Commit Graph

64 Commits

Author SHA1 Message Date
erik.corry@gmail.com
280af55178 Cosmetic changes to the ARM port.
Review URL: http://codereview.chromium.org/1790002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4492 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-04-26 11:41:39 +00:00
ager@chromium.org
c678e44805 Add missing smi check in IC for nonexistent properties.
Add regression test.

BUG=v8:681
Review URL: http://codereview.chromium.org/1673005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4446 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-04-20 10:20:39 +00:00
ager@chromium.org
afc15bb48f Reapply load ICs for nonexistent properties.
We need to be careful to check global property cells for the property
encountered during lookup.  Therefore, the ICs have to be specific to
the name of the property if global objects are involved.  In
principle, this means that we could get a large number of monomorphic
ICs for the same map if there is a global object in the prototype
chain.  However, since this is only done for normal load ICs and not
for keyed load ICs I do not expect this to be a problem.  I will
experiment with it once this goes in.

BUG=675
Review URL: http://codereview.chromium.org/1559033

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4426 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-04-15 11:25:41 +00:00
whesse@chromium.org
518481d5e6 Revert change 4401.
Review URL: http://codereview.chromium.org/1589035

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4416 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-04-14 11:45:03 +00:00
ager@chromium.org
aa8818560d Add inline caches for loading non-existing properties.
Review URL: http://codereview.chromium.org/1539034

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4401 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-04-14 06:25:11 +00:00
vitalyr@chromium.org
6fbad6e230 Simplify passing of AccessorInfo to interceptors:
* Use slots on the native stack when possible instead of Relocatable.
 * Got rid of a gap in AccessorInfo fields.
 * Added test for non-cacheable post-interceptor lookup.

Review URL: http://codereview.chromium.org/1327002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4288 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-25 17:08:22 +00:00
erik.corry@gmail.com
3672624b45 Fix pop push optimization to work with partial snapshots (correct
registration of external references in Proxy objects).
I moved the declaration of the two functions to stub-cache.h
because with all the types they use it's hard to declare them
anywhere else.  But the actual definition is still in runtime.cc
near to the place where they are used.
Review URL: http://codereview.chromium.org/1079012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4231 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-23 14:33:42 +00:00
kasperl@chromium.org
7d6a3b433f Merge the partial_snapshots branch back into bleeding_edge. For
now, the custom call generator stuff is disabled.
Review URL: http://codereview.chromium.org/1094014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4217 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-23 11:40:38 +00:00
antonm@chromium.org
7f1eda50cc Implement a custom call compiler for Array.pop.
Review URL: http://codereview.chromium.org/870007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4118 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-12 13:45:31 +00:00
antonm@chromium.org
de070ccfa6 First take on custom call generators.
Review URL: http://codereview.chromium.org/669061

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4108 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-11 16:24:31 +00:00
antonm@chromium.org
21af2c360a Reland 4068 which wrapped Array methods into native builtins (reverted in 4075).
Review URL: http://codereview.chromium.org/857002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4104 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-11 13:13:21 +00:00
ager@chromium.org
5fd33c7527 Revert change to always call builtins for Array functions.
It causes regressions on test shell tests.

TBR=antonm@chromium.org
Review URL: http://codereview.chromium.org/819001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4075 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-10 08:45:00 +00:00
antonm@chromium.org
52ec954dd2 Always invoke C++ ArrayPush builtin.
Now this builtin checks if it should go into fast case or resort to JS ArrayPush builtin.

Review URL: http://codereview.chromium.org/660298

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4068 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-09 15:43:04 +00:00
serya@chromium.org
43d074985a Review URL: http://codereview.chromium.org/652118
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3937 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-24 08:33:51 +00:00
ager@chromium.org
806c15e4b5 Remove the LookupResult IsValid method because it is confusing.
Replaced IsValid by IsPropertyOrTransition and used IsProperty in most
of the places where IsValid was used before.  Most of the time when
inspecting a lookup result we really want to know if there is a real
property present.  Only for stores are we interested in transitions.

BUG=http://crbug.com/20104
TEST=cctest/test-api/NamedInterceptorMapTransitionRead
Review URL: http://codereview.chromium.org/647015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3901 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-18 15:10:35 +00:00
kmillikin@chromium.org
a2bb215611 Change the interface of CallStubCompiler::CompileCallInterceptor.
Move responsibility for a typecast into the caller so that it can be
locally verified to succeed.

