Optimizations to use fast memmove to move elements are preserved, as
well as heuristics for bailout to the runtime if left or right
trimming is desired.
Bug: v8:7672
Change-Id: I01ffc1143b63d705d99a40eab3a7e873596d0aa4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1499495
Commit-Queue: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60118}
This updates the existing bootstrap.sh script for gcmole to work against
LLVM and Clang version 8.0 releases. This is a follow-up to a previous
change which adapted the gcmole plugin to compile against those same
versions.
R=mslekova@chromium.org
BUG=v8:8813
Change-Id: Id6052fb9a7ec8a63d205eab2d4e233e2121c733d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511275
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60116}
We have to create WasmExportedFunction objects for any WebAssembly
function which may escape a WebAssembly instance. Up until now we
created these WasmExportedFunction objects eagerly during instantiation
time: for any exported function, and any element in an exported table we
create such an object.
With the anyref proposal, the table.get instruction can allow any
function in a table to escape its instance. Therefore we would have to
create a WasmExportedFunction object for any function which is put into
a table.
With this CL we create WasmExportedFunctions for table entries lazily.
We initialize tables with placeholders consisting of the instance and
the function index. If we encounter a placeholder in table.get, we
create the WasmExportedFunction for the expected function to return it.
R=mstarzinger@chromium.orgCC=titzer@chromium.org
Bug: v8:7581
Change-Id: I4f32bd7433285d0b04a22c0fb70b736bac55b3f1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1505575
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Ben Titzer <titzer@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60115}
instead of forwarding template constructors for these classes introduced in
edab9a2021 commit.
TurboAssemblerBase constructors were declared as public to make the inherited
TurboAssembler, and MacroAssembler ctors also public.
This fixes Visual C++ 2017 compile error, when the template ctor in
TurboAssemblerBase class matches deleted copy ctor.
Bug: v8:8935
Change-Id: I1144a7025830c3a0ab86acaa8ea81def02d293b1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1496977
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60114}
RotateRight32 needs a "number of bits" operand in the range 0..31.
Thankfully that's how x86 shift instructions behave anyway, and
how the bitwise shift operators in JavaScript are spec'ed, so this
fix is unobservable in non-UBSan builds.
RemoveArrayHolesGeneric can be used for length values anywhere in
the uint32_t range, so it must not implicitly cast those to int.
That actually caused an observable bug where a proxy's traps would
not get called at all, but only for huge "length" properties, where
the entire operation would also be painfully slow.
Bug: chromium:935133, chromium:937652
Change-Id: I13f74ca27eae6b2b089d58217842b699b2574509
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1510272
Auto-Submit: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60112}
- Converts most integer vector tests to use globals (except Select)
so results can be checked in C++ code.
- Remove integer vector result checking macros.
- Add specializations of test CompareOps for floats, so we can use
BinOps for integer vector compare opcodes.
- Remove Run#format#CompareOpTests helper functions for integer vector
types. Use Run#BinOpTests helper function instead.
Bug: v8:6020
Change-Id: I968a71c874b028a750e1118cf51f6678cae90091
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1496281
Commit-Queue: Bill Budge <bbudge@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60111}
This is a reland of 821bc64951
Original change's description:
> [wasm simd] Fix F32x4 Min and Max
>
> - Fix F32x4 tests to save results in globals, so they can be checked
> in C++ code. Perform correct checks in case of NaNs.
> - Fix ia32, x64 implementations of F32x4Min, F32x4Max to correctly
> deal with NaNs.
> - Enable tests for all float values on all platforms, except skip
> denormalized results on ARM, and skip extreme values for reciprocal,
> reciprocal square root approximation opcodes.
> - Disable Min, Max test for interpreter (see v8:8425) since it doesn't
> handle NaNs correctly.
> - Fix vmin, vmax implementations in ARM simulator.
>
> Bug: v8:8639
> Change-Id: I87e188e3cb078f09fdacfd9955f426c20a11bf64
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1495897
> Commit-Queue: Bill Budge <bbudge@chromium.org>
> Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#60021}
Bug: v8:8639
Change-Id: Ic557aa1d323693eabf5885ff5eddc15e3174079b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1501279
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Bill Budge <bbudge@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60109}
V8's MemMove and libc's memset do not specify how the memory is filled,
but wasm requires the direction to be backward only when the regions are
overlapping and the source address is less than the destination address.
Bug: v8:8950
Change-Id: If392967b51b33ad65fe145660f44b857559e34b1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1507316
Commit-Queue: Ben Smith <binji@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60106}
... because the latter are not meant to be modified from non-main thread
and especially after V8 isolate is set up while the former are modified
cuncurrently by tracing API.
