This is a preliminary step before we stop swapping maps in the bootstrapper
(strict/sloppy map with writable prototype <-> readonly prototype).
Bug: v8:6459
Change-Id: I120550c10e98a234e283d79a8d408096601c92af
Reviewed-on: https://chromium-review.googlesource.com/558879
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46403}
Regexp code flushing support has been unintentionally disabled for quite a
while without any signals on our benchmarks. This CL completely removes
support.
BUG=v8:6569
Change-Id: Ic018a115c38387ff5610b34d3c09cb360e30ad3f
Reviewed-on: https://chromium-review.googlesource.com/559331
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46401}
Did only work with full MCs. Since the majority of GCs is now
incremental and will be concurrent in future this becomes obsolete.
Bug: v8:6569
Change-Id: I28280a71dd4e779742dd4d1b3dd01087cd4f8fe0
Reviewed-on: https://chromium-review.googlesource.com/558983
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46399}
Instead of having the "next step" state machine logic in the compiler
dispatcher, move it to the compiler dispatcher job. This is a
prerequisite to having more generic jobs with slightly different state
sets, such as an optimizing compile job.
Bug: v8:6537
Change-Id: Ib7319222f26dbc0e2afc95573b06d813825afde3
Reviewed-on: https://chromium-review.googlesource.com/554751
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46396}
Every, Some, Reduce, ReduceRight. Added a test that should improve
when TurboFan inlines these builtins. Updated Map and Filter tests
to include a TurboFan inline test.
Bug: v8:2229
Change-Id: Ie84d414fdcccea23c734caca55a3344f9442547f
Reviewed-on: https://chromium-review.googlesource.com/558935
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Michael Stanton <mvstanton@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46395}
v8.cpu_profiler has a long and variable warm up period. If you look in
tracing that time shows up under V8.Execute which is very misleading. By
adding a tracing category for the time spent in
CpuProfiler::StartProfiling the warm up time can be easily identified or
even filtered out.
With this tracing event the block in StartProfiling correctly described
the time spent prepare v8.cpu_profiler out from V8.Execute
Test by collecting a trace with V8.Execute and look at when a trace
starts generating v8 samples.
BUG=chromium:733853
Review-Url: https://codereview.chromium.org/2950543002
Cr-Commit-Position: refs/heads/master@{#46393}
This reverts commit 8840d622aa.
Reason for revert: This requires the infra-side to be adapted to look
for open rolls. There's no such feature in the gerrit recipe_module
or git-cl tooling yet.
Original change's description:
> [release] Make gerrit the default for all release tools
>
> This switches also the V8->Chromium rolls to Gerrit.
>
> NOTRY=true
> TBR=hablich@chromium.org
>
> Bug: chromium:685318
> Change-Id: Idc168f790541f09bd2f2d7c2f72806ac9e966843
> Reviewed-on: https://chromium-review.googlesource.com/558913
> Reviewed-by: Michael Achenbach <machenbach@chromium.org>
> Commit-Queue: Michael Achenbach <machenbach@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#46388}
TBR=machenbach@chromium.org,hablich@chromium.org,tandrii@chromium.org
Change-Id: I597538b6165b9952b5df9cde72466b95739cf56b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:685318
Reviewed-on: https://chromium-review.googlesource.com/558225
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46390}
The release tools used to reload CL descriptions from rietveld to consider
late edits. This makes no sense anymore with gerrit, so we drop the feature.
NOTRY=true
TBR=hablich@chromium.org
Bug: chromium:685318
Change-Id: I08231795ba3b25c0939aa2b4428973086548484d
Reviewed-on: https://chromium-review.googlesource.com/558915
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46389}
This switches also the V8->Chromium rolls to Gerrit.
NOTRY=true
TBR=hablich@chromium.org
Bug: chromium:685318
Change-Id: Idc168f790541f09bd2f2d7c2f72806ac9e966843
Reviewed-on: https://chromium-review.googlesource.com/558913
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46388}
On Loongson 3A, MADD/MSUB instructions are actually fused MADD/MSUB and
they can cause failure in some of the tests. Since this optimization is
rarely used, and not used at all on MIPS64R6, MADD/MSUB instructions
are removed from the source base.
TEST=
BUG=
Change-Id: Ifbb5508a62731bb061f332864ffd1e210e97f963
Reviewed-on: https://chromium-review.googlesource.com/558066
Reviewed-by: Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com>
Commit-Queue: Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com>
Cr-Commit-Position: refs/heads/master@{#46387}
The test runner automatically infers certain options from the build
product if it was generated with GN.
This CL makes the code for inferring the options more generic and makes
sure that:
1) boolean options can't be set on builds where they don't make sense
(e.g. can't pass --asan in a non-asan build)
2) string options are equal to the derived option from the build
(e.g. can't test arch x64 when build is x86)
Bug: v8:5533
Change-Id: I4badf3a17a2fc23ddb9d129602aa15f12665821f
Reviewed-on: https://chromium-review.googlesource.com/552542
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46385}
- Avoid static methods.
