mikhail.naganov@gmail.com [Mon, 13 Dec 2010 10:42:06 +0000 (10:42 +0000)]
New heap profiler: add support for progress reporting and control.
As taking a snapshot of a large heap takes noticeable time, it's
good to be able to monitor and control it.
The change itself is small, big code deletes and additions are in
fact moves. The only significant change is simplification of
approximated retained sizes calculation algorithm.
Review URL: http://codereview.chromium.org/5687003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5978
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Mon, 13 Dec 2010 10:41:50 +0000 (10:41 +0000)]
Revert r5970 and r5975.
Review URL: http://codereview.chromium.org/5717005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5977
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Mon, 13 Dec 2010 10:09:09 +0000 (10:09 +0000)]
Fix incorrect assumption about young/old space allocation in
JSON stringify (introduced in r5951).
Review URL: http://codereview.chromium.org/5746005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5976
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Mon, 13 Dec 2010 10:05:19 +0000 (10:05 +0000)]
Land Vitaly's change to fix compare IC performance.
Original change: http://codereview.chromium.org/5733004/
When we have inlined smi code can transition to heap number state before going to the generic state. Without inlined smi code the behaviour is unchanged.
Review URL: http://codereview.chromium.org/5689005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5975
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Mon, 13 Dec 2010 08:33:32 +0000 (08:33 +0000)]
Make RegExp character class match JSC.
See http://trac.webkit.org/changeset/73594
Review URL: http://codereview.chromium.org/5723002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5974
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Fri, 10 Dec 2010 16:33:36 +0000 (16:33 +0000)]
Fix x64 build after r5970, the same way as for ARM.
TBR=fschneider@chromium.org
Review URL: http://codereview.chromium.org/5709005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5973
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
karlklose@chromium.org [Fri, 10 Dec 2010 14:49:24 +0000 (14:49 +0000)]
Fix bug that disabled optimization when profiling.
Review URL: http://codereview.chromium.org/5720003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5972
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Fri, 10 Dec 2010 14:35:40 +0000 (14:35 +0000)]
Fix long line.
Review URL: http://codereview.chromium.org/5680005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5971
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Fri, 10 Dec 2010 14:33:20 +0000 (14:33 +0000)]
Improve our type feedback by recogizining never-executed IC calls for binary operations.
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
vegorov@chromium.org [Fri, 10 Dec 2010 14:25:10 +0000 (14:25 +0000)]
Fix issue 962.
SplitBetween (formely known as Split with 3 arguments) should select split position from [start, end] instead of [start, end[. This should also improve allocation quality (remove certain redundant move patterns).
Also some minor renaming and refactoring to make register allocator code more readable.
BUG=v8:962
TEST=test/mjsunit/regress/regress-962.js
Review URL: http://codereview.chromium.org/5720001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5969
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Fri, 10 Dec 2010 14:11:17 +0000 (14:11 +0000)]
Increase the size of the max object that can be new space allocated.
This is neutral on in-browser SunSpider, but beneficial on other
things, and is likely to lower memory use by collecting earlier.
Review URL: http://codereview.chromium.org/5753003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5968
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Fri, 10 Dec 2010 14:10:54 +0000 (14:10 +0000)]
ARM: Fix heap number allocation in lithium-codegen-arm that assumed
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
lrn@chromium.org [Fri, 10 Dec 2010 13:07:52 +0000 (13:07 +0000)]
Ensure that default value is explicitly initialized in DefineOwnProperty.
TBR: kmillikin
Review URL: http://codereview.chromium.org/5781002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5966
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Fri, 10 Dec 2010 12:58:18 +0000 (12:58 +0000)]
Add gyp target to build preparser as stand-alone library.
Likely only works on Linux yet.
Review URL: http://codereview.chromium.org/5716001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5965
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Fri, 10 Dec 2010 12:33:59 +0000 (12:33 +0000)]
Add missing include directory for shell sample.
BUG=http://code.google.com/p/v8/issues/detail?id=967
TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/5680004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5964
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Fri, 10 Dec 2010 12:12:06 +0000 (12:12 +0000)]
ARM: Fix missing SetCC in crankshaft code emitted when running with
the --debug-code flag.
Review URL: http://codereview.chromium.org/5736003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5963
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Fri, 10 Dec 2010 12:05:28 +0000 (12:05 +0000)]
Change test script to make it easy to run crankshaft tests on ARM and
x64 where crankshaft is not the default. Add ability to add custom
expectations for running in this special crankshaft mode.
