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
It seems that when calling a method that has two overloaded versions like this:
f(char* format, ...)
f(char* format, va_list args)
with a second pointer argument: f("format", pointer), the second version is picked up.
I've found a description of a similar issue here: http://bugs.gentoo.org/63112
So, to resolve this ambiguity, I've named such LogMessageBuilder's Append functions differently.
Review URL: http://codereview.chromium.org/125125
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2172 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Code addresses are now written as an offset from the previous address for ticks, code move and delete events. Employed backreference and RLE compression for code move and delete events. This gives additional 30% log size reduction for benchmarks run w/o snapshot.
Overall compression results (compared with the revision of V8 having no compression):
- V8: 70% size reduction for benchmarks run w/o snapshot (for reference, gzip gives 87%)
- Chromium: 65% size reduction for public html version of benchmarks (v4) (for reference, gzip gives 90%)
The one obvious opportunity for improving compression results in Chromium is to compress URLs of scripts.
Review URL: http://codereview.chromium.org/125114
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2162 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
For each frame it is now possible to request information on the scope chain. Each scope in the chain can have one of the types local, global, with and closure. For scopes of type global and with the mirror for the actual global or with object is available. For scopes of type local and closure a plain JavaScript object with the materialized content of the scope is created and its mirror is returned. Depending on the level of possible optimization the content of the materialized local and closure scopes might only contain the names which are actually used.
To iterate the scope chain an iterator ScopeIterator have been added which can provide the type of each scope for each part of the chain. This iterator creates an artificial local scope whenever that is present as the context chain does not include the local scope.
To avoid caching the mirror objects for the materialized the local and closure scopes transient mirrors have been added. They have negative handles and cannot be retrieved by subsequent lookup calls. Their content is part of a single response.
For debugging purposes an additional runtime function DebugPrintScopes is been added.
Added commands 'scopes' and 'scope' to the developer shell and fixed the dir command.
BUG=none
TEST=test/mjsunit/debug-scopes.js
Review URL: http://codereview.chromium.org/123021
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2149 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Two techniques are involved:
- compress repeated line ends (common stack beginnings) by using back references;
- do RLE compression of repeated tick events.
This gives only 5% size reduction on benchmarks run, but this is because tick events are only comprise 10% of file size. Under Chromium winnings are bigger because long repeated samples of idleness are now compressed into a single line.
Tickprocessor will be updated in the next patch.
Review URL: http://codereview.chromium.org/123012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2147 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
* Identify heap numbers that contain non-Smi int32s and do bit
ops on them without calling the fp hardware or emulation.
* Identify results that are non-Smi int32s and write them into
heap numbers without calling the fp hardware or emulation.
* Do unary minus on heap numbers without going into the runtime
system.
* On add, sub and mul if we have both Smi and heapnumber inputs
to the same operation then convert the Smi to a double and do
the op without going into runtime system. This also applies
if we have two Smi inputs but the result is not Smi.
Review URL: http://codereview.chromium.org/119241
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2131 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
instructions. The intention is that the snapshots generated
by the simulator should be usable on the hardware. Instead of
swi instructions we generate a branch to a swi instruction that
is not part of the snapshot. The call/jump is patched up in
the same way as other external references when the snapshot
is deserialized. This only works for EABI targets: on old ABI
targets we still emit some instructions not supported by the
simulator (fp coprocessor instructions).
Review URL: http://codereview.chromium.org/119036
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2127 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
Scripts now have a compilation type which can be host, eval or JSON. Host scripts are compiled through the API, eval scripts are compiled through call to evan and JSON scripts are compiled as a result of calling JSON.parse.
For scripts scripts compiled through eval the JavaScript function in top of the stack and the pc offset into the code is stored in the script object. This makes it possible to calculate the source position of the eval call later when requested. This information can be obtained through the script mirror object and is part of the script mirror JSON serialization for the debugger protocol.
Moved the enumeration ScripType into class Script and remamed to Type. The new compilation type enumeration is also inside the class Script.
This information is now shown when using the scripts command in he developer shell debugger.
Review URL: http://codereview.chromium.org/119108
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2119 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
interceptors and dont-delete attributes.
Minor change to the behavior of eval: throw exception when calling
eval in a context for which the global has been detached. This
matches the behavior of both Firefox and Safari post navigation in the
browser.
Review URL: http://codereview.chromium.org/118374
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2118 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
submitted in revisions 2093, 2094, 2099, and 2106.
There's no evidence that supports that these changes
should be the cause of the unexplained performance
regressions on the intl2 and DHTML page cyclers.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2109 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Change stack alignment on linux to 16 bytes to keep gcc 4.4 happy.
This fixes the mksnapshot segfault without requiring -fno-tree-vectorize
which just avoided the problem by not generating code with movdqa.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2107 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The problem was I incorrectly treated NULL result as failure to fetch
a property with a getter. However, if getter returns zero, it is
manifested as NULL pointer (see added test case).
Good news: that gives another boost as before this CL if getter returned
0, I did another slow lookup.
Review URL: http://codereview.chromium.org/119172
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2106 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
immediately if holder has this property or saves binary search on holder if
property doesn't belong to holder. Of course, in the cases when named getter
returns nothing.
That gives ~20% for dom benchmark/Document Object String Get, speeds up overall
dom_perf (not dramatically) and overall score for peacekeeper. Strange, but DOM
part of peacekeepr runs somewhat slower.
Review URL: http://codereview.chromium.org/118118
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2093 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
My assumption that log initialization happens somewhere near the stack's bottom is true for V8's sample shell but isn't true for Chromium, causing many otherwise valid stack addresses to be thrown out. The solution proposed is to save stack pointer value for the outermost JS function in ThreadLocalTop similar to c_entry_fp.
Implemented only for IA-32. Currently I'm not dealing with profiling on ARM and x86-64 anyway.
Review URL: http://codereview.chromium.org/112082
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2086 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
* Running "system" JavaScript with the debug break flag active leads to slow running code while waiting for the break in non "system" JavaScript (one exception to this it is to try to avoid breaks in the clear mirror cache JavaScript code called when leaving the debugger).
* If this happens while processing RegExp running in native code an infinite loop is created as the stack guard handler for RegExp does not move execution forward
Fixed a GC bug in the interrupt handling for RegExp running in native code.
Added test of debug break while in debug message handler callback and debug break while executing a RegExp.
Review URL: http://codereview.chromium.org/115262
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2074 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
When profiler's memory buffer is filled up, profiling is stopped and it is ensured that the last record in the buffer is "profiler,\"pause\"" thus making the end of profiling session explicit. Otherwise DevTools Profiler would need to guess whether the current profiling session has been stopped.
Tested with Chromium.
Review URL: http://codereview.chromium.org/115859
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2072 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
I suggest that the lack of initialization causes issue 358 to happen. In Profiler::Disengage an empty TickSample is inserted in order to wake up the Profiler thread. Issue reporter claims that crash happens in LogTickEvent function. My guess is that frames_couint receives a wild value.
Review URL: http://codereview.chromium.org/113939
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2068 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The goal of this change is to allow longer profiling sessions and preserve memory when profiler isn't started. The buffer starts with 64K and grows until it reaches the upper limit, which is currently set to 50MB --- according to my evaluations, this is enough for at least 20 minutes of GMail profiling. As we're planning to introduce compression for the profiler log, this time boundary will be significantly increased soon.
To make possible unit testing of the new component, I've factored out Logger's utility classes into a separate source file: log-utils.h/cc. Log and LogMessageBuilder are moved there from log.cc without any semantical changes.
Review URL: http://codereview.chromium.org/115814
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2067 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
