This shares patch libraries between vm and wasm.
The shared libraries are those previously shared as parts which
need private access to other patch libraries.
TEST=existing
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Change-Id: I750a20de5a78362e84b87b2bfe1e5395ca4d3769
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/262341
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Commit-Queue: Johnni Winther <johnniwinther@google.com>
Change-Id: I1566febad2b4e1399f50bb923ad0596f04aa1a40
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/262268
Reviewed-by: Joshua Litt <joshualitt@google.com>
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Commit-Queue: Jackson Gardner <jacksongardner@google.com>
Reviewed-by: Alexander Thomas <athom@google.com>
Change-Id: Iea75c14b6709f2347fe82af2ef226a1ed068d8a6
Cq-Include-Trybots: luci.dart.try:dart-sdk-linux-try,dart-sdk-mac-arm64-try,dart-sdk-mac-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/261502
Reviewed-by: Alexander Thomas <athom@google.com>
Commit-Queue: Joshua Litt <joshualitt@google.com>
Reviewed-by: Aske Simon Christensen <askesc@google.com>
This adds support for the experimental 'records' feature in the sdk
by settings its experimental release version to 2.19 and opting the sdk
into the feature the allowed_experiments.
Change-Id: I7a30212e5724e5d8ae3208f177103764b9aed737
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/262760
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Leaf Petersen <leafp@google.com>
This changes the way the CFE handled private injected members. Previously
these were "hidden" within the patch library in which they were declared.
Now they are, like with members injected in augmentation libraries, fully
integrated and accessible within the origin library and all its patches.
This change revealed that the vm and wasm platforms had some
inconsistently declared classes in which injected private members that
were not implemented by all implementing classes. For these, throwing
stubs have been added.
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Change-Id: I330eade944964ef43b83aa416baef75e3649d023
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/262340
Reviewed-by: Jens Johansen <jensj@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Commit-Queue: Johnni Winther <johnniwinther@google.com>
The correct normalization should only normalize FutureOr<Never>
when Never is non-nullable.
Change-Id: I592f3a4856c219b33a8f1ac8377567a956e1148c
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/261000
Commit-Queue: Nicholas Shahan <nshahan@google.com>
Reviewed-by: Mayank Patke <fishythefish@google.com>
Reviewed-by: Erik Ernst <eernst@google.com>
This shares patch libraries between vm and wasm.
The shared libraries are those previously shared as parts, except
the libraries that need private access to other patch libraries.
Change-Id: I69598a0d2ede5138e9ce33fb59dfa46c987eb38a
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/262320
Commit-Queue: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
This changes the VM patch files to use parts explicitly, making
handling of patch libraries similar to regular libraries.
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Change-Id: I6280d62edba3d12a1f77ae6a7f64e1e9b5d18227
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/224949
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Johnni Winther <johnniwinther@google.com>
DDC:
Adding return types avoids casts on the returned values and
using JS foreign functions avoids the cast when setting the
value during truncation as well as avoiding the extra
method dispatch for setting and getting elements.
Dart2js:
Updated the return types for consistency but the dynamic
return type was already being inferred to avoid casts.
Change-Id: Ic9abaffd8dee3d12793d5083c78d35a9ada47527
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/261787
Reviewed-by: Stephen Adams <sra@google.com>
Commit-Queue: Nicholas Shahan <nshahan@google.com>
`internal/patch.dart` is changed to be an actual part of
`dart:_internal` and backends are changed to all use the `patch`
constant from `dart:_internal` in `@patch` annotations.
This is done in preparation for changing the import uri of a
patch library builder to be its file uri, with which the
`internal/patch.dart` part can no longer declare itself as
`part of 'dart:_internal';` through a patch library.
Change-Id: I884e939d4a56782e64fb527cdf900769c2b2fcfd
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/224946
Reviewed-by: Jens Johansen <jensj@google.com>
Commit-Queue: Johnni Winther <johnniwinther@google.com>
This change is a small incremental step, but should allow us to delete
the public NoSuchMethod constructor in dart:core. The change simply creates
a private constructor in the dart2js patch file and uses that instead.
https://github.com/dart-lang/sdk/issues/49529
Change-Id: I51cb2ec3b7f169a6cb831e753a3e5e9bf0752dbd
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/260940
Reviewed-by: Stephen Adams <sra@google.com>
Commit-Queue: Sigmund Cherem <sigmund@google.com>
And all the other things it didn't mention.
And update the `==` documentation too.
Change-Id: Iccd4b046b0632db788e44b02799d1f8d5541d766
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/260881
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
Makes it clear that the original future can still complete,
it's just the new future which doesn't wait for it any longer.
Fixes#42318
Bug: http://dartbug.com/42318
Change-Id: I16c25e4e4588d444bd99796d25bdb92bea1467f0
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/260582
Commit-Queue: Nate Bosch <nbosch@google.com>
Auto-Submit: Lasse Nielsen <lrn@google.com>
Reviewed-by: Nate Bosch <nbosch@google.com>
"vm_common" set of core libraries is shared between various Dart VM
embedders: standalone VM, Flutter, Fuchsia Dart runner and Fuchsia
Flutter runner.
This change would simplify future changes of core library patch
files and will remove the need for additional changes in Flutter
engine when patch file is added or removed.
TEST=ci
Change-Id: Id9fb9dc369504fd4d7dab44de3799dbef594b12d
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/260900
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Johnni Winther <johnniwinther@google.com>
This reverts commit 959854769c.
Reason for revert: This CL appears to cause some performance regressions in Flutter hot reload, please see https://github.com/flutter/flutter/issues/112177
In addition the recently added benchmark IntegerSetLookup.DefaultHashSet also shows regressions.
Original change's description:
> [VM/Library] - Adjust next probe to be a quadratic probe in _HashBase
>
> Adjust next probe to a quadratic probe so that it matches up with the internal VM hash map implementation.This avoids some issues with hash code clustering and uses the probing strategy in the VM's HashTable.
