[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
// Copyright 2016 the V8 project authors. All rights reserved.
|
|
|
|
|
// Use of this source code is governed by a BSD-style license that can be
|
|
|
|
|
// found in the LICENSE file.
|
|
|
|
|
|
|
|
|
|
#include "test/cctest/interpreter/bytecode-expectations-printer.h"
|
|
|
|
|
|
2016-05-11 12:21:56 +00:00
|
|
|
#include <iomanip>
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
#include <iostream>
|
2016-02-19 12:36:26 +00:00
|
|
|
#include <vector>
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
|
|
|
|
#include "include/libplatform/libplatform.h"
|
|
|
|
|
#include "include/v8.h"
|
2019-05-17 12:13:44 +00:00
|
|
|
#include "src/api/api-inl.h"
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
#include "src/base/logging.h"
|
2019-05-21 09:30:15 +00:00
|
|
|
#include "src/codegen/source-position-table.h"
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
#include "src/interpreter/bytecode-array-iterator.h"
|
|
|
|
|
#include "src/interpreter/bytecode-generator.h"
|
|
|
|
|
#include "src/interpreter/bytecodes.h"
|
2016-06-21 14:37:16 +00:00
|
|
|
#include "src/interpreter/interpreter-intrinsics.h"
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
#include "src/interpreter/interpreter.h"
|
2018-12-17 17:01:48 +00:00
|
|
|
#include "src/objects/heap-number-inl.h"
|
2018-05-23 13:29:02 +00:00
|
|
|
#include "src/objects/module-inl.h"
|
2019-05-23 08:51:46 +00:00
|
|
|
#include "src/objects/objects-inl.h"
|
2018-05-23 13:29:02 +00:00
|
|
|
#include "src/runtime/runtime.h"
|
2019-05-23 13:27:57 +00:00
|
|
|
#include "src/utils/ostreams.h"
|
2018-08-07 13:14:23 +00:00
|
|
|
#include "test/cctest/cctest.h"
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
|
|
|
|
namespace v8 {
|
|
|
|
|
namespace internal {
|
|
|
|
|
namespace interpreter {
|
|
|
|
|
|
2016-12-21 16:41:11 +00:00
|
|
|
static const char* NameForNativeContextIntrinsicIndex(uint32_t idx) {
|
|
|
|
|
switch (idx) {
|
|
|
|
|
#define COMPARE_NATIVE_CONTEXT_INTRINSIC_IDX(NAME, Type, name) \
|
|
|
|
|
case Context::NAME: \
|
|
|
|
|
return #name;
|
|
|
|
|
|
|
|
|
|
NATIVE_CONTEXT_INTRINSIC_FUNCTIONS(COMPARE_NATIVE_CONTEXT_INTRINSIC_IDX)
|
|
|
|
|
|
|
|
|
|
default:
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return "UnknownIntrinsicIndex";
|
|
|
|
|
}
|
|
|
|
|
|
2016-02-19 12:36:26 +00:00
|
|
|
// static
|
|
|
|
|
const char* const BytecodeExpectationsPrinter::kDefaultTopFunctionName =
|
|
|
|
|
"__genbckexp_wrapper__";
|
2016-05-11 12:21:56 +00:00
|
|
|
const char* const BytecodeExpectationsPrinter::kIndent = " ";
|
2016-02-19 12:36:26 +00:00
|
|
|
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
v8::Local<v8::String> BytecodeExpectationsPrinter::V8StringFromUTF8(
|
|
|
|
|
const char* data) const {
|
|
|
|
|
return v8::String::NewFromUtf8(isolate_, data, v8::NewStringType::kNormal)
|
|
|
|
|
.ToLocalChecked();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
std::string BytecodeExpectationsPrinter::WrapCodeInFunction(
|
|
|
|
|
const char* function_name, const std::string& function_body) const {
