Commit Graph

9642 Commits

Author SHA1 Message Date
mstarzinger@chromium.org
81a43b5607 Make incremental marking clear ICs.
This extends the existing clearing of ICs during GC to incremental
marking in order to prevent cross-context retention that would last
until the next non-incremental GC.

R=erik.corry@gmail.com
TEST=cctest/test-heap/IncrementalMarkingClears[Mono,Poly]morhpicIC

Review URL: https://chromiumcodereview.appspot.com/10831123

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12259 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-06 10:35:48 +00:00
yangguo@chromium.org
e0954ca592 Take advantage of batched results when matching global regexp.
BUG=
TEST=regexp-global.js

Review URL: https://chromiumcodereview.appspot.com/10831126

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12258 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-06 09:06:27 +00:00
yangguo@chromium.org
4e82c3fb1a Allow SetHiddenValue to accept empty value.
R=mstarzinger@chromium.org
BUG=v8:2274

Review URL: https://chromiumcodereview.appspot.com/10825196

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12257 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-06 07:55:05 +00:00
yangguo@chromium.org
0a0b4f4f2f Fix crash bug when calling getV8Statistics().
BUG=v8:2270

Review URL: https://chromiumcodereview.appspot.com/10830160

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12255 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-03 14:03:30 +00:00
svenpanne@chromium.org
83fc4205f6 Fixed compound/count operations with getter-only accessor properties.
The underlying problem is that for compound/count operations we use the *load*
type feedback for storing, too. For normal properties this doesn't matter, but
for accessor properties we should better use the *store* type feedback, which
would be available, too. This consistent feedback usage could be guaranteed if
we removed the heavy copy-n-paste in the crankshaft code generation for
compound/count operations and assignments/property loads.

To be on the safe side, we postpone this refactoring and do a quick and easily
mergeable fix.

BUG=140083
TEST=mjsunit/regress/regress-crbug-140083.js

Review URL: https://chromiumcodereview.appspot.com/10828146

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12252 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-03 09:45:08 +00:00
mstarzinger@chromium.org
680c4a7a13 Make test-heap resilient against --stress-compaction.
R=ulan@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10843052

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12251 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-02 14:20:38 +00:00
svenpanne@chromium.org
930aaf4267 android: refine custom ucontext_t definitions.
This is a forward-compatible change to avoid type/naming
conflicts when the Android platform/NDK will update its
<signal.h> header to properly define 'struct sigcontext',
'mcontext_t' and 'ucontext_t'.

In particular:

  - Do not define 'struct sigcontext.h' to avoid
    conflicts with the C library definition (which
    is different, see below).

  - Only provide custom ucontext_t declarations if
    the Android <signal.h> doesn't provide it. This can
    be tested with a macro check (__BIONIC_HAVE_UCONTEXT_T)

+ Use 'gettid()' on Android since it is available (at all
  API levels).

See http://code.google.com/p/android/issues/detail?id=34784

Review URL: https://chromiumcodereview.appspot.com/10829122
Patch from David Turner <digit@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12250 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-02 10:03:21 +00:00
ulan@chromium.org
c6d4094e44 Android: increase default test timeout and skip time sensitive mjsunit/regress/regress-1969.
R=yangguo@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10836049

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12249 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 13:06:08 +00:00
mstarzinger@chromium.org
ec11473354 When specifying a new create histogram callback, reset all histograms so they recreate themselves on next use.
We only do this for histograms, and not for counters, as counters might be used directly from generated code

BUG=none
TEST=none

Review URL: https://chromiumcodereview.appspot.com/10828113
Patch from Jochen Eisinger <jochen@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12248 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 12:51:24 +00:00
erik.corry@gmail.com
5c6db6d081 Fix the 137002 fix (Don't generate ICs for accessors on slow
case objects).  We should be testing the holder for dictionary
mode, not the receiver.
Review URL: https://chromiumcodereview.appspot.com/10827113

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12247 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 12:35:33 +00:00
mstarzinger@chromium.org
d4e6aa33fc Refactor marking to share embedded pointer visitor.
R=ulan@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10826079

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12246 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 11:29:48 +00:00
mstarzinger@chromium.org
7dd2e9777c Make incremental marking clear type feedback cells.
This extends the existing clearing of type feedback cells during GC to
incremental marking in order to prevent cross-context retention that
would last until the next non-incremental GC.

