During parsing functions are analyzed for statements of the form this.x = ...;. These assignments are categorized in two types: simple and non simple. The simple ones are where the right hand side is known to be either a constant or an argument to the function. If a function only contains statements of this type the property names are collected and for the simple assignments the index of the argument or the constant value assigned are stored as well.
When the initial map for a function is created and the function consists of only this type of assignemnts the initial map is created with a descriptor array describing these properties which will be known to always exist in an object created from the function.
The information on this property assignments is not collected during pre-parsing so if compiling using pre-parse data these optimization hints are not available.
Next step will be to use the information collected for the simple assignments to generate constructor code which will create and initialize the object from this information without calling the code for the function.
Review URL: http://codereview.chromium.org/172088
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2710 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Only send the inscructions part of a code object to oprofile when reporting dynamically generated code. Before the code object header was also reported to oprofile as code which caused strange disassembly output when using opannotate.
Review URL: http://codereview.chromium.org/125126
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2174 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Marsaglia's multiply-with-carry instead of mixing the
bits obtained from calling the system random() twice.
This seems to be a bit faster and gives a better
distribution than the system random() in particular on
Windows.
Review URL: http://codereview.chromium.org/126113
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2159 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This is a trivial per-row compression:
- short aliases are introduced for events and code creation tags;
- in tick events, offsets are used instead of absolute addresses;
- removed 'code-allocation' event, as it seems not used.
The first two options are depend on the new flag: 'compress-log', which is off by default.
On benchmarks run w/o snapshot, this gives 45% log size reduction.
Review URL: http://codereview.chromium.org/119304
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2122 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
deferred code snippets are highly stylized. They always make a call
to a stub or the runtime and then return. This change takes advantage
of that.
Creating a deferred code object now captures a snapshot of the
registers in the virtual frame. The registers are automatically saved
on entry to the deferred code and restored on exit.
The clients of deferred code must ensure that there is no change to
the registers in the virtual frame (eg, by allocating which can cause
spilling) or to the stack pointer. That is currently the case.
As a separate change, I will add either code to verify this constraint
or else code to forbid any frame effect.
The deferred code itself does not use the virtual frame or register
allocator (or even the code generator). It is raw macro assembler
code.
Review URL: http://codereview.chromium.org/118226
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2112 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
called from within a loop or not. In the past we lost the
information if a call site went megamorphic before a lazily
compiled callee was called for the first time. Now we track
that correctly (this is an issue that affects richards).
We still don't manage to track the in-loop state through a
constructor call, since constructor calls use LoadICs instead
of CallICs. This issue affects delta-blue. So in this patch
we assume that lazy compilations that don't happen through a
CallIC happen from inside a loop. I have an idea to fix this
but this patch is big enough already.
With our improved tracking of in-loop state I have switched
off the inlining of in-object loads for code that is not in
a loop. This benefits compile speed. One issue is that
eagerly compiled code now doesn't get the in-object loads
inlined. We need to eagerly compile less code to fix this.
Review URL: http://codereview.chromium.org/115744
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2046 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
This issue was raised by Brett Wilson while reviewing my changelist for readability. Craig Silverstein (one of C++ SG maintainers) confirmed that we should declare one namespace per line. Our way of namespaces closing seems not violating style guides (there is no clear agreement on it), so I left it intact.
Review URL: http://codereview.chromium.org/115756
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2038 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
ENABLE_DEBUGGER_SUPPORT is enabled by default unless it is on Android platform.
On Android platform, it can also enabled by passing -DENABLE_DEBUGGER_SUPPORT flag to the compiler.
This should not affect any existing build (I hope, cross my fingers) except the build in real Android environment (in other word, it only affects me now).
There are lot of room for code refactoring in stead of using #ifdef all over the place. I will leave this to v8 folks.
Review URL: http://codereview.chromium.org/77035
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@1745 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Currently function name inference is wired with AST optimization pass to avoid introducing another pass over AST. A better solution would be to rewrite AST visitors so they can be naturally combined together in a single pass, as their current implementation doesn't allow it.
For examples of cases where function names can be inferred, see the tests file.
Review URL: http://codereview.chromium.org/62146
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@1696 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
counting the reference to the return value and passing it to the
return label. This requires threading it through try/catch and
try/finally. The return value is loaded into eax more lazily than
before.
Also, perform some related refactoring of jump targets.
Review URL: http://codereview.chromium.org/56172
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@1669 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
void CodeForStatement(Node* node)
void CodeForSourcePosition(int pos)
The first is used to indicate that code is about to be generated for the given statement and the second is used to indicate that code is about to be generated for the given source position.
Added position information for some statements which was missing whem.
Updated the code generator for ARM to emit source position the same way as for IA-32.
Added an assert to ensure that deferred code stubs will always have a source source position as if it has not it will take whatever source position before which makes no sense.
The passing test on ARM has only been tested using the simulator.
Review URL: http://codereview.chromium.org/14170
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@985 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
upper case. Moved it into the RelocInfo class together with the associated
is_xxx functions. Renamed is_xxx to IsXxx in the process.
Removed the exit_js_frame mode as it was no longer used.
Patch Set 2 renames RELOC_MODE_COUNT to NUMBER_OF_MODES and fixes a couple of lint errors.
Review URL: http://codereview.chromium.org/3186
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@354 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
until a possible debug break location is reached. Currently this is call sites
with calls to code objects and JS return. Source position information in the
code therefore no longer refers to the "first" instruction generated for a
given source position (which was not the case defered code anyway) but to the
first break location after that source position was passed (again defered code
always start with source position information). This doesn't make a difference
for the debugger as it will always be stopped only at debug break locations.
However, this makes the life of the peep-hole optimizer much easier as many
oportunities for posh/pop eliminations where previosly blocked by relocation
information already written to the code object.
Two types of source positions are still collected. Statement positions indicate
the position of the start of the statement leading to this code and (plain)
positions indicate other places typically call sites to help indicate current
position in backtraces. The two different types of positions are also used to
distinguish between step next and step in.
Runs all the tests (including debugger tests) as before.
Moved the checking for the FLAG_debug_info to one place.
I will do the same changes to the ARM codegenerator in a seperate changelist.
Review URL: http://codereview.chromium.org/2957
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@335 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