Review URL: http://codereview.chromium.org/594055

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3858 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-15 13:20:49 +00:00
antonm@chromium.org
3d4d5e9e34 Refactoring various pieces of post interceptor caching.
Review URL: http://codereview.chromium.org/596096

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3857 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-15 12:32:27 +00:00
whesse@chromium.org
58b04396bf Change CallIC interface on ARM. Remove name from the stack, and pass it in register r2.
Review URL: http://codereview.chromium.org/598065

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3856 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-15 12:26:07 +00:00
whesse@chromium.org
4507ca1523 Simplify compilation of CallIC interceptor stub.
Review URL: http://codereview.chromium.org/604054

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3854 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-15 11:52:18 +00:00
whesse@chromium.org
077d785842 Fix CallIC stub for CallInterceptor on ARM
Review URL: http://codereview.chromium.org/593080

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3848 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-12 16:01:38 +00:00
whesse@chromium.org
9344612bef Change StoreIC interface on ARM platform
Review URL: http://codereview.chromium.org/554152

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3830 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-10 15:06:08 +00:00
vitalyr@chromium.org
d1d56d9891 ia32: Fuse map and type checks in call ICs for API functions.
This uses the fact that if a map stayed the same then the object
still passes the type check. A new builtin is added to handle the
API call in this case.

Review URL: http://codereview.chromium.org/573003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3825 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-09 16:14:14 +00:00
whesse@chromium.org
ad83e411e1 Eliminate unused function KeyedStoreIC::GenerateExtendStorage, and unused parameter from StubCompiler::GenerateStoreField.
Review URL: http://codereview.chromium.org/575022

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3824 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-09 15:18:37 +00:00
whesse@chromium.org
c00dcbe5f2 Fix error introduced in version 3820. Add optimization suggested by the error.
Review URL: http://codereview.chromium.org/597006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3823 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-09 14:19:23 +00:00
whesse@chromium.org
503bbeb8c4 Change GenerateLoadStringLength to have two scratch registers on all platforms. This is needed so that the LoadIC can pass the receiver in a register, and not have it overwritten, and also simplifies the code.
Review URL: http://codereview.chromium.org/596011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3820 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-09 13:06:37 +00:00
antonm@chromium.org
3ec70cab81 Port caching of lookup followups for interceptors to ARM
Review URL: http://codereview.chromium.org/551191

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3766 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-02-02 11:32:17 +00:00
whesse@chromium.org
e9263697cb Remove some unneeded restorations of a register in StoreIC.
Review URL: http://codereview.chromium.org/548181

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3737 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-01-28 14:01:14 +00:00
sgjesse@chromium.org
562f90d375 Ensure correct boxing of values when calling functions on them
When a function is called with a value type as the receiver this is now boxed as an object.

This is a low-impact solution where the receiver is only boxed when required. For IC calls to the V8 builtins values are not boxed and as most of the functions on String.prototype, Number.prototype and Boolean.prototype are sitting there most IC calls on values will not need any boxing of the receiver.

For calls which are not IC calls but calls through the CallFunctionStub a flag is used to determine whether the receiver might be a value and only when that is the case will the receiver be boxed.

No changtes to Function.call and Function.apply - they already boxed values. According to the ES5 spec the receiver should not be boxed for these functions, but current browsers have not adopted that change yet.

BUG=223
TEST=test/mjsunit/value-wrapper.js
TEST=test/mjsunit/regress/regress-crbug-3184.js
Review URL: http://codereview.chromium.org/542087

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3617 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-01-15 13:42:32 +00:00
kasperl@chromium.org
51aa605e62 Fix ARM and x64 tests in debug mode after r3477.
TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/500090

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3480 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-12-17 07:35:12 +00:00
kmillikin@chromium.org
1d57c13b8c Fix arm and x64 build.
The ApiGetterEntryStub is not implemented on arm or x64, but the
platform-specific implementations need to change to match the change
to the platform-independent signatures.

TBR=ager@chromium.org

Review URL: http://codereview.chromium.org/471009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3441 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-12-09 15:34:57 +00:00
ager@chromium.org
0ee41eec90 Fix megamorphic load regression on ARM caused by the string
length/hash change.
Review URL: http://codereview.chromium.org/454018

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3388 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-12-01 09:54:47 +00:00
sgjesse@chromium.org
eb4d261e24 Remove the different length string types
The different length string types was used to encode the string length and the hash in one field. This is now split into two fields one for length and one for hash. The hash field still encodes the array index of the string if it has one. If an array index is encoded in the hash field the string length is added to the top bits of the hash field to avoid a hash value of zero.

On 32-bit this causes an additional 4 bytes to be used for all string objects. On 64-bit this will be half on average dur to pointer alignment.
Review URL: http://codereview.chromium.org/436001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3350 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-11-24 14:10:06 +00:00
sgjesse@chromium.org
b1a5bf4c5d Change the functions named AllocateObjectInNewSpace to AllocateInNewSpace.
These functions just moves forward the new space allocation top, and in some situations they are used to reserve space for several objects.
Review URL: http://codereview.chromium.org/219030

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2993 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-09-30 09:24:46 +00:00
lrn@chromium.org
8f60f1a46f Win64 - Allow returning two values from a runtime function.
(Not yet fully functional)

Review URL: http://codereview.chromium.org/201042


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2845 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-09-08 11:52:05 +00:00
sgjesse@chromium.org
b8b7832a61 Last step in allocating objects in generated code on ARM.
Port specialized constructor code for constructing simple objects to ARM.

