- Don't scavenge when there is no OOM reservation, so refilling the OOM reservation will get the first go once free space is available instead it being consumed by promotions.
- Increase the desired free space in new space needed to continue scavenging. Since new space became paged, 1 KB / 64 pages only guarentees the ability to allocate a 16 byte object.
- Stop issuing stress test reloads after an OOM is encountered.
- Add a HANDLESCOPE to reload to make memory collectible sooner after an error.
TEST=vm/out_of_memory
Bug: https://github.com/dart-lang/sdk/issues/48920
Bug: https://github.com/dart-lang/sdk/issues/53455
Bug: https://github.com/dart-lang/sdk/issues/43642
Change-Id: Ife8bed52e341648a66f78d55bfa60b3944be5d46
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/422324
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
HWASAN tags stack regions (and stack pointers) with 0,
but also selectively tags some stack allocations with
non-zero tags. These tags need to be cleared when
allocation goes out of scope so that stack can be
later safely reused - otherwise stale tags confuse
HWASAN instrumentation later.
This CL teaches Exceptions::JumpToFrame to do that.
CL also adds basic build changes to support building
with HWASAN, though I don't wire it into CI for now
as it can only really run on Android ARM64
TEST=manually on Android phone and on internal tests
Bug: b/374433249
Change-Id: I678443dc1de693c999a226bee0d71a6f3582d4f9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/421180
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Auto-Submit: Slava Egorov <vegorov@google.com>
This change makes it so that when the VM detects that a timer is at
least 100 ms overdue, it sends a warning in a VM Service event on the
'Timer' stream.
TEST=Built an Android Flutter app with a modified Engine that contained
a `sleep` call in `eventhandler_linux.cc` and confirmed that the warning
event got sent, built an iOS Flutter app with a modified Engine that
contained a `sleep` call in `eventhandler_macos.cc` and verified that
the warning event got sent,
pkg/vm_service/test/overdue_timer_detection_test
CoreLibraryReviewExempt: This CL does not include any core library API
changes, only VM Service implementation changes.
Change-Id: Ie8db047116b7f63cfb5413f763eaf56c7bdd6975
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/409500
Reviewed-by: Ben Konyi <bkonyi@google.com>
Commit-Queue: Derek Xu <derekx@google.com>
The way coverage collection was implemented for const constructors was a bit of a hack. After ordinary coverage collection was complete and all ordinary source ranges had been added to the report, SourceReport::CollectConstConstructorCoverageFromScripts would iterate through all the const constructors and add extra ranges to the source report that reported them as being hit. The old ranges were still in the report, reporting the constructors as missed, but the miss was overwritten by the hit when package:coverage turned the source report into a coverage report. That hacky approach didn't work for branch coverage.
I've refactored it to gather all those const constructor hits before ordinary coverage gathering, and store the hits in the script table. Then during the ordinary coverage collection flow, when we see one of those const constructors, we mark everything inside the function as hit.
This also fixes a related bug that we hadn't noticed before, where calls inside the const constructor were treated as missed.
Note: If there are dozens of const constructors in a single script, it might be worth sorting ScriptTableEntry.const_constructor_hits and then binary searching it, but I don't think that's worth doing atm.
Bug: https://github.com/dart-lang/tools/issues/513
Fixes: https://github.com/dart-lang/tools/issues/513
Change-Id: Ie66a153fbeaac5b3ecfc9a28a7f8d129ab61114a
TEST=source_report_test.cc
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/421720
Reviewed-by: Ben Konyi <bkonyi@google.com>
Commit-Queue: Liam Appelbe <liama@google.com>
This should help speeding up new isolate creation (or running dart code on mutator thread when it is supported).
TEST=ci
Change-Id: I88b5deeb33e72694cb3e2727cf5668ce15fd8c6b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/420361
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Alexander Aprelev <aam@google.com>
This reduces amount of handles and live data kept around
after each library is loaded which helps on large
apps.
TEST=ci
Cq-Include-Trybots: luci.dart.try:vm-aot-linux-debug-x64-try,vm-aot-linux-debug-x64c-try,vm-linux-debug-x64-try,vm-linux-debug-x64c-try,vm-mac-debug-arm64-try,vm-mac-debug-x64-try
Change-Id: I9ae8f31a5912c589b4668f0937e4744315c21d19
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/420321
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Slava Egorov <vegorov@google.com>
Change 8085a97a63 moved commonly
invoked StackResource::Unwind(...) from several different code
paths (simulators & interpreter) into Exceptions::JumpToFrame.