>
> Change-Id: I6af560b971c2e6fa745e3b3414e2d8a9075e7ff5
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/239004
> Reviewed-by: Ryan Macnak <rmacnak@google.com>
> Commit-Queue: Siva Annamalai <asiva@google.com>
# Not skipping CQ checks because original CL landed > 1 day ago.
Change-Id: I27955a4eb239cf9afcbd40989d570467b9eea063
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/260747
Reviewed-by: Siva Annamalai <asiva@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Siva Annamalai <asiva@google.com>
This CL introduces a new kernel-to-kernel pass in Wasm backend to
compile `FfiNative`-annotated external functions to external functions
with `wasm:import` pragmas.
The new pass `WasmFfiNativeTransformer` extends `FfiNativeTransformer`.
Some `FfiNativeTransformer` methods made public to allow reuse in the
pass.
The conversion works like this: when we see a member like:
@FfiNative<Int8 Function(Int8, Int8)>("ffi.addInt8")
external int addInt8(int a, int b);
We generate a Wasm import using Wasm ABI types according to emscripten
calling conventions:
@pragma('wasm:import', 'ffi.addInt8')
external static WasmI32 addInt8_$import(WasmI32 a, WasmI32 b);
and convert the original member (`addInt8()`) to a wrapper function that
calls the Wasm import, converting the arguments and return values to
Dart types:
static int addInt8(int a, int b) {
return WasmI32.toIntSigned(
addInt8_$import(
WasmI32::int8FromInt(a),
WasmI32::int8FromInt(b),
));
}
Build, test, and CI changes:
- Test runner now uses `// SharedObjects=...` lines in dart2wasm tests
to pass extra Wasm modules to d8 command used to run the tests.
Support for `// SharedObjects=...` lines were already used in
native/AOT FFI tests, not added in this CL. We just start to use those
lines in dart2wasm tests.
- dart2wasm gn file updated with a target "test_wasm_modules" that
builds FFI test Wasm modules to Wasm using emcc (emscripten C
compiler).
- CI dart2wasm_hostasserts configuration updated to copy the Wasm files
generated by the "test_wasm_modules" target to the shards.
TEST=tests/web/wasm/ffi_native_test
Change-Id: I6527fe4e2ca2b582e16d84fee244e9cbe6dee307
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/252822
Reviewed-by: Alexander Thomas <athom@google.com>
Commit-Queue: Ömer Ağacan <omersa@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Aske Simon Christensen <askesc@google.com>
!!! NOTE to Google3 rollers:
If this change causes test failures in google3 you can temporarily
set `null_safety_asserts = False` on the failing
`dart_browser_binary` targets until we can fix the mistyped code.
That should allow the test to pass with this change.
Of course feel free to revert this change again if the number of
failures is too high.
Re-land previously reverted change:
https://dart-review.googlesource.com/c/sdk/+/258780
Adds additional tests and more checks to ensure assertions are only
added in libraries that have been migrated to null safety.
Issue: https://github.com/dart-lang/sdk/issues/49918
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/259120
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Commit-Queue: Nicholas Shahan <nshahan@google.com>
Change-Id: Idcab0fe49a8dd98cdd73192eabae4439aebe2013
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/260070
Adjust next probe to a quadratic probe so that it matches up with the internal VM hash map implementation.This avoids some issues with hash code clustering and uses the probing strategy in the VM's HashTable.
Change-Id: I6af560b971c2e6fa745e3b3414e2d8a9075e7ff5
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/239004
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Siva Annamalai <asiva@google.com>
Registering custom elements are already broken -- see: https://github.com/dart-lang/sdk/issues/49536. There are plans on deleting this code at some point. This current change cleans up dynamic invocations in the deprecated code.
Removes cruft on already-deprecated code and doing it now because I assume dynamic clean up will finish before the issue resolves.
Change-Id: Ic6250c139c5d9b08d88650110f55442a7bf5dd42
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/259247
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Reviewed-by: Lasse Nielsen <lrn@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
Reviewed-by: Leaf Petersen <leafp@google.com>
Bug: https://github.com/dart-lang/sdk/issues/49351
Adds checks for the following cases:
- Type arguments to createStaticInteropMock are correct
- No missing members in implementing class
- Inherited and non-overridden @staticInterop members are implemented
- Dart class can implement through inheritance and mixins
- Implemented members are correct subtypes of @staticInterop members
- Potential extension member conflicts that are attempted to be
resolved through subtyping rules
Change-Id: Iacbe5846040ba7fab41459aa19be77351cf1efca
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255761
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Reviewed-by: Riley Porter <rileyporter@google.com>
There's one only one path for creating a _FileResourceInfo which comes from RandomAccessFile. Type the `file` field to avoid a dynamic call to `.path`.
Change-Id: Ia1041d5b221d0496af3dd177562263b31cdf4157
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/259106
Reviewed-by: Leaf Petersen <leafp@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
This function is used internally to abstract over null safety modes.
Since dart2wasm is always fully null safe, the function is trivial.
Change-Id: I8d636d4ce146a2a848dde736d8563dcb0028f92a
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/258928
Commit-Queue: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Leaf Petersen <leafp@google.com>
There's no LICENSE file that mentions the Go authors in this repository, but since this code is covered by the CLA we can just move the name to the AUTHORS file and rely on the existing statement that references the Dart LICENSE file.
Change-Id: I3e8e04ef39e7ed6de0134cc33bf8a22e3d33389b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/250792
Commit-Queue: Michael Thomsen <mit@google.com>
Reviewed-by: Michael Thomsen <mit@google.com>
Avoid removing dynamic invocations in sanitizing code as we don't know what type the element and its attributes may be.
Change-Id: I057d908027befe7fd284bf4ec6c68bd1d0e977f4
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/259108
Reviewed-by: Srujan Gaddam <srujzs@google.com>
Reviewed-by: Leaf Petersen <leafp@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
When using `Uri.parse` or `Uri(path:..)`, a `\` is treated as, and converted to, a `/`.
