Using a C++-style method PrintName (a.k.a. << ;-), things get a lot easier when
two unrelated concerns are separated. Stubs don't need a name cache anymore,
simpler code while generating the stub name, memory allocation is centralized,
etc.
Review URL: http://codereview.chromium.org/7342042
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8627 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This patch just adds a nop after the call to the binary operation stub in optimized code to avoid the patching for the inlined smi case used in the full code generator to kick in if the next instruction generated by the lithium code generator should accidentially enable that. For calls generated by CallCodeGeneric this was already handled on Intel platforms, but missing on ARM.
On IA-32 I did also try to check for whether the code containing the call was optimized (patch below), but that caused regressions on some benchmarks.
diff --git src/ia32/ic-ia32.cc src/ia32/ic-ia32.cc
index 5f143b1..f70e208 100644
--- src/ia32/ic-ia32.cc
+++ src/ia32/ic-ia32.cc
@@ -1603,12 +1603,18 @@ void CompareIC::UpdateCaches(Handle<Object> x, Handle<Object> y) {
// Activate inlined smi code.
if (previous_state == UNINITIALIZED) {
- PatchInlinedSmiCode(address());
+ PatchInlinedSmiCode(address(), isolate());
}
}
-void PatchInlinedSmiCode(Address address) {
+void PatchInlinedSmiCode(Address address, Isolate* isolate) {
+ // Never patch in optimized code.
+ Code* code = isolate->pc_to_code_cache()->GetCacheEntry(address)->code;
+ if (code->kind() == Code::OPTIMIZED_FUNCTION) {
+ return;
+ }
+
// The address of the instruction following the call.
Address test_instruction_address =
address + Assembler::kCallTargetAddressOffset;
diff --git src/ic.cc src/ic.cc
index f70f75a..62e79da 100644
--- src/ic.cc
+++ src/ic.cc
@@ -2384,7 +2384,7 @@ RUNTIME_FUNCTION(MaybeObject*, BinaryOp_Patch) {
// Activate inlined smi code.
if (previous_type == BinaryOpIC::UNINITIALIZED) {
- PatchInlinedSmiCode(ic.address());
+ PatchInlinedSmiCode(ic.address(), isolate);
}
}
diff --git src/ic.h src/ic.h
index 11c2e3a..9ef4b20 100644
--- src/ic.h
+++ src/ic.h
@@ -721,7 +721,7 @@ class CompareIC: public IC {
};
// Helper for BinaryOpIC and CompareIC.
-void PatchInlinedSmiCode(Address address);
+void PatchInlinedSmiCode(Address address, Isolate* isolate);
} } // namespace v8::internal
R=danno@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org//7350015
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8623 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The preprocessor defines ENABLE_LOGGING_AND_PROFILING and ENABLE_VMSTATE_TRACKING has been removed as these where required to be turned on for Crankshaft to work. To re-enable reducing the binary size by leaving out heap and CPU profiler a new set of defines needs to be created.
R=ager@chromium.org
BUG=v8:1271
TEST=all
Review URL: http://codereview.chromium.org//7350014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8622 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
architecture-independent.
jsregexp.h is itself included transitively quite a lot, and by getting rid of 19
of its dependencies (which even included things like src/cpu.h, the various
assemblers, etc.), the recompilation behaviour is a bit less funny than it was.
Review URL: http://codereview.chromium.org/7331014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8589 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The debugger can be entered from the deferred stack check in optimized code. This can cause both lazy deoptimization and debugger deoptimization (setting the first break point and inspecting the stack for optimized code respectively). This required deoptimization support from the deferred stack check.
The lazy deoptimiztion call is inserted when the deferred code is done including restoring the registers. The bailout to the full code is the begining of the loop body as that is where the stack check is sitting in the optimized code. The bailout is not to the stack check in the full code as that is sitting at the end of the loop.
R=kmillikin@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org//7212025
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8535 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Due to issues relating mostly to chrome extensions we have lately been
running into OOMs that are caused by our executable space running
out. This change introduces flushing of code from regexps if we have
not used the code for 5 mark sweeps.
The approach is different from the normal function code flusing. Here
we make a copy of the code inside the data array, and exchange the
original code with a smi determined by the sweep_generation (a new
heap variable increased everytime we do mark sweep/compact). If we
encounter a smi in EnsureCompiled we simply reinstate the code
object. If, in the marking phase of mark sweep, we find a regexp that
already have a smi in the code field, and this is more than 5
generations old we flush the code from the saved index.
Review URL: http://codereview.chromium.org/7282026
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8532 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Eliminates the enum flag RESTORE_CONTEXT and CONTEXT_ADJUSTED, and adds a context HValue and LOperand to many hydrogen and lithium instructions.
Context is still used from the stack from in CallKnownFunction (this seems safe), and in CallRuntimeFromDeferred in lithium-codegen-ia32.cc, which needs to be fixed.
BUG=
TEST=
Review URL: http://codereview.chromium.org/7132002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8529 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The catch variable is bound in the catch scope. For simplicity in this
initial implementation, it is always allocated even if unused and always
allocated to a catch context even if it doesn't escape. The presence of
catch is no longer treated as a with.
In this change, care must be taken to distinguish between the scope where a
var declaration is hoisted to and the scope where the initialization occurs.
R=ager@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/7280012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8496 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Optimized frames are now handled by the debugger. When discovering optimized frames during stack inspection in the debugger they are "deoptimized" using the normal deoptimization code and the deoptimizer output information is used to provide frame information to the debugger.
Before this change the debugger reported each optimized frame as one frame no matter the number of inlined functuions that might have been called inside of it. Also all locals where reported as undefined. Locals can still be reposted as undefined when their value is not "known" by the optimized frame.
As the structures used to calculate the output frames when deoptimizing are not GC safe the information for the debugger is copied to another structure (DeoptimizedFrameInfo) which is registered with the global deoptimizer data and processed during GC.
R=fschneider@chromium.org
BUG=v8:1140
TEST=test/mjsunit/debug-evaluate-locals-optimized*
Review URL: http://codereview.chromium.org//7230045
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8464 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Before: every context cached the nearest enclosing function context. This
assumed that for nested contexts (i.e., with and catch contexts) the
enclosing function had a materialized link in the context chain.
Now: when necessary, we loop up the context chain to find such a context.
This enables catch contexts without forcing the enclosing function to
allocate its own context.