The expectations are not updated in this change. There are a couple of
bugs that I would like to fix before doing that. Otherwise the lists
will be very long. :)
Review URL: http://codereview.chromium.org/5787001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5962
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
karlklose@chromium.org [Fri, 10 Dec 2010 12:00:26 +0000 (12:00 +0000)]
Fix a bug that caused the runtime profiler to sample huge amounts of stack frames in programs with recursively called optimized functions.
Review URL: http://codereview.chromium.org/5786001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5961
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Fri, 10 Dec 2010 11:27:15 +0000 (11:27 +0000)]
Fix issue 965.
Review URL: http://codereview.chromium.org/5773002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5960
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Fri, 10 Dec 2010 11:02:09 +0000 (11:02 +0000)]
Fix presubmit
Review URL: http://codereview.chromium.org/5772003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5959
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Fri, 10 Dec 2010 10:02:42 +0000 (10:02 +0000)]
Use correct calling convention for API calls on MinGW
BUG=v8:950
Review URL: http://codereview.chromium.org/5699003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5958
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kmillikin@chromium.org [Thu, 9 Dec 2010 14:09:50 +0000 (14:09 +0000)]
Garbage collection of unused setters in the AST classed.
Review URL: http://codereview.chromium.org/5717001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5957
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Thu, 9 Dec 2010 13:18:23 +0000 (13:18 +0000)]
Fix cast that fails on Win64.
Review URL: http://codereview.chromium.org/5712001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5956
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
karlklose@chromium.org [Thu, 9 Dec 2010 13:12:23 +0000 (13:12 +0000)]
Collect only optimizable function samples.
Keep track of the ratio between JS and non-JS ticks and use this ratio to adjust the lookup threshold. (Also add support to trace compilation statistics.)
Review URL: http://codereview.chromium.org/5633009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5955
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kmillikin@chromium.org [Thu, 9 Dec 2010 12:49:53 +0000 (12:49 +0000)]
Change the HGraphBuilder to dispatch on the context.
Before, expressions didn't take advantage of knowing their context in
the AST. Now, we use the context to decide what to do with a value at
the end of visiting an expression.
Review URL: http://codereview.chromium.org/5620007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5954
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Thu, 9 Dec 2010 12:07:52 +0000 (12:07 +0000)]
Revert 5911 (RegExp fail on invalid range syntax).
Review URL: http://codereview.chromium.org/5703001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5953
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Wed, 8 Dec 2010 18:08:23 +0000 (18:08 +0000)]
Landing for Martyn Capewell.
Fix detection of VFP support on Nexus One.
BUG=none
TEST=none
Code review URL: http://codereview.chromium.org/5664004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5952
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 8 Dec 2010 16:23:25 +0000 (16:23 +0000)]
Speed up quoting of JSON strings by allocating a string that is big enough
and then trimming it when the length is known. This way we only have to
traverse the input once.
Review URL: http://codereview.chromium.org/5556012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5951
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
whesse@chromium.org [Wed, 8 Dec 2010 15:03:08 +0000 (15:03 +0000)]
Fix compilation on ARM when adding Math.pow optimization in 5949.
Review URL: http://codereview.chromium.org/5546006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5950
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
whesse@chromium.org [Wed, 8 Dec 2010 14:32:40 +0000 (14:32 +0000)]
Allow the optimizing code generator to call Math.pow with untagged doubles.
Review URL: http://codereview.chromium.org/5640004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5949
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Wed, 8 Dec 2010 10:47:59 +0000 (10:47 +0000)]
Use the file opened from argv[1] in preparser-process.cc
Review URL: http://codereview.chromium.org/5612006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5948
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Wed, 8 Dec 2010 10:42:32 +0000 (10:42 +0000)]
Remove a unused function form the sample shell
Review URL: http://codereview.chromium.org/5668001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5947
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Wed, 8 Dec 2010 10:06:40 +0000 (10:06 +0000)]
Changed interface to preparser to not require pushback support.
Changed implementation of pushback in preparer character stream.
Removed assert that isn't satisfied by in test-cases, but only by the real code.
Make preparser compile again.
Review URL: http://codereview.chromium.org/5593004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5946
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Wed, 8 Dec 2010 10:05:10 +0000 (10:05 +0000)]
Update the Visual Studio 2005 project files to include the new crankshaft files
Tested With Visual Studio 2008 which converts the files and builds all targets.