This avoids a particular problematic difference in behavior between Dart and the browser's `URL` functionality. There are still examples where the two differ in interpretation of the same code, but in those cases, the Dart `Uri` will most likely end up without a host name, which should be easily detected.
Change-Id: I798df6c3c27c6d64fb9fc8dc30d90b06ba5a9004
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/258120
Reviewed-by: Nate Bosch <nbosch@google.com>
Reviewed-by: Michael Thomsen <mit@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
These methods create new sets, whose implementation must match the
target platform. Thus, they should be specialized for the VM and
dart2wasm instead of being implemented in the shared mixin.
Change-Id: I102a45686e3eb373463f2fcdf84ee861e202ecc0
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/258362
Commit-Queue: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Lasse Nielsen <lrn@google.com>
Rename `_CompactLinkedHashSet` to `_InternalLinkedHashSet` and
`_CompactImmutableLinkedHashSet` to `_InternalImmutableLinkedHashSet`
to make the names consistent with the corresponding `Map` class names
and to distinguish them from the non-VM-specific hash maps and sets,
which are generally named starting with `_Compact`.
Tested: ci
Change-Id: Ic1ebc44ba8beba34ceab382a4edddc6624fe3fb3
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-debug-x64-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/258002
Reviewed-by: Chloe Stefantsova <cstefantsova@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
This gets rid of some occurrences of `dynamic` and makes the
associated implicit conversions explicit.
Based on https://dart-review.googlesource.com/c/sdk/+/237582
Tested: ci
Change-Id: I7faeffab395c916800b29ea7e56bd3f7d15aac54
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/258242
Commit-Queue: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Lasse Nielsen <lrn@google.com>
Because the value type is dynamic, I noticed use sites passing in
`Map<String, String?>`. After inspecting the implementation, it looks
like this is by design: `writeParameter` in `_makeQuery` takes a
nullable value.
This CL updates the documented type to reflect this.
Change-Id: I7b6f1c4a59ec44282e5e8b4ada86becafb308d30
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/258240
Reviewed-by: Lasse Nielsen <lrn@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
This reverts commit f630a7c07d.
Reason for revert: Remove the `is` check since it's more costly than removing a dynamic call for web.
Original change's description:
> Eliminate dynamic call() in html_dart2js in _EventStreamSubscription.
>
> Change-Id: Ia4a300c210bc505f71cc3bfb1a1b911c96c1aa53
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/256703
> Commit-Queue: Kallen Tu <kallentu@google.com>
> Reviewed-by: Leaf Petersen <leafp@google.com>
> Reviewed-by: Lasse Nielsen <lrn@google.com>
# Not skipping CQ checks because original CL landed > 1 day ago.
Change-Id: I36b2f8cedc5f8b5235832ea29ea5eb40e2e8dc92
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/257421
Reviewed-by: Srujan Gaddam <srujzs@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
This reverts commit 9292afa710.
Reason for revert: b/244407576
Original change's description:
> Make `addStream` events be delivered synchronously.
>
> The asynchronous stream controllers were forwarding stream events
> from an `addStream` stream through the same path as user-added event,
> which added an extra asynchronous delay that is unnecessary for forwarding existing events.
>
> Change-Id: I8edd1936d0d155d051d3cb8f7f82c577334d5f8e
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/256560
> Reviewed-by: Nate Bosch <nbosch@google.com>
> Commit-Queue: Lasse Nielsen <lrn@google.com>
# Not skipping CQ checks because original CL landed > 1 day ago.
Change-Id: I0e4791f4d32616fee9772a2ab7015bac4e117312
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/257080
Auto-Submit: Ivan Inozemtsev <iinozemtsev@google.com>
Commit-Queue: Alexander Thomas <athom@google.com>
Reviewed-by: Alexander Thomas <athom@google.com>
Reviewed-by: Ivan Inozemtsev <iinozemtsev@google.com>
This reverts commit 3688a4723c.
Reason for revert: b/244409289
Original change's description:
> Fix bug in Completer.complete.
>
> When given a completed `_Future`, it would complete its own future synchronously.
>
> Fixes#48493.
>
> Bug: https://github.com/dart-lang/sdk/issues/48493
> Change-Id: Ic4a68389b5a009ef52b3875bf03a3bafb493f79a
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/234981
> Reviewed-by: Nate Bosch <nbosch@google.com>
> Commit-Queue: Lasse Nielsen <lrn@google.com>
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: https://github.com/dart-lang/sdk/issues/48493
Change-Id: I5ed360bebe160868f6152c72c82070c93188858e
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/257100
Reviewed-by: Ivan Inozemtsev <iinozemtsev@google.com>
Commit-Queue: Ivan Inozemtsev <iinozemtsev@google.com>
Reviewed-by: Lasse Nielsen <lrn@google.com>
The asynchronous stream controllers were forwarding stream events
from an `addStream` stream through the same path as user-added event,
which added an extra asynchronous delay that is unnecessary for forwarding existing events.
Change-Id: I8edd1936d0d155d051d3cb8f7f82c577334d5f8e
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/256560
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
There should be no control characters or delimiters in method name
provided to open/openUrl methods.
Fixes https://github.com/dart-lang/sdk/issues/45744
TEST=http_open_method_validate_test
Change-Id: I0db98f2376c980a054420fb447d4f7ef9321f1a9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/256429
Reviewed-by: Siva Annamalai <asiva@google.com>
Reviewed-by: Brian Quinlan <bquinlan@google.com>
Commit-Queue: Alexander Aprelev <aam@google.com>
Additionally, fixed the template for html_dart2js so the generator would generate the correct code.
Change-Id: I2172bb10db413169b86690faa027365ff72e3d68
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255763
Reviewed-by: Srujan Gaddam <srujzs@google.com>
Reviewed-by: Lasse Nielsen <lrn@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
This adds support for the newly added `extern.externalize` and
`extern.internalize` instructions and uses these for converting to/from
`externref` instead of a JS round-trip.