R=ager@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/7230047
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8452 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
* src/hydrogen.cc (HEnvironment::CopyForInlining): As the code for both
the ::HYDROGEN and ::LITHIUM compilation phases is the same, just use
one code path and remove the arg.
* src/hydrogen.h (HEnvironment): Remove now-unused CompilationPhase
enum type and arg to CopyForInlining.
* src/arm/lithium-arm.cc (LChunkBuilder::DoEnterInlined):
* src/ia32/lithium-ia32.cc (LChunkBuilder::DoEnterInlined):
* src/x64/lithium-x64.cc (LChunkBuilder::DoEnterInlined): Adapt
callers.
* AUTHORS: Add Igalia.
BUG=
TEST=I ran tools/test.py.
Review URL: http://codereview.chromium.org/7272002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8442 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The hydrogen stack check instruction is now added to each loop and the stack check handling on the back edge has been removed.
This change causes regression on small tight loops as the stack check is now at the top of the loop instead of at the bottom, and that requires one additional unconditional jump per loop iteration. However the reason for this change is to avoid worse regressions for upcoming changes to correctly support debugger break in optimized code.
R=fschneider@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org//7216009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8428 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Detect the pattern in both, the full compiler and crankshaft and generate direct pointer
comparisons. Along the way I cleaned up 'typeof <expression> == <string literal>' comparisons
as well by lifting platform independent code and checking the symmetric case.
BUG=v8:1440
TEST=cctest/test-api.cc
Review URL: http://codereview.chromium.org/7216008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8420 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This reverts r8381.
It was causing Mozilla test mozilla/ecma/Date/15.9.3.8-5 to fail and Sputnik tests S15.9.3.1_A5_T5, S15.9.3.1_A5_T1, S15.9.3.1_A5_T2, S15.9.3.1_A5_T4, S15.9.3.1_A5_T3 and S15.9.3.1_A5_T6 to timeout.
R=ager@chromium.org
BUG=none
TEST=mozilla/ecma/Date/15.9.3.8-5, S15.9.3.1_A5_T5, S15.9.3.1_A5_T1, S15.9.3.1_A5_T2, S15.9.3.1_A5_T4, S15.9.3.1_A5_T3 and S15.9.3.1_A5_T6
Review URL: http://codereview.chromium.org//7246004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8397 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The declaration of the ToBoolean class moved to the platform-independent part
and its implementations are now structurally very similar. This is just an
intermediate cleanup step to add type recording at the call site.
Note that the MIPS implementation has not really been touched, so it should
continue to work, too.
Review URL: http://codereview.chromium.org/7218012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8359 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
* AST Expression nodes get a separate testing ID to record type info in
ToBooleanStub later. This is necessary to avoid clashes with other uses of
already existing IDs.
* In order to avoid threading the condition expression through tons of places,
TestContexts carry it now with them. Note that we will probably only need the
testing ID of the expression, but having the whole thing at hand makes
debugging easier. Probably we will change this later...
Review URL: http://codereview.chromium.org/7046073
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8274 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This change makes the number of successors of a control instruction
configurable with a template parameter and changes the existing instructions
to use it.
To iterate over all successors I added an iterator instead of always calling
First- and SecondSuccessor.
Review URL: http://codereview.chromium.org/7114004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8262 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Introduce separate maps for function and with contexts. Use the function
context map for testing whether a context is a function context (global
contexts are no longer function contexts).
Split the paths for allocating with and catch contexts.
Rename some functions. Generally refactor code to make it simpler.
R=ager@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/7003058
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8231 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Mainly, there were errors concerning blank lines before and after class access
control sections [whitespace/blank_line].
BEFORE an access control section (e.g. public:, private:) there should be a
blank line (except for the section right after the class declaration).
AFTER an access control section there should be no blank line.
TBR=ager@chromium.org
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8193 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The Advance() function of the class responsible for iterating
environment uses didn't always advance as far as it could (relying on
the HasNext predicate to finish advancing). This is brittle.
The HasNext predicate also didn't advance as far as it could when it
was at the end of an environment level. This is a bug.
R=jkummerow@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/6993023
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8181 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
If type-feedback indicates that an expression was never executed in
the non-optimized code, we insert a forced deoptimization right away
to enable re-optimization if we ever hit this path.
With this change we still continue to build the graph. As a next step, we
should remove the dead code after the deoptimize.
I had to remove one assert about the optimization status in a test since
we now immediately deoptimize after exiting the loop that triggers OSR.
Also remove a restriction that control-flow from an inlined function in a
test context always reaches both true- and false-target.
Review URL: http://codereview.chromium.org/7105015
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8140 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
- Introduce a class JSReceiver, that is a common superclass of JSObject and
JSProxy. Use JSReceiver where appropriate (probably lots of places that we
still have to migrate, but we will find those later with proxy test suite).
- Move appropriate methods to JSReceiver class (SetProperty,
GetPropertyAttribute, Get/SetPrototype, Lookup, and so on).
- Introduce new JSFunctionProxy subclass of JSProxy. Currently only a stub.
- Overhaul enum InstanceType:
* Introduce FIRST/LAST_SPEC_OBJECT_TYPE that ranges over all types that
represent JS objects, and use that consistently to check language types.
* Rename FIRST/LAST_JS_OBJECT_TYPE and FIRST/LAST_FUNCTION_CLASS_TYPE
to FIRST/LAST_[NON]CALLABLE_SPEC_OBJECT_TYPE for clarity.
* Eliminate the overlap over JS_REGEXP_TYPE.
* Also replace FIRST_JS_OBJECT with FIRST_JS_RECEIVER, but only use it where
we exclusively talk about the internal representation type.
* Insert JS_PROXY and JS_FUNCTION_PROXY in the appropriate places.
- Fix all checks concerning classification, especially for functions, to
use the CALLABLE_SPEC_OBJECT range (that includes funciton proxies).
- Handle proxies in SetProperty (that was the easiest part :) ).
- A few simple test cases.
R=kmillikin@chromium.org
Review URL: http://codereview.chromium.org/6992072
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8126 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
* Centralized AND/OR handling, keeping related code together.
* Removed HandleExpression/HandleInNonTestContext and introduced VisitInSameContext instead, making it more obvious what's actually going on.
* Consistently use a new context when visiting the left sub-expression of an AND/OR. Note that the context stacks in the full code generator and crankshaft are still a bit out of sync for the right sub-expression.
Review URL: http://codereview.chromium.org/6976028
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8124 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The approach is to handle the common case in the optimizing
compiler and to bailout for the rare corner cases.