|
|
|
|
|
std::ostringstream program_stream;
|
|
|
|
|
program_stream << "function " << function_name << "() {" << function_body
|
|
|
|
|
<< "}\n"
|
|
|
|
|
<< function_name << "();";
|
|
|
|
|
|
|
|
|
|
return program_stream.str();
|
|
|
|
|
}
|
|
|
|
|
|
2016-10-04 18:42:13 +00:00
|
|
|
v8::Local<v8::Script> BytecodeExpectationsPrinter::CompileScript(
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
const char* program) const {
|
|
|
|
|
v8::Local<v8::String> source = V8StringFromUTF8(program);
|
2016-02-19 12:36:26 +00:00
|
|
|
return v8::Script::Compile(isolate_->GetCurrentContext(), source)
|
|
|
|
|
.ToLocalChecked();
|
|
|
|
|
}
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
2016-10-04 18:42:13 +00:00
|
|
|
v8::Local<v8::Module> BytecodeExpectationsPrinter::CompileModule(
|
|
|
|
|
const char* program) const {
|
2017-01-19 06:59:20 +00:00
|
|
|
ScriptOrigin origin(
|
|
|
|
|
Local<v8::Value>(), Local<v8::Integer>(), Local<v8::Integer>(),
|
|
|
|
|
Local<v8::Boolean>(), Local<v8::Integer>(), Local<v8::Value>(),
|
|
|
|
|
Local<v8::Boolean>(), Local<v8::Boolean>(), True(isolate_));
|
|
|
|
|
v8::ScriptCompiler::Source source(V8StringFromUTF8(program), origin);
|
2016-10-04 18:42:13 +00:00
|
|
|
return v8::ScriptCompiler::CompileModule(isolate_, &source).ToLocalChecked();
|
|
|
|
|
}
|
|
|
|
|
|
2016-02-19 12:36:26 +00:00
|
|
|
void BytecodeExpectationsPrinter::Run(v8::Local<v8::Script> script) const {
|
2018-04-13 22:28:05 +00:00
|
|
|
MaybeLocal<Value> result = script->Run(isolate_->GetCurrentContext());
|
|
|
|
|
USE(result);
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
i::Handle<v8::internal::BytecodeArray>
|
|
|
|
|
BytecodeExpectationsPrinter::GetBytecodeArrayForGlobal(
|
|
|
|
|
const char* global_name) const {
|
|
|
|
|
const v8::Local<v8::Context>& context = isolate_->GetCurrentContext();
|
|
|
|
|
v8::Local<v8::String> v8_global_name = V8StringFromUTF8(global_name);
|
|
|
|
|
v8::Local<v8::Function> function = v8::Local<v8::Function>::Cast(
|
|
|
|
|
context->Global()->Get(context, v8_global_name).ToLocalChecked());
|
|
|
|
|
i::Handle<i::JSFunction> js_function =
|
|
|
|
|
i::Handle<i::JSFunction>::cast(v8::Utils::OpenHandle(*function));
|
|
|
|
|
|
|
|
|
|
i::Handle<i::BytecodeArray> bytecodes =
|
2018-04-10 14:31:06 +00:00
|
|
|
i::handle(js_function->shared().GetBytecodeArray(), i_isolate());
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
|
|
|
|
return bytecodes;
|
|
|
|
|
}
|
|
|
|
|
|
2016-10-04 18:42:13 +00:00
|
|
|
i::Handle<i::BytecodeArray>
|
|
|
|
|
BytecodeExpectationsPrinter::GetBytecodeArrayForModule(
|
|
|
|
|
v8::Local<v8::Module> module) const {
|
|
|
|
|
i::Handle<i::Module> i_module = v8::Utils::OpenHandle(*module);
|
2019-06-19 17:07:58 +00:00
|
|
|
return i::handle(SharedFunctionInfo::cast(
|
|
|
|
|
Handle<i::SourceTextModule>::cast(i_module)->code())
|
|
|
|
|
.GetBytecodeArray(),
|
|
|
|
|
i_isolate());
|
2016-10-04 18:42:13 +00:00
|
|
|
}
|
|
|
|
|
|
2016-02-19 12:36:26 +00:00
|
|
|
i::Handle<i::BytecodeArray>
|
|
|
|
|
BytecodeExpectationsPrinter::GetBytecodeArrayForScript(
