Commit Graph

30853 Commits

Author SHA1 Message Date
yangguo
28df32d908 [serializer] do not cache resource data pointer for native source.
The cached resource data pointer is a source of non-determinism when
creating the snapshot.  Long-term we may not keep the native source in
memory anyways, so caching the resource data pointer will not be
possible.

R=ulan@chromium.org
BUG=v8:4886
LOG=N

Review-Url: https://codereview.chromium.org/1990183002
Cr-Commit-Position: refs/heads/master@{#36361}
2016-05-19 12:28:55 +00:00
rmcilroy
f241a61a34 [Interpreter] Support compiling for baseline on return from interpreted function.
We cannot tier up from interpreted to baseline code when there is an
activation of the function on the stack. This significantly regresses
the performance of recursive functions since they are unlikely to get
tiered up.

This CL adds the ability for a function to be marked for baseline
compilation when it returns. To do this we patch the
InterpreterEntryTrampoline return address to point to
InterpreterMarkBaselineOnReturn, which leaves the
interpreted frame and recompile the function for
baseline.

This improves the score of EarlyBoyer by ~8x for Ignition.

BUG=v8:4280
LOG=N

Review-Url: https://codereview.chromium.org/1965343002
Cr-Commit-Position: refs/heads/master@{#36360}
2016-05-19 12:28:54 +00:00
jgruber
c04d547298 Refactor script position calculation
Script position calculation logic (i.e. line & column numbers for a
given code position) is now based on a single method
Script::GetPositionInfo(). Refactored related code in isolate.cc and
js/messages.js to use the new method and removed the line_ends JS
accessor.

R=yangguo@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/1986173002
Cr-Commit-Position: refs/heads/master@{#36359}
2016-05-19 12:23:34 +00:00
hpayer
fbb221c48e Disable black allocation to investigate memory bloat.
BUG=chromium:613125
LOG=n

Review-Url: https://codereview.chromium.org/1992193002
Cr-Commit-Position: refs/heads/master@{#36358}
2016-05-19 11:19:22 +00:00
ulan
c3a906d53b Refactor PointerUpdatingVisitor.
This patch extracts typed slot processing logic from
the PointerUpdatingVisitor to a set of functions.

Each of these functions takes an untyped slot processing function
and applies it to a typed slot.

BUG=chromium:612847
LOG=NO

Review-Url: https://codereview.chromium.org/1991843003
Cr-Commit-Position: refs/heads/master@{#36357}
2016-05-19 11:06:57 +00:00
machenbach
5e1c87dd95 [gn] Port gyp/gn comparison script
BUG=chromium:474921
LOG=n
NOTRY=true

Review-Url: https://codereview.chromium.org/1988163002
Cr-Commit-Position: refs/heads/master@{#36356}
2016-05-19 10:44:53 +00:00
danno
78b1585f1d [turbofan] Add DebugBreak machine operator and support
Review-Url: https://codereview.chromium.org/1995543003
Cr-Commit-Position: refs/heads/master@{#36355}
2016-05-19 09:47:54 +00:00
clemensh
452b7f2483 [wasm] Clean up test case
R=titzer@chromium.org, yangguo@chromium.org

Review-Url: https://codereview.chromium.org/1961453002
Cr-Commit-Position: refs/heads/master@{#36354}
2016-05-19 09:20:49 +00:00
machenbach
479de28c3a Revert of Adding ia32 simd assembler changes. (patchset #2 id:20001 of https://codereview.chromium.org/1991713002/ )
Reason for revert:
Crashes on win32 debug:
https://build.chromium.org/p/client.v8/builders/V8%20Win32%20-%20debug/builds/2305/steps/Check/logs/stdio

Also, would be nice if the test output could be a bit shorter and only print what's necessary to trace a failure. Or split things into more smaller tests. Like that, these logs must be processed, json-encoded/decoded and sent around through the infrastructure.

Some chars in the output make the json encoder unhappy, therefore the infrastructure can't nicely display the failures.