This is done by initializing all local const-variables with
the hole value and disallowing any use of the hole value statically.
Review URL: http://codereview.chromium.org/6026006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8104 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Only IA32 version for now. I'll start porting.
Strict mode functions are to get 'undefined' as the receiver when
called with an implicit receiver. Modes are bad! It forces us to have
checks on all function calls.
This change attempts to limit the cost by passing information about
whether or not a call is with an implicit or explicit receiver in ecx
as part of the calling convention. The cost is setting ecx on all
calls and checking ecx on entry to strict mode functions.
Implicit/explicit receiver state has to be maintained by ICs. Various
stubs have to not clobber ecx or save and restore it.
CallFunction stub needs to check if the receiver is implicit when it
doesn't know from the context.
Review URL: http://codereview.chromium.org/7039036
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8040 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
KeyedLoadStubCompiler::GenerateLoadExternalArray is tail-calling a stub, which it is not allowed to, because its major key is higher than InstanceOf. Instead of changing the major key I decided to provide an inline implementation.
TEST=test-api/ExternalArrays with --noenable-vfp3
R=danno@chromium.org
Review URL: http://codereview.chromium.org/7060007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@8002 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
When the stub return a heap number we do a state transition to
a version HEAP_NUMBER that can handle -0.
There is room for further improvement in the typefeedback for the
case of -0. This change however does not address this and only fixes
the acute issue.
Review URL: http://codereview.chromium.org/7037025
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7944 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Better support for 'polymorphic' JS and external arrays
Allow keyed store/load stubs to switch between external array and fast JS arrays without forcing a state transition to the generic stub.
There CL consists of two pieces of functionality. First, code stubs for fast element arrays don't immediately transition to the MEGAMORPHIC state when there's a map mismatch. Second, two ICs are cached per map for fast elements, the MONOMORPHIC version, and a new MEGAMORPHIC version that handles two or more different maps and dispatches to shared stubs to perform the array operation.
Review URL: http://codereview.chromium.org/7036016
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7935 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
As a side note, this change has some positive impact on the performace, e.g. imaging-darkroom is 8.3% faster and ws-ieee754conv is even 19.9% faster. No idea why the speedup is so big in some cases, but never look a gift horse in the mouth... ;-)
Review URL: http://codereview.chromium.org/7042004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7926 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Allow keyed store/load stubs to switch between external array and fast JS arrays without forcing a state transition to the generic stub.
There CL consists of two pieces of functionality. First, code stubs for fast element arrays don't immediately transition to the MEGAMORPHIC state when there's a map mismatch. Second, two ICs are cached per map for fast elements, the MONOMORPHIC version, and a new MEGAMORPHIC version that handles two or more different maps. Currently, the only array types supported by the MEGAMORPHIC stub are fast elements for objects and JSArrays.
BUG=
TEST=
Review URL: http://codereview.chromium.org/6894003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7917 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The profiler did not correctly identify when there was a JS frame
on the stack after an exception was thrown.
I also refactored the code to use PopTryHandler macro-instructions on all platforms
BUG=73722
TEST=No more valgrind errors related to v8::internal::StackFrameIterator.
Review URL: http://codereview.chromium.org/7019010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7910 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This allows us to get rid of totally fake LAST_STRING_TYPE and makes
it possible to test for symbols.
I considered splitting HCheckInstanceType into two instructions, but
it seems nice to be able to hide the instance type implementation
details from the hydrogen level.
Review URL: http://codereview.chromium.org/6964011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7840 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
If the result of an fast elements load is converted to an untagged
representation we can omit the hole check if the value is not used
anywhere else except for HChange instructions converting it to
an untagged representation since those will deoptimize for the hole
value anyway.
Review URL: http://codereview.chromium.org/6964012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7827 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Use a special slot for HContext, and fetch the value from there each time it is used. Allocate space for special slots in every HEnvironment. Fill them with constant undefined. Do not copy them to LEnvironment.
BUG=
TEST=
Review URL: http://codereview.chromium.org/6929066
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7807 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
In my previous change I mixed up "compare" and "equals". This made us
miss the fast length check before comparing strings for equality.
Now we have a separate helper for "equals". It shares some of the
inner loop details with "compare". I'll see if this can be cleaned up
without making it unreadable.
Review URL: http://codereview.chromium.org/6928020
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7794 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The only difference to revision 7763 is the implementation in the
builtins file for arm and x64, plus a move of Array.prototype.toString
and Array.prototype.toLocaleString from should throw on null or
undefined to the non generic test cases in the function-call test (due
to us not currently supporting generic cases with these to functions)
Review URL: http://codereview.chromium.org/6928007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7786 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Always use CpuFeaturesImpliedByCompiler() when selecting CPU features. This checks both for CAN_USE_ARMV7_INSTRUCTIONS and CAN_USE_VFP_INSTRUCTIONS and for GCC preprocessor symbols. This will support using the CAN_USE_XXX for a simulator build used for generating a snapshot followed by a crosscompile using -march= and -mfpu= for selecting the (minimal) target device CPU features. The snapshot will use instructions based on the CAN_USE_XXX whereas the target will at least use features based on both CAN_USE_XXX and -march= and -mfpu=, but will try runtime CPU feature detection a well looking for somethis better.
Remove the compiler based CPU feature detection from the OS::CpuFeaturesImpliedByPlatform() as it did not belong there. Also was already in the CpuFeaturesImpliedByCompiler().
Add the variable 'v8_can_use_vfp_instructions' to the GYP file which can be used to turn on CAN_USE_VFP_INSTRUCTIONS when building V8. I did not add any -mfpu= cflags for this, as there are several options here (e.g. vfp and neon).
R=erik.corry@gmail.com, karlklose@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org//6904164
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7754 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Add option armeabi to the SCons build for selecting the floating point variant to use. Also add externally defined CCFLAGS environment for all targets. Run test.py with option -S armeabi=hardfloat to test with hardfloat enabled.
Make selecting hardfloat EABI variant a build-time option instead of a runtime option.
Add a simple check of the EABI variant during V8 initialization to exit if the compilation was not configured correctly. The reason for this is that GCC does not provide a compile time symbol defining the EABI variant. This check is not fool-proof as it cannot check the compilation configuration used for the snapshot if any.
