruni, 2016-05-11 14:35:23]

The CQ bit was checked by cbruni@chromium.org

[Camillo Bruni, 2016-05-11 14:35:23]

The patchset sent to the CQ was uploaded after l-g-t-m from
mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1923893002/#ps220001
(title: "using more consistent api log messages")

[commit-bot: I haz the power, 2016-05-11 14:35:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1923893002/220001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1923893002/220001

[commit-bot: I haz the power, 2016-05-11 14:36:44]

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

[commit-bot: I haz the power, 2016-05-11 14:36:45]

Try jobs failed on following builders:
  v8_linux64_avx2_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_avx2_rel_ng/buil...)
  v8_linux_arm64_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel_ng/build...)
  v8_linux_arm_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm_rel_ng/builds/...)
  v8_linux_gcc_compile_rel on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_gcc_compile_rel/bu...)
  v8_linux_nodcheck_rel_ng on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_nodcheck_rel_ng/bu...)

[Camillo Bruni, 2016-05-12 10:52:13]

cbruni@chromium.org changed reviewers:
+ ishell@chromium.org

[Camillo Bruni, 2016-05-12 10:52:13]

Igor can you have a look at ic.cc, I changed it to what we discussed offline.

[Camillo Bruni, 2016-05-12 11:13:20]

Description was changed from

==========
[counters] Annotate api.cc with more runtime call counters.

BUG=
==========

to

==========
[counters] Annotate api.cc with more runtime call counters.

By fully annotating the API with runtime counters we can properly measure
how much time we spend in total in v8. When --runtime-call-stats is specified
we now disable the fast-paths for callbacks to properly measure them.
As a drive-by-fix this CL unifies the LOG messages in api.cc.

BUG=
==========

[Igor Sheludko, 2016-05-12 11:19:37]

ic changes lgtm

[Camillo Bruni, 2016-05-12 11:25:01]

The CQ bit was checked by cbruni@chromium.org

[Camillo Bruni, 2016-05-12 11:25:02]

The patchset sent to the CQ was uploaded after l-g-t-m from jochen@chromium.org,
mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1923893002/#ps300001
(title: "directly return slow_stub")

[commit-bot: I haz the power, 2016-05-12 11:25:13]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1923893002/300001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1923893002/300001

[commit-bot: I haz the power, 2016-05-12 11:33:04]

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

[commit-bot: I haz the power, 2016-05-12 11:33:05]

Try jobs failed on following builders:
  v8_presubmit on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_presubmit/builds/15130)

[Camillo Bruni, 2016-05-12 15:37:43]

cbruni@chromium.org changed reviewers:
+ adamk@chromium.org

[Camillo Bruni, 2016-05-12 15:37:45]

adamk@ could have a look at the added timers in parser/*?

[adamk, 2016-05-12 20:44:50]

Besides the style nits, this lgtm structurally (given that these are being added
in exactly the same places we have TRACE macros).

But I don't understand why there are modifications in here given that the CL
description is just talking about api.cc. Can you explain in the CL description
what more this is doing? Perhaps the CL has been expanded since that was
initially written?

https://codereview.chromium.org/1923893002/diff/300001/src/parsing/parser.cc
File src/parsing/parser.cc (right):

https://codereview.chromium.org/1923893002/diff/300001/src/parsing/parser.cc#...
src/parsing/parser.cc:823: RuntimeCallTimerScope runtimeTimer(isolate,
&RuntimeCallStats::Parse);
Naming nit: this should be runtime_timer

https://codereview.chromium.org/1923893002/diff/300001/src/parsing/parser.cc#...
src/parsing/parser.cc:988: RuntimeCallTimerScope runtimeTimer(isolate,
&RuntimeCallStats::ParseLazy);
and here

[Camillo Bruni, 2016-05-13 07:22:23]

Description was changed from

==========
[counters] Annotate api.cc with more runtime call counters.

By fully annotating the API with runtime counters we can properly measure
how much time we spend in total in v8. When --runtime-call-stats is specified
we now disable the fast-paths for callbacks to properly measure them.
As a drive-by-fix this CL unifies the LOG messages in api.cc.

BUG=
==========

to

==========
[counters] Annotate v8 with more runtime call counters.

By fully annotating the API with runtime counters we can properly measure
how much time we spend in total in v8. When --runtime-call-stats is specified
we now disable the fast-paths for callbacks to properly measure them.
As a drive-by-fix this CL unifies the LOG messages in api.cc.
Additionally we added missing timers to gain better resolution in the parser
and callbacks.

