This reverts commit fbcdb28178.
Reason for revert: New test fails for multiple (concurrent) isolates: https://ci.chromium.org/ui/p/v8/builders/ci/V8%20Linux/45152/overview
Original change's description:
> [wasm] Lazy compilation after deserialization
>
> The serialization format contains one boolean flag per function which
> specifies whether the function code exists in the serialized module or
> not. With this CL, this boolean flag is extended to a three-value flag
> which indicates whether the function exists, and if not, whether the
> function was executed before serialization. This information can then be
> used upon deserialization to compile only those functions that were
> executed before serialization.
>
> Design doc: https://docs.google.com/document/d/1U3uqq4njqLqFhr1G2sU_bmpQxY-3bvfG55udSb-DvA4/edit?usp=sharing
>
> Bug: v8:12281
> Change-Id: I465e31e5422fa45163256be0e6594045865f0174
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3364089
> Reviewed-by: Clemens Backes <clemensb@chromium.org>
> Commit-Queue: Andreas Haas <ahaas@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#78545}
Bug: v8:12281
Change-Id: If0e327d02e8257a4d1cfcf8b82381af11f28e91c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3377126
Auto-Submit: Clemens Backes <clemensb@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@{#78546}
The serialization format contains one boolean flag per function which
specifies whether the function code exists in the serialized module or
not. With this CL, this boolean flag is extended to a three-value flag
which indicates whether the function exists, and if not, whether the
function was executed before serialization. This information can then be
used upon deserialization to compile only those functions that were
executed before serialization.
Design doc: https://docs.google.com/document/d/1U3uqq4njqLqFhr1G2sU_bmpQxY-3bvfG55udSb-DvA4/edit?usp=sharing
Bug: v8:12281
Change-Id: I465e31e5422fa45163256be0e6594045865f0174
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3364089
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78545}
When creating a new JSError object (or using the non-standard API
`Error.captureStackTrace`) V8 would previously capture the "simple stack
trace" (as FixedArray of CallSiteInfo instances) to be used for the non-
standard `error.stack` property, and if the inspector was active also
capture the "detailed stack trace" (as FixedArray of StackFrameInfo
instances). This turns out to be quite a lot of overhead, both in terms
of execution time as well as memory pressure, especially since the
information needed for the inspector is a proper subset of the
information needed by `error.stack`.
So this CL addresses the above issue by capturing only the "simple stack
trace" (in the common case) and computing the "detailed stack trace"
from the "simple stack trace" when on demand. This is accomplished by
introducing a new ErrorStackData container that is used to store the
stack trace information on JSErrors when the inspector is active. When
capturing stack trace for a JSError object while the inspector is
active, we take the maximum of the program controlled stack trace limit
and the inspector requested stack trace limit, and memorize the program
controlled stack trace limit for later formatting (to ensure that the
presence of the inspector is not observable by the program).
On the `standalone.js` benchmark from crbug.com/1283162 (with the
default max call stack size of 200) we reduce execution time by around
16% compared to ToT. And compared to V8 9.9.4 (the version prior to the
regression in crbug.com/1280831), we are 6% faster now.
Doc: https://bit.ly/v8-cheaper-inspector-stack-traces
Bug: chromium:1280831, chromium:1278650, chromium:1258599
Bug: chromium:1280803, chromium:1280832, chromium:1280818
Fixed: chromium:1283162
Change-Id: I57dac73e0ecf7d50ea57c3eb4981067deb28133e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3366660
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78542}
We should only check the "SIMD sibling" register if we are handling a
SIMD register. This avoids unneeded spills, and in this particular case
ran into a DCHECK because there are only 29 registers, but we tried
checking #29.
R=thibaudm@chromium.org
Bug: v8:12330, v8:1285007
Change-Id: Ife8b295ac958990611ca8816bbfbfb5124a4297d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3372916
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78532}
This should have been updated in https://crrev.com/c/3370408
Bug: chromium:1284506
Change-Id: Ie44d80b507c9a798ce6f4776672270f9d4b12195
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3371463
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Commit-Queue: Austin Sullivan <asully@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78519}
The test was added in https://crrev.com/c/3372910, but needs to be
skipped on non-SIMD hardware because it contains SIMD instructions.