R=karlklose@chromium.org, erik.corry@gmail.com
BUG=none
TEST=none
Review URL: http://codereview.chromium.org//6905098
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7715 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This reduces the binary size by making the Is* type-test functions non-virtual.
I had to change Gap and Label instructions to have a common abstract superclass because both act as gap-instructions for the register allocator.
Review URL: http://codereview.chromium.org/6880204
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7691 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Rather than representing a use as a pointer to an HValue and then searching
for the specific (ambiguous) operand, we now represent a use as a pair of an
HValue and the input operand index. Additionally, use a linked list instead
of a growable array list since we never use random access.
This allows us to remove a bunch of similarly named and subtly different
functions from the HValue API. The cost in extra zone allocation per use is
partially offset by reusing use list nodes when replacing a use of one value
with another.
R=danno@chromium.org,fschneider@chromium.org
Review URL: http://codereview.chromium.org/6881044
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7674 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
For instructions that are marked as calls we can use:
1. Fixed input registers
2. Use-at-start policy (register, memory or constant).
(Memory or constant would not need the use-at-start policy, but
not specifying use-at-start prevents the allocator from using
a register even if there is one available. That's why use-at-start
is required and guarded by assertion)
Review URL: http://codereview.chromium.org/6853010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7616 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The type recording binary operations stub contained a number of issues which broke V8 when running without VFPv3 supoort.
1. The order of the high/low words of doubles in core register pairs where wrong
2. Missing allocation of a heap number for the result
3. Missing call to runtime
Added the option for printing the double values in register pairs to the simulator.
R=karlklose@chromium.org
BUG=v8:1315
TEST=
Review URL: http://codereview.chromium.org//6834020
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7603 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
- mutual inlining strict and non-strict functions in crankshaft.
- assignment to undefined variable with eval in scope.
- propagation of strict mode through lazy compilation.
BUG=
TEST=test/mjsunit/strict-mode.js test/mjsunit/strict-mode-opt.js
Review URL: http://codereview.chromium.org/6814012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7561 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Crankshaft is now the default on all platforms. This is the first
patch on the way to removing the classic code generator from the
system.
This time with no removal of the crankshaft flag. --nocrankshaft is
not at all the same as --always-full-compiler which I had used instead
for testing. That was what caused timeouts on the buildbots because of
repeated attempts to optimize hot functions. It makes sense to keep
the crankshaft flag in case you want to run only with the full
compiler and with no adaptive compilation.
R=vitalyr@chromium.org
Review URL: http://codereview.chromium.org/6759070
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7486 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
In the cases where a global property cell cannot be used in the optimized code
use standard load ic to get the property instead of bailing out.
This is re-committing r7212 and r7215 which where reverted in r7239 with the addition of recoring the source position in the hydrogen code for the LoadGlobalCell instruction. To record that position an optional position field has been added to the variable proxy AST node.
Review URL: http://codereview.chromium.org/6758007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7474 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Fix %NewObjectFromBound to correctly handle optimized frames (including those with inlined functions).
Fix %_IsConstructCall handling in hydrogen: when called from inlined function return false constant directly instead of emiting HIsConstructCall.
Fix success case in TraceInline.
BUG=v8:1229
TEST=test/mjsunit/regress/regress-1229.js
Review URL: http://codereview.chromium.org/6740023
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7472 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The previous implementation attempted to keep track of the needed
relocation size for deoptimization while generating the optimized
code. That was error prone. This patch moves the relocation resizing
to the deoptimizer as the last step of creating an optimized code
object.
The down side to this approach is that two relocation information byte
arrays are created for all optimized functions that do not have enough
relocation space for lazy deoptimization.
R=sgjesse@chromium.org
Review URL: http://codereview.chromium.org/6730050
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7360 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This change adds a new IR instruction for polymorphic loads. It performs
map compares and loads in one IR instruction instead of splitting each
load into a graph of map-compares and field loads.
The advantage is a smaller IR and less basic blocks, plus it allows to
do GVN on polymorphic loads.
Review URL: http://codereview.chromium.org/6708085
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7336 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This allows fast calls and inlining of functions like:
var o = {f: function() { return "foo"; }}
o.f();
Object literals that contain function literals are initially created a dictionary mode
object and only transformed to fast properties once all properties are computed and
added. This allows us to create constant function properties for functions declared
inside the object literal. Function literals inside object literals are marked for
pretenuring so that they work as contant function properties.
Object literals without functions should just function as before.
Review URL: http://codereview.chromium.org/6240012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7283 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This patch
- removes the unimplemented code crash when rendering invalid/unknown instructions and prints "unknown" instead.
- prints the beginning of the constant pool marker.
- adds "da" as a shortcut for "disasm".
- print hexadecimal representation of double and single registers. This makes it easier to debug move/conversion code that uses temporary int32 values in floating point registers.
- annotates the stack with short prints of the values (HeapObjects and smis),
- makes disasm take an address or a register as second argument without a third argument, which defaults to printing ten instructions.
Review URL: http://codereview.chromium.org/6676042
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7279 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
On ARM the break at break slots wrongly used the break at return stub which assumes an object opinter in r0. At break slots there are no objects live in registers so a GC while at a break on a break stub could cause a crash as r0 was wrongly pushed on a internal frame to be handled by GC.
Review URL: http://codereview.chromium.org/6676037
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7203 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
we know that both sides are Smi and those where we don't. Fix inlined
symbol table probes to cope with strings, undefined and null (indicating
a deleted entry). Some changes to other architectures that were found
with the new asserts.
Review URL: http://codereview.chromium.org/6682026
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7172 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
For bit operations, values are truncated and converted to the range -2^31 through 2^31 - 1. This patch provides inlined handling of this conversion for all heap numbers.
Before, we were calling runtime for every heap number that did not fit into a signed int32.
Review URL: http://codereview.chromium.org/6658034
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7148 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Ensure that there is always enough bytes between consequtive calls in optimized code to write a call instruction at the return points without overlapping.
Add a call to deoptimize all functions after running tests with --stress-opt. This will catch some issues with functions which cannot be forcefully deoptimized. Some of the tests failed on ARM with that change without the rest of the changes in this change.
Review URL: http://codereview.chromium.org/6661022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7132 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Worth mentioning:
- Specialized versions of pixel array and store/loads inside the generic stubs have been removed, since to have parity for all external arrays, 8 different versions would have to be inlined/checked.
- There's a new constant in v8.h for external arrays with pixel array elements.
Review URL: http://codereview.chromium.org/6546036
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7106 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Refactor construction of switch statements so it doesn't use class
HSubgraph.