Original issue's description:
> Adding ia32 simd assembler support.
>
> Based on assembler changes from this patch:
> https://codereview.chromium.org/90643003/
>
> BUG=https://bugs.chromium.org/p/v8/issues/detail?id=4124
> R=titzer@chromium.org
> LOG=N
>
> Committed: https://crrev.com/fbf58a5af1d07a7fbb3763aa15f8ba26e2ce7d11
> Cr-Commit-Position: refs/heads/master@{#36349}

TBR=bbudge@chromium.org,titzer@chromium.org,gdeepti@chromium.org,aseemgarg@chromium.org,bradnelson@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=https://bugs.chromium.org/p/v8/issues/detail?id=4124

Review-Url: https://codereview.chromium.org/1992163002
Cr-Commit-Position: refs/heads/master@{#36353}
2016-05-19 08:58:18 +00:00
verwaest
155a392be1 Speed up common ObjectProtoToString cases
1) avoid the builder
2) by precomputing the result we're guaranteed to have a flat string
avoiding extra cost later of using the string

BUG=

Review-Url: https://codereview.chromium.org/1988023004
Cr-Commit-Position: refs/heads/master@{#36352}
2016-05-19 08:53:04 +00:00
danno
9c15c05596 Add a html-based visualizer for TurboFan graphs
Review-Url: https://codereview.chromium.org/729913004
Cr-Commit-Position: refs/heads/master@{#36351}
2016-05-19 08:18:01 +00:00
yangguo
5a88c04741 [serializer] do not copy code if snapshot is not required to be deterministic.
Code in V8 embeds memory addresses. Upon deserialization, those addresses will
be updated anyways, so it's not important whether the serializer records those
addresses. In order to have a deterministic build, we create a copy of the code
and null out memory addresses and serialize this copy.

For the code cache, we do not care about determinism. By avoiding the copy we
can save some time spent on serialization. For the mandreel script in Octane,
this reduces the serialization time from ~85ms to ~75ms.

R=vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/1991033003
Cr-Commit-Position: refs/heads/master@{#36350}
2016-05-19 08:18:00 +00:00
bradnelson
fbf58a5af1 Adding ia32 simd assembler support.
Based on assembler changes from this patch:
https://codereview.chromium.org/90643003/

BUG=https://bugs.chromium.org/p/v8/issues/detail?id=4124
R=titzer@chromium.org
LOG=N

Review-Url: https://codereview.chromium.org/1991713002
Cr-Commit-Position: refs/heads/master@{#36349}
2016-05-19 08:04:10 +00:00
clemensh
3596cac87e [wasm] Differentiate unnamed and empty names
Empty function names are allowed and are output as such, unnamed
functions or functions with no valid UTF-8 name are output as
"<WASM UNNAMED>", while the CallSite object returns null as the
function name.

R=titzer@chromium.org, yangguo@chromium.org

Review-Url: https://codereview.chromium.org/1970503004
Cr-Commit-Position: refs/heads/master@{#36348}
2016-05-19 07:54:07 +00:00
machenbach
2c95b572cc Revert of Reland changes to v8_snapshot GN build arg. (patchset #2 id:20001 of https://codereview.chromium.org/1979883002/ )
Reason for revert:
Blocks the roll again:
https://codereview.chromium.org/1993163002/

https://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_amd64-generic_chromium_compile_only_ng/builds/140169

Lets add chromeos_amd64-generic_chromium_compile_only_ng on a reland as well.

Original issue's description:
> Reland changes to v8_snapshot GN build arg.
>
> This patch re-lands #36193 with a couple of minor tweaks
> including updating the BUILD.gn file to use the new
> variable name and incorporating thakis' request to support
> cross-compiling the win snapshot from a mac.
>
> R=thakis@chromium.org, jochen@chromium.org, machenbach@chromium.org
> BUG=608596, 595653
> CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:chromeos_daisy_chromium_compile_only_ng
>
> Committed: https://crrev.com/b6168e1223d266dba1a357a921bc84ae97a1d803
> Cr-Commit-Position: refs/heads/master@{#36342}

TBR=jochen@chromium.org,thakis@chromium.org,stevenjb@chromium.org,adamk@chromium.org,dpranke@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=608596, 595653

Review-Url: https://codereview.chromium.org/1996533002
Cr-Commit-Position: refs/heads/master@{#36347}
2016-05-19 07:11:39 +00:00
littledan
7517f8776d Actually unship --intl_extra
The default value of the flag was true, so --intl_extra was not previously
unshipped. This patch sets it to false.

BUG=v8:3785
CQ_INCLUDE_TRYBOTS=tryserver.v8:v8_linux_noi18n_rel_ng

Review-Url: https://codereview.chromium.org/1990763003
Cr-Commit-Position: refs/heads/master@{#36346}
2016-05-19 06:44:38 +00:00
zhengxing.li
9cf93e1c6d X87: Add cmpxchg and lock instructions to x64 and ia32 {dis,}assemblers.
port 5c22cf5ae7 (r36341)

  original commit message:

BUG=

Review-Url: https://codereview.chromium.org/1990133002
Cr-Commit-Position: refs/heads/master@{#36345}
2016-05-19 05:58:55 +00:00
zhengxing.li
32ba3c91d1 X87: [Interpreter] Remove InterpreterExitTrampoline and replace with returning to the entry trampoline.
port 39738bc905 (r36310)

  original commit message:
  In order to support compiling to baseline on return we need to be able to
  return to the actual return address. With this change this is what the
  Return bytecode now does, removing the need for the
  InterpreterExitTrampoline.

