This delays removing dead loop's loop exits after we iterate all uses of
the loop. That way, we avoid mutating the use collection while iterating
it.
Bug: chromium:803022
Change-Id: I17462dd82c3cb78f2f630e5db81d8ccdcc517d83
Reviewed-on: https://chromium-review.googlesource.com/878329
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Commit-Queue: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50813}
This adds a tri-state flag --infra-staging and --no-infra-staging, which
can be used to explicitly opt in or out of the staging test runner.
If not specified, a new architecture whitelist will enable roll-out per
arch for manual (none CI) runs.
We'll start whitelisting archs in follow ups.
Bug: v8:7343
Change-Id: I1228e48969fd379f5c231a2b8fad4afc01da94c0
Reviewed-on: https://chromium-review.googlesource.com/881169
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Sergiy Byelozyorov <sergiyb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50812}
Introduce option to run fuzzer processor without analysis phase.
It will be used in fuzzing combined tests.
Bug: v8:6917
Change-Id: Ic96d6b8c5a35c81da80340555bdd75c0d518cb5a
Reviewed-on: https://chromium-review.googlesource.com/880948
Commit-Queue: Michał Majewski <majeski@google.com>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50810}
This CL moves allocations in array-multiple-receiver-maps.js
to prevent gc fuzzing from cleaning out code objects, which
will mess with assertOptimized in the test.
Bug: v8:7338
Change-Id: I9ee88cf5518307ff12302df2fdaca5258c23b779
Reviewed-on: https://chromium-review.googlesource.com/880957
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50809}
As per TODO, this is its rightful place, which gets the module-compiler
out of the business of doing finalization.
R=mstarzinger@chromium.org,clemensh@chromium.org
Bug: v8:7316
Change-Id: Ie419a1e348f14f2613f62fed7083e19365cd9347
Reviewed-on: https://chromium-review.googlesource.com/880950
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Ben Titzer <titzer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50805}
Currently, yields and awaits inside loops compile to bytecode which
switches to the top of the loop header, and switch again once inside the
loop. This is to make loops reducible.
This replaces this switching logic with a single switch bytecode that
directly jumps to the bytecode being resumed. Among other things, this
allows us to no longer maintain the generator state after the switch at
the top of the function, and avoid having to track loop suspend counts.
TurboFan still needs to have reducible loops, so we now insert loop
header switches during bytecode graph building, for suspends that are
discovered to be inside loops during bytecode analysis. We do, however,
do some environment magic across loop headers since we know that we will
continue switching if and only if we reached that loop header via a
generator resume. This allows us to generate fewer phis and tighten
liveness.
Change-Id: Id2720ce1d6955be9a48178322cc209b3a4b8d385
Reviewed-on: https://chromium-review.googlesource.com/866734
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50804}
Anonymous functions have no index, thus we cannot get their source
position table.
Technically, we are not even allowed to call {index()} on anonymous
functions, as this will DCHECK that {index_} contains a value.
R=mstarzinger@chromium.org
Change-Id: I9a8b07cf836671e080cc1784c1712ecd88778972
Reviewed-on: https://chromium-review.googlesource.com/880921
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50802}
It must be monotone.
R=bmeurer@chromium.org
Bug: v8:7354
Change-Id: I08dcd3333518029eef08c074c2b91b5c20ad699e
Reviewed-on: https://chromium-review.googlesource.com/880982
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50801}
Reduce the code duplication overhead in the InternalPerformPromiseThen
helper, which saves quite a bit of space and makes code more concise and
readable.
Bug: v8:7253, v8:7310
Change-Id: I64d11661d7258ced32df564d2e83c5ea45955415
Reviewed-on: https://chromium-review.googlesource.com/880722
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50800}
This fixes a corner-case where deserialization of a module containing
multiple exported functions of the same signature forgot to properly
unprotect the code-space. Test coverage has been added.