R=thibaudm@chromium.org
Bug: v8:12330, chromium:1284980
Change-Id: Ifaede466b24aea4f9ef6b062414a31698bcca864
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3372917
Auto-Submit: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Deepti Gandluri <gdeepti@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78517}
Spilling was already fixed if a fixed SIMD register overlaps with an
allocated FP register, but the other way around was missing: If an odd
FP register (in this case d1) is used as a fixed output register, but
this register is in use as the upper half of a SIMD register (in this
case q0), we did not detect this and would just use overwrite the SIMD
half.
This CL also fixes this case.
R=thibaudm@chromium.org
Bug: v8:12330, chromium:1284980
Change-Id: Id3f98b7accd77e38ab4cd5ff8910aaf5ad96a1ed
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3372910
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78514}
https://crrev.com/c/3297708 changed the serialization format for typed
arrays without bumping the format version. As a consequence, builds that
include that CL fail to deserialize typed arrays serialized by previous
V8 versions.
This CL reverts the serialization format change, and does minimal test
changes to reflect the revert. https://crbug.com/v8/12532 tracks
serializing typed array flags in a backwards-compatible manner.
Bug: chromium:1284506
Change-Id: Ib32e88c6383e0ad4ad1a9ff63f413a1eb123b1ef
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3370408
Reviewed-by: Adam Klein <adamk@chromium.org>
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Commit-Queue: Victor Costan <pwnall@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78507}
We implement loop peeling for wasm, currently available behind a flag.
Loops are peeled regardless of size.
Bug: v8:11510
Change-Id: Ia4c883abdee83df632b2611584d608c44e3295c8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3367615
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78496}
The V8InspectorSessionImpl constructor accepts a state, as either text
or CBOR encoded, and generally ignores all invalid inputs, except for
the case where it's a valid value, but not a dictionary value, in which
case it'll leak the value and crash upon casting to a `DictionaryValue`.
This is purely an issue with the test driver, so no security impact on
Chromium in the wild.
Fixed: chromium:1281031
Change-Id: I7b4d0aea83370499b1274d3fa214a14dc098d2f2
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3361838
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78490}
If a fixed register is defined for an input, we did only spill the
sibling SIMD register if the other sibling was allocated. This is not
correct. If only the sibling is in use (e.g. s1 colliding with q0) we
also have to spill that sibling.
R=mslekova@chromium.org
Bug: chromium:1283042, v8:12330
Change-Id: I6a22eaf461774a0b4603ec3ff17062134a528161
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3359615
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78483}
The mid-tier register allocator did not handle block merges correctly
where a SIMD register was partially overlapping with a non-SIMD
register. This CL fixes that, and reorders the code to allow for early
exits.
R=mslekova@chromium.org
Bug: chromium:1282224, v8:12330
Change-Id: I2e9275d5c1aaa764ecb63fbf8fa197b68d6b6c3c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3358294
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78482}
This change fixes the implementation of the previously introduced API
`Runtime.setMaxCallStackSizeToCapture` to work correctly and also apply
(consistently) to stack traces captured by V8 when exceptions are
thrown. It does so in a fully backwards compatible manner.
This change thus makes the previous fix for catapult (which landed in
http://crrev.com/c/3347789) effective, and therefore ensures that real
world performance benchmarks aren't affected by the use of the `Runtime`
domain in the catapult test framework.
Note this is basically a reland of crrev.com/c/3361839, but without
touching the stack traces for console messages (which led to the
regressions in crbug/1283516, crbug/1283523, etc.).
Fixed: chromium:1280831
Bug: chromium:1283162, chromium:1278650, chromium:1258599
Bug: chromium:1280803, chromium:1280832, chromium:1280818
Doc: https://bit.ly/v8-cheaper-inspector-stack-traces
Change-Id: I3dcec7b75d76ca267fac8bd6fcb2cda60d5e60dd
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3364086
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78479}
We unify the implementation of element segment expression entries with
other initializer expressions: we represent them with a {WireBytesRef}
and decode them with {InitExprInterface}. Except for reducing code
duplication, this also fixes a bug where {global.get} entries in element
segments could reference invalid globals.
Changes:
- Change {WasmElemSegment::Entry} to a union of a {WireBytesRef}
initializer expression and a {uint32_t} function index.
- In module-decoder, change parsing of expression entries to use
{consume_init_expr}. Add type checking to
{consume_element_func_index}, to complement type checking happening in
{consume_init_expr}.
- In module-instantiate.cc:
- Move instantiation of indirect tables before loading of element
segments. This way, when we call {UpdateDispatchTables} in
{SetTableEntry}, the indirect table for the current table will also
be updated.