R=erik.corry@gmail.com
TEST=cctest/test-heap/IncrementalMarkingClearsTypeFeedbackCells

Review URL: https://chromiumcodereview.appspot.com/10823082

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12245 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 11:24:14 +00:00
jkummerow@chromium.org
2b786204ee Prepare push to trunk. Now working on version 3.13.0.
R=yangguo@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10824125

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12242 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 11:10:32 +00:00
verwaest@chromium.org
bc86ec4d5f Don't leak inobject space when transforming to fast properties without descriptors.
Review URL: https://chromiumcodereview.appspot.com/10828112

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12241 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 10:50:26 +00:00
mstarzinger@chromium.org
3bb994c4e4 Make test-serialize resilient against --gc-interval.
R=erik.corry@gmail.com
TEST=cctest/test-serialize (--gc-interval=100)

Review URL: https://chromiumcodereview.appspot.com/10829097

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12240 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-08-01 09:43:05 +00:00
erik.corry@gmail.com
1d0f872ef9 Fix full code generator to not use --debug-code if it is in
mksnapshot or a VM that is booted from a snapshot.  --debug-code
can still have an effect on stub and optimized code and it still
works on the full code generator when running without snapshots.

The deoptimizer generates full-code-generator code and relies on it having
the same layout as last time.  This means that the code the full code
generator makes for the snapshot should be the same as the code it makes
later.  This change makes the full code generator create more consistent
code between mksnapshot time and run time.

This is a bug fix and a step towards making the snapshot code more robust.
Review URL: https://chromiumcodereview.appspot.com/10834085

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12239 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 14:59:32 +00:00
danno@chromium.org
cd95464dbc Fix write barrier for StoreKeyedFastElements on ARM
R=mstarzinger@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10824107

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12238 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 14:31:19 +00:00
erik.corry@gmail.com
b3e2440580 Speed up quicksort test to avoid timeouts on simulators.
Review URL: https://chromiumcodereview.appspot.com/10830093

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12237 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 13:36:24 +00:00
verwaest@chromium.org
ec2b5ab57c Use architecture-specific alignment for hashing in Descriptor lookup cache.
Review URL: https://chromiumcodereview.appspot.com/10825099

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12236 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 12:33:28 +00:00
mstarzinger@chromium.org
bf0322a861 Fix build for disassembler=on and objectprint=off.
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10830088

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12235 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 09:53:39 +00:00
danno@chromium.org
d884792dea Fix bogus AbortIfNotZeroExtended on x64 for bounds checks
R=mmassi@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10837019

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12234 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 09:47:43 +00:00
erik.corry@gmail.com
8d96a0190c Fix deserializer to understand direct pointers from code to cell payloads.
The deoptimizer generates full-code-generator code and relies on it having
the same layout as last time.  This means that the code the full code
generator makes for the snapshot should be the same as the code it makes
later.  This change makes the full code generator create more consistent
code between mksnapshot time and run time.

This is a bug fix and a step towards making the snapshot code more robust.
Review URL: https://chromiumcodereview.appspot.com/10824084

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12233 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 09:25:23 +00:00
danno@chromium.org
9641c25573 Improve constant element index access code generation
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10831049

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12232 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-31 08:44:51 +00:00
yangguo@chromium.org
1094c8f17c Fix regression test on x64.
R=verwaest@chromium.org
BUG=
TEST=

Review URL: https://chromiumcodereview.appspot.com/10827084

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12231 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 13:48:52 +00:00
yangguo@chromium.org
514fc74a47 Limit initial size of hidden properties and store identity hashes inline.
BUG=v8:2211
TEST=test-heap/Regress2211