Review URL: http://codereview.chromium.org/5660005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5945
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sandholm@chromium.org [Wed, 8 Dec 2010 09:52:48 +0000 (09:52 +0000)]
Fix sputnik regression introduced in r5943.
Review URL: http://codereview.chromium.org/5516013
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5944
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sandholm@chromium.org [Wed, 8 Dec 2010 09:10:36 +0000 (09:10 +0000)]
Use the PushIfAbsent function for the JSON stringify stack.
Optimize ConvertToString.
Review URL: http://codereview.chromium.org/5614004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5943
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 8 Dec 2010 08:31:10 +0000 (08:31 +0000)]
Fix compile error on gcc-4.5 bug 963
Review URL: http://codereview.chromium.org/5642004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5942
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Tue, 7 Dec 2010 15:47:17 +0000 (15:47 +0000)]
Fix arm gyp files, lithium files where not added.
Review URL: http://codereview.chromium.org/5558009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5938
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Tue, 7 Dec 2010 14:51:49 +0000 (14:51 +0000)]
Remove NearestNextGapPos. It is not used anymore.
Remove uses of NearestGapPos in splitting helpers. We can split at any position.
Review URL: http://codereview.chromium.org/5605004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5937
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vitalyr@chromium.org [Tue, 7 Dec 2010 14:10:41 +0000 (14:10 +0000)]
Fix RecordStackTraceAtStartProfiling test.
Review URL: http://codereview.chromium.org/5598008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5936
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Tue, 7 Dec 2010 14:03:59 +0000 (14:03 +0000)]
Optimized scanner to avoid virtual calls for every character read.
Review URL: http://codereview.chromium.org/5545006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5935
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Tue, 7 Dec 2010 13:24:22 +0000 (13:24 +0000)]
Remove log compression support.
This is no longer used in Chromium, and only pollutes code.
BUG=859
Review URL: http://codereview.chromium.org/5575006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5934
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Tue, 7 Dec 2010 12:43:23 +0000 (12:43 +0000)]
Fix ARM sim build.
Review URL: http://codereview.chromium.org/5618005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5931
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Tue, 7 Dec 2010 12:42:37 +0000 (12:42 +0000)]
Disable mozilla test that when timing out prints insane amounts of data.
I have filled a bug for this:
http://code.google.com/p/v8/issues/detail?id=960
Review URL: http://codereview.chromium.org/5662003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5930
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kasperl@chromium.org [Tue, 7 Dec 2010 12:26:36 +0000 (12:26 +0000)]
Prepare push to trunk. Now working on version 3.0.1.
Review URL: http://codereview.chromium.org/5586007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5929
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Tue, 7 Dec 2010 12:21:26 +0000 (12:21 +0000)]
Fix ARM build.
Review URL: http://codereview.chromium.org/5638003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5928
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kmillikin@chromium.org [Tue, 7 Dec 2010 12:07:40 +0000 (12:07 +0000)]
Be more careful about exiting inlined functions in a test context.
When falling off the end of a function inlined in a test context, we cannot
constant fold the test of undefined away. The graph builder assumes that
control flow always reaches both branches of a test.
Instead, explicitly test and branch on "undefined". Introduce a pair of
empty blocks to hold the necessary LeaveInlined instructions.
Review URL: http://codereview.chromium.org/5566005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5927
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Tue, 7 Dec 2010 11:59:50 +0000 (11:59 +0000)]
Disable tests failing due to sockets on arm.
This is a copy of http://codereview.chromium.org/5365005/ against the 3.0 branch.
TBR: eric.corry
Review URL: http://codereview.chromium.org/5624005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5926
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Tue, 7 Dec 2010 11:53:19 +0000 (11:53 +0000)]
Fix Win64 compilation.
Review URL: http://codereview.chromium.org/5597007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5925
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Tue, 7 Dec 2010 11:53:11 +0000 (11:53 +0000)]
Fix no sse3 support by correctly allocating temp register
Review URL: http://codereview.chromium.org/5534004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5924
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Tue, 7 Dec 2010 11:40:36 +0000 (11:40 +0000)]
Fix strict aliasing rule violation in runtime-profiler.cc
Review URL: http://codereview.chromium.org/5621005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5923
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kasperl@chromium.org [Tue, 7 Dec 2010 11:31:57 +0000 (11:31 +0000)]
Update V8 to version 3.0 (re-land r5920).