  This change also removes the InterpreterNotifyDeoptXXX builtins and
  unifies FCG and Igntion to both use NotifyDeoptXXX. As part of this
  change, FullCodegenerator::State is moved to Deoptimize::BailoutState.

BUG=

Review-Url: https://codereview.chromium.org/1987053006
Cr-Commit-Position: refs/heads/master@{#36344}
2016-05-19 04:03:34 +00:00
v8-autoroll
26d94abd29 Update V8 DEPS.
Rolling v8/build to b2d15686436cdc17f67c3621c314f8d96b5b6fd9

Rolling v8/tools/clang to 996bab489f816e51dde704bd215fb3403919f07e

TBR=machenbach@chromium.org,vogelheim@chromium.org,hablich@chromium.org

Review-Url: https://codereview.chromium.org/1992113002
Cr-Commit-Position: refs/heads/master@{#36343}
2016-05-19 03:41:27 +00:00
dpranke
b6168e1223 Reland changes to v8_snapshot GN build arg.
This patch re-lands #36193 with a couple of minor tweaks
including updating the BUILD.gn file to use the new
variable name and incorporating thakis' request to support
cross-compiling the win snapshot from a mac.

R=thakis@chromium.org, jochen@chromium.org, machenbach@chromium.org
BUG=608596, 595653
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:chromeos_daisy_chromium_compile_only_ng

Review-Url: https://codereview.chromium.org/1979883002
Cr-Commit-Position: refs/heads/master@{#36342}
2016-05-19 01:06:33 +00:00
binji
5c22cf5ae7 Add cmpxchg and lock instructions to x64 and ia32 {dis,}assemblers
Review-Url: https://codereview.chromium.org/1986113004
Cr-Commit-Position: refs/heads/master@{#36341}
2016-05-19 00:56:08 +00:00
gsathya
04aa9436ce array-iterator.js: use object_get_prototype_of instead of directly accessing the proto
Instead of directly looking up Uint8Array.__proto__, to get a hold of
the TypedArray prototype, use object_get_prototype_of. This is a
functionally equivalent, but cleaner approach.

Review-Url: https://codereview.chromium.org/1990983002
Cr-Commit-Position: refs/heads/master@{#36340}
2016-05-18 21:04:27 +00:00
caitpotter88
f6865cb142 [runtime] set AsyncFunctionNext/Throw to adapt arguments
Prevent crash/UB during stack frame iteration through functions, which occurs
when debugging, when building stacktraces, etc.

Also prevents these functions from appearing in stacktraces, by unsetting the "native" flag.

BUG=v8:4483, v8:5025
R=yangguo@chromium.org, littledan@chromium.org, adamk@chromium.org

Review-Url: https://codereview.chromium.org/1990803005
Cr-Commit-Position: refs/heads/master@{#36339}
2016-05-18 20:53:40 +00:00
bryleun
dc37f6e6ab S390: Added 74 new instructions to the simulator EVALUATE code.
R=joransiu@ca.ibm.com,michael_dawson@ca.ibm.com,mbrandy@us.ibm.com,jyan@ca.ibm.com

BUG=

Review-Url: https://codereview.chromium.org/1992703004
Cr-Commit-Position: refs/heads/master@{#36338}
2016-05-18 20:13:29 +00:00
ulan
dbca1feb4e Refactor ObjectVisitor functions to not mutate the slot.
BUG=chromium:612847
LOG=N

Review-Url: https://codereview.chromium.org/1989973002
Cr-Commit-Position: refs/heads/master@{#36337}
2016-05-18 19:25:56 +00:00
jyan
a207b6400e PPC/S390: [Interpreter] Remove InterpreterExitTrampoline and replace with returning to the entry trampoline.
port 39738bc905

Original Commit Message:
  In order to support compiling to baseline on return we need to be able to
  return to the actual return address. With this change this is what the
  Return bytecode now does, removing the need for the
  InterpreterExitTrampoline.

  This change also removes the InterpreterNotifyDeoptXXX builtins and
  unifies FCG and Igntion to both use NotifyDeoptXXX. As part of this
  change, FullCodegenerator::State is moved to Deoptimize::BailoutState.