- Consolidate table entry instantiation into {SetTableEntry}, which
handles lazily instantiated functions, or dispatches to
{WasmTableObject::Set}.
- Rename {InitializeIndirectFunctionTables} to
{InitializeNonDefaultableTables}.
- Change {InitializeNonDefaultableTables} and {LoadElemSegmentImpl}
to use {EvaluateInitExpression}.
- Add a test to exclude mutable/non-imported globals from the element
section.
- Update tests as needed.
- Update .js module emission in wasm-fuzzer-common.
Change-Id: I29c541bbca8531e8d0312ed95869c8e78a5a0c57
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3364082
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78476}
See related CL for context.
Changes:
- In InitExprInterface, add the ability to evaluate function references
as index only. Remove the global buffers and use the ones passed with
the instance object instead.
- In WasmElemSegment, add a field indicating if elements should be
parsed as expressions or indices. Change module-decoder.cc to reflect
this change.
- In module-instantiate, change the signatures of LoadElemSegment,
LoadElemSegmentImpl, and EvaluateInitExpr. Move the latter out of
InstanceBuilder.
Change-Id: I1df54393b2005fba49380654bdd40429bd4869dd
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3364081
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78470}
This reverts commit 34f73cc759.
Reason for revert: Performance regressions throughout a lot of
system health and browsing benchmarks.
Original change's description:
> [inspector] Fix `Runtime.setMaxCallStackSizeToCapture`.
>
> This change fixes the implementation of the previously introduced API
> `Runtime.setMaxCallStackSizeToCapture` to work correctly and also apply
> (consistently) to stack traces captured by V8 when exceptions are
> thrown. It does so in a fully backwards compatible manner.
>
> This change thus makes the previous fix for catapult (which landed in
> http://crrev.com/c/3347789) effective, and therefore ensures that real
> world performance benchmarks aren't affected by the use of the `Runtime`
> domain in the catapult test framework.
>
> Bug: chromium:1283162, chromium:1278650, chromium:1258599
> Bug: chromium:1280803, chromium:1280832, chromium:1280818
> Fixed: chromium:1280831
> Doc: https://bit.ly/v8-cheaper-inspector-stack-traces
> Change-Id: I4ec951a858317fa49096cd4023deb0104d92c9c9
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3361839
> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#78458}
Bug: chromium:1283162, chromium:1278650, chromium:1258599
Bug: chromium:1280803, chromium:1280832, chromium:1280818
Bug: chromium:1280831
Change-Id: Id1efaffa2f7f08c47f833f68b8a297494edee21e
Fixed: chromium:1283751, chromium:1283749, chromium:1283746
Fixed: chromium:1283729, chromium:1283700, chromium:1283700
Fixed: chromium:1283691, chromium:1283687, chromium:1283678
Fixed: chromium:1283677, chromium:1283676, chromium:1283675
Fixed: chromium:1283674, chromium:1283618, chromium:1283536
Fixed: chromium:1283523, chromium:1283516
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3364078
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78462}
This change fixes the implementation of the previously introduced API
`Runtime.setMaxCallStackSizeToCapture` to work correctly and also apply
(consistently) to stack traces captured by V8 when exceptions are
thrown. It does so in a fully backwards compatible manner.
This change thus makes the previous fix for catapult (which landed in
http://crrev.com/c/3347789) effective, and therefore ensures that real
world performance benchmarks aren't affected by the use of the `Runtime`
domain in the catapult test framework.
Bug: chromium:1283162, chromium:1278650, chromium:1258599
Bug: chromium:1280803, chromium:1280832, chromium:1280818
Fixed: chromium:1280831
Doc: https://bit.ly/v8-cheaper-inspector-stack-traces
Change-Id: I4ec951a858317fa49096cd4023deb0104d92c9c9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3361839
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78458}
Previously the `Debugger.CallFrame`s in `Debugger.paused` events would
report locations relative to the surrounding document in case of inline
scripts with `//@ sourceURL` annotations (while `Runtime.CallFrame` was
already fixed previously as part of crrev.com/c/3069289). With this CL
the locations in `Debugger.CallFrame` are also appropriately adjusted.
Drive-by-fix: Several inspector tests were (incorrectly) relying on this
wrong treatment, and were also unnecessarily using //# sourceURL
annotations. So part of this CL also addresses that problem and makes
the tests more robust, using addInlineScript() helper.