There are also a few improvements. We do not use an auxiliary list of
comparisons because they're embedded as a linked list in the graph
under construction. We share a common break block for all breaks from
the same switch. We do not insert empty blocks unless necessary to
maintain edge-split form.
There is also a bug fix. The entry to a clause body is a potential
join and must have a join ID set, otherwise deoptimization within the
body can go to an unpredictable place in the unoptimized code.
Review URL: http://codereview.chromium.org/6650021
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7099 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Non-moveable code-stube are allocated in large object space. They are only required on ARM where the different C-entry stubs are required to never move.
This gets rid of pre-computing these stubs and hope that they never move. Also for crankshaft the C-entry stub which saved doubles is not generated in the snapshot so it ends up being generated at runtime and potentially move.
Review URL: http://codereview.chromium.org/6626072
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7097 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Move strict mode flag from TemporaryScope to Scope so that it can be accessed from variable binding code.
Arguments do not alias in strict mode (ia32, x64 and arm, codegen and full codegen).
Hydrogen tolerates null arguments_shadow().
In codegen-<arch> arguments object is allocated eagerly to capture values before they get modified.
Review URL: http://codereview.chromium.org/6625048/
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7083 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Change the way we construct the graph for polymorphic loads to match that of
polymorphic stores.
Introduce a stack-allocated helper for saving and restoring all the
function-specific graph builder state that needs to change when we begin
translating an inlined function. Make this class authoritative by moving
redundant state out of the builder and deferring to the current function's
state.
Ensure that we always print a tracing message when abandoning an inlining
attempt.
Review URL: http://codereview.chromium.org/6628012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7074 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
In built-in code we use arrays for internal computations.
This makes it possible to affect the built-in code by putting getters
or setters on the Array prototype chain.
This adds a new internal Array constructor that creates Arrays with
a very simplistic prototype chain that doesn't include any publicly
visible objects. These Arrays shoudl ofcourse never leak outside the
builtins, since that would expose the prototype object.
The prototype object contains only the array functions that we use:
push, pop and join (and not even a toString, so it doesn't stringify
well).
Also change uses of .call to %_CallFunction.
BUG=1206
Review URL: http://codereview.chromium.org/6602081
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7040 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This change makes the full code generator always materialize smi constants to match the behavior
of the environment tracked in the optimizing compiler. The mismatch between environment
and what the non-optimized code causes problems and potentially crashes when doing deoptimization
and OSR.
BUG=67732
Review URL: http://codereview.chromium.org/6592064
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6982 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Use ToBooleanStub in DoTest instead of runtime call.
Try allocation in generated code in VisitRegExpLiteral.
Use TranscendentalCacheStub for sin, cos and log instead of calling runtime.
Minor "alignment" of the full codegen code across platforms.
Review URL: http://codereview.chromium.org/6597003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6953 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Having a string add stub which can handle a string in just one of the arguments is essencial for the type recording binary operation stub when expecting strings. Otherwise string added with e.g. smi will always call the runtime for a type transition which will be back to the same types as the transition code keeps it in string arguments when one argument is a string.
This fixes the regression (especially Delta Blue) caused by replacing the generic binary operation stub with the type recording binary operation stub in the ARM lithium code generator.
Review URL: http://codereview.chromium.org/6551008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6882 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Strict mode flag is passed to runtime DELETE function
and then to JSObject::Delete(Property/Element) as STRICT_DELETION enum.
When deleting non-configurable property/eleemnt, TypeError is thrown.
Adding mozilla test to .gitignore.
Incorporate CR feedback.
Review URL: http://codereview.chromium.org/6515005/
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6782 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Strict mode assignment to undefined reference.
Simple assignments (x = <value>) use CODE_TARGET_CONTEXT.
StoreIC stores its own strictness in extra_ic_state.
The strcitness is propagated as further ic stubs are generated.
Details:
* ReferenceError on assignment to non-resolvable reference in strict mode.
* Fix es5conform test expectation file.
* Add es5conform test suite into .gitignore.
* Fix Xcode project.
* Change implemented in virtual frame code generator, as well as full-codegen
for all architectures.
* Fix debugger test.
* Fix comment for CODE_TARGET_CONTEXT
* Implement remaining StoreIC stubs to be strict mode aware.
* Trace extra_ic_state() for ic code stubs.
Code Review URL: http://codereview.chromium.org/6474026/
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6760 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Strict mode assignment to undefined reference.
Simple assignments (x = <value>) use CODE_TARGET_CONTEXT.
StoreIC stores its own strictness in extra_ic_state.
The strcitness is propagated as further ic stubs are generated.
Details:
* ReferenceError on assignment to non-resolvable reference in strict mode.
* Fix es5conform test expectation file.
* Add es5conform test suite into .gitignore.
* Fix Xcode project.
* Change implemented in virtual frame code generator, as well as full-codegen
for all architectures.
* Fix debugger test.
* Fix comment for CODE_TARGET_CONTEXT
* Implement remaining StoreIC stubs to be strict mode aware.
* Trace extra_ic_state() for ic code stubs.
Code Review URL: http://codereview.chromium.org/6474026/
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6756 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Before, an attempt to delete a parameter in a function that used the
arguments object in any way would succeed with true and delete both
the parameter and the corresponding arguments object property.
Now, an attempt to delete such a parameter does not delete and
evaluates to false.
Parameters can be deleted, as before, from functions that use the
arguments object, by deleting the corresponding arguments object
property (this is a spec violation).
BUG=fixes v8:1136
Review URL: http://codereview.chromium.org/6484023
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6745 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
There were a couple of bug fixes to this code on IA32 which have not yet
been ported to ARM. They are: failure to correctly handle non-JSObject
receivers and failure to restore the context register after calling JS code.
Review URL: http://codereview.chromium.org/6479019
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6744 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Change the comparison in the full code generator to use CompareIC instead of the CompareStub to record the types. This also implements the patching in the full code generator where the inlined smi code is de-activated by default to call the CompareIC once and then activating the inlined smi code by patching the code.
Fixed the smi comparison in the ICCompareStub.
Fixed ToBooleanStub to ensure that the scratch register used is not the input. Use r9 as default as that will never be input with Crankshaft.
Implemented lithium instruction CmpTAndBranch.
Make sure that the lithium instruction CmpID have operands in registrers as the current optimized code expects that.