|
|
|
|
|
v8::Local<v8::Script> script) const {
|
|
|
|
|
i::Handle<i::JSFunction> js_function = v8::Utils::OpenHandle(*script);
|
2018-04-10 14:31:06 +00:00
|
|
|
return i::handle(js_function->shared().GetBytecodeArray(), i_isolate());
|
2016-02-19 12:36:26 +00:00
|
|
|
}
|
|
|
|
|
|
2018-08-07 13:14:23 +00:00
|
|
|
i::Handle<i::BytecodeArray>
|
|
|
|
|
BytecodeExpectationsPrinter::GetBytecodeArrayOfCallee(
|
|
|
|
|
const char* source_code) const {
|
|
|
|
|
i::Handle<i::Object> i_object =
|
|
|
|
|
v8::Utils::OpenHandle(*CompileRun(source_code));
|
|
|
|
|
i::Handle<i::JSFunction> js_function =
|
|
|
|
|
i::Handle<i::JSFunction>::cast(i_object);
|
|
|
|
|
CHECK(js_function->shared().HasBytecodeArray());
|
|
|
|
|
return i::handle(js_function->shared().GetBytecodeArray(), i_isolate());
|
|
|
|
|
}
|
|
|
|
|
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
void BytecodeExpectationsPrinter::PrintEscapedString(
|
|
|
|
|
std::ostream& stream, const std::string& string) const {
|
|
|
|
|
for (char c : string) {
|
|
|
|
|
switch (c) {
|
|
|
|
|
case '"':
|
|
|
|
|
stream << "\\\"";
|
|
|
|
|
break;
|
|
|
|
|
case '\\':
|
|
|
|
|
stream << "\\\\";
|
|
|
|
|
break;
|
|
|
|
|
default:
|
|
|
|
|
stream << c;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintBytecodeOperand(
|
2016-05-11 12:21:56 +00:00
|
|
|
std::ostream& stream, const BytecodeArrayIterator& bytecode_iterator,
|
2016-02-19 12:36:26 +00:00
|
|
|
const Bytecode& bytecode, int op_index, int parameter_count) const {
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
OperandType op_type = Bytecodes::GetOperandType(bytecode, op_index);
|
2016-03-21 17:08:21 +00:00
|
|
|
OperandSize op_size = Bytecodes::GetOperandSize(
|
2016-05-11 12:21:56 +00:00
|
|
|
bytecode, op_index, bytecode_iterator.current_operand_scale());
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
|
|
|
|
const char* size_tag;
|
|
|
|
|
switch (op_size) {
|
|
|
|
|
case OperandSize::kByte:
|
|
|
|
|
size_tag = "8";
|
|
|
|
|
break;
|
|
|
|
|
case OperandSize::kShort:
|
|
|
|
|
size_tag = "16";
|
|
|
|
|
break;
|
2016-03-21 17:08:21 +00:00
|
|
|
case OperandSize::kQuad:
|
|
|
|
|
size_tag = "32";
|
|
|
|
|
break;
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
default:
|
|
|
|
|
UNREACHABLE();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (Bytecodes::IsRegisterOperandType(op_type)) {
|
2016-05-11 12:21:56 +00:00
|
|
|
Register register_value = bytecode_iterator.GetRegisterOperand(op_index);
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
stream << 'R';
|
|
|
|
|
if (op_size != OperandSize::kByte) stream << size_tag;
|
2017-08-11 15:16:31 +00:00
|
|
|
if (register_value.is_current_context()) {
|
2016-02-19 12:36:26 +00:00
|
|
|
stream << "(context)";
|
|
|
|
|
} else if (register_value.is_function_closure()) {
|
|
|
|
|
stream << "(closure)";
|
|
|
|
|
} else if (register_value.is_parameter()) {
|
|
|
|
|
int parameter_index = register_value.ToParameterIndex(parameter_count);
|
|
|
|
|
if (parameter_index == 0) {
|
|
|
|
|
stream << "(this)";
|
|
|
|
|
} else {
|
|
|
|
|
