I replaced the hidden string with hidden_properties_symbol, so we don't
need the extra hash-check anymore. This is slightly faster anyway.
BUG=
Review URL: https://codereview.chromium.org/1707653003
Cr-Commit-Position: refs/heads/master@{#34086}
This intrinsic was only supported in fullcodegen, and is actually no
longer relevant for SunSpider peak performance it seems, so let's get
rid of it and maybe just implement Array.prototype.join with a fast
path at some point instead.
R=mstarzinger@chromium.org
Review URL: https://codereview.chromium.org/1708523002
Cr-Commit-Position: refs/heads/master@{#34084}
This CL introduces two new bytecodes TailCall and TailCallWide.
BUG=v8:4698,v8:4687
LOG=N
Review URL: https://codereview.chromium.org/1698273003
Cr-Commit-Position: refs/heads/master@{#34083}
This removes the CompilationInfo field from the BytecodeGraphBuilder
class. The intention is to reduce the risk of using uninitialized or
unavailable values from the CompilationInfo (e.g. values that are only
available after parsing).
R=rmcilroy@chromium.org
Review URL: https://codereview.chromium.org/1704913002
Cr-Commit-Position: refs/heads/master@{#34081}
There's no need to have inline platform code for this intrinsic, which
is only used when generating a descriptive string for an error, the
runtime call is just fine here.
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1706743002
Cr-Commit-Position: refs/heads/master@{#34080}
Frame slots indexes numbers are used more consistently for
computation in both TurboFan and Crankshaft. Specifically,
Crankshaft now uses frame slot indexes in LChunk, removing
the need for some special-case maths when building the
deoptimization translation table.
LOG=N
R=mstarzinger@chromium.org
Review URL: https://codereview.chromium.org/1702593002
Cr-Commit-Position: refs/heads/master@{#34078}
This removes the language mode parameter from all JSCall operators. The
information is no longer used anywhere and is not threaded through the
interpreter bytecode. We should only thread it through the bytecode if
it has a semantic impact on the compilation.
R=bmeurer@chromium.org
Review URL: https://codereview.chromium.org/1709493002
Cr-Commit-Position: refs/heads/master@{#34073}
Everything that HCallStub does can easily be done using the more general
HCallWithDescriptor, so there's no need to have this dedicated
instruction around.
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1705633004
Cr-Commit-Position: refs/heads/master@{#34072}
If sweeping is in progress then we need to filter out slots in free space after
array trimming, because the sweeper will add the free space into free list.
This CL also fixes a bug in SlotSet::RemoveRange.
BUG=chromium:587004
LOG=NO
TBR=hpayer@chromium.org
Review URL: https://codereview.chromium.org/1701963003
Cr-Commit-Position: refs/heads/master@{#34071}
There's no point in having the setup or the toString/valueOf methods in
JavaScript. The full setup can be done during bootstrapping when the
Boolean constructor is created, and the prototype methods don't benefit
from JS + %_ at all.
R=jarin@chromium.org
Review URL: https://codereview.chromium.org/1701273003
Cr-Commit-Position: refs/heads/master@{#34068}
Port fd8fd05cc5
Original commit message:
This functionality is useful for stubs that need to walk the stack. The new
machine operator, LoadParentFramePointer doesn't force the currently compiling
method to have a frame in contrast to LoadFramePointer. Instead, it adapts
accordingly when frame elision is possible, making efficient stack walks
possible without incurring a performance penalty for small stubs that can
benefit from frame elision.
BUG=
Review URL: https://codereview.chromium.org/1701933003
Cr-Commit-Position: refs/heads/master@{#34066}
port fd8fd05cc5 (r34014)
original commit message:
This functionality is useful for stubs that need to walk the stack. The new
machine operator, LoadParentFramePointer dosn't force the currently compiling
method to have a frame in contrast to LoadFramePointer. Instead, it adapts
accordingly when frame elision is possible, making efficient stack walks
possible without incurring a performance penalty for small stubs that can
benefit from frame elision.
BUG=
Review URL: https://codereview.chromium.org/1705673002
Cr-Commit-Position: refs/heads/master@{#34064}
port e0129d0f87 (r33986)
original commit message:
Turn the fast case of ArgumentsAccessStub into a new stub
FastNewSloppyArgumentsStub, which is similar to the existing
FastNewStrictArgumentsStub, although not polished yet, and the slow
case always went to the runtime anyway, so we can just directly emit
a runtime call there.
