Commit Graph

11827 Commits

Author SHA1 Message Date
yangguo@chromium.org
8db56239a0 Correcting mime type of tools/profviz/profviz.css.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15297 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:55:36 +00:00
yangguo@chromium.org
239fffdb6d Changing mime type of tools/profviz/profviz.css.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15296 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:53:17 +00:00
yangguo@chromium.org
64cdd5e42c Changing mime type of tools/profviz/profviz.html.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15295 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:52:09 +00:00
bmeurer@chromium.org
13a7c993d0 Add phase zone to CompilationInfo and use it in GVN pass.
The phase_zone of CompilationInfo is intended for local allocations that
are freed at the end of the phase.

R=danno@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17573003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15294 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:37:46 +00:00
yurys@chromium.org
264b7255e1 Move ExitFrame::ComputeStackPointer from frames-<arch>.cc to frames.cc
The method has identical implementation for all architectures. Moved it into frames.cc

Drive-by: deleted SafeStackFrameIterator::is_working_iterator_, SafeStackFrameIterator::iteration_done_ is used instead.

BUG=None
R=loislo@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17581004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15293 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:27:48 +00:00
yangguo@chromium.org
3aec8269f9 Add v8.log visualizer page.
R=jkummerow@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17592002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15292 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:25:58 +00:00
bmeurer@chromium.org
3481344df8 Drop (mis)use of HPhase in full code gen.
The full code generator is not a proper Hydrogen phase. The full code
gen ticks are tracked in a special member of HStatistics, so HPhase is
most probably misused at this point.

R=danno@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17590005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15291 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:25:44 +00:00
yangguo@chromium.org
c63a5255b4 Do not use deprecated MakeWeak in d8.
R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17589009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15290 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:15:07 +00:00
bmeurer@chromium.org
00f6a8903b Fix HStatistics::Print() output to fit into 80 columns.
BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/17572006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15289 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:07:20 +00:00
loislo@chromium.org
e04a874edd CPUProfiler: fix for test.
TBR= yangguo@chromium.org

Review URL: https://codereview.chromium.org/17589010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15288 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 13:03:19 +00:00
loislo@chromium.org
1a15f8f768 CPUProfiler: Simplify logging part of CreateCodeEvent functions.
We have 5 overloaded functions with name CreateCodeEvent.
All these functions have many common parts. I'd like to eliminate the difference between them.

TEST=existing tests
R=yangguo@chromium.org, yurys@chromium.org

Review URL: https://codereview.chromium.org/16901014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15287 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 12:55:19 +00:00
mstarzinger@chromium.org
b632a4be86 Deprecate one virtual marking visitor for mark-compact GC.
R=hpayer@google.com

Review URL: https://codereview.chromium.org/17546005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15286 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 12:39:02 +00:00
dslomov@chromium.org
4308fb462d API for DataView.
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17155014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15282 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 11:23:50 +00:00
mstarzinger@chromium.org
48c16454b1 Revert "Remove IsInitialized checks from inlined API functions."
This reverts r15277 due to failures in layout tests. Apparently Blink
still initializes the Isolate by calling v8::Null() as the first API
function on some paths.

TBR=svenpanne@chromium.org
TEST=webkit:crypto/worker-random-values-concurrent.html

Review URL: https://codereview.chromium.org/17577008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15281 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 11:10:40 +00:00
rossberg@chromium.org
007837365d Rename Literal::handle to Literal::value
Just a completely mechanical change...

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17576005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15280 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 10:37:59 +00:00
yurys@chromium.org
a8a9fa807a Update the TSan-specific atomics implementation to match the current API.
BUG=128314

Review URL: https://codereview.chromium.org/17591005

Patch from Alexander Potapenko <glider@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15279 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 09:35:30 +00:00
mstarzinger@chromium.org
1c22b2cd98 Make a Persistent ctor take const Persistent& instead of Persistent&.
BUG=
R=mstarzinger@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17580010

Patch from Marja Hölttä <marja@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15278 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 09:34:56 +00:00
mstarzinger@chromium.org
9ef18efff6 Remove IsInitialized checks from inlined API functions.
R=marja@chromium.org, svenpanne@chromium.org
TEST=cctest/test-api

Review URL: https://codereview.chromium.org/17068006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15277 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 08:58:01 +00:00
loislo@chromium.org
9d95db6fad CPUProfiler: remove copy&paste from CodeCreateEvent.
This is the next patch on the way of simplifying profiler API.
It is extracting the common parts of the serialization process.