R=clemensh@chromium.org
TEST=mjsunit/wasm/compiled-module-serialization
BUG=chromium:804767
Change-Id: I0082303db19bcc14c4de30f29d604665e281d79d
Reviewed-on: https://chromium-review.googlesource.com/880844
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50799}
As part of the effort to despecialize WASM code, convert many uses of
WasmInstanceObject which were simply indirecting through to either
the compiled module or the shared module data with helpers on
the respective Frame objects.
R=mstarzinger@chromium.org
Bug:
Change-Id: I05bd1a18b1d81cceef8a80d9f6988e4f5d537e66
Reviewed-on: https://chromium-review.googlesource.com/876125
Commit-Queue: Ben Titzer <titzer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50798}
When spilling a value to the stack, make sure to fill it as the same
type later. Otherwise, we might load garbage from the stack and violate
the assumption that the upper 32 bits of a 64 bit register are zero if
it currently holds a 32 bit value.
R=titzer@chromium.org
Bug: v8:7353, v8:6600
Change-Id: I7f2b1b31b7f3c13aa152c682cb59400fb5a3ebf0
Reviewed-on: https://chromium-review.googlesource.com/880682
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Ben Titzer <titzer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50797}
Now that SuspendGenerator returns, we have to update the interrupt
budget during that return to ensure that generators can be optimized.
Bug: chromium:804796
Change-Id: I8a9fa1c2399da81a3c2a7d8a07a774d5648d1c5e
Reviewed-on: https://chromium-review.googlesource.com/880821
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50796}
Use this in the PromiseThen operation to skip the (expensive) lookup in
the SpeciesConstructor operation. This yields in a nice 3-5% improvement
on the bluebird and wikipedia benchmarks, and paves the way for inlining
certain Promise operations into TurboFan optimized code later.
On the micro-benchmark mentioned in the bug (from the findings doc), we
reduce the overall execution time by 25%, which makes sense given that
Promise.prototype.then spends a significant portion of it's time just
figuring out the appropriate constructor.
Bug: v8:7253, v8:7349
Change-Id: Ia1577b59d1b7e4b8dbda83e2186583edab76695a
Reviewed-on: https://chromium-review.googlesource.com/880681
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50794}
The predicatable mode sets --single-threaded flag, which disables
--wasm-async-compilation. The test relies on async compilation.
Change-Id: I49dae829506c69f21f148cc9c9565c136abcda42
Reviewed-on: https://chromium-review.googlesource.com/880842
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50793}
Async compilation relies on background tasks, which are not available
in single-threaded mode.
Change-Id: Id47e7f5bd1406a5323d6a770936a23b6839e7936
Reviewed-on: https://chromium-review.googlesource.com/880583
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50789}
The extra output on mac was to investigate a bug that's now fixed.
TBR=sergiyb@chromium.org
NOTRY=true
Bug: v8:6927
Change-Id: Iac8074c7e89a987e164f676442da1ed4f93987cc
Reviewed-on: https://chromium-review.googlesource.com/880623
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50787}
Since we're not winning anything by changing the result between
processors on the main process, reduce is noop there and result is
immutable.
Bug: v8:6917
Change-Id: Ieb282e7abd4ab31162aee6b52493a6e1b6a25109
Cq-Include-Trybots: luci.v8.try:v8_linux64_fyi_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/878239
Commit-Queue: Michał Majewski <majeski@google.com>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50784}
Simplified lowering may loose feedback by inserting Checked
conversions for BoundsChecks in case the bounds check gets
optimized away later on.
Bug: v8:7127
Change-Id: I254a29ba4e578d653d1dee2d70582ce0a4b57789
Reviewed-on: https://chromium-review.googlesource.com/878743
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50783}
The common operator reducer was loosing feedback information
when replacing DeoptimizeIf/Unless with DeoptimizeUnless/If nodes.
Bug: v8:7127
Change-Id: I5d6f253ca9dfec04f4e7c8d1485f0ca668a8db95
Reviewed-on: https://chromium-review.googlesource.com/878781
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50782}
This change allows turbofan to store the index variable in
a Word32 integer; my last change caused the index to be
promoted to float64.
