Change Checkpoints from IR Nodes to just normal Zone objects, pointed to
by the deopting Node. Also merge Checkpoint and DeoptimizationInfo --
this has the side effect that multiple Nodes that share a checkpoint
will point to the exact same deopt call.
Bug: v8:7700
Change-Id: Ib36aa13afe3af6a0a22d2cfe80a13fef4bea1227
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545179
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79607}
Currently, OSB can not be safely accessed if sweeping is in progress.
This can, however, be easily lifted with atomic stores.
Having the consistent bitmap is needed for the generational barrier for
source objects (to retrieve the source object beginning).
Bug: chromium:1029379
Change-Id: I5fb8db579f881ddf240ce68ad51fa8264ee645dc
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545071
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Commit-Queue: Anton Bikineev <bikineev@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79606}
- When the property being defined with DefineKeyedOwnIC or
DefineNamedOwnIC already exists, we should use the slow path to
check if the operation is allowed in case the property is
non-configurable or Object.preventExtensions() has been called on
the property.
- Since KeyedStoreIC:Store() reuses StoreIC::Store() when the key is a
name, we should use Runtime::DefineObjectOwnProperty() for
DefineKeyedOwnIC too.
- When dealing with public fields, Runtime::DefineObjectOwnProperty()
should use JSReceiver::CreateDataProperty() instead of
Object::SetProperty() for the specified semantics. This patch also
adds JSReceiver::AddPrivateField() for it and StoreIC::Store to
define private fields without triggering traps or checking
extensibility.
- To emit a more specific error message when redefining properties
on non-extensible objects, Object::AddDataProperty() now also takes
a EnforceDefineSemantics enum to distinguish between set and define.
- Drive-by: fix JSReceiver::CheckIfCanDefine() which should check for
extensibility even if the configurability check passes.
Bug: chromium:1259950, v8:9888
Change-Id: Ib1bc851ffd4b9c3a0e98cac96dafe743c08ee37e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3517934
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Joyee Cheung <joyee@igalia.com>
Cr-Commit-Position: refs/heads/main@{#79603}
Remove StoreToFrame and the general diff encoding for checkpoints, and
instead make all Checkpoints immediately copy the live part of the
interpreter frame state.
This means that we don't need to recreate the frame state during graph
processing, and we don't have to copy the checkpoint's state for storing
in the deferred DeoptimizationInfo.
In theory the diff encoding was meant to save zone memory for unused
checkpoints, and for checkpoints that don't differ much from each other.
However,
a) We expect to do most checkpoint elimination during graph building,
so the assumption that many checkpoints will be unused seems less
probable, and
b) We need to copy the checkpoint's frame state for emitting deopts,
so we don't actually end up avoiding doing the copies.
So, we can simplify things by removing this complexity.
Bug: v8:7700
Change-Id: Iff9743fabbf7a017cccf0ece76a797c571764ea6
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545178
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79602}
This is a reland of commit d9e1f2aee5
Change: disable regression test on non-SIMD hardware
Original change's description:
> [wasm][liftoff] Spill regs for multi-value merges
>
> If there is more than one value in the merge region, a stack-to-stack
> move can overwrite the source of a stack-to-register move. To avoid
> this, spill all registers.
>
> R=clemensb@chromium.org
>
> Bug: chromium:1299183
> Change-Id: I10495434d0a18c9072ee3882e00a687edd8c592a
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3523044
> Reviewed-by: Clemens Backes <clemensb@chromium.org>
> Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79584}
Bug: chromium:1299183
Change-Id: I6f2af786ab91194a93945f5030575d1b8abee7fd
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3548716
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79601}
Rather than emitting eager deopt code inline, use the full
DeoptimizationData+TranslationArray mechanism in maglev, for consistency
with TurboFan and simplification of implementing lazy deopts in the
future.
Bug: v8:7700
Change-Id: I67282b27493772c78ad28feaa4ad6a2c35c8e239
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545169
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79600}
This reverts commit 802c7b3e16.
Reason for revert: Cluserfuzz found a reason to have this.
Original change's description:
> [string] DCHECK instead of test that ThinString's actuals have same length
>
> ThinStrings always forward to internalized strings that have the same
> character contents and thus the same length.