stream << "(arg" << (parameter_index - 1) << ')';
|
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
stream << '(' << register_value.index() << ')';
|
|
|
|
|
}
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
} else {
|
2016-03-21 17:08:21 +00:00
|
|
|
switch (op_type) {
|
|
|
|
|
case OperandType::kFlag8:
|
2017-01-25 17:39:24 +00:00
|
|
|
stream << 'U' << size_tag << '(';
|
2016-05-11 12:21:56 +00:00
|
|
|
stream << bytecode_iterator.GetFlagOperand(op_index);
|
2016-03-21 17:08:21 +00:00
|
|
|
break;
|
2016-12-21 16:41:11 +00:00
|
|
|
case OperandType::kIdx: {
|
2017-01-25 17:39:24 +00:00
|
|
|
stream << 'U' << size_tag << '(';
|
2017-09-05 11:23:26 +00:00
|
|
|
stream << bytecode_iterator.GetIndexOperand(op_index);
|
2016-03-21 17:08:21 +00:00
|
|
|
break;
|
2016-12-21 16:41:11 +00:00
|
|
|
}
|
2016-09-13 14:48:39 +00:00
|
|
|
case OperandType::kUImm:
|
2017-01-25 17:39:24 +00:00
|
|
|
stream << 'U' << size_tag << '(';
|
2016-09-13 14:48:39 +00:00
|
|
|
stream << bytecode_iterator.GetUnsignedImmediateOperand(op_index);
|
|
|
|
|
break;
|
2016-03-21 17:08:21 +00:00
|
|
|
case OperandType::kImm:
|
2017-01-25 17:39:24 +00:00
|
|
|
stream << 'I' << size_tag << '(';
|
2016-05-11 12:21:56 +00:00
|
|
|
stream << bytecode_iterator.GetImmediateOperand(op_index);
|
2016-03-21 17:08:21 +00:00
|
|
|
break;
|
|
|
|
|
case OperandType::kRegCount:
|
2017-01-25 17:39:24 +00:00
|
|
|
stream << 'U' << size_tag << '(';
|
2016-05-11 12:21:56 +00:00
|
|
|
stream << bytecode_iterator.GetRegisterCountOperand(op_index);
|
2016-03-21 17:08:21 +00:00
|
|
|
break;
|
|
|
|
|
case OperandType::kRuntimeId: {
|
2017-01-25 17:39:24 +00:00
|
|
|
stream << 'U' << size_tag << '(';
|
2016-06-21 14:37:16 +00:00
|
|
|
Runtime::FunctionId id =
|
|
|
|
|
bytecode_iterator.GetRuntimeIdOperand(op_index);
|
|
|
|
|
stream << "Runtime::k" << i::Runtime::FunctionForId(id)->name;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
case OperandType::kIntrinsicId: {
|
2017-01-25 17:39:24 +00:00
|
|
|
stream << 'U' << size_tag << '(';
|
2016-06-21 14:37:16 +00:00
|
|
|
Runtime::FunctionId id =
|
|
|
|
|
bytecode_iterator.GetIntrinsicIdOperand(op_index);
|
|
|
|
|
stream << "Runtime::k" << i::Runtime::FunctionForId(id)->name;
|
2016-03-21 17:08:21 +00:00
|
|
|
break;
|
2017-09-05 11:23:26 +00:00
|
|
|
}
|
|
|
|
|
case OperandType::kNativeContextIndex: {
|
|
|
|
|
stream << 'U' << size_tag << '(';
|
|
|
|
|
uint32_t idx = bytecode_iterator.GetNativeContextIndexOperand(op_index);
|
|
|
|
|
stream << "%" << NameForNativeContextIntrinsicIndex(idx);
|
|
|
|
|
break;
|
2016-03-21 17:08:21 +00:00
|
|
|
}
|
|
|
|
|
default:
|
|
|
|
|
UNREACHABLE();
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
stream << ')';
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintBytecode(
|
2016-05-11 12:21:56 +00:00
|
|
|
std::ostream& stream, const BytecodeArrayIterator& bytecode_iterator,
|
2016-02-19 12:36:26 +00:00
|
|
|
int parameter_count) const {
|
2016-05-11 12:21:56 +00:00
|
|
|
Bytecode bytecode = bytecode_iterator.current_bytecode();
|
|
|
|
|
OperandScale operand_scale = bytecode_iterator.current_operand_scale();
