This brings the graph builder in sync with the serializer (and
exponentiation in sync with the other binary operators).
Bug: chromium:995430, v8:7790
Change-Id: I809b6f3756f75392cdc6747f8bcee8cdf0ee0f74
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1762013
Commit-Queue: Georg Neis <neis@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63269}
... by making the operator have a control output, since we could deopt
after my last change.
Bug: chromium:995562, v8:7790
Change-Id: Ibc8c44708b4d43c4b2c3dfab2fd8fdf79c7ea671
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1762010
Commit-Queue: Georg Neis <neis@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63268}
Chromium has checks which don't like static initializers in binaries,
which fires on effect_control_linearizer.cc. We can remove these by
making kMinusZeroBits (and family) constexpr, but to do this we have to
avoid bit_cast. Instead, set the correct bit pattern manually (thankfully
IEEE 754 0.0 is just zero bits, and -0.0 is 0.0 with a set sign bit).
Change-Id: If1695ff715ad8f821e956757f8f9f7c850895011
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1762009
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63266}
This is a very small change which must go in before
https://chromium-review.googlesource.com/c/chromium/src/+/1731108
is able to land.
Certain WPT tests for synthetic modules were spawning DCHECK crashes by JSObject::SetNormalizedProperty.
Export names were previously failing:
DCHECK(name->IsUniqueName());
This small change corrects the issue and allows Module::GetModuleNamespace to run correctly.
This change aligns synthetic module behavior for export string storage with JS modules, as well as the spec.
chromium: 967018
Change-Id: I151e7150290bd72d4e4753c8c5be243eafae915f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1761583
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63264}
This change is very mechanical:
own<Foo*> → own<Foo>
vec<Foo*> → ownvec<Foo>
As usual, everything in third_party/ is straight-up copied from upstream.
Change-Id: If5fabda99e2b281da6f2e71ce23a2f5b68aaac86
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1760815
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63263}
Mozilla suite acts as the bottleneck for this bot.
This CL bumps the shards for the Mozilla suite.
TBR=machenbach@chromium.org
Bug: v8:9633
Change-Id: Ia73b2688b29a3387cc69fa8f79c008f6bf0114cd
No-Try: True
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1760824
Reviewed-by: Tamer Tas <tmrts@chromium.org>
Commit-Queue: Tamer Tas <tmrts@chromium.org>
Auto-Submit: Tamer Tas <tmrts@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63262}
Adds support to the CPU profiler for scraping the incumbent contexts of
V8 stack frames. While it is generally unsafe to access heap objects
during a profiling interrupt, the native context is uniquely usable due
to being guaranteed an alive root on the stack, as well as its slots
being immutable after context creation.
Change-Id: I2c3149c1302b74d2f13aa99d1fdd0cf006e0f9d1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1580020
Commit-Queue: Andrew Comminos <acomminos@fb.com>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Alexei Filippov <alph@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Peter Marshall <petermarshall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63258}
Rewrite the reducer in terms of the ordinary keyed-store reducer and
reuse the existing serializer machinery for that as well.
Bug: v8:7790
Change-Id: I5909739feee1d77dca1827166bad3d2a61561784
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1760807
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63254}
This reverts commit 60843b426b.
Reason for revert: TSAN detected issue between Scavenge workers.
One task could invoke RefillFreeList(), while the other task iterates the remembered set of a swept page.
Original change's description:
> Use list of invalidated objects for old-to-new refs
>
> Instead of inserting "deletion" entries into the store buffer, keep a
> list of invalidated objects to filter out invalid old-to-new slots.
>
> The first CL https://crrev.com/c/1704109 got reverted because both the
> sweeper and the main task were modifying the invalidated slots data
> structure concurrently. This CL changes this, such that the sweeper
> only modifies the invalidated slots during the final atomic pause when
> the main thread is not running. The sweeper does not need to clean this
> data structure after the pause, since the "update pointers" phase
> already removed all invalidated slots.
>
> The second CL https://crrev.com/c/1733081 got reverted because the
> sweeper might find more free space than the full GC before it. If an
> object shrinks after the pause but before the sweep, the invalidated
> object might span free memory and potentially new allocated objects.
> Therefore shrink invalidated objects when processing swept pages on
> the main thread. Also clean recorded slots in the gap.