BUG=
==========

[Camillo Bruni, 2016-05-13 13:59:46]

The CQ bit was checked by cbruni@chromium.org

[Camillo Bruni, 2016-05-13 13:59:47]

The patchset sent to the CQ was uploaded after l-g-t-m from ishell@chromium.org,
jochen@chromium.org, adamk@chromium.org, mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1923893002/#ps320001
(title: "nits")

[commit-bot: I haz the power, 2016-05-13 13:59:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1923893002/320001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1923893002/320001

[commit-bot: I haz the power, 2016-05-13 15:05:33]

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

[commit-bot: I haz the power, 2016-05-13 15:05:35]

Failed to apply patch for src/api.cc:
While running git apply --index -3 -p1;
  error: patch failed: src/api.cc:1782
  Falling back to three-way merge...
  Applied patch to 'src/api.cc' with conflicts.
  U src/api.cc

Patch:       src/api.cc
Index: src/api.cc
diff --git a/src/api.cc b/src/api.cc
index
49b078c51030f4ea7a0dada06e81707676567170..691da057f961cabcdcae0070902ca37fa8a70290
100644
--- a/src/api.cc
+++ b/src/api.cc
@@ -68,54 +68,54 @@
 
 namespace v8 {
 
-#define LOG_API(isolate, expr) LOG(isolate, ApiEntryCall(expr))
-
+#define LOG_API(isolate, class_name, function_name)                       \
+  i::RuntimeCallTimerScope _runtime_timer(                                \
+      isolate, &i::RuntimeCallStats::API_##class_name##_##function_name); \
+  LOG(isolate, ApiEntryCall("v8::" #class_name "::" #function_name))
 
 #define ENTER_V8(isolate) i::VMState<v8::OTHER> __state__((isolate))
 
-
-#define PREPARE_FOR_EXECUTION_GENERIC(isolate, context, function_name, \
-                                      bailout_value, HandleScopeClass, \
-                                      do_callback)                     \
-  if (IsExecutionTerminatingCheck(isolate)) {                          \
-    return bailout_value;                                              \
-  }                                                                    \
-  HandleScopeClass handle_scope(isolate);                              \
-  CallDepthScope call_depth_scope(isolate, context, do_callback);      \
-  LOG_API(isolate, function_name);                                     \
-  ENTER_V8(isolate);                                                   \
+#define PREPARE_FOR_EXECUTION_GENERIC(isolate, context, class_name,  \
+                                      function_name, bailout_value,  \
+                                      HandleScopeClass, do_callback) \
+  if (IsExecutionTerminatingCheck(isolate)) {                        \
+    return bailout_value;                                            \
+  }                                                                  \
+  HandleScopeClass handle_scope(isolate);                            \
+  CallDepthScope call_depth_scope(isolate, context, do_callback);    \
+  LOG_API(isolate, class_name, function_name);                       \
+  ENTER_V8(isolate);                                                 \
   bool has_pending_exception = false
 
-
-#define PREPARE_FOR_EXECUTION_WITH_CONTEXT(                                  \
-    context, function_name, bailout_value, HandleScopeClass, do_callback)    \
-  auto isolate = context.IsEmpty()                                           \
-                     ? i::Isolate::Current()                                 \
-                     : reinterpret_cast<i::Isolate*>(context->GetIsolate()); \
-  PREPARE_FOR_EXECUTION_GENERIC(isolate, context, function_name,             \
+#define PREPARE_FOR_EXECUTION_WITH_CONTEXT(context, class_name, function_name,
\
+                                           bailout_value, HandleScopeClass,   
\
+                                           do_callback)                       
\
+  auto isolate = context.IsEmpty()                                            
\
+                     ? i::Isolate::Current()                                  
\
+                     : reinterpret_cast<i::Isolate*>(context->GetIsolate());  
\
+  PREPARE_FOR_EXECUTION_GENERIC(isolate, context, class_name, function_name,  
\
                                 bailout_value, HandleScopeClass, do_callback);
 
+#define PREPARE_FOR_EXECUTION_WITH_ISOLATE(isolate, class_name, function_name,
\
+                                           T)                                 
\
+  PREPARE_FOR_EXECUTION_GENERIC(isolate, Local<Context>(), class_name,        
\
+                                function_name, MaybeLocal<T>(),               
\
+                                InternalEscapableScope, false);
 