TEST=current test set
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/16917008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15276 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 08:40:38 +00:00
yurys@chromium.org
f830fbea3c Simplify stack iterators implementation
In order to fix https://code.google.com/p/chromium/issues/detail?id=252097 I
need to change SafeStackTraceFrameIterator. Stack iterators hierarchy looks
excessively complicated and I'd like to flatten it a bit by removing some
intermediate classes. In particular there are two hierarchies sharing
JavaScriptFrameIteratorTemp<T> template for no good reason.

This change extracts some of JavaScriptFrameIteratorTemp functionality directly
into SafeStackTraceFrameIterator. This made it obvious that a few checks were
performed twice.

The rest of JavaScriptFrameIteratorTemp<T> is merged with
JavaScriptFrameIterator. Now that the class is not a template some of its
implementation is moved from frames-inl.h into frames.cc

So in this change I removed JavaScriptFrameIterator and
SafeJavaScriptFrameIterator. As the next step I'm going to merge
SafeStackFrameIterator into SafeStackTraceFrameIterator.

BUG=None
R=loislo@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/16917004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15275 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 08:38:37 +00:00
dslomov@chromium.org
b05628f540 Update boot-up memory use limit after adding DataView.
R=danno@chromium.org

Review URL: https://codereview.chromium.org/17589005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15274 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-24 07:02:53 +00:00
palfia@homejinni.com
ed4d3c8c66 MIPS: Use AST's type field and merge types for unary, binary & compare ICs.
Port r15264 (d4c3767)

Missing fixed_right_arg().has_value cases added also.

BUG=

Review URL: https://codereview.chromium.org/17552008
Patch from Balazs Kilvady <kilvadyb@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15273 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 20:31:06 +00:00
jkummerow@chromium.org
96e818136c Use kPointerSize to compute the size of StubCache::Entry for X64
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/17552004

Patch from Haitao Feng <haitao.feng@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15272 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 16:19:51 +00:00
dslomov@chromium.org
91eb5f8d25 DataView implementation.
R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/17153011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15269 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 13:02:38 +00:00
jkummerow@chromium.org
a4d2759d49 Distinguish times_8 and times_pointer_size for X64
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/17550005

Patch from Haitao Feng <haitao.feng@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15268 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 12:57:19 +00:00
jkummerow@chromium.org
cbc8ed7110 HBoundsCheck shouldn't ignore Tagged input representations unconditionally
BUG=v8:2740
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/16917003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15267 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 12:50:02 +00:00
yurys@chromium.org
7f1750580c V8 API: Add a missing NULL check into Isolate::GetCurrentContext().
There is a missing NULL check for: "internal_isolate->context() != NULL".
Right now before calling this method one should call v8::Context::InContext()
first to perform this check, otherwise we may crash. But this static method
will do this check on the current isolate, which may not be the same as a given one.

BUG=249655
R=yurys@chromium.org,mvstanton@chromium.org

Review URL: https://codereview.chromium.org/16943006

Patch from Andrey Adaykin <aandrey@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15266 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 12:34:47 +00:00
rossberg@chromium.org
04e1278529 Remove BinaryOp::result_type
R=jkummerow@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17261023

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15265 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 12:16:14 +00:00
rossberg@chromium.org
0977c60984 Use AST's type field and merge types for unary, binary & compare ICs
R=jkummerow@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17468003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15264 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 11:10:06 +00:00
yangguo@chromium.org
b7b92bd9ac Short-circuit embedded cons strings.
R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17418003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15263 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 09:24:30 +00:00
yangguo@chromium.org
cbf6244029 Use mutex instead of busy wait when installing optimized function.
R=jkummerow@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17099012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15262 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 08:38:12 +00:00
yangguo@chromium.org
2e10e3e336 Clean up queued compilation tasks when quitting.
R=jkummerow@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17543002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15261 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 08:37:05 +00:00
yurys@chromium.org
6f09cb471b Delete deprecated methods from v8-profiler.h
All these methods were already marked as V8_DEPRECATED in V8 Version 3.18.0 (see https://code.google.com/p/v8/source/browse/trunk/include/v8-profiler.h?r=14304) so it is time to delete them.

BUG=None
R=loislo@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17524007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15260 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 07:56:22 +00:00
plind44@gmail.com
0bc8ef88dd MIPS: Let NaN flow as double into HBranch + some minor improvements
Port r15246 (cb18dce2)

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/17198012

Patch from Akos Palfi <palfia@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15259 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-21 02:43:13 +00:00
palfia@homejinni.com
54483c4880 MIPS: Generators: Avoid calling into runtime if operand stack is empty
Port r15240 (909f716e)

Original commit message:
This patch makes yield sites save the resume continuation and context
inline.  If the operand stack is empty, we can avoid a call into the
runtime.  This also makes the SuspendJSGeneratorObject runtime function
less magical: it just has to save the operand stack and stack handlers.