>
> TBR=petermarshall@chromium.org
>
> Bug: v8:9454
> Change-Id: I80d1fa3bbc24e97f7c97a373aaad66f105456f12
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1751795
> Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63239}
TBR=ulan@chromium.org,hpayer@chromium.org,dinfuehr@chromium.org
Change-Id: I9c6a371ebe36a1873acbe0d6c6a75dd2f5a55f4e
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:9454
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1760817
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63253}
- Adds a histogram to measure time between streaming start and
deserialization finished.
Bug: chromium:719172
Change-Id: Ib4ce24bee05a0db7e1bbf50d3bf456af89dbd2a6
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1754721
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Bill Budge <bbudge@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63252}
The optional chaining bytecode in delete expressions was
unconditionally jumping if the receiver was nullish, instead
of just when the property was an actual optional chain link.
This change adds the missing check around the jump.
Change-Id: Ic7bed58be4ae62d157e63e4f77666b1abd1f802d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1755264
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63251}
Flag off by default now since it's causing crashes in the profiler.
R=rmcilroy
Bug: chromium:994673
Change-Id: I92b46e1f90819c0007106d843ecae2c3974eb3c8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1760814
Commit-Queue: Dan Elphick <delphick@chromium.org>
Commit-Queue: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63250}
This is a reland of 82111e2286
Relanding since we now have more shards:
https://chromium-review.googlesource.com/c/v8/v8/+/1760810
Original change's description:
> [CSA][cleanup] TNodify some methods related to prototype and property lookup
>
> This is a CL in a string of CLs that aims to TNodify CSA. In particular,
> there were some loads that were done in AnyTagged instead of
> TaggedPointer. TNode-ifying them brings improvement in pointer
> compression since we are able to decompress using the Pointer
> decompression.
>
> TNodified:
> * LoadJSFunctionPrototype
> * TryPrototypeChainLookup
> * OrdinaryHasInstance
>
> Also TNodified loads regarding:
> * FeedbackCell::kValueOffset
> * HeapObject::kMapOffset
> * JSFunction::kSharedFunctionInfoOffset
> * JSFunction::kFeedbackCellOffset
> * Map::kInstanceTypeOffset
> * Map::kInstanceDescriptorsOffset
> * Map::kPrototypeOffset
>
> Drive-by cleanup: StoreJSArrayLength and StoreElements were unused.
>
> Bug: v8:6949, v8:9396
> Change-Id: I89697b5c02490906be1eee63cf3d9e60a1094d48
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1755844
> Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63216}
Bug: v8:6949, v8:9396
Change-Id: I040aefcf8af60611f7b3c24f3bd5c661e03b6ada
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1760811
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63249}
The following methods didn't need the use of context, and could
be replaced by NoContextConstant():
* AllocateSeqOneByteString
* AllocateSeqTwoByteString
* StringBuiltinsAssembler::GenerateStringEqual
* StringBuiltinsAssembler::StringEqual_Core
* StringBuiltinsAssembler::GenerateStringRelationalComparison
Change-Id: I98068980377450daef7c999e3d413e839f66fda9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758321
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63246}
DoComputeInterpretedFrame and friends are long and complex functions.
It is often not clear which variables are constants and which are
later modified. This CL tries to clarify, mostly by marking variables
const when possible.
Bug: v8:9534
Change-Id: Ifa73402c392ad244ab5ea37262293f8d9db98be0
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1752848
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63245}
IsolateAllocator::InitReservation can fail with OOM if multiple V8
isolates are created simultaneously and race to reserve the same
memory region.
Now the function falls back to using overreserved region as the
last resort.
Bug: v8:9588
Change-Id: I9731e04181382f0c1401b2a78f3eba63a445bfc9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758315
Auto-Submit: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Santiago Aboy Solanes <solanes@chromium.org>
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63243}
- Eliminate unconditional heap reads in tracing code.
- Change operator<< on ObjectRef to additionally print
the Brief() output when the broker is disabled.
- Print line number in TRACE_BROKER_MISSING and make
some messages more consistent.
- Make PrintCandidates output clearer.
- Be more consistent about dereferencing optionals.
Bug: v8:7790, chromium:990478
Change-Id: I2917529d5138a0d63ad476d3f8fee6a963767b23
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758311
Commit-Queue: Georg Neis <neis@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63242}
This is a reland of 29585a06cc after
removing an incorrect DCHECK.