Change-Id: If18d8f44ddf013d4c26bf1597be91bcd0db41c5a
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255462
Commit-Queue: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Joshua Litt <joshualitt@google.com>
The API for the JS Promise integration API has changed to be part of
the `new WebAssembly.Function` constructor. This requires parameter
types to be specified explicitly. Apparently, these types currently
must be all `externref`.
Change-Id: Ifae634f65cfa377441be4b2dc0643c34b702fbe7
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255061
Reviewed-by: Stephen Adams <sra@google.com>
This splits the type hierarchy into three separate hierarchies with
the top types `any`, `func` and `extern`.
Update d8 to 10.6.91, which switches to the new type hierarchy.
Also, all ref shorthands for abstract heap types are now nullable, so
the type emitter in the `wasm_builder` is updated to follow that scheme.
To reduce confusion about the nullability of abstract reference types,
these now all require the nullability to be specified explicitly.
Change-Id: I4774d08cbed18307e481c466b2e3402a8d8fb6bd
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255060
Reviewed-by: Joshua Litt <joshualitt@google.com>
This implementation has a number of limitations:
- Returning a Future from an async function is not supported.
- Every async call switches to a new stack. A later optimization will
stay on the same stack if the call is directly awaited.
- Exceptions that cross async boundaries do not receive full stack
traces including the async suspension points. The stack trace of the
final exception just reflects the stack trace for the outermost
async call.
Change-Id: I2a5d5e30d6e955999ba55842c8b2ca3427cbf954
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/241010
Reviewed-by: William Hesse <whesse@google.com>
Reviewed-by: Joshua Litt <joshualitt@google.com>
Adds static method on `Isolate` to run an asynchronous function
in a separate isolate.
Change-Id: I673373fa02524f1d0b88099027cfaf1b796eb344
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/254960
Commit-Queue: Lasse Nielsen <lrn@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Statically type String variables to avoid dynamic calls for substring or concatenation.
Change-Id: If485243dc0580fd1f4e53ec4cf69cb3819a57b5d
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255301
Reviewed-by: Leaf Petersen <leafp@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
This reverts commit 4c7110332e.
Reason for revert: Still breaks the same tests, see b/242715525
Original change's description:
> Reapply "Avoid computing the URI scanner tables at runtime."
>
> This reverts commit 855e1cd975.
>
> The blocking issue in internal test code is assumed fixed.
>
> Change-Id: I74e0be130d149a45f77dc90c354916308b76b741
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255248
> Commit-Queue: Lasse Nielsen <lrn@google.com>
> Reviewed-by: Martin Kustermann <kustermann@google.com>
> Auto-Submit: Lasse Nielsen <lrn@google.com>
TBR=lrn@google.com,kustermann@google.com
Change-Id: I391d2eb6dd9ae7367f656eaaaa7f9aeb1c31e0f7
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255254
Reviewed-by: Alexander Thomas <athom@google.com>
Commit-Queue: Ilya Yanok <yanok@google.com>
Reviewed-by: Ilya Yanok <yanok@google.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
This reverts commit 855e1cd975.
The blocking issue in internal test code is assumed fixed.
Change-Id: I74e0be130d149a45f77dc90c354916308b76b741
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/255248
Commit-Queue: Lasse Nielsen <lrn@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Auto-Submit: Lasse Nielsen <lrn@google.com>
Builds on top of https://dart-review.googlesource.com/c/sdk/+/228660, introduces vm:vm:always-consider-inlining that tell inliner not to give up inlining of a function.
HashMap constructor got prefer-inline pragma to accommodate shift in inlining away from where it used to be inlined due to large size of inlined HashMap, greater than FLAG_inline_getters_setters_smaller_than heuristic.
Notable performance changes:
JIT (x64)
===
ListCopy.List.of.fixed.100 27.37%
ListCopy.List.of.fixed.2 16.31%
AOT (x64)
===
Empty -9.978%
InstantiateTypeArgs.Instantiate1 -8.262%
...
ListCopy.List.of.fixed.2 13.73%
ListCopy.spread.num.2 15.51%
ListCopy.List.of.2 15.08%
MapCopy.Map.String.of.Map.100 55.62%
MapCopy.Map.Thing.of.Map.100 56.06%
flutter release
===
flutter_gallery_apk_size (Pixel 2)
-0.0074% (1.0 noise)41809568.00 41812676.00
Addresses https://github.com/dart-lang/sdk/issues/49408
TEST=Inliner_always_consider_inlining, Inliner_List_of_inlined
Change-Id: I7f8fc7cb0ac4a69310c108cf519518c384dc0164
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/253740
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Alexander Aprelev <aam@google.com>
Changes `FfiNative` instance methods on `NativeFieldWrapperClass`es
that have `Pointer<Void>` conversion to check for `nullptr`.
If the native field has `0`/`nullptr` a `StateError` is thrown:
"Bad state: Native field is nullptr."
This only makes sense if the first native field is used to point to
the C++ object corresponding to the Dart object. As far as we know
all current use cases do so.
TEST=pkg/vm/testcases/transformations/ffi/ffinative.dart.expect
TEST=tests/ffi/vmspecific_ffi_native_test.dart
Closes: https://github.com/dart-lang/sdk/issues/49620
Change-Id: I92f760c33d391476010722358f9713fa4491ab61
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/254200
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Daco Harkes <dacoharkes@google.com>
The library wasn't taken into account when handling computing class
members in the K/J-model and not handled correctly when computing
applicable selectors. This made dart2js unable to handle members with
a name private to a different library than the enclosing library.
Fixes#33732Fixes#49226
Change-Id: I5ba143d87662bbd42e0ff02355054e4a937be8f8
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/252665
Commit-Queue: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Stephen Adams <sra@google.com>
These types now work with Dart_TypedDataAcquireData.
The presence of these types no longer degrades the performance of typed data indexed loads.
The presence of these types degrades the performance of typed data indexed stores much less. The performance of indexed stores is somewhat regressed if these types were not used.