|
2016-03-21 17:08:21 +00:00
|
|
|
if (Bytecodes::OperandScaleRequiresPrefixBytecode(operand_scale)) {
|
|
|
|
|
Bytecode prefix = Bytecodes::OperandScaleToPrefixBytecode(operand_scale);
|
|
|
|
|
stream << "B(" << Bytecodes::ToString(prefix) << "), ";
|
|
|
|
|
}
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
stream << "B(" << Bytecodes::ToString(bytecode) << ')';
|
|
|
|
|
int operands_count = Bytecodes::NumberOfOperands(bytecode);
|
|
|
|
|
for (int op_index = 0; op_index < operands_count; ++op_index) {
|
|
|
|
|
stream << ", ";
|
2016-05-11 12:21:56 +00:00
|
|
|
PrintBytecodeOperand(stream, bytecode_iterator, bytecode, op_index,
|
2016-02-19 12:36:26 +00:00
|
|
|
parameter_count);
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2016-05-11 12:21:56 +00:00
|
|
|
void BytecodeExpectationsPrinter::PrintSourcePosition(
|
|
|
|
|
std::ostream& stream, SourcePositionTableIterator& source_iterator,
|
|
|
|
|
int bytecode_offset) const {
|
|
|
|
|
static const size_t kPositionWidth = 4;
|
|
|
|
|
if (!source_iterator.done() &&
|
2016-06-28 05:51:09 +00:00
|
|
|
source_iterator.code_offset() == bytecode_offset) {
|
2016-05-11 12:21:56 +00:00
|
|
|
stream << "/* " << std::setw(kPositionWidth)
|
This CL enables precise source positions for all V8 compilers. It merges compiler::SourcePosition and internal::SourcePosition to a single class used throughout the codebase. The new internal::SourcePosition instances store an id identifying an inlined function in addition to a script offset.
SourcePosition::InliningId() refers to a the new table DeoptimizationInputData::InliningPositions(), which provides the following data for every inlining id:
- The inlined SharedFunctionInfo as an offset into DeoptimizationInfo::LiteralArray
- The SourcePosition of the inlining. Recursively, this yields the full inlining stack.
Before the Code object is created, the same information can be found in CompilationInfo::inlined_functions().
If SourcePosition::InliningId() is SourcePosition::kNotInlined, it refers to the outer (non-inlined) function.
So every SourcePosition has full information about its inlining stack, as long as the corresponding Code object is known. The internal represenation of a source position is a positive 64bit integer.
All compilers create now appropriate source positions for inlined functions. In the case of Turbofan, this required using AstGraphBuilderWithPositions for inlined functions too. So this class is now moved to a header file.
At the moment, the additional information in source positions is only used in --trace-deopt and --code-comments. The profiler needs to be updated, at the moment it gets the correct script offsets from the deopt info, but the wrong script id from the reconstructed deopt stack, which can lead to wrong outputs. This should be resolved by making the profiler use the new inlining information for deopts.
I activated the inlined deoptimization tests in test-cpu-profiler.cc for Turbofan, changing them to a case where the deopt stack and the inlining position agree. It is currently still broken for other cases.