Review URL: http://codereview.chromium.org/6461017
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6704 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Instead of constructing a temporary container for all LOperands of each
instruction, the register works directly on the LIR instructions that
provide an abstract interface for input/output/temp operands.
This saves allocation of zone memory and speeds up LIR construction,
but makes iterating over all uses in the register allocator slightly
more expensive because environment uses are stored in a linked list of
environments. We can fix this by using a flat representation of LOperands.
Review URL: http://codereview.chromium.org/6352006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6638 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Record a safepoint with a deoptimization id for throw in optimized code. We
don't seem to much care what the AST ID is because we will not be using it
for lazy deoptimization (throw doesn't return to the point of throw). For
hygiene we use the actual ID of the throw expression. Throw is no longer a
control-flow instruction, but it's followed by an unconditional abnormal
exit. This is required to insert a simulate between the throw and the exit.
Make our optimized treatment of Function.prototype.apply act like a call and
have side effects. This ensures that it will get a lazy deoptimization
environment. Use that deoptimization ID in the safepoint for the call.
Deleting a property was also missing a deoptimization ID, though there was a
deoptimization environment assigned to the instruction. Record the
environment and use the deoptimization ID at the safepoint.
Review URL: http://codereview.chromium.org/6250105
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6576 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Change OSR stack check patching to use the stack check table to iterate over the calls to stack guards platform independent. Introduce Deoptimizer::PatchStackCheckAt for each platform to perform the platform specific patch at a given pc.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6392027
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6551 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Deletion of global properties puts 'the hole' in the global property
cell and updates the property details in the property dictionary with
the information that the property has been deleted. When setting
global properties that have been deleted in generated code we just
store the new value in the global property cell. This does not update
the property details in the property dictionary. Therefore, it looks
like the property is not there eventhough it was just reintroduced.
Perform 'the hole' checks in generated code for global property stores
and bail out of ICs and optimized code if storing to a property cell
that contains 'the hole'.
Review URL: http://codereview.chromium.org/6306014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6508 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Refactor SafepointTableBuilder::DefineSafepoint and ARM LCodeGen::RecordSafepoint to use an enum for different kinds of safepoints. This change removes a lot of duplicated code and makes it easier to include new kinds of safepoints in the future. The remaining variants of LCodeGen::RecordSafepoint remain as a convinient way to record common safepoint kinds.
BUG=http://code.google.com/p/v8/issues/detail?id=1043
TEST=none
Review URL: http://codereview.chromium.org/6341008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6505 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Fix a bug in the --debug-code alignment check in the C entry stub.
Don't force the --debug-code flag in the ARM disassembler tests. The framework does support passing flags and the test runner will when running tests in debug mode.
Skip some deserialization tests which crashes from time to time.
Review URL: http://codereview.chromium.org/6393007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6484 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
To avoid deopts a few extra changes were needed:
o Enable megamorphic state for special property loads on
primitives. We used to flip between monomorphic stubs.
o Extract pure string (no string wrapper support) version of the
string length stub.
Review URL: http://codereview.chromium.org/6334015
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6472 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This implements the type recording binary operation stub for ARM. This first iteration only supports ADD. Handling of 32-bit integers is currently not implemented but just transitions. The generic case for now delegates to the generic binary operation stub.
Review URL: http://codereview.chromium.org/6342019
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6471 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Change the ARM exit frame to have the same layout as the IA32 exit frame. This basically re-arranges the order of fp and sp and changes the sp location of the entry frame to hold the sp used by the gc and not the sp for popping the arguments. This removes the option of tearing down the frame and returning using one ldm instruction.
The main motivation for this is to avoid pushing an alignment word before generating the entry frame. The GC handling of optimized frames process the registers pushed as part of a safepoint and asumes that these are at the top of the frame, so if an alignment word is pushed this processing will be one off.
The alignment handling in the C entry stub have also been simplified. Now the value of lr is stored to a stack slot already reserved avoiding pushing it and keeping track of "frame skew".
This does result in more instructions in the exit frame on ARM, but we can look into improving this later.
Review URL: http://codereview.chromium.org/6247019
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6448 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
files. On ARM it is not enough to iterate the relocation information
because that will only give us access to the constant pool and not to
the place in the instruction stream where the target in the constant
pool is called.
Review URL: http://codereview.chromium.org/6343005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6444 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The function Runtime_LookupContext searches the context chain for a
LOOKUP slot and returns the object holding the slot. It returned the
global context if the slot was not found or if it was found in a
function's context or arguments object. This is not the correct
object to use for 'delete'.
Since this lookup function is only ever used when deleting LOOKUP
slots (those that have to go through a with or a scope with eval), it
is simply replaced with a Runtime_DeleteContextSlot function that does
the appropriate thing for all kinds of context lookups.
This fixes Chromium bug 70066.
http://code.google.com/p/chromium/issues/detail?id=70066
Review URL: http://codereview.chromium.org/6280013
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6442 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
conversions for external array types, which implement the Typed Array
spec. (Revision of http://codereview.chromium.org/6315004 .)
Prefer SSE2 code path on x86 processors. Non-SSE2 processors now make
a slow runtime call for float-to-int conversions. Use SSE3 for 32-bit
signed and unsigned int array types where possible.
The movement of code from ic-arm.cc to stub-cache-arm.cc caused the
VFP3 code path to be tested for the first time. Fixed bugs in the
register usage and in the constant value stored into integer arrays
for NaN and +/-Infinity.
Added new truncation test to test-api.cc. Storage of NaN and +/-Inf
was already covered. Ran unit tests on x86, x64 and ARM simulator.
Tested ia32 and x64 code in Chromium on Mac and Linux respectively
with Typed Array unit tests and WebGL content.
BUG=http://code.google.com/p/chromium/issues/detail?id=50972
TEST=test-api/ExternalArrays
Review URL: http://codereview.chromium.org/6303012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6431 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This is svn r6415 with a bugfix.
The bug was that we can have scopes with 0 heap slots (functions with
such scopes do not need a local context) and scopes with at least
Context::MIN_CONTEXT_SLOTS. The code generator only allocated a local
context if there were strictly more than Context::MIN_CONTEXT_SLOTS.
Before the change to with/arguments, it was impossible to actually get
a function scope with exactly Context::MIN_CONTEXT_SLOTS because every
such scope included a heap-allocated arguments (and arguments shadow)
slot.
