kmillikin@chromium.org
dde853a4ad
Small change to stack checks in unoptimized code.
...
The stack checks include a loop nesting depth encoded in the code stream
after the call to the stack check stub. Change the code to jump around this
instruction as well as the call when the stack check is OK.
Review URL: http://codereview.chromium.org/6005014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6204 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-01-06 13:48:12 +00:00
whesse@chromium.org
2655060546
Fix errors in x64 crankshaft port, add failing tests to test expectations.
...
Review URL: http://codereview.chromium.org/6104004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6203 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-01-06 13:29:22 +00:00
sgjesse@chromium.org
7311e10fdb
Optimize instanceof further
...
If the instance of is performed against what is beliwed to be a constant global function inline the instance of check and have the call to the instanceof stub in deferred code. The inlined check will be patched by the instanceof stub when called from deferred code. This is indicated by the lithium instruction LInstanceOfKnownGlobal.
To help the patching the delta from the return address to the patch site is placed just below the return address in the edi slot of the pushad/popad ares. This is safe because the edi register (which is pushed last) is a temporary for the lithium instruction.
As the instanceof stub can call other JavaScript an additional marking for saving all double registers have been added.
Also tweaked the instanceof stub to produce true/false objects instead of 0/1 for the case with deferred code.
Review URL: http://codereview.chromium.org/5990005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6173 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-01-05 11:17:37 +00:00
whesse@chromium.org
4330070e32
Use Assembler::Set() to assign constants to registers in more places. Commit of http://codereview.chromium.org/6016007/ .
...
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6104 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-22 09:55:19 +00:00
whesse@chromium.org
5656d64ece
Add untagged double versions of Math.sin and Math.cos. Merge classes TranscendentalCacheStub and TranscendentalCacheSSE2Stub.
...
Review URL: http://codereview.chromium.org/5996002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6083 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-20 13:52:14 +00:00
sgjesse@chromium.org
d559d8c6c0
IA32: Optimize instanceof
...
The instanceoff stub now supports arguments in registers and checks for null,
smi and string values without going to runtime.
Needs porting to X64 and ARM.
Review URL: http://codereview.chromium.org/5828004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6042 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-16 08:58:42 +00:00
vitalyr@chromium.org
7a3fd2aa52
Deoptimize non-smi switch cases if they are reached.
...
This way if the type oracle says an unreachable clause has a non-smi
type, we can still emit optimized code instead of doing an early
bailout.
This change depends of Florian's r5970.
Review URL: http://codereview.chromium.org/5812005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6037 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-15 17:11:11 +00:00
kmillikin@chromium.org
ace6290452
Fix issue 974.
...
When entering a finally block in unoptimized code, we unconditionally
save the accumulator register in the stack in case it holds a return
value or an exception. In the case of a break, continue, or falling
off the end of the try or catch block, this value is unpredictable and
not necessarily safe for GC.
Review URL: http://codereview.chromium.org/5883003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6035 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-15 16:14:29 +00:00
fschneider@chromium.org
b93a0f5d9a
Second attempt: Improve our type feedback by recogizining never-executed IC calls for binary operations.
...
This is an improved version of my earlier change r5970. It avoids degrading the
non-optimized code.
Initially we emit a conditional branch that is either always- or never-taken
after a smi-check (depending on whether we test for smi for for non-smi)
Since test-eax always sets the carry-flag to 0 we use jump-if-carry and
jump-if-not-carry.
The first invocation of the stub patches a jc with a jz and
jnc with a jnz-instruction so that the code looks exactly as it was
without patching. The only difference is the test- or nop-instruction
after the IC-call.
Review URL: http://codereview.chromium.org/5763004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6030 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-15 13:56:41 +00:00
karlklose@chromium.org
5f962f2eb2
Add array bound checks to code generated for SwapElements. This fixes a bug that lead to a segfault when an array was modified while it was sorted.
...
Review URL: http://codereview.chromium.org/5686006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6020 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-15 09:52:58 +00:00
kmillikin@chromium.org
49f4c39d6d
Deoptimize to the proper target after assignment side effects.
...
This fixes V8 issue 989.
Before, assignments used the AST ID of the assignment expression to
mark the side effect of the store, which became a target for
deoptimization bailout for code after the assignment. In effect
contexts this environment included the value of the assignment, which
was unexpected by the unoptimized code.
Now we introduce a new assignment ID for AST node types that include
an assignment (Assignment, CountOperation, and ForInStatement) and use
it for the side effect of the store.