The following additional changes were necessary:
- The source position table (internal::SourcePositionTableBuilder etc.) supports now 64bit source positions. Encoding source positions in a single 64bit int together with the difference encoding in the source position table results in very little overhead for the inlining id, since only 12% of the source positions in Octane have a changed inlining id.
- The class HPositionInfo was effectively dead code and is now removed.
- SourcePosition has new printing and information facilities, including computing a full inlining stack.
- I had to rename compiler/source-position.{h,cc} to compiler/compiler-source-position-table.{h,cc} to avoid clashes with the new src/source-position.cc file.
- I wrote the new wrapper PodArray for ByteArray. It is a template working with any POD-type. This is used in DeoptimizationInputData::InliningPositions().
- I removed HInlinedFunctionInfo and HGraph::inlined_function_infos, because they were only used for the now obsolete Crankshaft inlining ids.
- Crankshaft managed a list of inlined functions in Lithium: LChunk::inlined_functions. This is an analog structure to CompilationInfo::inlined_functions. So I removed LChunk::inlined_functions and made Crankshaft use CompilationInfo::inlined_functions instead, because this was necessary to register the offsets into the literal array in a uniform way. This is a safe change because LChunk::inlined_functions has no other uses and the functions in CompilationInfo::inlined_functions have a strictly longer lifespan, being created earlier (in Hydrogen already).
BUG=v8:5432
Review-Url: https://codereview.chromium.org/2451853002
Cr-Commit-Position: refs/heads/master@{#40975}
2016-11-14 17:21:37 +00:00
|
|
|
<< source_iterator.source_position().ScriptOffset();
|
2016-05-11 12:21:56 +00:00
|
|
|
if (source_iterator.is_statement()) {
|
|
|
|
|
stream << " S> */ ";
|
|
|
|
|
} else {
|
|
|
|
|
stream << " E> */ ";
|
|
|
|
|
}
|
|
|
|
|
source_iterator.Advance();
|
|
|
|
|
} else {
|
|
|
|
|
stream << " " << std::setw(kPositionWidth) << ' ' << " ";
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
void BytecodeExpectationsPrinter::PrintV8String(std::ostream& stream,
|
2018-11-27 00:48:42 +00:00
|
|
|
i::String string) const {
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
stream << '"';
|
|
|
|
|
for (int i = 0, length = string.length(); i < length; ++i) {
|
|
|
|
|
stream << i::AsEscapedUC16ForJSON(string.Get(i));
|
|
|
|
|
}
|
|
|
|
|
stream << '"';
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintConstant(
|
|
|
|
|
std::ostream& stream, i::Handle<i::Object> constant) const {
|
2016-09-06 16:10:19 +00:00
|
|
|
if (constant->IsSmi()) {
|
|
|
|
|
stream << "Smi [";
|
|
|
|
|
i::Smi::cast(*constant).SmiPrint(stream);
|
|
|
|
|
stream << "]";
|
|
|
|
|
} else {
|
|
|
|
|
stream << i::HeapObject::cast(*constant).map().instance_type();
|
|
|
|
|
if (constant->IsHeapNumber()) {
|
|
|
|
|
stream << " [";
|
|
|
|
|
i::HeapNumber::cast(*constant).HeapNumberPrint(stream);
|
|
|
|
|
stream << "]";
|
|
|
|
|
} else if (constant->IsString()) {
|
|
|
|
|
stream << " [";
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
PrintV8String(stream, i::String::cast(*constant));
|
2016-09-06 16:10:19 +00:00
|
|
|
stream << "]";
|
|
|
|
|
}
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintFrameSize(
|
|
|
|
|
std::ostream& stream, i::Handle<i::BytecodeArray> bytecode_array) const {
|
2019-05-30 19:55:07 +00:00
|
|
|
int32_t frame_size = bytecode_array->frame_size();
