When the CPU profiler receives a bytecode flush event, ensure that we
clear the appropriate CodeEntry.
Bug: v8:11054
Change-Id: I94e771e42192b75ea6d317738e4f2d5b76533dc8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2691826
Reviewed-by: Peter Marshall <petermarshall@chromium.org>
Commit-Queue: Andrew Comminos <acomminos@fb.com>
Cr-Commit-Position: refs/heads/master@{#72684}
Extract v64x2.alltrue code sequence into macro-assembler for sharing
between TurboFan and Liftoff.
Bug: v8:11347,v8:11348
Change-Id: Ica436178b4f92ea0ed574010cd74f1babf66680f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2686013
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72683}
- Add a no-simd-sse flag to skip SIMD tests on bots with no
hardware support.
Change-Id: I4efdbb5ee39c2e10ea8776a1f1e536ac96823efe
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2629465
Commit-Queue: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72682}
Extract code sequence for v64x2.alltrue into macro-assembler for sharing
between TurboFan and Liftoff.
Bug: v8:11347,v8:11348
Change-Id: I8119f5425c8cf11ddac77f69ed9e62a408f7049d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2686011
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72681}
In https://chromium-review.googlesource.com/c/v8/v8/+/2641180 , we are
discussing renaming AccumulatorUse. To avoid polluting that change with
a large mechanical find&replace, I've created a separate change for the
renaming.
Change-Id: Ibc7e438f9e719571c9237e7e08ba86562a3c679f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2684923
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Seth Brenith <seth.brenith@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#72679}
This speeds up sparkplug by >20%.
This reland fixes the OffHeapBytecodeArray to also register a GC
callback. Turns out off-heap here doesn't mean that the underlying
bytecode array is off-heap and it can in fact move.
Change-Id: I7c6e82abd2a7be08ead537ab84855e76edc3b290
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2688400
Auto-Submit: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
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@{#72677}
Port 66964c519a
Original Commit Message:
Extract codegen into macro-assembler functions for reuse in Liftoff.
Some minor tweaks in I32x4TruncSatF64x2SZero and I32x4TruncSatF64x2UZero
to check dst and src overlap and move to scratch/dst accordingly. In
TurboFan we can set these restrictions in the instruction-selector, but
not in Liftoff. This doesn't make TurboFan codegen any worse, since
those restrictions are still in place.
R=zhin@chromium.org, joransiu@ca.ibm.com, junyan@redhat.com, midawson@redhat.com
BUG=
LOG=N
Change-Id: Iae59472a5e77c1becc5ff880081f2c0c8c149630
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2690828
Reviewed-by: Junliang Yan <junyan@redhat.com>
Commit-Queue: Milad Fa <mfarazma@redhat.com>
Cr-Commit-Position: refs/heads/master@{#72676}
If the accessor pair is available, thread it through to the
TryLookupCachedProperty function rather than looking it up again.
On a simple microbenchmark[0] with --no-opt and --no-use-ic this
provides a 5-10% improvement.
[0]: https://gist.github.com/gsathya/c47da0a15be08062c12cda9b0887de3d
Bug: v8:9805
Change-Id: I5b2d0c5e27c49a1d39a99dc63c3b0809bca4d6a7
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2685178
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Mythri Alle <mythria@chromium.org>
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72675}
Other archs will come later.
Bug: v8:11347,v8:11348
Change-Id: I9ea656b9c7ce03c9dafb631dd67f6e2f7d4346a2
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2686312
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72674}
Reasons:
* We disabled it more than a year ago for all configs
* Not easy to re-enable
* Not compatible with pointer compression as-is
* Not compatible with concurrent TP/TF as-is
* No concrete plans to re-enable it
Also remove Map's layout_descriptor since it was only used for double
field unboxing.
Bug: v8:11422
Change-Id: I9260906eac199213b3210712e9903f1ecf1d7979
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2676637
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72671}
cppgc/testing.h is already part of a testonly gn target which only can
be included from other test targets. This prevents any production
target to depend on cppgc/testing.h.
Bug: chromium:1056170
Change-Id: I51f6c47ffac2a05c8c63d7b4663c456a64fe75b4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689196
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72670}
Currently %OptimizeFunctionOnNextCall returns if there is the function
is already optimized. This cl changes this function to allow tiering up
till we reach top tier. That allows us to tier up from Turboprop to
Turbofan using intrinsics. This cl also introduces a runtime-test
function to check if turboprop-as-toptier or turboprop-as-midtier is
enabled.