Fixed: chromium:1283049
Bug: chromium:1183990, chromium:578269
Change-Id: I6e3b215d951c3453c0a9cfc9bccf3dc3d5e92fd6
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3359619
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78450}
Included in this CL:
(*) Introduce CppMarkingState that V8 should use to push references to
Oilpan. CppMarkingState allocates its own Worklist::Locals to
support concurrent updates from V8.
(*) Split Oilpan MarkingWorklist object to form a base class used by
CppMarkingState.
(*) Remove MarkerFactory and split marking initialization. Marking
worklists should already be initialized when V8 initializes
visitors. For incremental marking, this requires splitting
marking initialization and marking start.
(*) Drive-by: Mark JSObject::IsApiWrapper and
JSObject::IsDroppableApiWrapper as const.
Bug: v8:12407
Change-Id: I35cc816343da86f69a68306204675720e9b3913f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3293410
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78446}
This CL splits the TF type for JSFunction into CallableFunction and
ClassConstructor. This differentiation allows us to lower calls to the
CallFunction Builtin only for functions that we can actually call.
Class Constructors are special, as they are callable but should raise
an exception if called.
By not lowering class constructors to calls to CallFunction (but the
more generall Call) builtin, we can remove the checks for class
constructors from CallFunction (in a follow-up CL).
Bug: chromium:1262750
Change-Id: I399967eb03b2f20d2dcb67aef2243b32c9d3174e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3350457
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Commit-Queue: Patrick Thier <pthier@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78445}
Wasm values are stored in memory in little endian order even
on BE machines and as a result they need to be manually reversed
after a load.
Other such atomic ops get patched during Wasm compilation or
during code-gen, this is one of the few places where a runtime call is
made to C++ which requires this fix.
As the the runtime stub is used on both TurboFan and Liftoff this
patch will fix both cases.
Up until now the cctest was passing incorrectly as it's mixing the
Wasm memory buffer with TypedArrays. TypedArrays don't have the
LE enforcement and use the native byte order.
With this patch the test is now failing as expected
and is being skipped for now.
Bug: v8:12505
Change-Id: I49fac208f1fab7396b7d9911e803bc047b3b8263
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3350744
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Milad Farazmand <mfarazma@redhat.com>
Cr-Commit-Position: refs/heads/main@{#78433}
This updates the following set of console builtins in V8 to match the
Console Standard (https://console.spec.whatwg.org) with respect to
(potentially side effecting) type conversions:
- console.debug
- console.error
- console.info
- console.log
- console.trace
- console.warn
- console.group
- console.groupCollapsed
- console.assert
The V8 implementation only performs the type conversions and updates
the arguments in-place with the results from the %String% constructor,
%parseInt%, or %parseFloat% invocations. The actual formatting is
still left completely to the debugger front-end.
To give a concrete example, the following code
```js
const msgFmt = {
toString() { return 'Message %i' }
};
console.log('LOG: %s`, msgFmt, 42);
```
sends the following parameters to the debugger front-end
```js
["LOG: %s", "Message %i", 42]
```
and it's then the job of the front-end to perform the actual string
substitutions.
It's also worth calling out that the console builtins are only
concerned with %s, %f, %d, and %i formatting specifiers, since
these are the only ones that trigger type conversions, and %o, %O,
and %c can only be implemented in a meaningful way at a higher
level.
Fixed: chromium:1277944
Bug: chromium:1282076
Doc: https://bit.ly/v8-proper-console-type-conversions
Spec: https://console.spec.whatwg.org
Change-Id: I0996680811aa96236bd0d879e4a11101629ef1a7
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3352118
Reviewed-by: Kim-Anh Tran <kimanh@chromium.org>
Auto-Submit: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78432}
Private method loads are compiled to a named load of a private brand,
which always loads a BlockContext. This BlockContext holds the private
methods common to all instances of a class. TurboFan currently considers
JSLoadNamed to be of Type::NonInternal(). Private methods break this
assumption, since BlockContext is of Type::OtherInternal().
This CL changes the typing of JSLoadNamed of private brands to be
Type::OtherInternal().
Bug: v8:12500
Change-Id: I91f39747bf9422bd419d299f44152f567d8be8db
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3351167
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Shu-yu Guo <syg@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78431}
This commit allows using unaligned load/store, which is more efficient
for 2 bytes,4 bytes and 8 bytes memory access.
Use RISCV_HAS_NO_UNALIGNED to control whether enable the fast path or not.