R=rmcilroy@chromium.org, joransiu@ca.ibm.com, michael_dawson@ca.ibm.com, mbrandy@us.ibm.com
BUG=v8:4280
LOG=N

Review-Url: https://codereview.chromium.org/1989983002
Cr-Commit-Position: refs/heads/master@{#36336}
2016-05-18 19:25:55 +00:00
machenbach
0aa3707dc4 Revert of [heap] Do not invoke GC to make heap iterable. (patchset #5 id:80001 of https://codereview.chromium.org/1961373003/ )
Reason for revert:
Breaks https://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20gc%20stress/builds/3551

Original issue's description:
> [heap] Do not invoke GC to make heap iterable.
>
> Remove kMakeHeapIterableMask since the heap is always iterable.
>
> BUG=chromium:580959
> LOG=n
>
> Committed: https://crrev.com/7c1cac4888a248fda3fa6de3624f32a6babb37e9
> Cr-Commit-Position: refs/heads/master@{#36333}

TBR=ulan@chromium.org,yangguo@chromium.org,hpayer@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:580959

Review-Url: https://codereview.chromium.org/1987363002
Cr-Commit-Position: refs/heads/master@{#36335}
2016-05-18 19:23:07 +00:00
mlippautz
768e9c509b [heap] Replace RELOCATED_CODE_OBJECT with proper recording
Properly records entries individually instead of using the hammer that requires
visiting the whole code object during pointer updating.

BUG=chromium:612847
LOG=N

Review-Url: https://codereview.chromium.org/1994653002
Cr-Commit-Position: refs/heads/master@{#36334}
2016-05-18 18:31:42 +00:00
hpayer
7c1cac4888 [heap] Do not invoke GC to make heap iterable.
Remove kMakeHeapIterableMask since the heap is always iterable.

BUG=chromium:580959
LOG=n

Review-Url: https://codereview.chromium.org/1961373003
Cr-Commit-Position: refs/heads/master@{#36333}
2016-05-18 18:03:48 +00:00
titzer
3fef34e023 [wasm] Refactor WASM run tests to allow them to run in the interpreter too.
R=mstarzinger@chromium.org,ahaas@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/1990923002
Cr-Commit-Position: refs/heads/master@{#36332}
2016-05-18 15:57:00 +00:00
nikolaos
067f90c6a2 Add posix/BUILD.gn to exec_script_whitelist
BUG=v8:5023
R=mythria@chromium.org,machenbach@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/1993663002
Cr-Commit-Position: refs/heads/master@{#36331}
2016-05-18 15:02:12 +00:00
littledan
29bd8286ee Use stricter type checks in Intl's bound methods
This patch ensures that Intl bound method getters can't be retargeted
to other Intl types. If that were to happen, then a RUNTIME_ASSERT
would trigger later. This patch throws a TypeError instead, as the
specification requires.

BUG=v8:4870

Review-Url: https://codereview.chromium.org/1986763003
Cr-Commit-Position: refs/heads/master@{#36330}
2016-05-18 14:57:58 +00:00
machenbach
517b659928 [gn] Port icu_use_data_file_flag in gn
BUG=chromium:474921
LOG=n
NOTRY=true

Review-Url: https://codereview.chromium.org/1988023003
Cr-Commit-Position: refs/heads/master@{#36329}
2016-05-18 14:44:38 +00:00
ivica.bogosavljevic
4a96bc2a94 Fix MIPS64 compilation issue with GN build system
NOTRY=true

Review-Url: https://codereview.chromium.org/1995473002
Cr-Commit-Position: refs/heads/master@{#36328}
2016-05-18 14:44:37 +00:00
yangguo
2b60b3580e [serializer] cosmetic changes to SerializerReference.
R=vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/1991793002
Cr-Commit-Position: refs/heads/master@{#36327}
2016-05-18 14:39:42 +00:00
mythria
b498f0c93b [Interpreter] Removes failure expectation for gc related blink_tests.
Updates blink_tests/TestExpecations by removing failure expectation for
gc related tests. These tests are modified to work with ignition by
the following cls:
https://codereview.chromium.org/1972943002/
https://codereview.chromium.org/1950613005/

BUG=v8:4280,chromium:595672
LOG=N

Review-Url: https://codereview.chromium.org/1985673002
Cr-Commit-Position: refs/heads/master@{#36326}
2016-05-18 14:15:39 +00:00
mstarzinger
6046e4a10a [turbofan] Remove verbose escape analysis helper methods.
This is a pure refactoring that removes unnecessary wrapper methods from
the EscapeAnalysis class. This completely removes the notion of "Alias"
from the interface.

R=jarin@chromium.org

Review-Url: https://codereview.chromium.org/1989243002
Cr-Commit-Position: refs/heads/master@{#36325}
2016-05-18 14:05:50 +00:00
mstarzinger
d5aa995e3c [turbofan] Make escape analysis non-experimental.
This promotes the escape analysis from an experimental feature to be a
fully supported feature. The main goal is to unleach ClusterFuzz on the
implementation so that we can stabilize it.

