This splits the work to be done for dead and life maps
and avoids unnecessary operations in the hot loop.
Results for v8.infinite_scroll:
Before:
nonlive_refs
len: 93
min: 0.0
max: 6.3
avg: 1.35268817204
[0,5[: 91
[5,10[: 2
After:
nonlive_refs
len: 91
min: 0.0
max: 4.2
avg: 0.968131868132
[0,5[: 91
BUG=chromium:554488
LOG=no
Review URL: https://codereview.chromium.org/1441633002
Cr-Commit-Position: refs/heads/master@{#32042}
Also properly support loading from holey double element backing stores
in JSNativeContextSpecialization. This adds a new simplified operator
NumberIsHoleNaN, which checks whether a certain value is the special NaN
that we use to encode "the hole" in holey double element backing stores.
R=jarin@chromium.org
BUG=v8:4470
LOG=n
Review URL: https://codereview.chromium.org/1448343002
Cr-Commit-Position: refs/heads/master@{#32039}
Reason for revert:
Regression in thread_times.key_idle_power_cases (crbug.com/555153)
Original issue's description:
> Remove redundant activation threshold for incremental marking.
>
> The minimum allocation limit already enforces this constraint for normal GCs.
>
> GCs triggered by the memory reducer and external limit should work for all heap sizes.
>
> BUG=chromium:552305
> LOG=NO
>
> Committed: https://crrev.com/e28e4d5f52852e2f14facc37a02069b6bfc82c4e
> Cr-Commit-Position: refs/heads/master@{#31921}
TBR=hpayer@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:552305
Review URL: https://codereview.chromium.org/1456523002
Cr-Commit-Position: refs/heads/master@{#32038}
Adds support for visiting the Call bytecode to the bytecode graph builder.
This change also adds the call type feedback slot to the Call bytecode.
This is not currently used by the interpreter, but is used by the
graph builder.
Also adds a CallWide varient of the Call bytecode, and adds the kCount16
operand type.
Landed on behalf of rmcilroy.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1456453002
Cr-Commit-Position: refs/heads/master@{#32033}
Adds implementation and tests for following operators in bytecode graph builder:
-VisitLoadICSloppy
-VisitLoadICStrict
-VisitLoadICSloppyWide
-VisitLoadICStrictWide
The current implementation introduces empty frame states for frame state inputs expected by these operations.
BUG=v8:4280
LOG=N
Review URL: https://codereview.chromium.org/1419373007
Cr-Commit-Position: refs/heads/master@{#32026}
port e5edd66d07 (r32022)
original commit message:
This is the first part to refactoring the JSNativeContextSpecialization
class, which has grown way too big recently.
Also don't collect cross context feedback for the CallIC in general.
Neither TurboFan nor Crankshaft can make any use of cross context
JSFunction feedback that is collected by the CallIC, so there's no
point in gathering that feedback at all (it just complicates the
checking that is necessary in the compilers). What we should do
instead at some point (when Crankshaft becomes less important) is
to collect the SharedFunctionInfo as feedback for those cases.
BUG=
Review URL: https://codereview.chromium.org/1453033002
Cr-Commit-Position: refs/heads/master@{#32024}
This is the first part to refactoring the JSNativeContextSpecialization
class, which has grown way too big recently.
Also don't collect cross context feedback for the CallIC in general.
Neither TurboFan nor Crankshaft can make any use of cross context
JSFunction feedback that is collected by the CallIC, so there's no
point in gathering that feedback at all (it just complicates the
checking that is necessary in the compilers). What we should do
instead at some point (when Crankshaft becomes less important) is
to collect the SharedFunctionInfo as feedback for those cases.
R=yangguo@chromium.org
BUG=v8:4470
LOG=n
Review URL: https://codereview.chromium.org/1451273002
Cr-Commit-Position: refs/heads/master@{#32022}
- Removed PASS/FAIL line for arrow/rest params test which should now pass.
- Moved soon-to-be-changed tests about built-in prototypes as plain objects
to INVALID section (and noted that they'll be fixed with the
next test262 roll).
R=littledan@chromium.org
Review URL: https://codereview.chromium.org/1450243002
Cr-Commit-Position: refs/heads/master@{#32020}
All uses of Token::INIT also have access to the relevant VariableMode,
so there's no reason to have more than one token representing an
initializing assignment.
Review URL: https://codereview.chromium.org/1431873006
Cr-Commit-Position: refs/heads/master@{#32016}
While all allocated objects need to have correct alignment, this requirement
doesn't make a lot of sense for the allocation limit. Removing this assertion
would makes it possible to make the inline-allocation-observers accurate. This
could be done in other ways, but removing this assertion is the most elegant way
to go about it.
R=hpayer@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1444883003
Cr-Commit-Position: refs/heads/master@{#32013}
This adds support for load/store access to FAST_HOLEY_ELEMENTS and
FAST_HOLEY_SMI_ELEMENTS backing stores. We don't currently fully support
the double holes, because we make some difficult assumptions in
Crankshaft there.