Review URL: http://codereview.chromium.org/6248014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6426 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Rename HBranch (the instruction that coerces an arbitrary HValue to
control flow) to HTest to free up the term Branch to refer to any
control instruction with two successors.
Change the virtual FirstSuccessor and SecondSuccessor functions on
control instructions to a pair of data members.
Review URL: http://codereview.chromium.org/6366002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6417 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This patch adds H- and L-variants of StringCharCodeAt and StringLength.
StringCharCodeAt is used to inline a constant function call of
String.prototype.charCodeAt and to implement the corresponding inline
runtime function. It does not yet use the recently introduced extra IC
state. (We can specialize on string encoding and avoid deopts because
of out of bounds accesses.)
StringLength needs more work because the stub version of it also
supports strings wrappers and it matters in some cases. (We have to
separate the string only case.)
Review URL: http://codereview.chromium.org/6243008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6408 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
We hit an assertion failure when we tried to record the AST ID of
the (shared) .arguments variable proxy more than once. This was hit
when we had multiple calls to the same parameter in a function that
used the arguments object. The fix is to not visit the subexpressions
of the (shared) property access expression.
BUG=1060
Review URL: http://codereview.chromium.org/6368007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6404 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Extra IC state is only two bits and only supported for call IC-s for
now. To change its extra state an IC stub jumps to a new miss stub
that goes to runtime as usual but then instead of going megamorphic
generates a monomorphic stub with the updated state.
Review URL: http://codereview.chromium.org/6344005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6370 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
emitted as part of the safepoint tables.
Always emit the constant pool as part of generating deferred code
(whether or not there actually is any deferred code) and ASSERT that
there is no pending relocation info while using db and dd to write
tables.
Review URL: http://codereview.chromium.org/6360003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6347 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Instead of spilling and then immediately restoring eax to resolve
memory to memory moves, the gap move resolver now tracks registers
that are known to be free and uses one if available. If not it spills
but restores lazily when the spilled value is needed or at the end of
the algorithm.
Instead of using esi for resolving cycles and assuming it is free to
overwrite because it can be rematerialized, the gap move resolver now
resolves cycles using swaps, possibly using a free register as above.
The algorithm is also changed to be simpler: a recursive depth-first
traversal of the move dependence graph. It uses a list of moves to be
performed (because it mutates the moves themselves), but does not use
any auxiliary structure other than the control stack. It does not
build up a separate list of scheduled moves to be interpreted by the
code generate, but emits code on the fly.
Review URL: http://codereview.chromium.org/6263005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6344 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This puts us very close to being able to compile the empty function.
This changes only has a small number of 64 bit specific assembler instructions.
The remaining changes are much more platform specific and will go in another change.
Review URL: http://codereview.chromium.org/6247005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6306 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This change provides fast code for a few special cases and calls the GenericBinaryOpStub for the rest.
It also changes the register allocation in the generation of lithium instructions to use fixed registers that are compatible with the generic stub. This allocation can be change once we use a more flexible implementation.
Finally, this change provides infrastructure to save double registers at safepoints, which is need to call the stub in deferred code.
BUG=
TEST=
Review URL: http://codereview.chromium.org/6164005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6304 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This instruction only depends on the prototype and the holder and can
completely ignore the receiver and its map.
This change also fixes a small bug on arm where a cell was loaded
instead of the prototype from new space.
Review URL: http://codereview.chromium.org/6094020
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6290 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Remove unused LOperands from keyed-loads. We do not have multiple representations
for load instructions anymore.
Correct number of output operands as for a couple of instructions form 1 to 0
because they do not produce a result (e.g. PushArgument)
Review URL: http://codereview.chromium.org/6158004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6258 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
I'm using the post call generator for the macro assembler
InvokeBuiltin call to be consistent with the ia32 version. It only
generates one call, but using the post call generator it should be
easier to catch if we change that at some point.
Review URL: http://codereview.chromium.org/6119004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6247 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The the call of the builtin in InstanceofStub was not correctly protected with an internal frame leading to the return address being handled as a pointer during GC.
Marked the Instanceof stub as allowing stub calls (the RecordWriteStub was removed some days ago).
This issue was not caught by the assertion designed for this when debug mode is run with --debug-code (which out tests always does) as generating code for Abort set the allow stub calls flag to true. This has been fixed by restoring the allow stub calls flag correctly.
Review URL: http://codereview.chromium.org/6097010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6218 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Currently, the bind for the slow label is after the pushing of the
args registers which renders the args pushing dead code because there
is a Ret preceeding it. The comments also imply that the arg pushing
should be part of the InvokeBuiltins tail call.
Patch by Mark Lam from Hewlett-Packard Development Company, LP
Review URL: http://codereview.chromium.org/6107001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6198 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The function HEnvironment::SetExpressionStackAt did not update the
environment's history. This function is used to patch the bailout
environment for count operations and global function calls.
Reorganize class HEnvironment to make it fit V8's style a bit better
and to try to add some sanity to which C++ functions are intended to
be inlined.
Remove the flag --trace-environment which merely duplicated data in
the hydrogen.cfg file except without enough context to be useful.
BUG=1004
Review URL: http://codereview.chromium.org/5992011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6137 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
1. Separating out the instance-type check from the array-length operation.
2. I also changed the bounds-check on keyed loads to use the length property
for JS arrays (like we do for array stores).
The new pattern should use less registers and allow more checks to be eliminated.
Review URL: http://codereview.chromium.org/5961016
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6125 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
When entering a finally block in unoptimized code, we unconditionally
save the accumulator register in the stack in case it holds a return
value or an exception. In the case of a break, continue, or falling
off the end of the try or catch block, this value is unpredictable and
not necessarily safe for GC.
Review URL: http://codereview.chromium.org/5883003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6035 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This is an improved version of my earlier change r5970. It avoids degrading the
non-optimized code.
Initially we emit a conditional branch that is either always- or never-taken
after a smi-check (depending on whether we test for smi for for non-smi)
Since test-eax always sets the carry-flag to 0 we use jump-if-carry and
jump-if-not-carry.
The first invocation of the stub patches a jc with a jz and
jnc with a jnz-instruction so that the code looks exactly as it was
without patching. The only difference is the test- or nop-instruction
after the IC-call.
Review URL: http://codereview.chromium.org/5763004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6030 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
SVN revision 5993 ported changes to the unoptimizing compiler from
ia32 to ARM and x64. A call to AstContext::Plug was moved out of
FullCodeGenerator::EmitVariableAssignment and into the callers. One
call site was missed in both ports.