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
2019-02-14 13:01:52 +00:00
|
|
|
DCHECK(IsAligned(frame_size, kSystemPointerSize));
|
|
|
|
|
stream << "frame size: " << frame_size / kSystemPointerSize
|
2016-02-25 12:07:07 +00:00
|
|
|
<< "\nparameter count: " << bytecode_array->parameter_count() << '\n';
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintBytecodeSequence(
|
|
|
|
|
std::ostream& stream, i::Handle<i::BytecodeArray> bytecode_array) const {
|
2016-02-25 12:07:07 +00:00
|
|
|
stream << "bytecode array length: " << bytecode_array->length()
|
|
|
|
|
<< "\nbytecodes: [\n";
|
2016-05-11 12:21:56 +00:00
|
|
|
|
|
|
|
|
SourcePositionTableIterator source_iterator(
|
2017-04-12 18:33:20 +00:00
|
|
|
bytecode_array->SourcePositionTable());
|
2016-05-11 12:21:56 +00:00
|
|
|
BytecodeArrayIterator bytecode_iterator(bytecode_array);
|
|
|
|
|
for (; !bytecode_iterator.done(); bytecode_iterator.Advance()) {
|
|
|
|
|
stream << kIndent;
|
|
|
|
|
PrintSourcePosition(stream, source_iterator,
|
|
|
|
|
bytecode_iterator.current_offset());
|
|
|
|
|
PrintBytecode(stream, bytecode_iterator, bytecode_array->parameter_count());
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
stream << ",\n";
|
|
|
|
|
}
|
|
|
|
|
stream << "]\n";
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintConstantPool(
|
2018-11-25 02:24:43 +00:00
|
|
|
std::ostream& stream, i::FixedArray constant_pool) const {
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
stream << "constant pool: [\n";
|
|
|
|
|
int num_constants = constant_pool.length();
|
|
|
|
|
if (num_constants > 0) {
|
|
|
|
|
for (int i = 0; i < num_constants; ++i) {
|
2016-05-11 12:21:56 +00:00
|
|
|
stream << kIndent;
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
PrintConstant(stream, i::FixedArray::get(constant_pool, i, i_isolate()));
|
|
|
|
|
stream << ",\n";
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
stream << "]\n";
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintCodeSnippet(
|
|
|
|
|
std::ostream& stream, const std::string& body) const {
|
|
|
|
|
stream << "snippet: \"\n";
|
|
|
|
|
std::stringstream body_stream(body);
|
|
|
|
|
std::string body_line;
|
|
|
|
|
while (std::getline(body_stream, body_line)) {
|
2016-05-11 12:21:56 +00:00
|
|
|
stream << kIndent;
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
PrintEscapedString(stream, body_line);
|
|
|
|
|
stream << '\n';
|
|
|
|
|
}
|
|
|
|
|
stream << "\"\n";
|
|
|
|
|
}
|
|
|
|
|
|
2016-02-19 12:36:26 +00:00
|
|
|
void BytecodeExpectationsPrinter::PrintHandlers(
|
|
|
|
|
std::ostream& stream, i::Handle<i::BytecodeArray> bytecode_array) const {
|
|
|
|
|
stream << "handlers: [\n";
|
2018-02-27 09:23:48 +00:00
|
|
|
HandlerTable table(*bytecode_array);
|
|
|
|
|
for (int i = 0, num_entries = table.NumberOfRangeEntries(); i < num_entries;
|
2016-02-19 12:36:26 +00:00
|
|
|
++i) {
|
2018-02-27 09:23:48 +00:00
|
|
|
stream << " [" << table.GetRangeStart(i) << ", " << table.GetRangeEnd(i)
|
|
|
|
|
<< ", " << table.GetRangeHandler(i) << "],\n";
|
2016-02-19 12:36:26 +00:00
|
|
|
}
|
|
|
|
|
stream << "]\n";
|
|
|
|
|
}
|
|
|
|
|
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
void BytecodeExpectationsPrinter::PrintBytecodeArray(
|
2016-02-19 12:36:26 +00:00
|
|
|
std::ostream& stream, i::Handle<i::BytecodeArray> bytecode_array) const {
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
PrintFrameSize(stream, bytecode_array);
|
|
|
|
|
PrintBytecodeSequence(stream, bytecode_array);
|
|
|
|
|
PrintConstantPool(stream, bytecode_array->constant_pool());
