This adds microbenchmarks for:
* the RegExp constructor,
* flag accessors,
* and RegExp functions: exec, @@match, @@search, test, @@split, @@replace.
Each benchmark is further split to measure fast and slow paths.
BUG=v8:5339
Review-Url: https://codereview.chromium.org/2521263003
Cr-Commit-Position: refs/heads/master@{#41285}
This code is no longer used in full-codegen or ast-graph-builder since all
functions which have class literals go through Ignition first.
BUG=v8:5657
Review-Url: https://codereview.chromium.org/2534463002
Cr-Commit-Position: refs/heads/master@{#41282}
This removes lots of boilerplate typedefs, with the caveat of renaming
CSA::Label and Variable to CLabel and CVariable within builtins-regexp.
BUG=v8:5339
Review-Url: https://codereview.chromium.org/2524363002
Cr-Commit-Position: refs/heads/master@{#41281}
This removes support for try-catch as well as try-finally constructs
from the {FullCodeGenerator}. Consequently optimized code containing
such constructs must use the {BytecodeGraphBuilder} and can no longer
use the {AstGraphBuilder} for graph building.
R=jarin@chromium.org
BUG=v8:5657
Review-Url: https://codereview.chromium.org/2521233002
Cr-Commit-Position: refs/heads/master@{#41279}
Reason for revert:
Seems to cause compile errors on Android. Will investigate on Monday.
Original issue's description:
> [base] Pass scalar arguments by value in CHECK/DCHECK
>
> This not only potentially improves performance, but also avoids weird
> linker errors, like the one below, where I used Smi::kMinValue in a
> DCHECK_EQ.
>
> > [421/649] LINK ./mksnapshot
> > FAILED: mksnapshot
> > src/base/logging.h|178| error: undefined reference to
> 'v8::internal::Smi::kMinValue'
>
> R=bmeurer@chromium.org, ishell@chromium.org
>
> Committed: https://crrev.com/76723502528c5af003fdffc3520632ea2a13fef3
> Cr-Commit-Position: refs/heads/master@{#41273}
TBR=bmeurer@chromium.org,ishell@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review-Url: https://codereview.chromium.org/2527883004
Cr-Commit-Position: refs/heads/master@{#41278}
Reason for revert:
Need to revert previous CL because of Android compile error, and this one depends in it.
Original issue's description:
> [base] Define CHECK comparison for signed vs. unsigned
>
> The current CHECK/DCHECK implementation fails statically if a signed
> value is compared against an unsigned value. The common solution is to
> cast on each caller, which is tedious and error-prone (might hide bugs).
> This CL implements signed vs. unsigned comparisons by executing up to
> two comparisons. For example, if i is int32_t and u is uint_32_t, a
> DCHECK_LE(i, u) would create the check
> i <= 0 || static_cast<uint32_t>(i) <= u.
> For checks against constants, at least one of the checks can be removed
> by compiler optimizations.
>
> The tradeoff we have to make is to sometimes silently execute an
> additional comparison. And we increase code complexity of course, even
> though the usage is just as easy (or even easier) as before.
>
> The compile time impact seems to be minimal:
> I ran 3 full compilations for Optdebug on my local machine, one time on
> the current ToT, one time with this CL plus http://crrev.com/2524093002.
> Before: 143.72 +- 1.21 seconds
> Now: 144.18 +- 0.67 seconds
>
> In order to check that the new comparisons are working, I refactored
> some DCHECKs in wasm to use the new magic.
>
> R=bmeurer@chromium.org, titzer@chromium.org
>
> Committed: https://crrev.com/5925074a9dab5a8577766545b91b62f2c531d3dc
> Cr-Commit-Position: refs/heads/master@{#41275}
TBR=ishell@chromium.org,titzer@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review-Url: https://codereview.chromium.org/2531533003
Cr-Commit-Position: refs/heads/master@{#41277}
This code is no longer used by full-codegen since all functions which use with statments,
call eval or have lookup variable access now go through Ignition first.