Bug: chromium:1172797, v8:9684
Change-Id: Idbd99b816d4b93e4e619be5d4ccdfe89fc561a9e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2682638
Commit-Queue: Mythri Alle <mythria@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72668}
This reverts commit b471bc9318.
Reason for revert: Seems like we don't reliably deliver scriptParsed events on reload after this CL.
Original change's description:
> [wasm] Send a single scriptParsed event per script
>
> If a script was shared between multiple modules (because they used the
> same wire bytes) it could happen that we still triggered multiple
> "scriptParsed" events via CDP. This was because
> {WasmEngine::GetOrCreateScript} did not communicate back whether it
> used a cached script or whether it created a new one.
>
> This CL moves the call to {Debug::OnAfterCompile} (which triggers the
> "scriptParsed" event) to the {WasmEngine::GetOrCreateScript} method,
> such that we only call it once per script.
> Since the engine only holds a weak reference to the script, we would
> still trigger multiple events if the script is garbage-collected in the
> meantime. In this case there is no way around this, as the new script
> would have a new ID, hence we need to emit a new event to make it
> public to the debugger.
>
> R=thibaudm@chromium.org
> CC=bmeurer@chromium.org
>
> Bug: chromium:1151211
> Change-Id: I1a7986514fd708680541a0e5dc24e60f01f42c28
> Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_isolates_rel_ng
> Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel_ng
> Cq-Include-Trybots: luci.v8.try:v8_mac64_gc_stress_dbg_ng
> Cq-Include-Trybots: luci.v8.try:v8_linux_gc_stress_dbg_ng
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2687755
> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
> Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
> Commit-Queue: Clemens Backes <clemensb@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#72648}
TBR=clemensb@chromium.org,bmeurer@chromium.org,thibaudm@chromium.org
Change-Id: I6cc299734e4fcff29289355973e7660b60b49a25
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:1151211
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_isolates_rel_ng
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel_ng
Cq-Include-Trybots: luci.v8.try:v8_mac64_gc_stress_dbg_ng
Cq-Include-Trybots: luci.v8.try:v8_linux_gc_stress_dbg_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689199
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72667}
Did not factor out the codegen because it is short enough (1 or 2
instructions) and will unlikely be changed (for optimization reasons).
Bug: v8:11265
Change-Id: Ia79c8553ad4b3924d21f77a6064c9003dfcaeb7a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689001
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72665}
Did not factor out the codegen because it is short enough (1 or 2
instructions) and will unlikely be changed (for optimization reasons).
Bug: v8:11265
Change-Id: Ic5e5bc7642e80448bdaa6d130dfe7c12018eb481
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2683209
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72664}
Torque generates runtime accessor member functions for most class fields
that are defined in .tq files, but fields with struct types are
currently omitted. This change adds those accessors. As an example, if a
.tq file defines the following:
struct InternalClassStructElement {
a: Smi;
b: Smi;
}
class InternalClassWithStructElements extends HeapObject {
const count: Smi;
entries[count]: InternalClassStructElement;
}
Then the following accessors are generated to get and set each struct
field within the 'entries' field:
inline int entries_a(int i) const;
inline void set_entries_a(int i, int value);
inline int entries_b(int i) const;
inline void set_entries_b(int i, int value);
Bug: v8:7793
Change-Id: Ia40b5918e9d09f53ad8e78bc33f8629b8d6a79fe
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2676926
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Seth Brenith <seth.brenith@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#72662}
In the latest spec, catch can take an exception index immediate, and
control-flow jumps to the appropriate catch handler depending on the
thrown exception.
Do this by allowing multiple jump targets for the same pc in labels and
in the control transfer map. At runtime, the unwinder will choose the
appropriate control transfer entry based on the exception tag, unpack
the exception and jump to the handler.
Enable the exception cctests that were currently disabled for the
interpreter, fix some issues and add tests for the new behaviors.
R=clemensb@chromium.org
Bug: v8:8091
Change-Id: I30cb8f9459647a7c6f7bfd9785b238a9c9e9fc10
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2690587
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72661}
HeapBase::CollectStatistics returns a HeapStatistics struct that can be
used by blink to populate a memory dump.