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5922
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kasperl@chromium.org [Tue, 7 Dec 2010 11:01:02 +0000 (11:01 +0000)]
Revert r5920. Will re-land shortly.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5921
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kasperl@chromium.org [Tue, 7 Dec 2010 09:11:56 +0000 (09:11 +0000)]
Update V8 to version 3.0.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5920
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sandholm@chromium.org [Mon, 6 Dec 2010 15:41:07 +0000 (15:41 +0000)]
Improved JSON stringify.
Review URL: http://codereview.chromium.org/5578004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5919
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sandholm@chromium.org [Mon, 6 Dec 2010 11:44:16 +0000 (11:44 +0000)]
JSON stringify collects substrings in one builder array rather than using regular
string cons.
Review URL: http://codereview.chromium.org/5567005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5918
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Mon, 6 Dec 2010 09:59:08 +0000 (09:59 +0000)]
Improve code generated for AllocInNewSpace. This is a commit of codereview.chromium.org/5512004 for Rodolph Perfetta.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5917
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sandholm@chromium.org [Fri, 3 Dec 2010 11:12:02 +0000 (11:12 +0000)]
Simplify JSON stringify and add special case for default replacer and space.
Review URL: http://codereview.chromium.org/5551002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5916
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Fri, 3 Dec 2010 09:54:06 +0000 (09:54 +0000)]
Irregexp: Preload more characters when we are not at the
start of the input and some alternations in the disjunction
are anchored.
Review URL: http://codereview.chromium.org/5524006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5915
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Thu, 2 Dec 2010 15:38:51 +0000 (15:38 +0000)]
New Heap Profiler: add API method for finding a graph node by id.
TEST=cctest/test-heap-profiler/HeapSnapshotGetNodeById
Review URL: http://codereview.chromium.org/5537001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5914
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Thu, 2 Dec 2010 15:37:45 +0000 (15:37 +0000)]
nGW build
- add missing functions SignalCodeMovingGC() and MemoryBarrier()
- avoid pointer conversion/comparison warnings
- don't attempt to hide symbols with -fvisibility, MinGW doesn't support it
BUG=http://code.google.com/p/v8/issues/detail?id=949
Patch by Bert Belder <bertbelder@gmail.com>
Review URL: http://codereview.chromium.org/5471001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5913
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
whesse@chromium.org [Thu, 2 Dec 2010 11:20:44 +0000 (11:20 +0000)]
Add generated code to calculate Math.log and to search Transcendental cache for logs. Implemented on all platforms.
Review URL: http://codereview.chromium.org/5437002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5912
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Thu, 2 Dec 2010 08:02:37 +0000 (08:02 +0000)]
Change RegExp syntax to fail on invalid ranges like [\d-x], [x-\d] and [\d-\d].
The previous behavior was to treat the "-" as verbatim if the range was invalid.
This change matches the JSC changeset http://trac.webkit.org/changeset/72813/
Review URL: http://codereview.chromium.org/5464001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5911
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Wed, 1 Dec 2010 13:11:28 +0000 (13:11 +0000)]
Fix compile problem on ARM. Remove unused argument.
Review URL: http://codereview.chromium.org/5455001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5910
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Wed, 1 Dec 2010 10:04:34 +0000 (10:04 +0000)]
Move quoting of a JSON string to a specialized runtime function.
Previously used string replace regexp with function replacement.
Review URL: http://codereview.chromium.org/5443001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5909
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vitalyr@chromium.org [Tue, 30 Nov 2010 13:17:36 +0000 (13:17 +0000)]
Save full source position state to avoid forced positions.
Review URL: http://codereview.chromium.org/5277008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5908
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sandholm@chromium.org [Tue, 30 Nov 2010 13:16:36 +0000 (13:16 +0000)]
Maintain the invariant that the pattern string in an atom regexp is always
a flat non-cons string.
Review URL: http://codereview.chromium.org/5270006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5907
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vitalyr@chromium.org [Tue, 30 Nov 2010 11:44:51 +0000 (11:44 +0000)]
Fix mac build.
Review URL: http://codereview.chromium.org/5333007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5906
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vitalyr@chromium.org [Tue, 30 Nov 2010 10:55:24 +0000 (10:55 +0000)]
RelocInfo: fix source position decoding.
We used to rely on reading both POSITION and STATEMENT_POSITION to get
correct decoding of positions. This was error prone and made liveedit
unhappy.