>
> Change-Id: I5929d266f96b23029f4786baf993a431cf4ad38d
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3541522
> Reviewed-by: Leszek Swirski <leszeks@chromium.org>
> Commit-Queue: Shu-yu Guo <syg@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79582}
Change-Id: Ib8a7b962242f8b8ab3ffc5c70ed1c84d187e5b8e
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3548459
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#79599}
In LowerStringCodePointAt, rather than call StringCodePointAt
builtin, we could inline it using StringCharCodeAt.
Bug: v8:11743
Change-Id: I924f4180ffcfd583cfcbba57b2e0cf114adef068
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3517935
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79598}
This reverts part of crrev.com/c/2413252 since std::unique_ptr does
not satisfy is_standard_layout; and is_standard_layout is needed
to use offsetof:
error: offset of on non-standard-layout type [-Werror,-Winvalid-offsetof]
Fixed: v8:12721
Change-Id: Ifbb1235fc3b8d1d855d41a226117fed88c506078
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3540141
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79597}
Report fine-grain incremental mark/sweep statistics to the Recorder API.
These will be used by Blink to populate UMA histograms such as
V8.GC.Event.MainThread.Full.Incremental.(Mark|Sweep).
Bug: chromium:1154636
Change-Id: I1cbdcb2ffa49bd01d04a2e1d43921cebf956ac84
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545070
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Nikolaos Papaspyrou <nikolaos@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79596}
The debugger maintains a stack of promises used for catch prediction
with promise builtins and async functions. Previously this stack would
hold on to the individual promises strongly, and subtle bugs that lead
to not properly cleaning up the stack in some corner cases would often
lead to significant memory issues (e.g. leaking whole iframes).
This refactors the PromiseOnStack to be
(a) on the V8 heap, rather than allocating C++ structs with global
handles pointing to the promises, and
(b) hold on to the promises only weakly.
While this will not guarantee proper promise stack management, it will
at least ensure that edge cases don't lead to catastrophic (debugger
only) leaks.
Bug: chromium:1292063
Change-Id: I9c293ca2032de3a59e1e9624f132d37187805567
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545176
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Simon Zünd <szuend@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79594}
Before productionizing this, we probably want to just ignore the whole
section if it contains invalid data, but for now failing with a decode
error is more consistent with existing checks.
R=ecmziegler@chromium.org
Bug: v8:12537
Change-Id: I7fc5933573a4d6eddd039bf51361c5bee5c5170d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545177
Reviewed-by: Emanuel Ziegler <ecmziegler@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79593}
Accidentally, remembered sets for slots and source objects were passed
by value when visiting.
Bug: chromium:1029379
Change-Id: I1942a13605990c0b13c378d33d3ace602daddf0d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545318
Commit-Queue: Anton Bikineev <bikineev@chromium.org>
Auto-Submit: Anton Bikineev <bikineev@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79591}
This reverts commit 164a040a2a.
Reason for revert: roll failure: https://ci.chromium.org/ui/p/chromium/builders/try/cast_shell_linux/1164753/overview
Original change's description:
> cppgc: Add regression test and check for object start bitmap
>
> Access to the object start bitmap is only safe during marking until
> sweeping is started as the concurrent sweeper may clear and rebuild
> the bitmap at any time during sweeping.
>
> Adds a DCHECK and an additional test for a previously broken
> pre-finalizer scenario.
>
> Bug: chromium:1307471
> Change-Id: If67ade43f7cdad6de4720c0efeac11bfe8c22b3c
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3535782
> Reviewed-by: Nikolaos Papaspyrou <nikolaos@chromium.org>
> Reviewed-by: Omer Katz <omerkatz@chromium.org>
> Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79550}
Bug: chromium:1307471
Change-Id: I181e63a34eae9369184fb86112bc64e53b8bfad5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545317
Owners-Override: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79590}
Remove deprecated TracedGlobal<>, greatly simplifying handling of
traced references in general.
Also saves a word per v8::TracedReference as there's no need to keep a
possible callback around.
Bug: v8:12603
Change-Id: Ice35d7906775b912d02e97a27a722b3e1cec28d9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3532251
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79589}
Currently, we assume:
allocated-object-size = marked-bytes-on-previous-cycle + k,
which is not the case for the sticky bits (with the young generation
enabled). The CL introduces a variable
StatsCollector::marked_bytes_so_far_ that keeps track of marked bytes
across GCs.