Review URL: http://codereview.chromium.org/5682010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5990 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-13 16:29:47 +00:00
fschneider@chromium.org
b1a2cc1e48
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
2010-12-13 10:41:50 +00:00
fschneider@chromium.org
b16df72475
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
2010-12-10 14:33:20 +00:00
kasperl@chromium.org
90b3370374
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
2010-12-07 11:31:57 +00:00
kasperl@chromium.org
51b494d096
Revert r5920. Will re-land shortly.
...
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5921 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-07 11:01:02 +00:00
kasperl@chromium.org
e5860bd6a8
Update V8 to version 3.0.
...
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5920 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-12-07 09:11:56 +00:00
whesse@chromium.org
66836efbba
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
2010-12-02 11:20:44 +00:00
vitalyr@chromium.org
b0e4479e70
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
2010-11-30 13:17:36 +00:00
ager@chromium.org
1e8413e188
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
2010-11-22 09:57:21 +00:00
whesse@chromium.org
1d11e32a01
Add a fast case to Array.join when all the elements and the separator are flat ascii strings.
...
Review URL: http://codereview.chromium.org/5122005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5860 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-11-19 09:25:46 +00:00
sgjesse@chromium.org
010f35f478
Change the order of evaluation of sub-expressions for keyed call
...
The expression of the key is now evaluated before the arguments, so all expressions in a keyed call are evaluared from left to right.
BUG=http://code.google.com/p/v8/issues/detail?id=931
TEST=test/mjsunit/regress/regress-931.js
Review URL: http://codereview.chromium.org/5161002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5842 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-11-17 13:59:07 +00:00
fschneider@chromium.org
582a3bde4f
Move ComputeCallInitialize from the codegen.cc to stub-cache.cc.
...
Review URL: http://codereview.chromium.org/4760003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5812 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-11-11 10:33:51 +00:00
fschneider@chromium.org
8673e6497c
Move common static helpers from codegen to the macro-assembler files.
...
Review URL: http://codereview.chromium.org/4654002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5807 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-11-10 17:00:20 +00:00
vegorov@chromium.org
9b96a9d592
Fix presubmit errors introduced by r5768.
...
Review URL: http://codereview.chromium.org/4409003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5770 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-11-04 15:39:06 +00:00
vegorov@chromium.org
746d72420c
Improve positions recording for calls.
...
Review URL: http://codereview.chromium.org/4469002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5768 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-11-04 15:12:03 +00:00
karlklose@chromium.org
1b0e2597e8
Fix a bug that prevents constants from overwriting function values in object literals.
...
BUG=http://code.google.com/p/v8/issues/detail?id=907
Review URL: http://codereview.chromium.org/4004006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5718 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-10-27 11:37:59 +00:00
whesse@chromium.org
9aec312468
Add stop-at flag implementation to full compiler on all platforms.
...
Review URL: http://codereview.chromium.org/3791009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5629 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-10-15 11:45:05 +00:00
kmillikin@chromium.org
830185b175
Clean up some messiness in Scopes.
...
For some reason, the scope's arguments and arguments shadow were
variable proxies, which resulted in all references to the arguments
shadow being shared in the AST. This makes it hard to put per-node
state on the AST nodes.
I took the opportunity to remove Variable::AsVariable which has
confused people in the past, and to rename Variable::slot to the more
accurate Variable::AsSlot.
Review URL: http://codereview.chromium.org/3432022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5517 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-24 07:53:59 +00:00
erik.corry@gmail.com
94cf27373c
Cleanup of contexts in the full code generator.
...
Review URL: http://codereview.chromium.org/3449004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5511 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-23 09:22:45 +00:00
kasperl@chromium.org
b2306a41d0
Clean up the insertion of nops (signalling non-inlined loads and
...
stores) in the full codegens.
Review URL: http://codereview.chromium.org/3431010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5470 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-16 09:10:34 +00:00
ricow@chromium.org
240cee91e4
Add support for near labels.
...
This change introduces near labels in the assembler, allowing us to
uptimize forward jumps (conditional and unconditional) if we can
guarantee that the jump is witin range -128 to +127.
I changed a large fractions of the existing Labels to NearLabels, and
left out cases where it was not immediately clear if it could be used
or not (not immediately clear means labels covering a large code
block, or used in function calls which we could potentially change to
accept near labels).
Review URL: http://codereview.chromium.org/3388004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5460 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-15 11:43:12 +00:00
fschneider@chromium.org
bd6a8c1e4e
Make the CompareStub and the UnaryOpStub accept smi inputs.
...
The stubs get an additional flag for including the smi code
inside the stub. This allows us to generate more compact code
if we don't want to inline the smi case outside the stub.
Review URL: http://codereview.chromium.org/3388005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5456 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-15 10:14:25 +00:00
vitalyr@chromium.org
c3f04bf46f
Add a few missing nops to signal no inlined code was generated.