Original change's description:
> [turbofan] Various serializer/broker improvements
>
> They are all somewhat entangled, sorry for the big CL.
>
> - Brokerize remaining feedback vector slots.
> - Introduce Hints::SingleConstant helper.
> - Introduce SerializationPolicy enum.
> - Eliminate use of nullptr for megamorphic load/store ic feedback.
> Instead use the corresponding ProcessedFeedback with an empty list
> of maps or the like. new class MegamorphicFeedback.
> - Separate processing of feedback from serialization. This eliminates
> code duplication.
> - Be very careful when clearing hints not to overwrite hints that are
> being processed.
> - Move AccessInfos out of NamedAccessFeedback. Always store them in
> property_access_infos_ map on broker. (This was actually unused
> before, somewhat by mistake.)
> - Support map inference in concurrent inlining. Rewrite
> ElementAccessFeedback such that we can refine it with the set of
> inferred maps.
>
> TBR: mvstanton@chromium.org
> Change-Id: I05e9eb250bdffc6dff29db01742550a86a41cb31
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1752853
> Commit-Queue: Georg Neis <neis@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63232}
TBR: mvstanton@chromium.org
Bug: v8:7790
Change-Id: Ia4acd31b339a941ee065e1ae4835bb7b85d5685e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758319
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63241}
Test mjsunit/regress/regress-992389 explicitly sets the jitless flag
when run.
Skip this test when run on builds without embedded-builtins.
Bug: v8:9632, chromium:992389
Change-Id: Ieb52a33006b1104080d8f5adb8c4f2c36e4413af
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758317
Commit-Queue: Patrick Thier <pthier@google.com>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63240}
Instead of inserting "deletion" entries into the store buffer, keep a
list of invalidated objects to filter out invalid old-to-new slots.
The first CL https://crrev.com/c/1704109 got reverted because both the
sweeper and the main task were modifying the invalidated slots data
structure concurrently. This CL changes this, such that the sweeper
only modifies the invalidated slots during the final atomic pause when
the main thread is not running. The sweeper does not need to clean this
data structure after the pause, since the "update pointers" phase
already removed all invalidated slots.
The second CL https://crrev.com/c/1733081 got reverted because the
sweeper might find more free space than the full GC before it. If an
object shrinks after the pause but before the sweep, the invalidated
object might span free memory and potentially new allocated objects.
Therefore shrink invalidated objects when processing swept pages on
the main thread. Also clean recorded slots in the gap.
TBR=petermarshall@chromium.org
Bug: v8:9454
Change-Id: I80d1fa3bbc24e97f7c97a373aaad66f105456f12
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1751795
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63239}
Information required for deoptimization is passed from codegen to the
deoptimizer through so-called translations. Translations contain,
among many other things, a 'height' field. It is used during deopts to
calculate the unoptimized frame height (but note that it does not
correspond exactly to the frame height itself - further calculations
on the deopt side are needed to get to the real frame height).
The height field has roughly the following data flow:
1. During codegen, we serialize whatever
FrameStateDescriptor::GetHeight() returns.
2. During deopts, serialized translations are converted into
TranslatedFrame objects in TranslatedState::CreateNextTranslatedFrame.
3. These are later used to arrive at the real frame height in multiple
spots, e.g. in DoComputeInterpretedFrame and friends.
Prior to this CL, we were adding and subtracting 1 in basically random
spots. For example, for interpreted and construct stub frames we added
1 in step 1 and subtracted 1 in step 3. For continuation frames, we
added 1 in step 2 and subtracted it in step 3. Argument adaptor frames
were left untouched.
This CL removes all these +-1's. The height field now contains
locals_count() for interpreted frames, and parameters_count() for
everything else. I also tried to make the meaning of adds/subs clearer
through use of named constants like kTheReceiver.
Bug: v8:9534
Change-Id: I6fd26886ff5aa63930f413d879d5480578d9dc7e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1751724
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63238}
This reverts commit 29585a06cc.
Reason for revert: Breaks GC stress bots -
https://ci.chromium.org/p/v8/builders/ci/V8%20Linux%20-%20gc%20stress/24009https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20GC%20Stress%20-%20custom%20snapshot/27281
Original change's description:
> [turbofan] Various serializer/broker improvements
>
> They are all somewhat entangled, sorry for the big CL.