TEST=ci
Bug: https://github.com/dart-lang/sdk/issues/32028
Bug: https://github.com/dart-lang/sdk/issues/40924
Bug: https://github.com/dart-lang/sdk/issues/42785
Change-Id: I05ac5c9543f6f61ac37533b9efe511254778caed
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/253700
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
It appears that the environment was intended to be nullable in
some type constructions.
DDC running in weak mode can fail when null is passed as an argument
that is typed to be non-nullable.
Change-Id: I2c67efacc1166e6053ef8d8d7820de2c022c5e1b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/253706
Reviewed-by: Mayank Patke <fishythefish@google.com>
Commit-Queue: Nicholas Shahan <nshahan@google.com>
Heap snapshots currently produced don't visit isolate stacks. As such
analyzing such snapshots may lead one to conclude there is a lot of
garbage while objects are actually reachable.
=> This CL makes us visit isolate stacks when building heap snapshots.
Furthermore we add a new `VMInternals.writeHeapSnapshotToFile` helper
that can be used to programmatically write snapshots and can be handy
for internal use at times. (We also use this helper in a test)
TEST=vm/dart{,_2}/heap_snapshot_test
Change-Id: I976544b7f6d20863764af9a40bf1ffb3c319bbce
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/253785
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Martin Kustermann <kustermann@google.com>
This was requested by internal teams for easier querying/processing.
Change-Id: I59fea565f79bf92d4cf808cc0466b4e9cfa37fe8
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/253707
Commit-Queue: Mark Zhou <markzipan@google.com>
Reviewed-by: Joshua Litt <joshualitt@google.com>
The `registerElement` APIs in `dart:html` are legacy APIs based on a
deprecated Web Components v0.5 specification. These methods don't work
on modern browsers and can only be used with a polyfill.
The latest Web Components specification is supported indirectly via
JSInterop and doesn't have an explicit API in the `dart:html` library.
This change marks these APIs as deprecated. We intend to remove them in
the future (see https://github.com/dart-lang/sdk/issues/49536)
Fixes https://github.com/dart-lang/sdk/issues/48973
Change-Id: I2e96d36e95c9971b59cde80bc4da49b63d12b17c
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/252840
Commit-Queue: Sigmund Cherem <sigmund@google.com>
Reviewed-by: Srujan Gaddam <srujzs@google.com>
Closes https://github.com/dart-lang/sdk/pull/49478
TEST=Manual
GitOrigin-RevId: f4c9c6869dfe73639295e86574a021523b3d374d
Change-Id: I134a97caed4eec59d70e9cbca16b7e9a472cf2c1
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/251902
Reviewed-by: Michael Thomsen <mit@google.com>
Commit-Queue: Alexander Thomas <athom@google.com>
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Kevin Chisholm <kevinjchisholm@google.com>
Reviewed-by: Alexander Thomas <athom@google.com>
Move the method body out of one giant JS foreign function and into
Dart code with small JS foreign function calls inside.
The generous use of small JS foreign function calls ensures almost
identical compiled output with the exception of indentation, curly
braces on if statements, and single vs double quote differences.
Issue: https://github.com/dart-lang/sdk/issues/46002
Change-Id: Ice039fd096893e4f8869696ca46651b3d2e88add
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/252410
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Commit-Queue: Nicholas Shahan <nshahan@google.com>
This is a reland of commit 08c13f173c
Fixes test failures on non-x64 architectures, both in the test
harness and due to DWARF5 line number program headers having a
non-backwards compatible format. (We generate DWARF2 in the
ELF snapshot writer, but the assembler used for assembly snapshots
may generate DWARF5.)
TEST=vm/dart{,_2}/use_dwarf_stack_traces_flag
Original change's description:
> [pkg/native_stack_traces] Support Mach-O dSYM debugging information.
>
> TEST=vm/dart{,_2}/use_dwarf_stack_traces_flag
>
> Bug: https://github.com/dart-lang/sdk/issues/43612
> Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-dwarf-linux-product-x64-try,vm-kernel-precomp-linux-product-x64-try,vm-kernel-precomp-mac-product-x64-try,pkg-mac-release-arm64-try,vm-kernel-mac-release-arm64-try,vm-kernel-precomp-nnbd-mac-release-arm64-try
> Change-Id: Icda21bb14dcc0cf4784cea118e6ba7dd4edd35aa
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/250381
> Commit-Queue: Tess Strickland <sstrickl@google.com>
> Reviewed-by: Slava Egorov <vegorov@google.com>
Bug: https://github.com/dart-lang/sdk/issues/43612
Change-Id: I8a9cb70e78bc8594bcae004809c5a1be778d691d
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-dwarf-linux-product-x64-try,vm-kernel-precomp-linux-product-x64-try,vm-kernel-precomp-mac-product-x64-try,pkg-mac-release-arm64-try,vm-kernel-mac-release-arm64-try,vm-kernel-precomp-nnbd-mac-release-arm64-try,vm-kernel-precomp-linux-debug-x64c-try,vm-kernel-nnbd-linux-release-simarm64-try,vm-kernel-precomp-linux-release-simarm_x64-try,vm-kernel-precomp-nnbd-mac-release-simarm64-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/251464
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
This fixes the breakage of barista3 introduced in
https://dart-review.googlesource.com/c/sdk/+/250480
Change-Id: I5ebca17ab35464e0c452d2fbe1b0f6574e05b6cd
Cq-Include-Trybots: luci.dart.try:dart2wasm-linux-x64-d8-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/252381
Reviewed-by: Joshua Litt <joshualitt@google.com>
Commit-Queue: Joshua Litt <joshualitt@google.com>
Auto-Submit: Aske Simon Christensen <askesc@google.com>
Move the method body out of one giant JS foreign function and into
Dart code with small JS foreign function calls inside.
Issue: https://github.com/dart-lang/sdk/issues/46002
Change-Id: I978f752c8bb6f09ed3c2ca7d149561f488034509
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/251851
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Commit-Queue: Nicholas Shahan <nshahan@google.com>
Previously there was a window between the decision to resume generator
and actually running the micro-task. The stream could be canceled
during that window, making running the generator incorrect.