-#define PREPARE_FOR_EXECUTION_WITH_ISOLATE(isolate, function_name, T)     \
-  PREPARE_FOR_EXECUTION_GENERIC(isolate, Local<Context>(), function_name, \
-                                MaybeLocal<T>(), InternalEscapableScope,  \
-                                false);
-
-
-#define PREPARE_FOR_EXECUTION(context, function_name, T)                      \
-  PREPARE_FOR_EXECUTION_WITH_CONTEXT(context, function_name, MaybeLocal<T>(), \
-                                     InternalEscapableScope, false)
-
+#define PREPARE_FOR_EXECUTION(context, class_name, function_name, T)          \
+  PREPARE_FOR_EXECUTION_WITH_CONTEXT(context, class_name, function_name,      \
+                                     MaybeLocal<T>(), InternalEscapableScope, \
+                                     false)
 
-#define PREPARE_FOR_EXECUTION_WITH_CALLBACK(context, function_name, T)        \
-  PREPARE_FOR_EXECUTION_WITH_CONTEXT(context, function_name, MaybeLocal<T>(), \
-                                     InternalEscapableScope, true)
-
-
-#define PREPARE_FOR_EXECUTION_PRIMITIVE(context, function_name, T)         \
-  PREPARE_FOR_EXECUTION_WITH_CONTEXT(context, function_name, Nothing<T>(), \
-                                     i::HandleScope, false)
+#define PREPARE_FOR_EXECUTION_WITH_CALLBACK(context, class_name,              \
+                                            function_name, T)                 \
+  PREPARE_FOR_EXECUTION_WITH_CONTEXT(context, class_name, function_name,      \
+                                     MaybeLocal<T>(), InternalEscapableScope, \
+                                     true)
 
+#define PREPARE_FOR_EXECUTION_PRIMITIVE(context, class_name, function_name, T)
\
+  PREPARE_FOR_EXECUTION_WITH_CONTEXT(context, class_name, function_name,      
\
+                                     Nothing<T>(), i::HandleScope, false)
 
 #define EXCEPTION_BAILOUT_CHECK_SCOPED(isolate, value) \
   do {                                                 \
@@ -657,7 +657,7 @@ void SetResourceConstraints(i::Isolate* isolate,
 