>
> - Brokerize remaining feedback vector slots.
> - Introduce Hints::SingleConstant helper.
> - Introduce SerializationPolicy enum.
> - Eliminate use of nullptr for megamorphic load/store ic feedback.
> Instead use the corresponding ProcessedFeedback with an empty list
> of maps or the like. new class MegamorphicFeedback.
> - Separate processing of feedback from serialization. This eliminates
> code duplication.
> - Be very careful when clearing hints not to overwrite hints that are
> being processed.
> - Move AccessInfos out of NamedAccessFeedback. Always store them in
> property_access_infos_ map on broker. (This was actually unused
> before, somewhat by mistake.)
> - Support map inference in concurrent inlining. Rewrite
> ElementAccessFeedback such that we can refine it with the set of
> inferred maps.
>
> TBR: mvstanton@chromium.org
> Change-Id: I05e9eb250bdffc6dff29db01742550a86a41cb31
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1752853
> Commit-Queue: Georg Neis <neis@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63232}
TBR=mvstanton@chromium.org,neis@chromium.org
Change-Id: I88625d92fddf993db63661666c59af05a47b2b58
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758314
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63237}
The --jitless and --interpreted-frames-native-stack flags are
incompatible since the latter requires code generation while the
former prohibits code generation.
Bug: v8:9619
Change-Id: Ic954724edd6a2d28e1bf2f6a79649f86e812abcf
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758312
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63235}
This removes the last dependencies on {JSGraph} from the wasm compiler.
Since the pipeline does not run appropriate lowerings, access to the
simplified or the javascript operator level is disallowed. With this
change that rule is statically enforced.
R=thibaudm@chromium.org
Change-Id: I4a97e63ffc5df1b7704477ac5c68707f5b195166
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758305
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63234}
This reverts commit 82111e2286.
Reason for revert: Speculative revert, could be causing timeouts - https://ci.chromium.org/p/v8/builders/ci/V8%20Linux%20-%20arm%20-%20sim%20-%20debug/17567
Original change's description:
> [CSA][cleanup] TNodify some methods related to prototype and property lookup
>
> This is a CL in a string of CLs that aims to TNodify CSA. In particular,
> there were some loads that were done in AnyTagged instead of
> TaggedPointer. TNode-ifying them brings improvement in pointer
> compression since we are able to decompress using the Pointer
> decompression.
>
> TNodified:
> * LoadJSFunctionPrototype
> * TryPrototypeChainLookup
> * OrdinaryHasInstance
>
> Also TNodified loads regarding:
> * FeedbackCell::kValueOffset
> * HeapObject::kMapOffset
> * JSFunction::kSharedFunctionInfoOffset
> * JSFunction::kFeedbackCellOffset
> * Map::kInstanceTypeOffset
> * Map::kInstanceDescriptorsOffset
> * Map::kPrototypeOffset
>
> Drive-by cleanup: StoreJSArrayLength and StoreElements were unused.
>
> Bug: v8:6949, v8:9396
> Change-Id: I89697b5c02490906be1eee63cf3d9e60a1094d48
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1755844
> Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#63216}
TBR=rmcilroy@chromium.org,solanes@chromium.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: v8:6949, v8:9396
Change-Id: Ib6ae8fe86a598ed1066894595565e1162cf7dd1f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758310
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Reviewed-by: Santiago Aboy Solanes <solanes@chromium.org>
Commit-Queue: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63233}
They are all somewhat entangled, sorry for the big CL.
- Brokerize remaining feedback vector slots.
- Introduce Hints::SingleConstant helper.
- Introduce SerializationPolicy enum.
- Eliminate use of nullptr for megamorphic load/store ic feedback.
Instead use the corresponding ProcessedFeedback with an empty list
of maps or the like. new class MegamorphicFeedback.
- Separate processing of feedback from serialization. This eliminates
code duplication.
- Be very careful when clearing hints not to overwrite hints that are
being processed.
- Move AccessInfos out of NamedAccessFeedback. Always store them in
property_access_infos_ map on broker. (This was actually unused
before, somewhat by mistake.)
- Support map inference in concurrent inlining. Rewrite
ElementAccessFeedback such that we can refine it with the set of
inferred maps.