Change-Id: Ia60d16461991ae941229f45d393e924d0dca5bbe
Reviewed-on: https://chromium-review.googlesource.com/878160
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50780}
* Set gdeepti@ as a compiler backend owner.
* Set sigurds@ as a compiler owner.
Change-Id: I0ceb300b6452a36a7f945e92193a3946604c4e51
Reviewed-on: https://chromium-review.googlesource.com/878542
Reviewed-by: Ben Titzer <titzer@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Brad Nelson <bradnelson@chromium.org>
Commit-Queue: Brad Nelson <bradnelson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50777}
Deoptimization may materialize values on the heap, which may get sampled
by the heap profiler. Such samples have imprecise stack. Indicate this.
BUG=v8:7314
Change-Id: I21ab079c36fc0492b05b546cc1d6a8e6c042aeb8
Reviewed-on: https://chromium-review.googlesource.com/877119
Commit-Queue: Ali Ijaz Sheikh <ofrobots@google.com>
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50775}
Port 5cef3ddd5f
Original Commit Message:
Instead of requiring the pattern that a SuspendGenerator must be
followed by a Return, make SuspendGenerator return directly. This can,
in the future, simplify some of the reasoning around generator suspends.
R=leszeks@chromium.org, joransiu@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=
LOG=N
Change-Id: Ic967164b40433ed64df5996560490b13d6159a94
Reviewed-on: https://chromium-review.googlesource.com/879223
Reviewed-by: Joran Siu <joransiu@ca.ibm.com>
Commit-Queue: Junliang Yan <jyan@ca.ibm.com>
Cr-Commit-Position: refs/heads/master@{#50774}
This removes the field in question to make it simpler to serialize and
deserialize modules without having to worry about the state of lazy
compilation. It is always possible to clone a non-anonymous builtin,
even without having this module-wide field.
R=clemensh@chromium.org
TEST=mjsunit/regress/wasm/regress-803427
BUG=chromium:803427
Change-Id: I72041e314eb6ee92859d45f1db0ed8500003edc4
Reviewed-on: https://chromium-review.googlesource.com/878581
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50771}
Now that write-protection of code memory is enabled everywhere and V8 is
fully W^X compliant, we can remove the permission mode in question.
R=hpayer@chromium.org
BUG=v8:6792
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I80fe95ac6bb0e2d1ad6d993154ce45d492d941be
Reviewed-on: https://chromium-review.googlesource.com/866855
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Reviewed-by: Bill Budge <bbudge@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50770}
During GC fuzzing we combine multiple tests and run them inside
a wrapper that needs to ignore all errors/exceptions/asserts to
keep the combined tests running. We will use this flag to ignore
%AbortJS calls.
Bug: v8:6917
Change-Id: Ib426a68228cadbea8364c5e1d29c39dd53129481
Reviewed-on: https://chromium-review.googlesource.com/857514
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michał Majewski <majeski@google.com>
Cr-Commit-Position: refs/heads/master@{#50768}
- JSObject: Record elements and properties
- JSCollecton: Record table
- Record global caches
Bug: v8:7266
Change-Id: I16b2eb511bed3dc0fb6f7af0e7037c6d42f03885
Reviewed-on: https://chromium-review.googlesource.com/878326
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50766}
When a tab in Chrome gets refreshed, the refreshed page reuses the
isolate of the original page. This means that at the moment,
AsyncCompileJobs which were stared on the original page do not get
aborted and will therefore eventually finish and resolve their promise.
With this CL I abort all running AsyncCompileJobs when V8 gets the tab
refresh signal, i.e. Isolate::ContextDisposedNotification. Note that I
cannot just call CompilationManager::TearDown because it assumes that
there are no pending tasks anymore.
R=clemensh@chromium.org, hpayer@chromium.org
Bug: chromium:803476
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: I88d28fdaba6f55b7aa7379c4b5338ae62134fc8a
Reviewed-on: https://chromium-review.googlesource.com/875923
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50765}