The solution is to check if the stream was canceled right before
resuming the generator in yield/yield*.
This change also unifies yield and yield* to check if the stream
was canceled even before suspending the generator.
TEST=language/async_star/cancel_while_paused_test,
language/async_star/cancel_while_paused_at_yield_test
Fixes https://github.com/dart-lang/sdk/issues/49451
Change-Id: Ib8ff3a2da9a6a7da8766a7dbf28e2c9d618728f9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/251562
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
This change introduces separate stubs for suspending sync* functions
at start and at yield/yield*. Suspend stub for yield/yield*
no longer calls Dart callback (in order to make it faster).
Also, ReturnSyncStar stub is removed - sync* functions now directly
return false instead of going through the stub.
TEST=ci
Issue: https://github.com/dart-lang/sdk/issues/48378
Change-Id: Iee9a1f48cab2812cf0f9f0e4e6d8e847547e49f7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/250420
Reviewed-by: Slava Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
This is a reland of commit bc8afad855
On top of the original commit, this change fixes incorrect
propagation of async/async*/sync* modifiers from a function to
its dynamic invocation forwarder.
TEST=ci, runtime/tests/vm/dart/regress_b_238653741_test.dart
Fixes b/238653741
Issue: https://github.com/dart-lang/sdk/issues/48378
Original change's description:
> [vm] Cleanup old async/async*/sync* implementation from the VM
>
> TEST=ci
>
> Issue: https://github.com/dart-lang/sdk/issues/48378
> Change-Id: I089ba4ed5613f30eec29f0db4ac6d5d8fbffd185
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249980
> Reviewed-by: Martin Kustermann <kustermann@google.com>
> Reviewed-by: Slava Egorov <vegorov@google.com>
> Commit-Queue: Alexander Markov <alexmarkov@google.com>
Change-Id: Iaad033d974a23fc6c5880a3d7f41818eb117f839
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/251300
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
Remove call to FATAL on a socket error event when we get back a null
error from the call to getsockopt with SO_ERROR, instead we just close
the socket.
(Should address the increased crash reports that apps are seeing after the FATAL call was added to see if we could get a stack trace when this condition happens).
TEST=ci
Change-Id: I39a399a7f60b218c665d30d6c43e7c18173c0e80
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/250346
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Siva Annamalai <asiva@google.com>
os_signposts needs explicit ids to match begin and end events, whereas all other trace systems do this implicitly.
TEST=Instruments
Bug: https://github.com/dart-lang/sdk/issues/49178
Change-Id: Ifeeaa1dfe823a0d5668f48b8c584f8bdeef23212
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249951
Reviewed-by: Siva Annamalai <asiva@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
There is only one use of it in the SDK, `RuneIterator`,
one use of it outside, `TreeSet` from `package:quiver`,
and one known use in Google internal code.
There are other bi-directional iteration interfaces,
like `CharacterRange` from `package:characters`,
which can move a cursor in two directions,
but which didn't loke the `movePrevious` name and therefore
are not `BidirectionalIterator`s.
(There is also a Google internal class which mentions explicitly
why they're not `BidirectionalIterator`.)
There is no real need for this "shared" interface.
It doesn't carry its own weight.
We have no abstractions which work on the *interface*,
only code that works on one specific concrete implementation.
I'd recommend:
* `quiver` introduce a `TreeSetIterator` for their use-case
(the actual class has more public members than `BidirectionalIterator`,
which cannot be accessed through the declared type.)
* the Google internal code introduces its own interface
(which is just `Iterable` + `movePrevious`.
* The SDK will just make `RuneIterator` not implement
`BidirectionalIterator` at some (breaking) point.
Maybe just when we remove the type.
FLUTTER: Needs to land https://github.com/flutter/flutter/pull/106471 before merging this CL into Flutter.
Change-Id: Iaaa6b0c428d30eb3b68898c077d265ac0a622805
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249486
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
This reverts commit c97f7b7fad.
Reason for revert: Still breaking google3 (b/236665701)
Original change's description:
> Make `nullFuture` be per-zone.
>
> We introduced a `nullFuture` during the null-safety migration where
> we changed some methods to no longer allow returning `null`,
> and they therefore had to return a `Future`.
> That affected timing, because returning `null` was processed
> synchronously, and that change in timing made some tests fail.
> Rather that fix the fragile tests, we made the function return
> a recognizable future, a canonical `Future<Null>.value(null)`,
> and then recognized it and took a synchronous path for it.
>
> That caused other issues, because the future was created in the
> root zone. (Well, originally, it was created in the first zone
> which needed one, that was worse. Now it's created in the root zone.)
> Some code tries to contain asynchrony inside a custom zone, and
> then the get a `nullFuture` and calls `then` on it, and that
> schedules a microtask in the root zone.
> (It should probably have used the listener's zone, and not store
> a zone in the future at all, but that's how it was first done,
> and now people rely on that behavior too.)
>
> This change creates a `null` future *per zone* (lazily initialized
> when asked for). That should be sufficient because the code recognizing
> a returned `null` future is generally running in the same zone,
> but if any other code gets the `nullFuture`, it will be in the
> expected zone for where it was requested.
>
> This is a reland of commit a247b158d6
>
> Change-Id: Ieec74d6f93c57175c357ec18889144635f5bdca6
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249490
> Commit-Queue: Lasse Nielsen <lrn@google.com>
> Reviewed-by: Erik Ernst <eernst@google.com>
TBR=lrn@google.com,eernst@google.com,nbosch@google.com
Change-Id: I870285b03ec05803c5aaa6b66f9a6e6ea360d732
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249609
Reviewed-by: Lasse Nielsen <lrn@google.com>
Commit-Queue: Emmanuel Pellereau <emmanuelp@google.com>
Reviewed-by: Emmanuel Pellereau <emmanuelp@google.com>
We introduced a `nullFuture` during the null-safety migration where
we changed some methods to no longer allow returning `null`,
and they therefore had to return a `Future`.
That affected timing, because returning `null` was processed
synchronously, and that change in timing made some tests fail.