BUG=v8:5657
Review-Url: https://codereview.chromium.org/2514393002
Cr-Commit-Position: refs/heads/master@{#41276}
The current CHECK/DCHECK implementation fails statically if a signed
value is compared against an unsigned value. The common solution is to
cast on each caller, which is tedious and error-prone (might hide bugs).
This CL implements signed vs. unsigned comparisons by executing up to
two comparisons. For example, if i is int32_t and u is uint_32_t, a
DCHECK_LE(i, u) would create the check
i <= 0 || static_cast<uint32_t>(i) <= u.
For checks against constants, at least one of the checks can be removed
by compiler optimizations.
The tradeoff we have to make is to sometimes silently execute an
additional comparison. And we increase code complexity of course, even
though the usage is just as easy (or even easier) as before.
The compile time impact seems to be minimal:
I ran 3 full compilations for Optdebug on my local machine, one time on
the current ToT, one time with this CL plus http://crrev.com/2524093002.
Before: 143.72 +- 1.21 seconds
Now: 144.18 +- 0.67 seconds
In order to check that the new comparisons are working, I refactored
some DCHECKs in wasm to use the new magic.
R=bmeurer@chromium.org, titzer@chromium.org
Review-Url: https://codereview.chromium.org/2526783002
Cr-Commit-Position: refs/heads/master@{#41275}
If code is flushed on a SFI, we can still use the bytecode if it was compiled,
since this never gets flushed.
This fixes a DCHECK where we were trying to compile the bytecode multiple
times after the baseline code was flushed.
BUG=chromium:668133
Review-Url: https://codereview.chromium.org/2526243002
Cr-Commit-Position: refs/heads/master@{#41274}
This not only potentially improves performance, but also avoids weird
linker errors, like the one below, where I used Smi::kMinValue in a
DCHECK_EQ.
> [421/649] LINK ./mksnapshot
> FAILED: mksnapshot
> src/base/logging.h|178| error: undefined reference to
'v8::internal::Smi::kMinValue'
R=bmeurer@chromium.org, ishell@chromium.org
Review-Url: https://codereview.chromium.org/2524093002
Cr-Commit-Position: refs/heads/master@{#41273}
For dictionary-mode receivers, the KeyedStoreGeneric stub can store
properties directly in most cases. Doing so avoids the need to have
an entry in the stub cache for every map/property combination.
Original review: https://codereview.chromium.org/2504403005/
Review-Url: https://codereview.chromium.org/2528883003
Cr-Commit-Position: refs/heads/master@{#41272}
Make apply calls with double array arguments such as
Math.min.apply(Math, [1.1, 2.2]) 1.6x faster.
Drive-by-fix: pass in the isolate to ElementsAccessor::GetImpl.
BUG=v8:4826
Review-Url: https://codereview.chromium.org/2521043005
Cr-Commit-Position: refs/heads/master@{#41268}
This makes sure call-site rendering for certain {TypeError} messages is
based on the correct underlying {JSFunction}, even when inlined frames
are present. Only the {FrameSummary} knows the exact function.
R=verwaest@chromium.org
TEST=message/regress/regress-crbug-661579
BUG=chromium:661579
Review-Url: https://codereview.chromium.org/2521103005
Cr-Commit-Position: refs/heads/master@{#41264}
Reason for revert:
Fails on perf.
Original issue's description:
> [test] Add performance test for closures.
>
> Short living closures are very common in Node.js. This benchmark tracks progress
> as we move the optimizations that are currently only behind
> --mark_shared_functions_for_tier_up to the default settings.
>
> BUG=v8:5512
>
> Committed: https://crrev.com/f277da2a00cfd27d44a33a70213a65bd82d0bc95
> Cr-Commit-Position: refs/heads/master@{#41246}
TBR=bmeurer@chromium.org,leszeks@chromium.org,franzih@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=v8:5512
Review-Url: https://codereview.chromium.org/2526163003
Cr-Commit-Position: refs/heads/master@{#41261}
This has two parts:
- in redundancy elimination, if we see addition with left hand side that
was bounds-checked, we reconnect the lhs to the bounds check if it has better
type.