Review URL: http://codereview.chromium.org/5277007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5905
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Mon, 29 Nov 2010 16:38:05 +0000 (16:38 +0000)]
Fix windows build.
TBR=lrn
Review URL: http://codereview.chromium.org/5367007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5904
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Mon, 29 Nov 2010 14:28:06 +0000 (14:28 +0000)]
Revert seeding the random number generator with rand_s on Windows. It
makes browser_tests fail.
TBR=jschuh@chromium.org
Review URL: http://codereview.chromium.org/5284006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5900
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Mon, 29 Nov 2010 13:24:37 +0000 (13:24 +0000)]
Preparser extracted into separate files that can be compiled to a library.
No scons target yet.
Review URL: http://codereview.chromium.org/5295004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5899
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Mon, 29 Nov 2010 07:47:34 +0000 (07:47 +0000)]
Prepare push to trunk. Now working on version 2.6.0.
Review URL: http://codereview.chromium.org/5381004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5895
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Fri, 26 Nov 2010 12:45:41 +0000 (12:45 +0000)]
Fix number parsing to not allow space between sign and digits.
Affects both parseFloat and ToNumber conversion.
Fix issue 946.
Review URL: http://codereview.chromium.org/5338005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5894
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Fri, 26 Nov 2010 11:48:35 +0000 (11:48 +0000)]
make DateParser::TimeComposer handle 1-2 digits millisecond values
see http://code.google.com/p/v8/issues/detail?id=944
This patch makes DateParser::TimeComposer process times that have
millisecond values with only 1 or 2 digits.
Without this patch, Date.parse("2010-11-25T22:02:30.5") returns
1290690150005 and
Date.parse("2010-11-25T22:02:30.5") == Date.parse("2010-11-25T22:02:30.005")
evaluates to true.
With this patch, Date.parse("2010-11-25T22:02:30.5") returns
1290690150500 instead, and
Date.parse("2010-11-25T22:02:30.5") == Date.parse("2010-11-25T22:02:30.005")
evaluates to false.
Review URL: http://codereview.chromium.org/5336005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5893
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Fri, 26 Nov 2010 08:43:34 +0000 (08:43 +0000)]
Landing for Martyn Capewell.
Implement string constructor stub on ARM.
BUG=none
TEST=none
Codereview URL: http://codereview.chromium.org/5322009/
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5892
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Thu, 25 Nov 2010 15:54:52 +0000 (15:54 +0000)]
Simplify ProfLazyMode test on Linux.
Instead of installing signal handler, count samples taken.
Review URL: http://codereview.chromium.org/5325003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5891
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yurys@chromium.org [Thu, 25 Nov 2010 08:04:12 +0000 (08:04 +0000)]
Expose a method for getting JSObject constructor name
Review URL: http://codereview.chromium.org/5256004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5890
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Thu, 25 Nov 2010 07:39:17 +0000 (07:39 +0000)]
Landing for Justin Schuh.
Seed the random number generator in Windows with rand_s
This is a quick fix for m9. It works on Windows Chrome because the random device is already initialized before permissions are dropped for the Chrome sandbox. The same trick isn't possible on Linux or Mac.
I think the long-term solution is to provide an interface for supplying v8 with a true random number generator. Then Chrome can just hook up the generator from base/rand_util.h
BUG=http://code.google.com/p/v8/issues/detail?id=936
TEST=None.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5889
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Wed, 24 Nov 2010 10:47:18 +0000 (10:47 +0000)]
New Heap profiler: fix JSON serialization of aggregated profiles.
Serialization was failing due to unset dominator pointers.
TEST=test-heap-snapshot/AggregatedHeapSnapshotJSONSerialization
Review URL: http://codereview.chromium.org/5314003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5888
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Wed, 24 Nov 2010 10:28:22 +0000 (10:28 +0000)]
Fix break of build.
Mental note: When you make a fix, remember to save it before committing.
Review URL: http://codereview.chromium.org/5330005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5887
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Wed, 24 Nov 2010 09:57:06 +0000 (09:57 +0000)]
Working stand-alone preparser.
BUG=
TEST=
Review URL: http://codereview.chromium.org/5302003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5885
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 24 Nov 2010 09:55:58 +0000 (09:55 +0000)]
ARM: Defer the prefix/postfix code generation. This is a fixed
version of http://codereview.chromium.org/3666001/ by ZhangK with
an added call to ForgetTypeInfo to fix a debug mode assert.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5884
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Wed, 24 Nov 2010 09:40:58 +0000 (09:40 +0000)]
Make randomized allocations along 64k granularity boundaries to avoid comitting unused memory.