- Instantiate the actual visitor on the stack.
- Get rid of unnecessary abstractions.
Bug: chromium:738865
Change-Id: I4115d7b88f17a7118aa9ee129eb39a28ec413696
Reviewed-on: https://chromium-review.googlesource.com/558878
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46384}
This uploads roll-branch CLs before landing and fixes an
ambiguity bug when pushing tags.
NOTRY=true
TBR=hablich@chromium.org
Bug: chromium:738679
Change-Id: I51e03d714ec97ee2d9bca4ecaf753d66038891c1
Reviewed-on: https://chromium-review.googlesource.com/558415
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46382}
When providing scope information (containing the value of local
variables of live stack frames), decode the local variable names of all
functions in a wasm module and store this in the WasmDebugInfo
structure.
Use these names to actually name the reported locals, instead of using
the default names "param#<d>" and "local#<d>". These names are only used
as fallbacks for locals which were not assigned a name.
R=titzer@chromium.org,kozyatinskiy@chromium.org
BUG=v8:6245
Change-Id: Ibf7d30e392248ef5590177cd8b6329239b45e018
Reviewed-on: https://chromium-review.googlesource.com/548495
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Ben Titzer <titzer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46379}
Instructions ins and ext didn't work properly when size = 32 because of
incorrect mask initialization, this CL fixes this. A test for Ins is
also added.
BUG=
Change-Id: I95cc8e13aaa2341b34ae59dae1eefb64c551b8b4
Reviewed-on: https://chromium-review.googlesource.com/558872
Commit-Queue: Miran Karić <Miran.Karic@imgtec.com>
Reviewed-by: Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com>
Cr-Commit-Position: refs/heads/master@{#46378}
Removes the ability of allocating dominators and folded allocations
which was tied to Crankshaft's allocation folding.
Bug: v8:6408
Change-Id: Id2e1b5445c8357ac770c88e734b6c50d5f6c5eae
Reviewed-on: https://chromium-review.googlesource.com/558093
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46377}
The jsapi-harness test runs the JS-API spec tests of WebAssembly, which
get fetched from github when 'gclient sync' is executed. Without
'gclient sync' the harness may executed a version of the tests which is
older than required by the harness. In this CL I add a suggestion to
the test to run 'gclient sync' which is shown when the test is failing.
R=marja@chromium.org
Change-Id: I36d03bebc4d6cc554eefd4eb376c3d309b7ee5b9
Reviewed-on: https://chromium-review.googlesource.com/558419
Reviewed-by: Marja Hölttä <marja@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46375}
(The test that catches the bug was test-bytecode-generator/LookupSlot)
BUG=v8:5516
Change-Id: I00a02c5326b2a132383a9d72b5b894fade53bbf2
Reviewed-on: https://chromium-review.googlesource.com/558864
Commit-Queue: Marja Hölttä <marja@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46374}
This sometimes caused problems with bots (for node too) because the
allocation could fail.
Bug: v8:6452
Change-Id: I346a9117eba8b6ed41566efeceaf7fb190784d76
Reviewed-on: https://chromium-review.googlesource.com/558075
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Commit-Queue: Peter Marshall <petermarshall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46373}
Reason for revert:
Speculative revert for tiny fire on Canary (crbug.com/738781)
Original issue's description:
> [turbofan] Extend escape analysis to reduce certain CheckMaps.
>
> Enable the experimental support in escape analysis to deal with
> constant-foldable CheckMaps nodes and remove them from the effect
> chain w/o blocking the scalar replacement of the object.
>
> BUG=v8:4586,v8:5267
> R=tebbi@chromium.org
>
> Review-Url: https://codereview.chromium.org/2964473002
> Cr-Commit-Position: refs/heads/master@{#46311}
> Committed: adf7f54e24TBR=tebbi@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=v8:4586,v8:5267
Review-Url: https://codereview.chromium.org/2970663002
Cr-Commit-Position: refs/heads/master@{#46372}
These were originally written as part of
https://chromium-review.googlesource.com/c/550396/. I've separated them
out into a separate CL with the intent of landing it first, so that it's
easier to see the difference these CLs will have on generated bytecode.
BUG=v8:5855
TBR=tebbi@chromium.org, rmcilroy@chromium.org
Change-Id: Ib84e65847d7396e31b0e38d28f59454cf7c58fc1
Reviewed-on: https://chromium-review.googlesource.com/558221
Reviewed-by: Caitlin Potter <caitp@igalia.com>
Commit-Queue: Caitlin Potter <caitp@igalia.com>
Cr-Commit-Position: refs/heads/master@{#46370}
`IsHoleyElementsKind` doesn’t just check for holeyness — it checks for
dictionary elements as well. Its name should reflect that.