Tbr: verwaest@chromium.org
Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel
Bug: v8:8929, v8:8834
Change-Id: I44d3da2f388bb8bb8d0365ac6354e761bf92b936
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1505581
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60104}
This change enables clean-up of weak map entries in the
scavenger of the weak map is in the young generation.
With this change, the scavenger treats keys in ephemerons as
weak instead of strong, but does not implement full ephemeron
semantics: Values are treated always as strong, independently
of whether the key is live or not.
This approach ensures that no value is cleaned up accidentally.
After scavenging, all entries with dead keys are removed from
weak maps. After that, some values that are not referenced anywhere
anymore might still be in the heap, and those can be cleaned up
in the next scavenge.
What the scavenger does, amounts to one iteration of the
fixed-point algorithm required to implement ephemeron semantics.
We hope that this is a reasonable trade-off between time spent
tracing and cleaned-up entries.
This change does not affect weak maps that reside in old space.
Bug: v8:8557
Change-Id: Ic5618b3b863ad8c314c87449571150e756fecbf0
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1467182
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60101}
After introducing the new pointer-containing Object class in V8 (see
https://docs.google.com/document/d/1_w49sakC1XM1OptjTurBDqO86NE16FH8LwbeUAtrbCo/edit),
gcmole stopped finding errorneous usage of raw pointers in functions that could
trigger GC. This CL modifies the heuristics of the tool to classify Object and
MaybeObject instances as raw pointers, thus giving back the missing warnings.
Updated the gcmole implementation to support modern llvm (tested with llvm 8.0)
for which additional support for MaterializeTemporaryExpr, ExprWithCleanups and
UnaryExprOrTypeTraitExpr was needed.
Basic tests are added to make it harder to introduce such errors without
noticing in the future.
This version gives a lot of false positives when ran on the whole project, see
https://docs.google.com/document/d/1K7eJ0f6m9QX6FZIjZnt_GFtUsjEOC_LpiAwZbcAA3f8/editR=jkummerow@chromium.org,mstarzinger@chromium.org
Bug: v8:8813
Change-Id: Ic0190a4bc2642eda8880d9f7b30b5145a76a7d89
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1494754
Commit-Queue: Maya Lekova <mslekova@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60099}
smi size is sill 8 bytes when V8_COMPRESS_POINTERS is undefined.
Bug: v8:7703
Change-Id: I0d1e757e42e8b1e6b10960420135245e24553175
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1508572
Commit-Queue: Junliang Yan <jyan@ca.ibm.com>
Auto-Submit: Junliang Yan <jyan@ca.ibm.com>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60097}
Emitting bytes to the Uint8Array directly speeds up generation of
binaries enormously.
On the limits-any.js spec test (which creates huge modules), the
execution time of an optdebug build reduces from 286 seconds to 61
seconds.
R=titzer@chromium.org
CC=ahaas@chromium.org, ssauleau@igalia.com
Change-Id: I5b473b7dc7b0853e54d2406f3db3658bb2abed40
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1508352
Reviewed-by: Ben Titzer <titzer@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60096}
When running mksnapshot standalone, no callback is registered to
serialize embedder fields. In this case they are most probably empty as
there is no embedder to set them, so provide a default serialization
that keeps them as nullptr.
Bug: v8:8822
Change-Id: I9219f2755488aa1473959c3ee75e9d4f47202359
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1507677
Auto-Submit: Peter Marshall <petermarshall@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60094}
Most test-log tests create their own isolates, so initializing the cctest
isolate is not necessary.
Bug: v8:7464
Change-Id: Icc3cd9d5053c4070833709cbe28e5bde2cee9708
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1507719
Reviewed-by: Dan Elphick <delphick@chromium.org>
Commit-Queue: Maciej Goszczycki <goszczycki@google.com>
Cr-Commit-Position: refs/heads/master@{#60093}
This makes sure an exception raised while compiling a module via the
embedder API is properly returned as a "scheduled exception" and hence
propagates to surrounding {v8::TryCatch} scopes.
R=clemensh@chromium.org
TEST=cctest/test-api/WasmModuleObjectCompileFailure
BUG=v8:8908
Change-Id: I52b21fbe5a4548aa346fd6c9b5bac061613db487
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1507673
Commit-Queue: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60092}
This umblocks switching this builder to Xenial. The corresponding gcc
version is not yet supported by goma.