Bug: chromium:1029379
Change-Id: I749c8c0174889c10093c75f88c096c68905bf36e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545167
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Anton Bikineev <bikineev@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79588}
This fixes a general race with stack-scoped `TestPlatform` which
may go out of scope while tasks on workers are still running.
Add a barrier for workers, implemented through tasks, to synchronize
destruction of `TestPlatform`.
While this fixes general races, such short-lived platforms still
break if tasks cache the global platform pointer.
Bug: v8:12635
Change-Id: Ifc6ecc29f0e2b7297ca52051eae9bd81013b60ce
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3536651
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79587}
This reverts commit 9e1db51817.
Reason for revert: https://chromium-review.googlesource.com/c/v8/v8/+/3535782 causes roll failures, this needs to be reverted too because it's based on it
Original change's description:
> cppgc: Add DCHECK that object start bitmap is safe to use
>
> During sweeeping/compaction the bitmap is being reconstructed and
> should not be relied on for finding object start.
> Add a DCHECK that the bitmap is fully populated.
>
> Bug: chromium:1307471
> Change-Id: I4aa414722262bb6fb169123a49fce1510a60d3ef
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3540680
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Commit-Queue: Omer Katz <omerkatz@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79575}
Bug: chromium:1307471
Change-Id: I377b8737609fff33199776dce3d997f31074c59b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545316
Auto-Submit: Tobias Tebbi <tebbi@google.com>
Owners-Override: Tobias Tebbi <tebbi@google.com>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#79586}
This reverts commit d9e1f2aee5.
Reason for revert: Linux test failures: https://ci.chromium.org/ui/p/v8/builders/ci/V8%20Linux/45960/overview
Original change's description:
> [wasm][liftoff] Spill regs for multi-value merges
>
> If there is more than one value in the merge region, a stack-to-stack
> move can overwrite the source of a stack-to-register move. To avoid
> this, spill all registers.
>
> R=clemensb@chromium.org
>
> Bug: chromium:1299183
> Change-Id: I10495434d0a18c9072ee3882e00a687edd8c592a
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3523044
> Reviewed-by: Clemens Backes <clemensb@chromium.org>
> Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79584}
Bug: chromium:1299183
Change-Id: I465129695cfc1c5678923f7eefe5b91e31383798
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3546745
Auto-Submit: Shu-yu Guo <syg@chromium.org>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Shu-yu Guo <syg@chromium.org>
Owners-Override: Shu-yu Guo <syg@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79585}
If there is more than one value in the merge region, a stack-to-stack
move can overwrite the source of a stack-to-register move. To avoid
this, spill all registers.
R=clemensb@chromium.org
Bug: chromium:1299183
Change-Id: I10495434d0a18c9072ee3882e00a687edd8c592a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3523044
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79584}
Since assignments to read-only private references can be skipped due
to short-circuiting in logical assignments, we should not eagerly
emit the error of invalid writes, and should instead load the values
as usual, only emitting an error when the assignment happens,
which can be handled by BytecodeGenerator::BuildAssignment().
Bug: v8:12680, v8:8330, v8:10372
Change-Id: Ia5fea9090bc48b0af8a9c8d6f95174f7aa2d86f8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3509298
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Reviewed-by: Marja Hölttä <marja@chromium.org>
Commit-Queue: Joyee Cheung <joyee@igalia.com>
Cr-Commit-Position: refs/heads/main@{#79583}
ThinStrings always forward to internalized strings that have the same
character contents and thus the same length.
Change-Id: I5929d266f96b23029f4786baf993a431cf4ad38d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3541522
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Shu-yu Guo <syg@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79582}
CreateInterpreterDataForDeserializedCode checks for bytecode, copies the
InterpreterEntryTrampoline, and installs that and the bytecode on
InterpreterData. However, the bytecode can be flushed when the IET is
copied, which results in a failure to read it afterward.
Add an IsCompiledScope to guard against this.
As a drive-by, guard against baseline code being installed on the
function. This shouldn't happen in normal execution, but could
theoretically happen with some extra support for --always-sparkplug.
Bug: chromium:1308178
Change-Id: Ia5e81b376bff2aaa19e9c6007242629ab8b0d4a8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545171
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79580}
Instead of manually ensuring checkpoints and marking that operations had
side effects, do this in AddNewNode based on OpProperties.
Bug: v8:7700
Change-Id: I1e2699af537056d066e7f919abe5e7479bd3af91
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545174
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Toon Verwaest <verwaest@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79578}
During sweeeping/compaction the bitmap is being reconstructed and
should not be relied on for finding object start.