This patch renames `IsHoleyElementsKind` to
`IsHoleyOrDictionaryElementsKind`, which makes it possible to rename
`IsFastHoleyElementsKind` to `IsHoleyElementsKind` in a future patch.
R=jkummerow@chromium.org, cbruni@chromium.org
BUG=v8:6548
Change-Id: Id799fe396442e9810426145359254d60990f8492
Reviewed-on: https://chromium-review.googlesource.com/558344
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Mathias Bynens <mathias@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46367}
This CL updates the wasm spec tests. In addition it adds an
assertNotEquals function to mjsunit.js, and it fixes the test harness
to not call quit() because it causes a dead-lock in combination with
async compilation.
R=rossberg@chromium.org
Change-Id: I50cf737993adb3e2bd27977efe7e20e304b89078
Reviewed-on: https://chromium-review.googlesource.com/558077
Reviewed-by: Andreas Rossberg <rossberg@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46366}
Replace the second level visitation with a much simpler logic that
just separately dispatches the special cases. All other cases can
use a dispatch that just evacuates an object based on size.
This is similar to the logic used in the mark-compact collector. The
goal is to align behaviors as much as possible, highlighting and
fixing performance issues in the different behaviors.
This CL is mechanical as possible. A followup will clean
up the naming scheme and dispatching.
Bug: chromium:738368
Change-Id: Ia5a426c5ebb25230000b127580c300c97cff8b1b
Reviewed-on: https://chromium-review.googlesource.com/558060
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46364}
The `FAST_` prefix doesn’t make much sense — they’re all just different cases
with their own optimizations. Packedness being implicit (e.g. `FAST_ELEMENTS`
vs. `FAST_HOLEY_ELEMENTS`) is not ideal, either.
This patch renames the FAST elements kinds as follows:
- e.g. FAST_ELEMENTS => PACKED_ELEMENTS
- e.g. FAST_HOLEY_ELEMENTS => HOLEY_ELEMENTS
The following exceptions are left intact, for lack of a better name:
- FAST_SLOPPY_ARGUMENTS_ELEMENTS
- SLOW_SLOPPY_ARGUMENTS_ELEMENTS
- FAST_STRING_WRAPPER_ELEMENTS
- SLOW_STRING_WRAPPER_ELEMENTS
This makes it easier to reason about elements kinds, and less confusing to
explain how they’re used.
R=jkummerow@chromium.org, cbruni@chromium.org
BUG=v8:6548
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Change-Id: Ie7c6bee85583c3d84b730f7aebbd70c1efa38af9
Reviewed-on: https://chromium-review.googlesource.com/556032
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Mathias Bynens <mathias@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46361}
Allocates the Await success/failure closures, their context, and
the two required JSPromise objects all at once in a single call,
rather than performing multiple allocations throughout the function.
Saves about 2kb of snapshot space on an x64.release build.
Performance impact of this change has not been measured yet.
BUG=v8:4483
R=ishell@chromium.org, jgruber@chromium.org
Change-Id: I8d911cb91f5d0e00544ad3ba608aa170f6b2f704
Reviewed-on: https://chromium-review.googlesource.com/549999
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Caitlin Potter <caitp@igalia.com>
Cr-Commit-Position: refs/heads/master@{#46360}
Catch prediction for optimized frames had two issues:
Inlined frames were iterated from caller-to-callee (which could
result in incorrect predictions if one frame predicted CAUGHT and
another predicted PROMISE).
When encountering a builtin frame, we'd unconditionally return its
prediction (which is wrong if it predicted UNCAUGHT and another inlined
frame predicted either CAUGHT or PROMISE).
This CL fixes both issues and refactors the function to reduce nesting.
BUG=v8:6536
Change-Id: I764a4ec033e4476bd840134b5eacfe0e08b3c1a4
Reviewed-on: https://chromium-review.googlesource.com/555519
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46358}
V8's catch prediction mechanism tries to predict whether a thrown
exception will be caught, just by looking at the current call stack.
At the time when catch prediction was first introduced, only a few
builtins (mostly related to Promise and Generator) could end up being
fed into the catch prediction mechanism. This is no longer the case now
that builtins are used in new ways, e.g. Array.p.forEach's continuation
builtins.
This CL removes the need to explicitly mark all builtins visible to the
StackFrameIterator as CAUGHT/UNCAUGHT/PROMISE, and instead defaults to
treating unmarked builtins as UNCAUGHT.
BUG=v8:6536
Change-Id: Ibdc106a91b2b0ffb93099433077642cad02c71e2
Reviewed-on: https://chromium-review.googlesource.com/555518
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46357}
... and set the instance class name in a bootstrapper instead.
Change-Id: Ie8a9a0e7cdc22ca19616b4a0d09665e059cd4d3e
Reviewed-on: https://chromium-review.googlesource.com/557864
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#46356}