Review URL: https://chromiumcodereview.appspot.com/10827040

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12230 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 13:09:09 +00:00
mstarzinger@chromium.org
249f29f22b Make AlwaysAllocateScope imply DisallowAllocationFailure.
R=erik.corry@gmail.com
TEST=cctest/test-api,cctest/test-heap (--gc-interval=100)

Review URL: https://chromiumcodereview.appspot.com/10824082

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12229 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 13:07:29 +00:00
yangguo@chromium.org
bfd51f29dc R=yangguo@chromium.org,svenpanne@chromium.org
x64 BinaryOpStub::GenerateSmiCode use wrong registers in one code path.

e.g. d8 --trace_ic
var dd = new Float64Array(2);
dd[1] = 1;
(function fn() {
  for (var i = 0; i < 1000; i++)
  dd[0] = 2 / dd[1];
})();
It keeps falling into runtime call and patching.

No regression test, because I don't how to test it.

Review URL: https://chromiumcodereview.appspot.com/10834064
Patch from Zheng Liu <shdwthr@gmail.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12226 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 13:04:10 +00:00
ulan@chromium.org
16f9ce7c2f Prepare push to trunk. Now working on version 3.12.19.
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10825076

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12225 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 13:00:52 +00:00
mstarzinger@chromium.org
76fe67b092 Implement verification of context separation.
This adds the --verify-global-context-separation flag which can be used
to verify that no code object embeds pointers to more than one global
context after a full GC. It uses an object visitor that just performs
shallow traversal of the object graph spanned by one code object, and
breaks at points where application objects are encountered. So it will
not trip on cross-context leaks introduced by the application itself.

R=verwaest@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10830049

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12224 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 10:47:04 +00:00
svenpanne@chromium.org
cb4840c0e5 Inline simple getter calls.
Currently only simple getter calls are handled (i.e. no calls in count
operations or compound assignments), and deoptimization in the getter is not
handled at all. Because of the latter, we temporarily hide this feature behind a
new flag --inline-accessors, which is false by default.

Review URL: https://chromiumcodereview.appspot.com/10828066

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 10:42:21 +00:00
mstarzinger@chromium.org
fdb22ea2ed Implement TracePathToObjectFrom helper for debugging.
R=verwaest@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10823054

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12222 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 10:32:54 +00:00
yangguo@chromium.org
d078747ac0 Actually fix build.
R=svenpanne@chromium.org
BUG=
TEST=

Review URL: https://chromiumcodereview.appspot.com/10825074

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12221 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 09:55:26 +00:00
yangguo@chromium.org
e622892fac Fix build.
R=jkummerow@chromium.org
BUG=
TEST=

Review URL: https://chromiumcodereview.appspot.com/10829075

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12220 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 09:36:03 +00:00
yangguo@chromium.org
21871da998 Force signed zero double.
R=svenpanne@chromium.org
BUG=v8:2239
TEST=

Review URL: https://chromiumcodereview.appspot.com/10821088

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12219 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 09:28:20 +00:00
jkummerow@chromium.org
0f7288aa21 Remove unused fields in test.py
Review URL: https://chromiumcodereview.appspot.com/10833002
Patch from Haitao Feng <haitao.feng@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12218 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 09:09:13 +00:00
yangguo@chromium.org
cfc5cf0b76 Fix build errors.
R=mstarzinger@chromium.org
BUG=
TEST=

Review URL: https://chromiumcodereview.appspot.com/10832061

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12217 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 08:54:51 +00:00
yangguo@chromium.org
0acc9e13af Make sure double to int conversion is correct.
R=svenpanne@chromium.org
BUG=v8:2260
TEST=test-utils/Utils1

Review URL: https://chromiumcodereview.appspot.com/10820047

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12216 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-30 08:34:26 +00:00
verwaest@chromium.org
a448d42428 Removed recording of Target Slot in transition arrays since they only contain maps.
Review URL: https://chromiumcodereview.appspot.com/10802048