Bug: chromium:1056170
Change-Id: Ic147a02ba6b4aa77bf92cfca067da70b7e1af55b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689181
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72660}
WrapperDescriptor is used to describe how JS wrapper objects can be
inspected to find C++ wrappable objects. In addition, to specifying
which embedder fields are used to find type and instance, the
descriptor also provides and embedder id that identifies
garbage-collected objects. It is expected that the first field of the
type is a uint16_t with that id.
Bug: chromium:1056170
Change-Id: I9cf8d79db972f2dea023114fd5a567e89a3bf373
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2688399
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72657}
This bug was flushed out while working on refactoring the stack traces
(as part of https://crrev.com/c/2689183).
Bug: v8:8742
Change-Id: I5bbd4066cc464b71f4d9a7c90acc35e8cef7afb3
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689193
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Simon Zünd <szuend@chromium.org>
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Simon Zünd <szuend@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72653}
This is a reland of da785659be
The reland overrides ShouldHaveBeenSerialized for typed array
refs to avoid disabling related optimizations when direct heap
access is enabled.
Original change's description:
> [compiler] Don't serialize JSTypedArray fields
>
> This CL removes serialization of JSTypedArray fields when direct heap
> reads are enabled. Invariants we rely on:
>
> - Of the underlying interesting fields,
> - base_pointer and external_pointer are set either during
> initialization, or in a one-time on-to-off-heap transition in
> GetBuffer.
> - length and buffer are immutable after initialization.
> - is_on_heap and DataPtr derive from base_pointer and
> external_pointer s.t. is_on_heap == (base_pointer != 0) and
> DataPtr == external_pointer in the off-heap case.
>
> In this CL we add one new invariant:
>
> - For all base_pointer and external_pointer mutations after
> initialization, base_pointer is guaranteed to be release-stored
> after external_pointer has been written.
>
> With these invariants, concurrent access to off-heap typed arrays is
> trivial as long as is_on_heap (= base_pointer) is read before other
> relevant fields.
>
> Note that JSTypedArray remains a kSerializedHeapObject due to the
> serialized superclass JSObject.
>
> Drive-by: Remove unused Torque operators and empty TODOs.
>
> Bug: v8:7790
> Change-Id: I3c4327318f94e4e6083d4e87476069aad2649386
> Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_isolates_rel_ng
> Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel_ng
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2679689
> Commit-Queue: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#72572}
Bug: v8:7790
Change-Id: I87b37de983e8cf89ca53b5efae7ab195781f3df5
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_isolates_rel_ng
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689182
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72652}
This reverts commit 2a446a9058.
Reason for revert: This caused regressions on Octane / Jetstream2 /
Sunspider and a couple of regressions on memory usage on mobile:
https://bugs.chromium.org/p/chromium/issues/detail?id=1177124https://bugs.chromium.org/p/chromium/issues/detail?id=1177241
Original change's description:
> Enable FLAG_feedback_allocation_on_bytecode_size
>
> This flag enables feedback allocation heuristics to be based on the
> function size. The threshold for feedback allocation is set to
> 4 * bytecode size to roughly mimic the allocation after 4 invocations.
>
> Change-Id: Ia840cd526e3718d4267e01c688c6c6467e352d72
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2685175
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Commit-Queue: Mythri Alle <mythria@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#72631}
TBR=rmcilroy@chromium.org,mythria@chromium.org,verwaest@chromium.org
Change-Id: Ib756116aa38117c06e95c7f68d8f9ba0acd34084
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689194
Reviewed-by: Mythri Alle <mythria@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72651}
Additional arguments are allowed along with the reference, but the
targeted branch must have at least one output (corresponding to the
cast reference).
Bug: v8:7748
Change-Id: I17383165e4bae1cada1676c6282437e1fa71905d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2685161
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72649}
If a script was shared between multiple modules (because they used the
same wire bytes) it could happen that we still triggered multiple
"scriptParsed" events via CDP. This was because
{WasmEngine::GetOrCreateScript} did not communicate back whether it
used a cached script or whether it created a new one.
This CL moves the call to {Debug::OnAfterCompile} (which triggers the
"scriptParsed" event) to the {WasmEngine::GetOrCreateScript} method,
such that we only call it once per script.
Since the engine only holds a weak reference to the script, we would
still trigger multiple events if the script is garbage-collected in the
meantime. In this case there is no way around this, as the new script
would have a new ID, hence we need to emit a new event to make it
public to the debugger.