This speeds up the following case by a factor of 3 or so:

  function* until(n) {
    for (var i = 0; i < n; i++)
      yield i;
  }

  function sum(iter) {
    var sum = 0;
    for (var x of iter) sum += x;
    return sum;
  }

  for (var i = 0; i < 10000; i++) sum(until(1000))

Also, there is no more sentinel value as the generators will resume in
the right place already, allowing me to remove the hack added to the
--debug-code check in r14437.

BUG=

Review URL: https://codereview.chromium.org/17291016

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15258 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 23:43:10 +00:00
palfia@homejinni.com
b58d0d2af7 MIPS: Lithium codegen should not pass around block_ids.
Port r15235 (95a844b)

Original commit message:
Rather encapsulate the basic block to assembly label mapping in the LInstruction.

BUG=

Review URL: https://codereview.chromium.org/17504002
Patch from Balazs Kilvady <kilvadyb@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15257 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 20:18:58 +00:00
danno@chromium.org
fd08ac7c26 Handle MONOMORPIC loads where type feedback is code stub
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/17451019

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15254 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 16:54:09 +00:00
jkummerow@chromium.org
0e54a418db Add counters for requested/inserted/executed soft deopts
R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/17498003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15253 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 16:53:22 +00:00
palfia@homejinni.com
6a91645a84 MIPS: Fix stack frame reconstruction for generators with formal arguments.
Port r15230 (e2bbe01)

Original commit message:
The formal parameter count was always being treated as an untagged
integer, but it is actually a Smi on ia32 and arm.

BUG=v8:2355
TEST=mjsunit/harmony/generators-iteration

Review URL: https://codereview.chromium.org/17261028
Patch from Balazs Kilvady <kilvadyb@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15250 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 15:29:22 +00:00
yurys@chromium.org
c7b3ef0e05 Make sure ExternalCallbackScope is always created when VM state changes to EXTERNAL
ExternalCallbackScope is used to let CPU profiler know which API callback is being executed. Whenever such callback is called we should create VMState<ETERNAL> and ExternalCallbackScope. This patch fixes several places where VMState<ETERNAL> went without ExternalCallbackScope.

BUG=244580
R=dcarney@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17059005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15249 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 14:47:35 +00:00
verwaest@chromium.org
6ae544eca7 Also apply InferControlFlowRange to smi-based tests.
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/16870011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15248 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 14:20:15 +00:00
verwaest@chromium.org
917ac7d413 Improve range analysis for smi phi
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17498002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15247 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 14:11:52 +00:00
olivf@chromium.org
29d3461529 Let NaN flow as double into HBranch + some minor improvements
BUG=
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/17082003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15246 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 13:51:03 +00:00
verwaest@chromium.org
2b485416ed Properly assign heapobject to constant functions.
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17406009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15245 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 13:46:29 +00:00
verwaest@chromium.org
4922e3d206 Flexible representation for BuildIncrement, but CannotBeTagged.
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17028017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15244 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 13:22:46 +00:00
verwaest@chromium.org
b17e759dc1 Add soft-deopt for uninitialized assignment
R=hpayer@google.com

Review URL: https://chromiumcodereview.appspot.com/17071002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15241 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 13:09:43 +00:00
wingo@igalia.com
646a34e1e0 Generators: Avoid calling into runtime if operand stack is empty
This patch makes yield sites save the resume continuation and context
inline.  If the operand stack is empty, we can avoid a call into the
runtime.  This also makes the SuspendJSGeneratorObject runtime function
less magical: it just has to save the operand stack and stack handlers.

This speeds up the following case by a factor of 3 or so:

  function* until(n) {
    for (var i = 0; i < n; i++)
      yield i;
  }

  function sum(iter) {
    var sum = 0;
    for (var x of iter) sum += x;
    return sum;
  }

  for (var i = 0; i < 10000; i++) sum(until(1000))

Also, there is no more sentinel value as the generators will resume in
the right place already, allowing me to remove the hack added to the
--debug-code check in r14437.

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/15990004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15240 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 12:59:45 +00:00
dcarney@chromium.org
8202410cca deprecate old style callbacks
R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17069003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15239 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 12:53:19 +00:00
mstarzinger@chromium.org
697bee6157 Use +kCallerSPOffset instead of -kMarkerOffset to compute the args pointer
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/17451014

Patch from Haitao Feng <haitao.feng@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15238 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 12:51:36 +00:00
dcarney@chromium.org
a74f511e61 remove all old style callbacks - patch 3
TBR=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17336003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15237 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-20 12:28:27 +00:00