|
2016-02-19 12:36:26 +00:00
|
|
|
PrintHandlers(stream, bytecode_array);
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void BytecodeExpectationsPrinter::PrintExpectation(
|
|
|
|
|
std::ostream& stream, const std::string& snippet) const {
|
2016-02-19 12:36:26 +00:00
|
|
|
std::string source_code =
|
|
|
|
|
wrap_ ? WrapCodeInFunction(test_function_name_.c_str(), snippet)
|
|
|
|
|
: snippet;
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
2018-08-07 13:14:23 +00:00
|
|
|
i::FLAG_enable_one_shot_optimization = oneshot_opt_;
|
2018-08-10 12:41:06 +00:00
|
|
|
i::FLAG_compilation_cache = false;
|
2016-10-04 16:29:59 +00:00
|
|
|
i::Handle<i::BytecodeArray> bytecode_array;
|
2016-10-04 18:42:13 +00:00
|
|
|
if (module_) {
|
|
|
|
|
CHECK(top_level_ && !wrap_);
|
|
|
|
|
v8::Local<v8::Module> module = CompileModule(source_code.c_str());
|
|
|
|
|
bytecode_array = GetBytecodeArrayForModule(module);
|
2018-08-07 13:14:23 +00:00
|
|
|
} else if (print_callee_) {
|
|
|
|
|
bytecode_array = GetBytecodeArrayOfCallee(source_code.c_str());
|
2016-10-04 16:29:59 +00:00
|
|
|
} else {
|
2016-10-04 18:42:13 +00:00
|
|
|
v8::Local<v8::Script> script = CompileScript(source_code.c_str());
|
|
|
|
|
if (top_level_) {
|
|
|
|
|
bytecode_array = GetBytecodeArrayForScript(script);
|
|
|
|
|
} else {
|
|
|
|
|
Run(script);
|
|
|
|
|
bytecode_array = GetBytecodeArrayForGlobal(test_function_name_.c_str());
|
|
|
|
|
}
|
2016-10-04 16:29:59 +00:00
|
|
|
}
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
|
2016-02-19 12:36:26 +00:00
|
|
|
stream << "---\n";
|
|
|
|
|
PrintCodeSnippet(stream, snippet);
|
|
|
|
|
PrintBytecodeArray(stream, bytecode_array);
|
[Interpreter] Change the output format of generate-bytecode-expectations.
Now the tool produces a far more readable output format, which bears a
lot of resemblance to YAML. In fact, the output should be machine
parseable as such, one document per testcase. However, the output format
may be subject to changes in future, so don't rely on this property.
In general, the output format has been optimized for producing a meaningful
textual diff, while keeping a decent readability as well. Therefore, not
everything is as compact as it could be, e.g. for an empty const pool we get:
constant pool: [
]
instead of:
constant pool: []
Also, trailing commas are always inserted in lists.
Additionally, now the tool accepts its output format as input. When
operating in this mode, all the snippets are extracted, processed and
the output is then emitted as usual. If nothing has changed, the output
should match the input. This is very useful for catching bugs in the
bytecode generation by running a textual diff against a known-good file.
The core (namely bytecode-expectations.cc) has been extracted from the
original cc file, which provides the utility as usual. The definitions
in the matching header of the library have been moved into the
v8::internal::interpreter namespace.
The library exposes a class ExpectationPrinter, with a method
PrintExpectation, which takes a test snippet as input, and writes the
formatted expectation to the supplied stream. One might then use a
std::stringstream to retrieve the results as a string and run it through
a diff utility.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1688383003
Cr-Commit-Position: refs/heads/master@{#33997}
2016-02-15 15:20:19 +00:00
|
|
|
stream << '\n';
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
} // namespace interpreter
|
|
|
|
|
} // namespace internal
|
|
|
|
|
} // namespace v8
|