R=thibaudm@chromium.orgCC=bmeurer@chromium.org
Bug: chromium:1151211
Change-Id: I1a7986514fd708680541a0e5dc24e60f01f42c28
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_isolates_rel_ng
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel_ng
Cq-Include-Trybots: luci.v8.try:v8_mac64_gc_stress_dbg_ng
Cq-Include-Trybots: luci.v8.try:v8_linux_gc_stress_dbg_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2687755
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72648}
For a long time, V8 had two distinct ways to capture and store a stack
trace, one where we'd just collect and symbolize the information for the
v8::StackTrace API (script id, name, line and colum information mostly),
and one where V8 would also memorize the closures, receivers, and
optionally the parameters of the stack frame, which we use for
Error.stack and the non-standard CallSite APIs. Those two were often out
of sync and suffered from various different issues. Eventually they were
refactored into a single captureStackTrace() bottleneck that would
produce a FrameArray.
This CL is a logical continuation of the refactorings. It repairs a
regression where we'd compute the method name (as part of the
cached StackFrameInfo) even if we don't need them (as is the case for
the inspector and any other use of the v8::StackTrace API).
Everytime a method was invoked on StackTraceFrame, it'd call into
StackTraceFrame::GetInfo(), which would lazily setup the StackFrameInfo
like this:
1. Create a FrameArrayIterator and point it to the FrameArray at the
index stored in the StackTraceFrame.
2. Invoke FrameArrayIterator::Frame(), which copies the information
from the FrameArray into a temporary JSStackFrame, AsmJsStackFrame
or WasmStackFrame C++ object, and use the StackFrameBase virtual
methods to transfer all information to a newly created
StackFrameInfo object.
3. Kill the link to the FrameArray and put a link to the
StackFrameInfo object into the StackTraceFrame.
This caching turned out to be extremely costly, since beyond other
things, it'd always invoke JSStackFrame::GetMethodName(), which is
extremely costly (the execution time is linear in the number of
properties on the receiver and it's prototype chain). The cost was so
high that several work-arounds had been added, which would avoid
triggering the eager construction of the StackFrameInfo object (i.e.
https://crrev.com/c/2080663, https://crrev.com/c/2550504 or
https://crrev.com/c/2261736, but also https://crrev.com/c/1688927).
This CL removes the StackFrameInfo caching completely, since neither the
inspector nor Error.stack benefit from the caching at all. It's only the
first part in a series of refactorings that will significantly reduce
the complexity and overhead of the stack trace collection.
Doc: https://bit.ly/2wkbuIy
Bug: chromium:1057211, chromium:1077657, chromium:1069425, v8:8742
Bug: chromium:1127391, chromium:1098530, chromium:981541
Change-Id: I8edb8ff48b620eb3043ae51ab4ea27146ef0a5a2
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689185
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Simon Zünd <szuend@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72647}
Make main-thread handle allocation fail more gracefully when run on
background threads.
Bug: v8:10315
Change-Id: Iece9215aed21020b97fede40d78ea56b9baffac4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2689184
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72646}
V8 implements a fast-path for RegExp.prototype.split which diverges
from the spec: instead of creating a new sticky regexp instance
`splitter` and running it in a loop, we reuse the existing non-sticky
regexp without looping through each character.
This works fine in most cases, but we run into issues when matching at
the very end of the string. According to the spec, matches at the end
of the string are impossible in @@split, but in our fast-path
implementation they can happen.
The obvious fix would be to remove our fast-path but this comes with
high performance costs. The fix implemented in this CL adds a special
flag to `exec` s.t. matches at the end of the string can be treated as
failures. This is only relevant for @@split.
Bug: chromium:1075514
Change-Id: Ifb790ed116793998d7aeb37e307f3f3f764023d3
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2681950
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72644}
Also move it from post-mvp to mvp, since it is now in the proposal.
Bug: v8:11002
Change-Id: I711ee7a92e6937948c93e6028ef018188ea4c976
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2676937
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72643}
Vector register lane numbers on IBM machines are reversed
compared to x64. For example, doing an I32x4 extract_lane with
lane number 0 on x64 will be equal to lane number 3 on IBM machines.
Vector registers are only used for compiling Wasm code at the moment.