R=jarin@chromium.org
BUG=v8:4470
LOG=n
Review URL: https://codereview.chromium.org/1448903002
Cr-Commit-Position: refs/heads/master@{#32012}
This way we can properly cancel them on the main thread in case they never get
scheduled by the platform.
BUG=chromium:524425
LOG=N
CQ_EXTRA_TRYBOTS=tryserver.v8:v8_linux_arm64_gc_stress_dbg;tryserver.v8:v8_linux_gc_stress_dbg;tryserver.v8:v8_mac_gc_stress_dbg;tryserver.v8:v8_linux64_msan_rel;tryserver.v8:v8_linux64_tsan_rel;tryserver.v8:v8_mac64_asan_rel
Review URL: https://codereview.chromium.org/1446523003
Cr-Commit-Position: refs/heads/master@{#32007}
The TiesEven rounding mode rounds float64 numbers to the nearest
integer. If there are two nearest integers, then the number is rounded
to the even one. This is the default rounding mode according to
IEEE~754.
I implemented the operator on ia32, x64, arm, arm64, mips, and mips64.
I think there is a bug in the current implementation of the ppc
simulator, which kept me from implementing the operator on ppc.
According to my understanding of the ppc instruction manual, the FRIN
instruction provides the right behavior for Float64RoundTiesEven. In the
simulator, however, FRIN provides a different semantics. If there are
two nearest integers, then the simulator returns the one which is
further away form 0.
Review URL: https://codereview.chromium.org/1440293002
Cr-Commit-Position: refs/heads/master@{#32005}
Extend the ElementAccessInfo machinery with support for elements kind
transitions, which can be either in-place updates of the map (i.e. when
going from FAST_SMI_ELEMENTS to FAST_ELEMENTS) or instance migrations
utilizing the TransitionElementsKindStub.
R=jarin@chromium.org
BUG=v8:4470
LOG=n
Review URL: https://codereview.chromium.org/1452563002
Cr-Commit-Position: refs/heads/master@{#32003}
port 34b7b21d1d (r32000)
original commit message:
If inline allocation fails, we can just use the %NewObject fallback,
which will do the right thing. We don't need a dedicated fallback to
%AllocateInNewSpace.
BUG=
Review URL: https://codereview.chromium.org/1451603002
Cr-Commit-Position: refs/heads/master@{#32002}
port ff283f7ded (r31987)
original commit message:
* Limit triggering of tail calls to explicit use of a new inline runtime
function %_TailCall. %_TailCall works just like %_Call except for using
tail-calling mechanics (currently only in TF).
* Remove hack that recognized some specific usages of %_Call and converted them
into tail calls.
* Support tail calls for all calls where the number of callee stack parameters
is less than or equal to the number of caller stack parameters.
* Use the gap resolver to swizzle parameters and registers to tail calls.
BUG=
Review URL: https://codereview.chromium.org/1445023002
Cr-Commit-Position: refs/heads/master@{#32001}
If inline allocation fails, we can just use the %NewObject fallback,
which will do the right thing. We don't need a dedicated fallback to
%AllocateInNewSpace.
R=verwaest@chromium.org
Review URL: https://codereview.chromium.org/1440193003
Cr-Commit-Position: refs/heads/master@{#32000}
This change binds each {Cancelable} task to a so-called {CancelableTaskManager},
which is then used to handle concurrent cancelation as well as synchronizing
shutdown for already running tasks. Since ownership of tasks is transferred to
the platform executing a task (destructor), handling in the manager uses integer
ids. Note that this also mitigates (modulo integer size) the ABA problem.
All handling of {Cancelable} tasks is now encapsulated into the corresponding
manager, which is instantiated for each isolate.
R=hpayer@chromium.org
BUG=chromium:524425
LOG=N
CQ_EXTRA_TRYBOTS=tryserver.v8:v8_linux_arm64_gc_stress_dbg;tryserver.v8:v8_linux_gc_stress_dbg;tryserver.v8:v8_mac_gc_stress_dbg;tryserver.v8:v8_linux64_msan_rel;tryserver.v8:v8_linux64_tsan_rel;tryserver.v8:v8_mac64_asan_rel
Review URL: https://codereview.chromium.org/1409993012
Cr-Commit-Position: refs/heads/master@{#31997}
The previous code did not properly check for harmony const when
doing the dynamic redeclaration check. This was masked in the
test because each eval had an initializer, and the initializer was what
triggered the exception.
This patch tightens the test by removing initializers and fixes the bug in
DeclareLookupSlot.
Also change the test to use assertThrows where possible.
BUG=v8:4550
LOG=n
Review URL: https://codereview.chromium.org/1437003006
Cr-Commit-Position: refs/heads/master@{#31995}