NOTRY=true
Bug: chromium:933093
Change-Id: Ib00d191c229bb598b209dcaf447321b874571e62
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1505577
Reviewed-by: Tamer Tas <tmrts@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60087}
Updates a bunch of links from https://github.com/v8/v8/wiki pages to the
appropriate v8.dev page that it redirected to anyway.
Bug: v8:8834
Change-Id: I5b37996900eb779753d97e487d16e1489f54d391
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1503473
Auto-Submit: Dan Elphick <delphick@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Dan Elphick <delphick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60086}
These changes fix Python 3 compatibility errors in third-party libraries
I'm not sure if these problems can be fixed this way.
But! (1) In deps/inspector_protocol no file roll.py
it looks like the repository v8 it shouldn't be, either.
(2) I have not found a separate repository for binutils.
Bug: chromium:935197
Change-Id: Ied2bb83e0539aa392f8fc7c3aeef662612f37d06
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1484302
Reviewed-by: Sergiy Belozorov <sergiyb@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Sergiy Belozorov <sergiyb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60085}
Execute foreground tasks triggered by the {CompilationStateCallback}
via the {CompileStep} mechanism of {AsyncCompileJob} such that they get
cancelled when the AsyncCompileJob dies.
R=ahaas@chromium.org
Bug: chromium:938311
Change-Id: I2082f93f47988c014c8dee3ddf3e9b2940f6f531
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1507674
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60082}
This adds the entrypoint to MicrotaskQueue, which used to miss the
implementation.
Bug: v8:8124
Change-Id: I114fb69d975ee75c86b19349ca76789e425ea910
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1505232
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Taiju Tsuiki <tzik@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60076}
Fixing a few float and int overflows.
Drive-by fix: with --experimental-wasm-bigint, Number values
may not be used to initialize i64-typed globals. The existing
code for doing that relied on UB; since it's a spec violation
the fix is to throw instead.
No regression test for 933103 because it will OOM anyway.
No regression test for 932896 because it would be extremely slow.
Bug: chromium:927894, chromium:927996, chromium:930086, chromium:932679, chromium:932896, chromium:933103, chromium:933134
Change-Id: Iae1c1ff1038af4512a52d3e56b8c4b75f2233314
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1495911
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60075}
... and fix header includes to please the respective bot.
Drive-by-fix: decompression implementation is now MSVC friendly.
Bug: v8:7703, v8:8834
Change-Id: Iaf589138e5bafb32b0d9feab5cf074b71f241a3c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1505579
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60074}
ReduceJSHasProperty was always passing STANDARD_LOAD instead of using the load mode from
the feedback verctor. This was causing deopt loops when checking for out-of-bounds
indices.
chromium: 937734
Change-Id: I6de29f2c6a80bcc171cf027d47a2d1af1414b76c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1501975
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Matt Gardner <magardn@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#60073}
...mainly by giving a more precise type to global_proxy getters.
Change-Id: If4aef6b25baa2c641a45b177c59690e3ebfc3985
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1505578
Commit-Queue: Georg Neis <neis@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60072}
This is a reland of 80f06d6fb3
Original change's description:
> [wasm] Lazy update instances on a shared Memory.Grow
>
> - Introduce a GROW_SHARED_MEMORY interrupt, and handler
> - Memory objects for isolates are updated on a stack check, add
> tracking for isolates that hit the stack check
> - When enough memory is not reserved ahead of time, fail to grow
> - Add tracking for externalized buffers in the MemoryTracker so
> that the MemoryTracker will know when backing_stores can be freed.
> - For shared buffer, do not always allocate a new buffer when
> growing an externalized buffer
>
>
> Change-Id: I9cf1be19f2f165fa6ea4096869f7d6365304c8c4
> Bug: v8:8564
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1472430
> Commit-Queue: Deepti Gandluri <gdeepti@chromium.org>
> Reviewed-by: Ben Smith <binji@chromium.org>
> Reviewed-by: Andreas Haas <ahaas@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#60064}
Bug: v8:8564
Change-Id: Id0cf8e42a9d54ac702dba351e248a1b92713c98a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1506357
Reviewed-by: Bill Budge <bbudge@chromium.org>
Commit-Queue: Deepti Gandluri <gdeepti@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60071}
This fixes a corner case during young generation evacuation:
1) Sweeping is in progress, but all pages are swept.
2) The main thread has acquired all swept pages.
3) A compaction space cannot get any swept page and has to allocate a
new page.
This reduces memory usage of ArrayLiteralInitialSpreadLargeDoubleHoley
microbenchmark by 2x.
Bug: chromium:934453
Change-Id: I146d22a408910ae89adbfbac5822448952f5f0c7
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1505798
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60069}