R=jarin@chromium.org

Review-Url: https://codereview.chromium.org/1989833002
Cr-Commit-Position: refs/heads/master@{#36324}
2016-05-18 13:51:01 +00:00
jyan
246d5bba8f PPC/S390: Fix %FunctionGetName and %_ClassOf for bound functions.
port 8e303dd0b3

R=bmeurer@chromium.org, yangguo@chromium.org, joransiu@ca.ibm.com, michael_dawson@ca.ibm.com, mbrandy@us.ibm.com
BUG=v8:5010

Review-Url: https://codereview.chromium.org/1986343002
Cr-Commit-Position: refs/heads/master@{#36323}
2016-05-18 13:48:28 +00:00
mstarzinger
84a8e817a5 [turbofan] Hide escape analysis internals from interface.
This is pure refactoring that moves many internal details of the escape
analysis from the header into the compilation unit. The goal of this is
to simplify the interface for readability.

R=jarin@chromium.org

Review-Url: https://codereview.chromium.org/1991723003
Cr-Commit-Position: refs/heads/master@{#36322}
2016-05-18 13:45:45 +00:00
cbruni
e4c3be5af3 [keys] KeyAccumulator cleanups
Move most of JSReceiver::GetKeys to the KeyAccumulator

BUG=

Review-Url: https://codereview.chromium.org/1989263002
Cr-Commit-Position: refs/heads/master@{#36321}
2016-05-18 13:43:08 +00:00
akos.palfi
233ca4cee3 MIPS64: Make GCC happy.
BUG=

Review-Url: https://codereview.chromium.org/1987253002
Cr-Commit-Position: refs/heads/master@{#36320}
2016-05-18 13:30:31 +00:00
machenbach
64a518fa8e [gn] Make gyp include dirs equal to gn
BUG=chromium:474921
LOG=n

Review-Url: https://codereview.chromium.org/1989193002
Cr-Commit-Position: refs/heads/master@{#36319}
2016-05-18 12:02:05 +00:00
yangguo
735fa0c478 [serializer] prepare attached references for general use.
Currently attached references are only used for global proxy, source
string and code stubs. Mid-term future we want to use attached
references for arbitrary objects (in fixed order) provided from outside.

This change renames BackReference to SerializerReference to include both
back references and attached references.

R=mtrofin@chromium.org, vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/1992723002
Cr-Commit-Position: refs/heads/master@{#36318}
2016-05-18 11:30:33 +00:00
jkummerow
e83ac2342d [arm][arm64] Remove COHERENT_CACHE support
We have evidence (see r34896) that this avoids crashes.

BUG=chromium:524337

Review-Url: https://codereview.chromium.org/1995483002
Cr-Commit-Position: refs/heads/master@{#36317}
2016-05-18 11:08:52 +00:00
balazs.kilvady
c7f5419fa5 MIPS64: Add a check to addiu simulation according to the specification.
BUG=

Review-Url: https://codereview.chromium.org/1984313002
Cr-Commit-Position: refs/heads/master@{#36316}
2016-05-18 11:05:38 +00:00
akos.palfi
253050cbf7 MIPS64: Fix '[Interpreter] Remove InterpreterExitTrampoline and replace with returning to the entry trampoline.'
Port 39738bc905

BUG=

Review-Url: https://codereview.chromium.org/1990823002
Cr-Commit-Position: refs/heads/master@{#36315}
2016-05-18 10:41:25 +00:00
rmcilroy
776d4d87a9 [Interpreter] Inline Inc/Dec code stubs into bytecode handlers.
BUG=v8:4280
LOG=N

Review-Url: https://codereview.chromium.org/1980463003
Cr-Commit-Position: refs/heads/master@{#36314}
2016-05-18 09:45:13 +00:00
neis
3560d9bd58 [runtime] Fix effect of setting .prototype on generator functions.
When setting a generator function's  "prototype" property to a non-object, the
prototype of new generator instances should be %GeneratorPrototype%, not
%ObjectPrototype%.

R=verwaest@chromium.org
BUG=v8:5011

Review-Url: https://codereview.chromium.org/1982203003
Cr-Commit-Position: refs/heads/master@{#36313}
2016-05-18 09:18:12 +00:00
neis
75140f39dd [compiler] Remove obsolete JSYield operator.
R=bmeurer@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/1981323002
Cr-Commit-Position: refs/heads/master@{#36312}
2016-05-18 08:55:33 +00:00