Add a DCHECK that the bitmap is fully populated.
Bug: chromium:1307471
Change-Id: I4aa414722262bb6fb169123a49fce1510a60d3ef
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3540680
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79575}
The code is dead since migrating to jobs API.
Change-Id: Icdcc3523ffe5830ef5851cf4ea86e579841f543c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3540103
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79573}
The lookup is safe during executing pre-finalizers which requires that
the GC is put in the atomic pause.
Bug: chromium:1307471
Change-Id: I4e61573f5cadf2585daab24d73ad798004d54188
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3545173
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Auto-Submit: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79572}
Some edges are self looping because of incorrect `horizontalPos`. This is occuring because of an unexpected scenario caused due to incorrect calculation of `inputApproch` and `outputApproach`. And all of this is occuring because of insufficient distance between two nodes.
An example of the problem is shown in the image: https://imgur.com/aAmnzaK.
Change-Id: I056e1fbcc420ce65a3ae9201e187b22ad3fbaaba
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3535791
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79570}
Following up on https://crrev.com/c/3540145, this also changes local
debug evaluate scripts to be marked as shared-cross-origin.
Drive-by-fix: This also updates the test for global debug evaluate to
use the official (debug) API instead of peaking into the V8 internals
unnecessarily.
Bug: chromium:1295750
Change-Id: Ief0bc76a4333671f8db761d1f6a5fb740aae698e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3541780
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79568}
This reverts commit aaedd8b788.
Changes in the reland:
The inital problem was caused by nodes that were removed during SL
because they are no-ops but have an effect on typing (in the repro, this
was e.g. PlainPrimitiveToNumber). The reland introdocues a new operator
SLVerifierHint that is used exclusively in SL to provide hints to the
verifier and that solves this problem. SLVerifierHint also replaces the
previous use of TypeGuard to type constant nodes for the verifier.
Bug: v8:12619, chromium:1302572
Change-Id: I0957645c03d8b7c26cd6d630a1ecbd0a6a8223ce
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3512574
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Nico Hartmann <nicohartmann@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79564}
CallProperty IR has inputs 0 (function), 1 (context), and the
remaining are variable according to the register count.
Bug: v8:7700
Change-Id: Ie9d8785bd5c1ae878c664683f8ebcff4ce6c408a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3541924
Auto-Submit: Victor Gomes <victorgomes@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79562}
This is a reland of 6b690a6b48.
The previous version of this CL was a bit too aggressive in the
duplication of branch conditions. This caused an increase in
register pressure in some cases, thus reducing performance.
In fact, duplicating branch conditions that require an "== 0" to be
added provides no benefits. We are thus now a bit less aggressive, and
only duplicate comparisons.
Original change's description:
> Reland [compiler] Simplify "==0" branches in MachineOperatorReducer
>
> This is a reland of 48b443f692.
>
> While fixing the initial CL, we stumbled upon a few bugs that
> we had to fix:
>
> - CommonOperatorReducer and SimplifiedOperatorReducer were applied
> before and after SimplifiedLowering, but always assumed that it
> was before SimplifiedLowering, and thus had the wrong semantics
> for branches in some cases. They now have an added parameter to
> know which semantics of branch they should use.
>
> - The lowering of StaticAssert was wrong and could leave kHeapConstant
> in the assert (instead of machine Booleans).
>
> Original change's description:
> > [compiler] Simplify "==0" branches in MachineOperatorReducer
> >
> > Bug: v8:12484
> > Change-Id: I0667c7464c0dd71338bc199a24a69248a7a0a525
> > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3497303
> > Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> > Owners-Override: Tobias Tebbi <tebbi@chromium.org>
> > Commit-Queue: Darius Mercadier <dmercadier@chromium.org>
> > Cr-Commit-Position: refs/heads/main@{#79379}
>
> Bug: v8:12484
> Change-Id: Ibbf5df96fce5ccb04868dc517539479bf69f5703
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3516869
> Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> Commit-Queue: Darius Mercadier <dmercadier@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79528}
Bug: v8:12484
Change-Id: I31f575a59811a83c7c1acb4c14bf5ded63a8f536
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3540102
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Darius Mercadier <dmercadier@chromium.org>
Cr-Commit-Position: refs/heads/main@{#79560}