Review URL: http://codereview.chromium.org/5781008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6003 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
In the case of inlined smi code in non-optimzied code we could not
distinguish between the smi-only case and the case that the operation was
never executed.
With this change the first execution of a binary operation always jumps
to the stub which in turn patches the smi-check into the correct
conditional branch, so that we benefit from inlined smi code after the
first invocation.
A nop instruction after the call to the BinaryOpIC indicates that no
smi code was inlined. A "test eax" instruction says that there was smi
code inlined and encodes the delta to the patch site and the condition
code of the branch at the patch site to restore the original jump.
Review URL: http://codereview.chromium.org/5714001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5970 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
that ip can be used as a scratch register. This is not true because
ip is already used for something else in AllocateInNewSpace in the
macro assembler.
Explicitly allocate a temp register for use in the heap number
allocation instead.
Review URL: http://codereview.chromium.org/5788001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5967 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
inline allocation code used the expected number of properties to
calculate the number of inobject properties for an object instead of
getting the actual number from the initial map.
It is safer to use the inobject property count from the initial map in
any case because that is the amount the instances will get. I think
this disconnect got introduced when adding shrinking of objects.
Unfortuntely I haven't been able to create a simple reproduction for a
test case but this fixes the webpage that exhibits the crash. I'll see
if I can create a reproduction tomorrow.
Review URL: http://codereview.chromium.org/5278003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5879 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Fix the ARM simulator, the ARM disassembler and extend the stop feature.
The stop feature in the simulator now support enabling, disabling and
counting.
BUG=None
TEST=None
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5723 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Change more functions used by the Compiler class to have a uniform
interface: they get passed as argument an input/output pointer to a
CompilationInfo that they mutate if they succeed, and they return a
flag telling whether they succeeded.
Also, remove some unnecessary timers.
Review URL: http://codereview.chromium.org/3561012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5583 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The plan is to use the CompilationInfo class to communicate inputs and
outputs to compilation pipeline phases, which each return a boolean
success/failure flag.
The intent is to make it easier to compose small pieces of the
pipeline without having to grow a custom function each time, each
taking a half dozen arguments.
This change modifies the very front end (the parser).
Review URL: http://codereview.chromium.org/3586006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5581 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
For some reason, the scope's arguments and arguments shadow were
variable proxies, which resulted in all references to the arguments
shadow being shared in the AST. This makes it hard to put per-node
state on the AST nodes.
I took the opportunity to remove Variable::AsVariable which has
confused people in the past, and to rename Variable::slot to the more
accurate Variable::AsSlot.
Review URL: http://codereview.chromium.org/3432022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5517 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The number of inobject properties used to be derived from the number
of this property assignments in the constructor (and increased by 2 to
allow for properties added later). This very often leads to wasted inobject
slots.
This patch reclaims some of the unused inobject space by the following method:
- for each constructor function the first several objects are allocated using the initial
("generous) instance size estimation (this is called 'tracking phase').
- during the tracking phase map transitions are tracked and actual property counts are collected.
- at the end of the tracking phase instance sizes in the maps are decreased if necessary
(starting with the function's initial map and traversing the transition tree).
- all further allocation use more realistic instance size estimation.
Shrinking generously allocated objects without costly heap traversal is made possible
by initializing their inobject properties with one_pointer_filler_map (instead of undefined).
The initial slack for the generous allocation is increased from 2 to 6 which really helps some tests.
Review URL: http://codereview.chromium.org/3329019
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5510 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Contextual load requires only a map check followed by a cell hole
check so we can generate pretty compact code for that. The fact that
we have inlined code is marked by mov ecx, offset instruction after
the IC call. Inlining is only enabled inside loops and in non-builtin
functions.
The generated code size increase is about 3%. This descreased the
pc-to-code cache hit rate in some of the benchmarks that trigger
GC. To compensate we now have 4 times as much entries in the cache.
Review URL: http://codereview.chromium.org/3402014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5497 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
When running profiling in debug mode, several assertions in frame
iterators that are undoubtedly useful when iterator is started from a
VM thread in a known "good" state, may fail when running over a stack
of a suspended VM thread. This patch makes SafeStackFrameIterator
to proactively check addresses and bail out from iteration early,
before an assertion will be triggered.
BUG=crbug/55565
Review URL: http://codereview.chromium.org/3436006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5467 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
the code object's instructions.
This allows us to find a code object using just the pc. This approach
uses a cache (PcToCodeCache) to make sure we don't continuously have
to iterate heap pages.
This change eliminates the need for cooking and uncooking of stack frames.
Review URL: http://codereview.chromium.org/3226014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5369 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The live registers are now only stored to the expression stack with the non pointer values being stored as smis (on the 32-bit platforms these values are assumed to be 31-bit max).
This makes the CEntryStub entry/exit code much simpler, and there is no longer any need for a mode (debug or normal) on it.
Fix a missing live register when breaking at ARM keyed load.
Review URL: http://codereview.chromium.org/3141047
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5358 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Object model changes
----------------------------------------
New fixed_cow_array_map is used for the elements array of a JSObject
to mark it as COW. The JSObject's map and other fields are not
affected. The JSObject's map still has the "fast elements" bit set. It
means we can do only the receiver map check in keyed loads and the
receiver and the elements map checks in keyed stores. So introducing
COW arrays doesn't hurt performance of these operations. But note that
the elements map check is necessary in all mutating operations because
the "has fast elements" bit now means "has fast elements for reading".
EnsureWritableFastElements can be used in runtime functions to perform
the necessary lazy copying.
Generated code changes
----------------------------------------
Generic keyed load is updated to only do the receiver map check (this
could have been done earlier). FastCloneShallowArrayStub now has two
modes: clone elements and use COW elements. AssertFastElements macro
is added to check the elements when necessary. The custom call IC
generators for Array.prototype.{push,pop} are updated to avoid going
to the slow case (and patching the IC) when calling the builtin should
work.
COW enablement
----------------------------------------
Currently we only put shallow and simple literal arrays in the COW
mode. This is done by the parser.
Review URL: http://codereview.chromium.org/3144002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5275 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Upgraded the CodeGenerator::ToBoolean() function in the ARM backend to use complete JIT code generation and not make runtime calls to ToBool (when VFP is enabled).
This change also includes the vcmp VFP instruction that supports a constant 0.0 as the second operand.
Patch by Subrato K De <subratokde@codeaurora.org>
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5267 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