 
 i::Object** V8::GlobalizeReference(i::Isolate* isolate, i::Object** obj) {
-  LOG_API(isolate, "Persistent::New");
+  LOG_API(isolate, Persistent, New);
   i::Handle<i::Object> result = isolate->global_handles()->Create(*obj);
 #ifdef VERIFY_HEAP
   if (i::FLAG_verify_heap) {
@@ -1141,7 +1141,7 @@ Local<FunctionTemplate> FunctionTemplate::New(Isolate*
isolate,
   // Changes to the environment cannot be captured in the snapshot. Expect no
   // function templates when the isolate is created for serialization.
   DCHECK(!i_isolate->serializer_enabled());
-  LOG_API(i_isolate, "FunctionTemplate::New");
+  LOG_API(i_isolate, FunctionTemplate, New);
   ENTER_V8(i_isolate);
   return FunctionTemplateNew(i_isolate, callback, nullptr, data, signature,
                              length, false);
@@ -1154,7 +1154,7 @@ Local<FunctionTemplate>
FunctionTemplate::NewWithFastHandler(
     v8::Local<Signature> signature, int length) {
   i::Isolate* i_isolate = reinterpret_cast<i::Isolate*>(isolate);
   DCHECK(!i_isolate->serializer_enabled());
-  LOG_API(i_isolate, "FunctionTemplate::NewWithFastHandler");
+  LOG_API(i_isolate, FunctionTemplate, NewWithFastHandler);
   ENTER_V8(i_isolate);
   return FunctionTemplateNew(i_isolate, callback, fast_handler, data,
signature,
                              length, false);
@@ -1338,7 +1338,7 @@ static Local<ObjectTemplate> ObjectTemplateNew(
   // Changes to the environment cannot be captured in the snapshot. Expect no
   // object templates when the isolate is created for serialization.
   DCHECK(!isolate->serializer_enabled());
-  LOG_API(isolate, "ObjectTemplate::New");
+  LOG_API(isolate, ObjectTemplate, New);
   ENTER_V8(isolate);
   i::Handle<i::Struct> struct_obj =
       isolate->factory()->NewStruct(i::OBJECT_TEMPLATE_INFO_TYPE);
@@ -1716,7 +1716,7 @@ int UnboundScript::GetId() {
   i::Handle<i::HeapObject> obj =
       i::Handle<i::HeapObject>::cast(Utils::OpenHandle(this));
   i::Isolate* isolate = obj->GetIsolate();
-  LOG_API(isolate, "v8::UnboundScript::GetId");
+  LOG_API(isolate, UnboundScript, GetId);
   i::HandleScope scope(isolate);
   i::Handle<i::SharedFunctionInfo> function_info(
       i::SharedFunctionInfo::cast(*obj));
@@ -1729,7 +1729,7 @@ int UnboundScript::GetLineNumber(int code_pos) {
   i::Handle<i::SharedFunctionInfo> obj =
       i::Handle<i::SharedFunctionInfo>::cast(Utils::OpenHandle(this));
   i::Isolate* isolate = obj->GetIsolate();
-  LOG_API(isolate, "UnboundScript::GetLineNumber");
+  LOG_API(isolate, UnboundScript, GetLineNumber);
   if (obj->script()->IsScript()) {
     i::Handle<i::Script> script(i::Script::cast(obj->script()));
     return i::Script::GetLineNumber(script, code_pos);
@@ -1743,7 +1743,7 @@ Local<Value> UnboundScript::GetScriptName() {
   i::Handle<i::SharedFunctionInfo> obj =
       i::Handle<i::SharedFunctionInfo>::cast(Utils::OpenHandle(this));
   i::Isolate* isolate = obj->GetIsolate();
-  LOG_API(isolate, "UnboundScript::GetName");
+  LOG_API(isolate, UnboundScript, GetName);
   if (obj->script()->IsScript()) {
     i::Object* name = i::Script::cast(obj->script())->name();
     return Utils::ToLocal(i::Handle<i::Object>(name, isolate));
@@ -1757,7 +1757,7 @@ Local<Value> UnboundScript::GetSourceURL() {
   i::Handle<i::SharedFunctionInfo> obj =
       i::Handle<i::SharedFunctionInfo>::cast(Utils::OpenHandle(this));
   i::Isolate* isolate = obj->GetIsolate();
-  LOG_API(isolate, "UnboundScript::GetSourceURL");
+  LOG_API(isolate, UnboundScript, GetSourceURL);
   if (obj->script()->IsScript()) {
     i::Object* url = i::Script::cast(obj->script())->source_url();
     return Utils::ToLocal(i::Handle<i::Object>(url, isolate));
@@ -1771,7 +1771,7 @@ Local<Value> UnboundScript::GetSourceMappingURL() {
   i::Handle<i::SharedFunctionInfo> obj =
       i::Handle<i::SharedFunctionInfo>::cast(Utils::OpenHandle(this));
   i::Isolate* isolate = obj->GetIsolate();
-  LOG_API(isolate, "UnboundScript::GetSourceMappingURL");
+  LOG_API(iso…
(message too large)

[Camillo Bruni, 2016-05-13 15:22:47]

The CQ bit was checked by cbruni@chromium.org

[Camillo Bruni, 2016-05-13 15:22:48]

The patchset sent to the CQ was uploaded after l-g-t-m from ishell@chromium.org,
jochen@chromium.org, adamk@chromium.org, mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1923893002/#ps340001
(title: "merging master")

[commit-bot: I haz the power, 2016-05-13 15:22:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1923893002/340001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1923893002/340001

[commit-bot: I haz the power, 2016-05-13 15:53:36]

Committed patchset #18 (id:340001)

[commit-bot: I haz the power, 2016-05-13 15:54:54]

Description was changed from

==========
[counters] Annotate v8 with more runtime call counters.

By fully annotating the API with runtime counters we can properly measure
how much time we spend in total in v8. When --runtime-call-stats is specified
we now disable the fast-paths for callbacks to properly measure them.
As a drive-by-fix this CL unifies the LOG messages in api.cc.
Additionally we added missing timers to gain better resolution in the parser
and callbacks.

BUG=
==========

to

==========
[counters] Annotate v8 with more runtime call counters.

By fully annotating the API with runtime counters we can properly measure
how much time we spend in total in v8. When --runtime-call-stats is specified
we now disable the fast-paths for callbacks to properly measure them.
As a drive-by-fix this CL unifies the LOG messages in api.cc.
Additionally we added missing timers to gain better resolution in the parser
and callbacks.

BUG=

Committed: https://crrev.com/407d9fcec722d21e886eb581173240887fdd9c4c
Cr-Commit-Position: refs/heads/master@{#36248}
==========

[commit-bot: I haz the power, 2016-05-13 15:54:56]

Patchset 18 (id:??) landed as
https://crrev.com/407d9fcec722d21e886eb581173240887fdd9c4c
Cr-Commit-Position: refs/heads/master@{#36248}