Wasm is also little endian enforced. On s390 native, we manually do
a reverse byte whenever values are loaded/stored from memory to
a Simd register. On the simulator however, we do not reverse the bytes
and data is just copied as is from one memory location to another
location which represents a register. To keep the Wasm simulation
accurate, we need to make sure accessing a lane is correctly simulated
and as such we reverse the lane number on the getters and setters.
We need to be careful when getting/setting values on the Low
or High side of a simulated register. In the simulation, "Low" is
equal to the MSB and "High" is equal to the LSB on memory.
As a result, many of the "#ifdef V8_TARGET_BIG_ENDIAN" blocks on
Simd opcodes are not needed anymore as we are now simulating
native behaviour.
Change-Id: Idfa80cdef7382febb4311c75eb6d3e1d110141fa
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2687756
Commit-Queue: Milad Fa <mfarazma@redhat.com>
Reviewed-by: Junliang Yan <junyan@redhat.com>
Reviewed-by: Joran Siu <joransiu@ca.ibm.com>
Reviewed-by: Milad Fa <mfarazma@redhat.com>
Cr-Commit-Position: refs/heads/master@{#72642}
Extract codegen into macro-assembler functions for reuse in Liftoff.
Some minor tweaks in I32x4TruncSatF64x2SZero and I32x4TruncSatF64x2UZero
to check dst and src overlap and move to scratch/dst accordingly. In
TurboFan we can set these restrictions in the instruction-selector, but
not in Liftoff. This doesn't make TurboFan codegen any worse, since
those restrictions are still in place.
Bug: v8:11265
Change-Id: I48f354c5ff86809bb3ddc38eca6dc8990b9b7d61
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2683208
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72641}
This reverts commit a16add806d.
Reason for revert: Broke Win32 debug https://ci.chromium.org/ui/p/v8/builders/ci/V8%20Win32%20-%20debug/29653/overview
Original change's description:
> [wasm-simd][ia32] Implement i64x2 signed compares
>
> The code sequence is exactly the same as x64.
>
> Bug: v8:11415
> Change-Id: I53ed2723eda29c0a250cff514372a3d45b203476
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2683495
> Reviewed-by: Bill Budge <bbudge@chromium.org>
> Commit-Queue: Zhi An Ng <zhin@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#72637}
TBR=bbudge@chromium.org,zhin@chromium.org
Change-Id: Idbfc8cd0fbbff607cff76953c53d0c149b87b573
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:11415
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2688074
Reviewed-by: Zhi An Ng <zhin@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72640}
Extract codegen into macro-assembler functions for reuse in Liftoff.
Some minor tweaks in I32x4TruncSatF64x2SZero and I32x4TruncSatF64x2UZero
to check dst and src overlap and move to scratch/dst accordingly. In
TurboFan we can set these restrictions in the instruction-selector, but
not in Liftoff. This doesn't make TurboFan codegen any worse, since
those restrictions are still in place.
Bug: v8:11265
Change-Id: Ib6b3ebeb5fed99eddd0700fb4aba91d4168c3213
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2683206
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72638}
The code sequence is exactly the same as x64.
Bug: v8:11415
Change-Id: I53ed2723eda29c0a250cff514372a3d45b203476
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2683495
Reviewed-by: Bill Budge <bbudge@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72637}
This reverts commit 60748ee2df.
Reason for revert: Broke Linux64 ASAN https://ci.chromium.org/ui/p/v8/builders/ci/V8%20Linux64%20ASAN/38792/overview.
There are 4 changes in that range causing the failure, I found that this change caused the failure by running locally `./tools/run-tests.py --outdir=out/repro mjsunit/wasm/gc-stress --variant turboprop_as_toptier --random-seed-stress-count 100`.
Original change's description:
> Reland "[interpreter] Speed up the BytecodeArrayAccessor through direct memory access"
>
> Tbr: ulan@chromium.org, neis@chromium.org, leszeks@chromium.org
> No-Presubmit: true
> Change-Id: I4ceb9e21ac7d78a87776b4be174772539d2da8d9
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2685173
> Commit-Queue: Toon Verwaest <verwaest@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Reviewed-by: Toon Verwaest <verwaest@chromium.org>
> Reviewed-by: Leszek Swirski <leszeks@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#72632}
TBR=ulan@chromium.org,neis@chromium.org,leszeks@chromium.org,verwaest@chromium.org
Change-Id: I441ddfda5d852b7a01f38a9e60edc56f40ae626a
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2686266
Reviewed-by: Zhi An Ng <zhin@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72635}