TBR: mvstanton@chromium.org
Change-Id: I05e9eb250bdffc6dff29db01742550a86a41cb31
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1752853
Commit-Queue: Georg Neis <neis@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63232}
- Move undef closer to end of usage
- Move I64x2ExtractWithF64x2 closer to Extract tests, and into ifdef
scope so it runs on arm64 builds
Change-Id: I7138c44097975d02e97f4b2b9bfcddd8eb9735c8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1754544
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63227}
This CL adds additional information in PropertyAccessInfos and FieldAccesses about the map that introduced the accessed field. We use this information to prevent load elimination from incorrectly optimizing certain accesses marked const.
Prior to this CL, load elimination simply stored information about eliminatable field accesses based on objects (identified by nodes in the graph) and offsets (i.e., statically known ones). In the presence of const stores and loads this is insufficient, since a single object (in the above sense) may contain distinct *const* properties at the same offset throughout its lifetime. As an example, consider the following piece of code:
let obj = {};
obj.a = 0;
obj[1024] = 1; // An offset of >=1024 forces an elements-kind transition
delete obj.a;
obj.b = 2;
assertEquals(obj.b, 2);
In this scenario, *both* the first ('obj.a = 0') and the second ('obj.b = 2') store to a field will be marked const by the runtime. The reason that storing to 'a' above ends up being marked const, is that 'a' before and after the elements-kind transition is encoded in separate transition trees. Removing 'a' ('delete obj.a') only invalidates const-ness in the dictionary-elements transition tree; not the holey-elements one used at the time of 'obj.a = 0'.
The above situation on its own violates an invariant in load elimination. Namely, we assume that for the same object and offset, we will never encounter two const stores. One can extend the above snippet to coax load-elimination into producing incorrect results. For instance, by "hiding" 'obj.b = 2' in an unoptimized function call, the consecutive load from 'b' will incorrectly produce 0, violating the assert.
R=neis@chromium.org, tebbi@chromium.org
Bug: chromium:980183, chromium:983764
Change-Id: I576a9c7efd416fa9db6daff1f42d483e4bd369b4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1751346
Commit-Queue: Georg Schmid <gsps@google.com>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63226}
V8_EXPORT is not required for public class templates, and using it may
lead to linker errors when users attempt to dllimport V8.
Change-Id: I0b2db2371d0fcbf50af94ba629670486b8f2bc00
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1757284
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63225}
GetIterator currently acts as a property load of the iterator symbol
(soon it will also call it). It makes sense to apply the same early
lowering logic as we do for property loads in the bytecode graph
builder. This also brings our treatment of the bytecode in-line with
the way it's treated in the serializer, which already respects the
early-lowering semantics.
Bug: v8:7790
Change-Id: Ieadc4b307b9f6d9a5aa77ca10c7c818026776f33
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1758304
Commit-Queue: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63224}
There are some accesses that we know that are pointers, and we can
specialize them to have Pointer as their Machine Type and Write Barrier.
Drive-by fix: ForJSGeneratorObjectParametersAndRegisters can be changed
to use Pointer.
Bug: v8:7703
Change-Id: I88e371746fc5f08c84795c95b1885264ef6c067c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1741658
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63223}
This removes the restriction on load elimination to only track fields
of representations with kTaggedSize, and instead also allows fields with
representations using multiples of kTaggedSize (that is, Float64 and Word64
on pointer-compressed or 32-bit platforms).
In order not to regress JIT-compile time for the common case of
kTaggedSize-sized fields, we maintain information for bigger fields multiple
times, once for each kTaggedSize-multiple offset that covers it. By checking
that all copies of this information are still there when reading from the
load elimination state, updates to the load elimination state don't need to
take special care of bigger fields.
Change-Id: I9b5f3d2d6e3b4f145c20d33fbc764869bf50a365
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1752843
Commit-Queue: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63222}
Create a new function NewHeapNumberForCodeAssembler used only by
CodeAssembler that allocates in read_only_space whenever it is writable
(e.g. when constructing builtins). This can allocate in old_space for
CSA tests that run after read_only_space is sealed.
This move 512 bytes from old_space to read_only_space.
Bug: v8:7464
Change-Id: I3e29b38a8c062fa74feed63ecf0d07625c04347f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1752855
Commit-Queue: Dan Elphick <delphick@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63220}