- in representation inference, eliminate overflow checks if the input types
guarantee no overflow.
Review-Url: https://codereview.chromium.org/2527083002
Cr-Commit-Position: refs/heads/master@{#41260}
In the ParseInfo constructor that takes a SharedFunctionInfo, we must
set the module flag when the function represents a module.
BUG=v8:1569
Review-Url: https://codereview.chromium.org/2525013002
Cr-Commit-Position: refs/heads/master@{#41257}
The old algorithm produces unnecessary decimal digits. The new one
converts the significand of the input double into an uint64_t to be
just as precise as necessary.
R=tebbi@chromium.org
BUG=chromium:658712,chromium:666376
Review-Url: https://codereview.chromium.org/2520363002
Cr-Commit-Position: refs/heads/master@{#41255}
RuntimeTimerScopes always subtract their own time from the parent timer's
counter to properly account for the own time. Once a scope is destructed it
adds it own timer to the current active counter. However, if the current
counter is changed with CorrectCurrentCounterId we will attribute all the
subtimers to the previous counter, and add the own time to the new counter.
This way it is possible to end up with negative times in certain counters but
the overall would still be correct.
BUG=
Review-Url: https://codereview.chromium.org/2511093002
Cr-Commit-Position: refs/heads/master@{#41254}
Reason for revert:
Secondary dependency to revert https://codereview.chromium.org/2522393002/
Original issue's description:
> [cleanup] Refactor builtins-number.cc
>
> Introducing a TF_BUILTIN macro that wraps CodeStubAssembler usage
> into a convenient interface (using a subclass under the hood).
>
> Committed: https://crrev.com/cbf59c4c704e83a43b52c5ba6825df576fdaece8
> Cr-Commit-Position: refs/heads/master@{#41236}
TBR=ishell@chromium.org,jkummerow@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review-Url: https://codereview.chromium.org/2529873002
Cr-Commit-Position: refs/heads/master@{#41252}
Presubmit on upload and on trybot will use the affected files as
determined by the presubmit input_api.
The continuously run presubmit will use the old method and search all
files.
BUG=v8:5603
NOTRY=true
Review-Url: https://codereview.chromium.org/2523993003
Cr-Commit-Position: refs/heads/master@{#41248}
Short living closures are very common in Node.js. This benchmark tracks progress
as we move the optimizations that are currently only behind
--mark_shared_functions_for_tier_up to the default settings.
BUG=v8:5512
Review-Url: https://codereview.chromium.org/2525053002
Cr-Commit-Position: refs/heads/master@{#41246}
The deprecated pipeline is used for asm.js only, where we forcibly
disable inlining anyways (for performance reasons), so inlining via
the AstGraphBuilder is essentially dead code by now, thus there's no
point in trying to keep that around in the code base.
Also nuke the test-run-inlining.cc file, which would require some heavy
surgery (for probably little benefit), and move the useful tests for
mjsunit tests instead.
BUG=v8:2206,v8:5657
R=yangguo@chromium.org
Review-Url: https://codereview.chromium.org/2527053002
Cr-Commit-Position: refs/heads/master@{#41245}
This exposes scopes for suspended generator objects by adding a
[[Scopes]] internal property to generator objects, similar to how
scopes for functions currently not on the stack are handled.
BUG=chromium:667286
Review-Url: https://codereview.chromium.org/2516973003
Cr-Commit-Position: refs/heads/master@{#41244}
Also lower JSToBoolean(x) where x is either some detectable receiver or
null, or any kind of receiver, null or undefined. Also fix a couple of
minor issues with the JSToBoolean lowering and tests.
R=yangguo@chromium.org
BUG=v8:5267
Review-Url: https://codereview.chromium.org/2530773002
Cr-Commit-Position: refs/heads/master@{#41241}