BUG=
Review URL: https://codereview.chromium.org/1701983004
Cr-Commit-Position: refs/heads/master@{#34062}
CPU profiler probes stack frame to find if it's in a valid
state. Under simulator it might happen the frame is not
initialized and MSAN repors that. That's totally ok.
Suppress the warning.
BUG=v8:4751
LOG=N
Review URL: https://codereview.chromium.org/1700533003
Cr-Commit-Position: refs/heads/master@{#34061}
port 09d845354742a90fc0596262eb5cbaac169e5ed9(r33925)
original commit message:
The FastNewStrictArgumentsStub is very similar to the recently added
FastNewRestParameterStub, it's actually almost a copy of it, except that
it doesn't have the fast case we have for the empty rest parameter. This
patch improves strict arguments in TurboFan and fullcodegen by up to 10x
compared to the previous version.
Also introduce proper JSSloppyArgumentsObject and JSStrictArgumentsObject
for the in-object properties instead of having them as constants in the
Heap class.
Drive-by-fix: Use this stub and the FastNewRestParameterStub in the
interpreter to avoid the runtime call overhead for strict arguments
and rest parameter creation.
BUG=
Review URL: https://codereview.chromium.org/1706703002
Cr-Commit-Position: refs/heads/master@{#34055}
port e519e6fadfaf7242231f4dcb45910f60304e26fc(r33886)
original commit message:
1) Update profiling counters in Full codegen.
2) Call Runtime::kTraceTailCall when tracing is on
test/mjsunit/es6/tail-call-simple.js is disabled for now, because Turbofan does not fully support TCO yet.
BUG=
Review URL: https://codereview.chromium.org/1706673002
Cr-Commit-Position: refs/heads/master@{#34054}
Port fd8fd05cc5
Original commit message:
This functionality is useful for stubs that need to walk the stack. The new
machine operator, LoadParentFramePointer dosn't force the currently compiling
method to have a frame in contrast to LoadFramePointer. Instead, it adapts
accordingly when frame elision is possible, making efficient stack walks
possible without incurring a performance penalty for small stubs that can
benefit from frame elision.
R=danno@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=
Review URL: https://codereview.chromium.org/1700323002
Cr-Commit-Position: refs/heads/master@{#34049}
This avoids spending lots of time in Scope::RemoveUnresolved for very long
variable declaration lists.
BUG=v8:4699
LOG=n
Review URL: https://codereview.chromium.org/1655313003
Cr-Commit-Position: refs/heads/master@{#34047}
This enables downloading all gcmole dependencies from
google storage on demand, controlled by a gyp flag
"gcmole=1". This makes the analysis portable to any linux64
host. The archive contains a prebuilt clang in the needed
version, the gcmole plugin and a lua binary.
The tool can be run through a new wrapper that sets up the
environment. This'll ease running it on swarming.
BUG=chromium:535160
LOG=n
Review URL: https://codereview.chromium.org/1703533002
Cr-Commit-Position: refs/heads/master@{#34046}
Removes some cctest and mjsunit test skips on Ignition for tests that now pass.
BUG=v8:4680
LOG=N
Review URL: https://codereview.chromium.org/1703563002
Cr-Commit-Position: refs/heads/master@{#34045}
Port 09d8453547
More fix after 3b980234c8
Original commit message:
The FastNewStrictArgumentsStub is very similar to the recently added
FastNewRestParameterStub, it's actually almost a copy of it, except that
it doesn't have the fast case we have for the empty rest parameter. This
patch improves strict arguments in TurboFan and fullcodegen by up to 10x
compared to the previous version.
Also introduce proper JSSloppyArgumentsObject and JSStrictArgumentsObject
for the in-object properties instead of having them as constants in the
Heap class.
Drive-by-fix: Use this stub and the FastNewRestParameterStub in the
interpreter to avoid the runtime call overhead for strict arguments
and rest parameter creation.
BUG=
Review URL: https://codereview.chromium.org/1699183002
Cr-Commit-Position: refs/heads/master@{#34043}
Various places assume that GetExpression returns the locals for a frame.
Modify InterpretedFrames such that GetExpression(0) returns the first
local, not the fixed parts of the interpreter frame.
BUG=v8:4690,v8:4680
LOG=N
Review URL: https://codereview.chromium.org/1697223003
Cr-Commit-Position: refs/heads/master@{#34040}