Rather that fix the fragile tests, we made the function return
a recognizable future, a canonical `Future<Null>.value(null)`,
and then recognized it and took a synchronous path for it.
That caused other issues, because the future was created in the
root zone. (Well, originally, it was created in the first zone
which needed one, that was worse. Now it's created in the root zone.)
Some code tries to contain asynchrony inside a custom zone, and
then the get a `nullFuture` and calls `then` on it, and that
schedules a microtask in the root zone.
(It should probably have used the listener's zone, and not store
a zone in the future at all, but that's how it was first done,
and now people rely on that behavior too.)
This change creates a `null` future *per zone* (lazily initialized
when asked for). That should be sufficient because the code recognizing
a returned `null` future is generally running in the same zone,
but if any other code gets the `nullFuture`, it will be in the
expected zone for where it was requested.
This is a reland of commit a247b158d6
Change-Id: Ieec74d6f93c57175c357ec18889144635f5bdca6
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249490
Commit-Queue: Lasse Nielsen <lrn@google.com>
Reviewed-by: Erik Ernst <eernst@google.com>
This reverts commit 4d750a862d.
Reason for revert: breaks google3 (b/236665701)
Original change's description:
> Make `nullFuture` be per-zone.
>
> We introduced a `nullFuture` during the null-safety migration where
> we changed some methods to no longer allow returning `null`,
> and they therefore had to return a `Future`.
> That affected timing, because returning `null` was processed
> synchronously, and that change in timing made some tests fail.
> Rather that fix the fragile tests, we made the function return
> a recognizable future, a canonical `Future<Null>.value(null)`,
> and then recognized it and took a synchronous path for it.
>
> That caused other issues, because the future was created in the
> root zone. (Well, originally, it was created in the first zone
> which needed one, that was worse. Now it's created in the root zone.)
> Some code tries to contain asynchrony inside a custom zone, and
> then the get a `nullFuture` and calls `then` on it, and that
> schedules a microtask in the root zone.
> (It should probably have used the listener's zone, and not store
> a zone in the future at all, but that's how it was first done,
> and now people rely on that behavior too.)
>
> This change creates a `null` future *per zone* (lazily initialized
> when asked for). That should be sufficient because the code recognizing
> a returned `null` future is generally running in the same zone,
> but if any other code gets the `nullFuture`, it will be in the
> expected zone for where it was requested.
>
> This is a reland of commit a247b158d6
>
> Change-Id: Ia113756de1f6d50af4b1abfec219d6b4dcd5d59b
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249488
> Reviewed-by: Nate Bosch <nbosch@google.com>
> Commit-Queue: Lasse Nielsen <lrn@google.com>
TBR=lrn@google.com,nbosch@google.com
Change-Id: I02d62d58bae33d6a606a80eb3eee2e8e721a8e20
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249620
Commit-Queue: Emmanuel Pellereau <emmanuelp@google.com>
Reviewed-by: Emmanuel Pellereau <emmanuelp@google.com>
Reviewed-by: Nate Bosch <nbosch@google.com>
We introduced a `nullFuture` during the null-safety migration where
we changed some methods to no longer allow returning `null`,
and they therefore had to return a `Future`.
That affected timing, because returning `null` was processed
synchronously, and that change in timing made some tests fail.
Rather that fix the fragile tests, we made the function return
a recognizable future, a canonical `Future<Null>.value(null)`,
and then recognized it and took a synchronous path for it.
That caused other issues, because the future was created in the
root zone. (Well, originally, it was created in the first zone
which needed one, that was worse. Now it's created in the root zone.)
Some code tries to contain asynchrony inside a custom zone, and
then the get a `nullFuture` and calls `then` on it, and that
schedules a microtask in the root zone.
(It should probably have used the listener's zone, and not store
a zone in the future at all, but that's how it was first done,
and now people rely on that behavior too.)
This change creates a `null` future *per zone* (lazily initialized
when asked for). That should be sufficient because the code recognizing
a returned `null` future is generally running in the same zone,
but if any other code gets the `nullFuture`, it will be in the
expected zone for where it was requested.
This is a reland of commit a247b158d6
Change-Id: Ia113756de1f6d50af4b1abfec219d6b4dcd5d59b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249488
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
This reverts commit a247b158d6.
Reason for revert: breaks google3. See b/236665701.
Original change's description:
> Make `nullFuture` be per-zone.
>
> We introduced a `nullFuture` during the null-safety migration where
> we changed some methods to no longer allow returning `null`,
> and they therefore had to return a `Future`.
> That affected timing, because returning `null` was processed
> synchronously, and that change in timing made some tests fail.
> Rather that fix the fragile tests, we made the function return
> a recognizable future, a canonical `Future<Null>.value(null)`,
> and then recognized it and took a synchronous path for it.
>
> That caused other issues, because the future was created in the
> root zone. (Well, originally, it was created in the first zone
> which needed one, that was worse. Now it's created in the root zone.)
> Some code tries to contain asynchrony inside a custom zone, and
> then the get a `nullFuture` and calls `then` on it, and that
> schedules a microtask in the root zone.
> (It should probably have used the listener's zone, and not store
> a zone in the future at all, but that's how it was first done,
> and now people rely on that behavior too.)
>
> This change creates a `null` future *per zone* (lazily initialized
> when asked for). That should be sufficient because the code recognizing
> a returned `null` future is generally running in the same zone,
> but if any other code gets the `nullFuture`, it will be in the
> expected zone for where it was requested.
>
> Change-Id: Ibe204eaabe175cbcbcb7822ca1f2bcdda72a478c
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/248660
> Reviewed-by: Nate Bosch <nbosch@google.com>
> Commit-Queue: Lasse Nielsen <lrn@google.com>
TBR=lrn@google.com,nbosch@google.com
Change-Id: Ic765eb1401dbff4dea6c08c007c0bf9aff186684
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249188
Reviewed-by: Oleh Prypin <oprypin@google.com>
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Emmanuel Pellereau <emmanuelp@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Commit-Queue: Emmanuel Pellereau <emmanuelp@google.com>
We introduced a `nullFuture` during the null-safety migration where
we changed some methods to no longer allow returning `null`,
and they therefore had to return a `Future`.