...
Review URL: http://codereview.chromium.org/3399005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5454 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-14 16:38:02 +00:00
kasperl@chromium.org
dffb7533aa
Implement for-in cache validity checking in the full codegen on IA-32.
...
Review URL: http://codereview.chromium.org/3376003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5445 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-13 06:13:20 +00:00
ager@chromium.org
666b98fb4b
Handle both global and local variables potentially shadowed by
...
eval-introduced variables in full-codegen.
Make sure that x64 assembler records source positions for calls.
Review URL: http://codereview.chromium.org/3357022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5441 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-10 10:57:07 +00:00
ager@chromium.org
e5b63e7213
Handle global variables potentially shadowed by eval-introduced
...
variables in full-codegen.
Review URL: http://codereview.chromium.org/3295022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5430 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-09 10:44:24 +00:00
whesse@chromium.org
a18f9d1114
Remove dependence of code-stubs on codegen, the virtual frame code generator. Move some functions used by code-stubs and full-codegen from codegen to macro-assembler.
...
Review URL: http://codereview.chromium.org/3169049
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5370 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-30 11:48:07 +00:00
serya@chromium.org
71548e969b
Using array index hash code for string-to-number conversion.
...
Review URL: http://codereview.chromium.org/3141022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5362 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-27 11:47:12 +00:00
kasperl@chromium.org
6d5451d685
Add inlining of binary smi operations in the full codegens on IA32
...
and x64 (on IA32 we even fold constants into the instructions for
a more compact representation) and prepare the ARM full codegen for
the doing the same there.
Review URL: http://codereview.chromium.org/3195028
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5351 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-26 08:50:38 +00:00
kasperl@chromium.org
1816e05dc2
Prepare for optionally inlining smi cases in the code generated
...
by the full codegens.
Review URL: http://codereview.chromium.org/3116042
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5343 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-25 14:22:03 +00:00
kasperl@chromium.org
3c26a55def
Start using the overwrite mode from the full codegens to generate
...
slightly better code and allow passing arguments to binary op stubs
in registers on the platforms that support it.
Review URL: http://codereview.chromium.org/3203005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5339 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-25 11:10:05 +00:00
whesse@chromium.org
82309445ad
Move code stub implementations from codegen-[platform].cc files to new code-stub-[platform].cc files, and declarations to new code-stub-[platform].h files.
...
Remaining work is to do the same for platform-independent code stub declarations, and to remove all dependencies on codegen header files from code stub files.
Review URL: http://codereview.chromium.org/3195022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5338 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-25 09:44:44 +00:00
kasperl@chromium.org
59b0bb1e2a
Improve the code generated by the full codegen by keeping
...
track of the fall-through label and avoiding silly jumps.
Review URL: http://codereview.chromium.org/3166033
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5337 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-25 08:57:21 +00:00
kasperl@chromium.org
02a842daae
Simplify the way constructors are called and avoid pushing
...
a dummy receiver that we end up ignoring anyway.
Review URL: http://codereview.chromium.org/3136032
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5335 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-25 07:18:32 +00:00
kasperl@chromium.org
a342aa03d5
Add position information for compares, binary ops, and count
...
operations.
Review URL: http://codereview.chromium.org/3120027
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5332 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-24 13:51:23 +00:00
kasperl@chromium.org
46c0c0f44f
Optimize 'typeof <expr> == <string literal>' in the full codegen.
...
Review URL: http://codereview.chromium.org/3110034
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5326 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-24 11:41:26 +00:00
kasperl@chromium.org
7b51dc7edb
Cleanup the AST code by removing unused parts and get rid of the
...
flow graph code completely. Add new AST node for null compares
and use it to make the full codegens better at generating code
for null comparisons.
Review URL: http://codereview.chromium.org/3146037
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5323 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-24 07:26:49 +00:00
lrn@chromium.org
0d12f8f4e5
Fixed remaining bugs in COW-regexp-results.
...
Review URL: http://codereview.chromium.org/3197011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5322 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-23 13:58:56 +00:00
kasperl@chromium.org
d0bdc7956e
Simplified the full codegens by removing the Expression::kTestValue
...
and Expression::kValueTest. At the same time, prepare the code for
keeping track of fall through labels to avoid unnecessary jumps.
Review URL: http://codereview.chromium.org/3152042
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5320 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-23 12:55:29 +00:00
lrn@chromium.org
0210df5848
Use Copy-on-write arrays for cached regexp results.
...
Review URL: http://codereview.chromium.org/3158020
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5310 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-20 09:37:22 +00:00