See http://codereview.chromium.org/173470 for description.
Review URL: http://codereview.chromium.org/196024

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2834 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-09-07 10:31:02 +00:00
sgjesse@chromium.org
6cdd89b165 hange bad choice for scratch register for native code counters on ARM
This caused running with native code counters on ARM to crash.
Review URL: http://codereview.chromium.org/181031



git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2792 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-09-01 11:26:48 +00:00
sgjesse@chromium.org
717394492d Generate specialized constructor code for constructing simple objects.
For objects which only have simple assignments of the form this.x = ...; a
specialized constructor stub is now generated. This generated code allocates the
object and fills in the initial properties directly. If this fails for some
reason code continues in the generic constructor stub which in turn might pass
control to the runtime system.

Added counter to see how many objects are constructed using a specialized stub.

The specialized stub is only implemented for ia32 architecture in this change.
For x64 and ARM the generic construct stub is used.

This is change is identical to http://codereview.chromium.org/174392 (committed in r2753 and reverted in r2754) except that a few parts have already been committed from http://codereview.chromium.org/173469 (committed in r2762).
Review URL: http://codereview.chromium.org/173470

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2764 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-08-26 12:22:44 +00:00
sgjesse@chromium.org
3fdbc0848c Reverting 2753.
Review URL: http://codereview.chromium.org/173349

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2754 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-08-25 15:00:07 +00:00
sgjesse@chromium.org
53531ed48a Generate specialized constructor code for constructing simple objects.
For objects which only have simple assignments of the form this.x = ...; a specialized constructor stub is now generated. This generated code allocates the object and fills in the initial properties directly. If this fails for some reason code continues in the generic constructor stub which in turn might pass control to the runtime system.

Added counter to see how many objects are constructed using a specialized stub.

The specialized stub is only implemented for ia32 architecture in this change. For x64 and ARM the generic construct stub is used.
Review URL: http://codereview.chromium.org/174392

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2753 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-08-25 12:23:58 +00:00
erik.corry@gmail.com
038f9074bc Reapply ARM root array change to reduce const pool size
by using r10 to point to a list of common root objects.
This time we also disable a debugger test on ARM that
has never worked, but with this change sometimes
crashes.
Review URL: http://codereview.chromium.org/174317

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2744 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-08-24 11:57:57 +00:00
feng@chromium.org
33fb11c12f Revert commit 2701 per Erik Corry's request.
Original CL:

http://codereview.chromium.org/171041
Review URL: http://codereview.chromium.org/171089

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2702 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-08-18 00:12:26 +00:00
whesse@chromium.org
349dc04a5d Use root array to load roots in generated ARM code.
This cuts down on code size and reloc work.
Review URL: http://codereview.chromium.org/171041

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2701 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-08-17 14:31:49 +00:00
antonm@chromium.org
0d2c81584b Compile precanned answers for the case of failed interceptor for some combinations.
Review URL: http://codereview.chromium.org/140069

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2577 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-29 12:34:21 +00:00
antonm@chromium.org
4354661a49 Get rid of unnecessary handle management when invoking interceptors.
Review URL: http://codereview.chromium.org/155682

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2562 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-28 14:46:06 +00:00
kasperl@chromium.org
516d47c73b Fix pixel array support for x64 and make the fast Array functions
that use JSARRAY_HAS_FAST_ELEMENTS_CHECK a bit safer in the presence
of pixel arrays.
Review URL: http://codereview.chromium.org/159500

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2556 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-28 12:46:08 +00:00
sgjesse@chromium.org
422b02717e Change the updating of the counters for global stub use to be consistent. Now counters are updated when they are sure to count to avoid decrementing them when there is a miss.
Review URL: http://codereview.chromium.org/160265

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2554 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-28 11:42:18 +00:00
vitalyr@chromium.org
57e8217c10 Stub Cache: speed up load callback accessor by allocating data handle on stack.
Review URL: http://codereview.chromium.org/160041

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2538 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-27 09:54:59 +00:00
whesse@chromium.org
f8e622a620 Added inline caching for calls to a field to X64. Changed a use of Array::kHeaderSize to FixedArray::kHeaderSize on all platforms.
Review URL: http://codereview.chromium.org/149798

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2496 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-17 08:55:01 +00:00
ager@chromium.org
7a80be473d Firefox and Safari both allow calling regular expression objects as
functions (as an alias for calling the exec method).  For
compatibility make call_regexp the default and remove the flag.

Review URL: http://codereview.chromium.org/155453

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2451 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-13 23:41:17 +00:00
ager@chromium.org
55050383e1 Port the function prototype load stub to ARM.
Review URL: http://codereview.chromium.org/155373

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2430 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-07-10 23:26:33 +00:00