Change-Id: I1d321e6e5fa5bc31541c8dbfe582881d80743483
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3329803
Reviewed-by: ji qiu <qiuji@iscas.ac.cn>
Commit-Queue: Yahan Lu <yahan@iscas.ac.cn>
Cr-Commit-Position: refs/heads/main@{#78427}
Shared array buffers are not tracked by the garbage collector, which
makes the fuzzer run out of memory pretty quickly. Since shared memory
is not needed any more for testing atomics, we can just make the memory
non-shared again.
This also improves the performance of the fuzzer (execs/s) by more than
2x locally.
R=ahaas@chromium.org
Bug: chromium:1281419
Change-Id: Ic7803617d6a14aaa698d9181327ec20b21d29faa
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3350764
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78418}
Without simple FP aliasing, a SIMD register will overlap with two
floating-point registers. If we spill an FP register to use it for a
SIMD operation, we need to make sure to also spill the "sibling" FP
register.
R=leszeks@chromium.org
Bug: v8:12330, chromium:1271244
Change-Id: I7fdc6cb8da35d66b4862a8a913ba4ff906cf05aa
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3347576
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78414}
Initialize the (thread-local) memory protection key permissions for any
isolate that joins the wasm engine. Otherwise it can happen that an
isolate gets Wasm code from the cache without ever compiling anything
(hence without ever changing memory protection key permissions), and
then it would not be allowed to access (read or execute) the code.
I tested this change manually on a PKU-enabled devices. The new test
crashed before the fix, and completes successfully afterwards.
R=ahaas@chromium.org
Bug: v8:11974, chromium:1280451
Change-Id: I90dded8b4fdaa8cf34b44107291d3f525ce16335
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3347563
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78413}
After https://crrev.com/c/3315446 we allocate the memory protection key
unconditionally, so the method is redundant.
R=ahaas@chromium.org
Bug: v8:11974
Change-Id: I205a0cda86dfaf394c68788a662241d76a3f8510
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3347562
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78412}
The mid-tier register allocator could not handle the case that the same
virtual register was used for
- the input corresponding to the 'same-as-input' output, and
- another 'unique register' input.
In this case, it cannot choose the already assigned register for the
'unique' register. Instead, it needs to allocate a new register and
introduce a gap move to duplicate the input value in two different
registers.
FYI, the instruction where the current logic failed was:
(v5(0), v6(R)) = IA32AddPair v7(R) v7(*) v8(R) v7(R)
(where the last input was marked 'unique').
R=leszeks@chromium.orgCC=thibaudm@chromium.org
Bug: v8:12330, chromium:1272204
Change-Id: Ie4843aa9f5e027afe503e0481a4acdfa325dfe0e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3347821
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78411}
A SIMD register can "block" more than one FP register. In that case, no
virtual register will be assigned for one of the FP registers. This is
fine, we just need to detect and handle that case correctly.
R=thibaudm@chromium.org
CC=leszeks@chromium.org
Bug: chromium:1271538, v8:12330
Change-Id: I7ec19229445c5ace0782f63945acb89322816540
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3293082
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78400}
We switch the order of inlining and loop unrolling optimizations. This
gives small improvements to wasm-gc benchmarks.
Changes:
- Change the loop analysis algorithm to accept loops directly connected
to the graph's end. This is required because some nodes in an inlined
function, such as tail calls, might be directly connected to the outer
function's end without an intervening LoopExit node.
- Based on the above, skip emitting loop exits for some Throw nodes in
WasmGraphBuildingInterface.
- Introduce WasmInliningPhase, add it before loop unrolling. Remove
inlining from WasmOptimizationPhase.
- Handle graph terminators in loop unrolling.
- Add loops in the inlined function to the callers loop_infos.
Drive-by:
- Allow more wasm builtins in unrolled loops.
- Reduce inlining parameters to reflect that functions are now slightly
smaller during inlining, as no unrolling has taken place yet.
Bug: v8:12166
Change-Id: Iadd6b2f75170aa153ca1efb47fbb0d185c2b8371
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3329783
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78394}
... in order to avoid Code <-> CodeT conversions in builtins.
This CL changes the meaning of RelocInfo::CODE_TARGET which now expects
CodeT objects as a code target.
In order to reduce code churn this CL makes BUILTIN_CODE and friends
return CodeT instead of Code. In the follow-up CLs BUILTIN_CODET and
friends will be removed.
Bug: v8:11880
Change-Id: Ib8f60973e55c60fc62ba84707471da388f8201b4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3338483
Reviewed-by: Patrick Thier <pthier@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#78393}