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12215 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 17:03:12 +00:00
mstarzinger@chromium.org
cf6d80e154 Fix minor handle unsafety in exception throwing code.
R=yangguo@chromium.org
TEST=mjsunit/harmony/proxies (--gc-interval=100)

Review URL: https://chromiumcodereview.appspot.com/10827072

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12214 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 15:33:06 +00:00
svenpanne@chromium.org
7fd21aa7a1 Move some Pop/Drop calls directly to the places where they are needed.
This is a refactoring-only CL and the fourth one in a series for enabling
inlining of accessors. Later when we try to inline accessor calls, their
arguments must still be on the expression stack, so we must not remove them too
early.

Review URL: https://chromiumcodereview.appspot.com/10828049

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12213 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 12:35:44 +00:00
mstarzinger@chromium.org
bff5cc9e94 Print type feedback info together with code objects.
R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10833049

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12212 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 12:06:26 +00:00
yangguo@chromium.org
49400e22b6 Rewrite Makefile rules for Android to allow parallel execution
BUG=v8:2257

Review URL: https://chromiumcodereview.appspot.com/10824039
Patch from Haitao Feng <haitao.feng@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12211 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 11:14:44 +00:00
yangguo@chromium.org
b58cfd485e Avoid GC when printing shared function info.
R=mstarzinger@chromium.org
BUG=
TEST=

Review URL: https://chromiumcodereview.appspot.com/10828048

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12210 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 09:54:56 +00:00
svenpanne@chromium.org
0923caccee Move BuildCallGetter/BuildCallSetter up in the call chain.
This is a refactoring-only CL and the third one in a series for enabling
inlining of accessors. The goal of this CL is to move the builders for accessors
to the places where we might be able to inline them later, i.e. the VisitFoo and
HandleBar member functions of HGraphBuilder.

Extracted duplicate code into LookupAccessorPair.

Review URL: https://chromiumcodereview.appspot.com/10831047

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12209 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 09:43:06 +00:00
verwaest@chromium.org
ce23fe9721 Prepare push to trunk. Now working on version 3.12.18.
R=danno@chromium.org

Review URL: https://chromiumcodereview.appspot.com/10829055

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12206 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 07:57:43 +00:00
svenpanne@chromium.org
eddf5edb62 Cleaned up BuildLoadNamed/BuildStoreNamed.
This is a refactoring-only CL and the second one in a series for enabling
inlining of accessors. It makes BuildLoadNamed and BuildStoreNamed a bit more
uniform and makes it clear that they both handle monomorphic accesses only.

Review URL: https://chromiumcodereview.appspot.com/10820028

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12205 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-27 06:27:47 +00:00
verwaest@chromium.org
bcc175ad43 Extending map verification to also verify LastAdded/DescriptorArray consistency.
Removed ASSERT now subsumed by extended map verification.

Review URL: https://chromiumcodereview.appspot.com/10827037

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12204 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-26 14:55:22 +00:00
mmassi@chromium.org
3eb1df6580 Re-enable array index dehoisting.
BUG=
TEST=

Review URL: https://chromiumcodereview.appspot.com/10826031

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12203 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-26 14:21:40 +00:00
verwaest@chromium.org
1c7aa4d172 Set LastAdded to kNoneAdded in RawCopy.
This ensures it is properly initialized if no descriptors are later set.

Review URL: https://chromiumcodereview.appspot.com/10833033

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12202 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-26 13:48:34 +00:00
jkummerow@chromium.org
80c35c6522 Always set the callee's context when calling a function from optimized code.
This is necessary even for recursive calls because we're sharing optimized code among closures, which could call each other and have distinct contexts.

BUG=138887
TEST=mjsunit/regress/regress-crbug-138887

Review URL: https://chromiumcodereview.appspot.com/10834031

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@12201 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2012-07-26 12:49:08 +00:00