That affected timing, because returning `null` was processed
synchronously, and that change in timing made some tests fail.
Rather that fix the fragile tests, we made the function return
a recognizable future, a canonical `Future<Null>.value(null)`,
and then recognized it and took a synchronous path for it.
That caused other issues, because the future was created in the
root zone. (Well, originally, it was created in the first zone
which needed one, that was worse. Now it's created in the root zone.)
Some code tries to contain asynchrony inside a custom zone, and
then the get a `nullFuture` and calls `then` on it, and that
schedules a microtask in the root zone.
(It should probably have used the listener's zone, and not store
a zone in the future at all, but that's how it was first done,
and now people rely on that behavior too.)
This change creates a `null` future *per zone* (lazily initialized
when asked for). That should be sufficient because the code recognizing
a returned `null` future is generally running in the same zone,
but if any other code gets the `nullFuture`, it will be in the
expected zone for where it was requested.
Change-Id: Ibe204eaabe175cbcbcb7822ca1f2bcdda72a478c
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/248660
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
Correctness changes:
Fix SEGV by checking for overflow in computing the length of the
repeated string.
Performance changes:
Use unnested loops to fill the string instead of nested loops.
Implement `operator *` for _TwoByteString in the same way.
(The default implementation using StringBuffer causes a lot of
allocations. For example, `"α" * 10000` repeatedly adds the small
string to the StringBuffer, which repeatedly compresses a sequence of
small strings to make a bigger string to stop the list of parts
becoming too large. This compression creates a lot of small strings
with the same contents.)
Bug: 49289
Change-Id: I06c3d91b531d7e4fffc8de9f3bada3eb62ad185f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/249122
Reviewed-by: Lasse Nielsen <lrn@google.com>
Commit-Queue: Stephen Adams <sra@google.com>
The ! operator causes _CastErrors to be thrown without the
_CastError.{_url,_line,_column} fields to be set (see
runtime/vm/runtime_entry.cc:NullErrorHelper).
This CL makes the fields nullable. An alternative would be to remove the
fields entirely since the fields are not used and the position of the
failed null check is visible from top frame in the stack frame as well.
Though the fields are visible in the debugger and there are other
situations where they are correctly set (e.g. in type errors that cause
_CastError to be thrown). To avoid changing this behavior, we make them
nullable.
Fixes https://github.com/dart-lang/sdk/issues/49279
TEST=vm/dart{,_2}/regress_49279_test
Change-Id: Ieff9e84819d5afa83c8950ad1f99d684d184340a
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/248980
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Tess Strickland <sstrickl@google.com>
Relanding after https://dart-review.googlesource.com/c/sdk/+/244200.
Original change's description:
> [vm] Only call `.hashCode` once when adding to `Map` and `Set`
>
> The methods to add to hash maps and hash sets are recursive: if the
> index needs to be rehashed then the same method is called again
> after rehashing.
>
> This CL nests the actual implementation in a private method that takes
> the full hashCode as an extra argument.
>
> No significant code size or run time changes are reported on our
> benchmarks. (Our benchmarks do not contain purposefully slow hashCodes.)
>
> Note that hashCode can be called again later if rehashing of the index
> is required on adding subsequent elements.
>
> Bug: https://github.com/dart-lang/sdk/issues/48948
> Change-Id: Ia3ccff9e592d675b4922ac78c4aa7ee0287ecb50
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/243623
> Reviewed-by: Martin Kustermann <kustermann@google.com>
> Commit-Queue: Daco Harkes <dacoharkes@google.com>
Change-Id: I033bd7cc29fc812dccb6dccf0c3dca6e22cae2ca
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/248802
Commit-Queue: Tess Strickland <sstrickl@google.com>
Auto-Submit: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Tess Strickland <sstrickl@google.com>
This reverts commit 0f3dea33f9.
Reason for revert: breaks dart->engine roller as flutter still uses NullThrownError, fails analysis step
Original change's description:
> Deprecate outdated errors, drop `CastError` and `NullThrownError`.
>
> Both `CastError` and `NullThrownError` becomes just (deprecated) aliases for `TypeError`.
>
> `FallThroughError` becomes deprecated. Fall-through was made a compile-time error in Dart 2.0, the error should no longer be used.
>
> `CyclicInitializationError` is deprecated. Null safe Dart doesn't specify which error a late initialization error throws. We use internal errors now.
>
> These errors should not be needed in sound null-safe mode (if they are even needed now), and so should be removed in Dart 3.0.
>
> TEST= No new tests, should not change behavior in a significant way.
>
> Bug: https://github.com/dart-lang/sdk/issues/49141
> Change-Id: I636e9a0d0c32021d40bb819a88a1f57db6efc5a9
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/247384
> Reviewed-by: Nate Bosch <nbosch@google.com>
> Reviewed-by: Brian Quinlan <bquinlan@google.com>
> Commit-Queue: Lasse Nielsen <lrn@google.com>
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: https://github.com/dart-lang/sdk/issues/49141
Change-Id: I1b2802ec69fe654525e683527ff3554ff972f0c9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/248741
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
Reviewed-by: Alexander Thomas <athom@google.com>
Commit-Queue: Alexander Aprelev <aam@google.com>
The current implementation uses a specialized version of the
pending-event queue used by stream subscriptions to remember
pending events.
That makes the queue polymorphic, and slightly more complicated than
necessary, and that again makes further refactorings of the
Stream implementation harder.
This change moves the logic from the specialized pending-queue
into a simple function instead, so you only pay for it if you
actually use `Stream.fromIterable`.
Also allows `Stream.fromIterable` to be listened to more than once.
(It uses `Stream.multi` for the general async+sync controller API,
and it would cost extra code to make it only work once.)
Change-Id: I44b2010225cd3d32c2bcdb8a315c94881331bdae
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/248146
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