BUG=56036
TEST=None.
Patch by Justin Schuh <jschuh@chromium.org>
Review URL: http://codereview.chromium.org/3849004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5883
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Wed, 24 Nov 2010 09:21:29 +0000 (09:21 +0000)]
Fix the process sample to actually dispose the contexts used for
processors.
Review URL: http://codereview.chromium.org/5302004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5882
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Wed, 24 Nov 2010 06:35:35 +0000 (06:35 +0000)]
Delete empty directory.
TBR=ricow@chromium.org
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5880
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Wed, 24 Nov 2010 06:26:36 +0000 (06:26 +0000)]
Fix crashes during GC caused by partially initialized objects. The
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
lrn@chromium.org [Tue, 23 Nov 2010 12:11:07 +0000 (12:11 +0000)]
Added files to gyp-config.
Review URL: http://codereview.chromium.org/5276004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5878
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Tue, 23 Nov 2010 11:46:36 +0000 (11:46 +0000)]
Untemplated preparser.h and made it depend on virtual types.
Extracted preparse-data specification and logging classes.
Review URL: http://codereview.chromium.org/5166006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5877
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Tue, 23 Nov 2010 10:07:02 +0000 (10:07 +0000)]
Prepare push to trunk. Now working on version 2.5.9.
Review URL: http://codereview.chromium.org/5302002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5873
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Tue, 23 Nov 2010 09:52:52 +0000 (09:52 +0000)]
Fix again HeapEntry size problem, now platform-independent way.
Rico noticed that V8 ARM builder also fails on HeapEntry size
assertion. As MSVC-specific way of fixing the problem causes
aliasing problems on G++, I re-implemented conversion using
unions. And #ifdefs are gone!
TBR=sgjesse@chromium.org
Review URL: http://codereview.chromium.org/5328001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5872
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Mon, 22 Nov 2010 16:09:14 +0000 (16:09 +0000)]
Fix windows compilation after r5867. Now linux-friendly.
TBR=sgjesse@chromium.org
Review URL: http://codereview.chromium.org/5216008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5871
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Mon, 22 Nov 2010 15:38:12 +0000 (15:38 +0000)]
Revert "Fix compilation on Win after r5867.". This breaks Linux.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5870
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Mon, 22 Nov 2010 15:31:43 +0000 (15:31 +0000)]
Fix compilation on Win after r5867.
TBR=sgjesse@chromium.org
Review URL: http://codereview.chromium.org/5242003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5869
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
floitschV8@gmail.com [Mon, 22 Nov 2010 14:42:07 +0000 (14:42 +0000)]
Remove Gay's dtoa from sources.
Farewell.
Review URL: http://codereview.chromium.org/5195003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5868
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Mon, 22 Nov 2010 14:00:40 +0000 (14:00 +0000)]
New heap profiler: implement fast retaining sizes approximation.
Approximation is done by building a dominators tree for the heap graph.
Dominator nodes and retained sizes are serialized into JSON.
Removed:
- reachable size (it is useless, after all);
- HeapEntryCalculatedData (size is now stored in the node, retaining
paths in a hash map);
Review URL: http://codereview.chromium.org/5154007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5867
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Mon, 22 Nov 2010 09:57:21 +0000 (09:57 +0000)]
Force pretenuring of closures that are immediately assigned to
properties. For these closures we would like to be able to use
constant functions and for that we need the closures allocated in old
space.
Review URL: http://codereview.chromium.org/5220007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5866
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
lrn@chromium.org [Fri, 19 Nov 2010 12:36:39 +0000 (12:36 +0000)]
Made some functions virtual that were previously hit by templating.
Currently means that preparsing doesn't record/export symbol data.
Review URL: http://codereview.chromium.org/5122007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5865
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Fri, 19 Nov 2010 12:08:52 +0000 (12:08 +0000)]
Add more tests of breaks in infinite loops
Move stack check in do while loops to before the continue target to enable breaks even if continue was always used in the loop.
Review URL: http://codereview.chromium.org/5184007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5862
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ager@chromium.org [Fri, 19 Nov 2010 11:24:51 +0000 (11:24 +0000)]
Remove unused variable to please Clang.
Review URL: http://codereview.chromium.org/5151010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5861
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00