However the change did not account for the order of operations
in Exceptions::JumpToFrame which broke interpreter. This CL
fixes that by moving Interpreter::JumpToFrame under
StackResource::Unwind(...).
We also forgot to remove StackResource::Unwind from RISC-V
simulator - but that's benign because calling it the second
time simply does nothing.
TEST=vm-aot-dyn-linux-debug-x64
Change-Id: Ia93fc6997b768180fbbb5e54bb0c0fd5c713a244
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/421060
Commit-Queue: Slava Egorov <vegorov@google.com>
Auto-Submit: Slava Egorov <vegorov@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Instead of handling FFI related execution state and safepoint
in assembly handle it in runtime code.
The transition needs to be done before JumpToFrame unwinds
stack because unwinding destroys exit frame and this can't
be done at safepoint as GC might be traversing the stack.
An incorrect order of operation was manifesting as crashes in
GC when one isolate in a group was encountering a lot of
exceptions thrown from an FFI call and another isolate is
triggering GCs.
To catch this in the future added a bit of validation to
ExitSafepoint runtime call which triggers when --use-slow-path
is enabled. Though after refactoring this code does not
trigger this code path anymore because it was completely
removed - but it is better than nothing.
This CL also removes a lot of unnecessary complexity which
was associated with handling this transition in the stub
itself.
TEST=ffi/vmspecific_handle_test
Bug: b/408377905
Cq-Include-Trybots: dart/try:vm-aot-android-release-arm64c-try,vm-aot-android-release-arm_x64-try,vm-aot-asan-linux-release-x64-try,vm-aot-linux-debug-x64-try,vm-aot-linux-debug-x64c-try,vm-aot-mac-release-arm64-try,vm-aot-msan-linux-release-x64-try,vm-aot-obfuscate-linux-release-x64-try,vm-aot-optimization-level-linux-release-x64-try,vm-aot-tsan-linux-release-x64-try,vm-aot-ubsan-linux-release-x64-try,vm-aot-win-debug-x64-try,vm-aot-win-debug-x64c-try,vm-appjit-linux-debug-x64-try,vm-asan-linux-release-arm64-try,vm-asan-linux-release-x64-try,vm-checked-mac-release-arm64-try,vm-eager-optimization-linux-release-ia32-try,vm-eager-optimization-linux-release-x64-try,vm-ffi-android-debug-arm-try,vm-ffi-android-debug-arm64c-try,vm-ffi-mac-debug-simarm64_arm64-try,vm-ffi-qemu-linux-release-arm-try,vm-ffi-qemu-linux-release-riscv64-try,vm-fuchsia-release-x64-try,vm-linux-debug-ia32-try,vm-linux-debug-x64-try,vm-linux-debug-x64c-try,vm-mac-debug-arm64-try,vm-mac-debug-x64-try,vm-msan-linux-release-arm64-try,vm-msan-linux-release-x64-try,vm-reload-linux-debug-x64-try,vm-reload-rollback-linux-debug-x64-try,vm-tsan-linux-release-arm64-try,vm-tsan-linux-release-x64-try,vm-ubsan-linux-release-arm64-try,vm-ubsan-linux-release-x64-try,vm-win-debug-x64-try,vm-win-debug-x64c-try,vm-win-release-ia32-try
Change-Id: Ia073cb6bb9e1b5a0ea8514c7e048cee6019b84d6
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/420324
Commit-Queue: Slava Egorov <vegorov@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
This will play better with native tooling compared to Dart_LoadELF.
Cq-Include-Trybots: luci.dart.try:vm-fuchsia-release-arm64-try,vm-fuchsia-release-x64-try
Change-Id: If5265d079200d7be754b82a9775d977aa66c15f9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/419684
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Zijie He <zijiehe@google.com>
This should improve DumpStackTrace from assertion failures, though x64 still doesn't work for lack of frame pointers in the runtime code.
TEST=manually add assertion failure.
Change-Id: Ie632f79b21fd14a2ece3cde23736adc097938315
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/419800
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
- Add `target-arch` option to `dart compile exe|aot-snapshot`
- Add artifacts cache in `~/.dart`, and download gen_snapshot and
dartaotruntime from cloud storage if necessary
I've manually built an aot snapshot and exe on macOS ARM64 for Linux
X64, copied them over to a linux machine, and made sure they work.
Change-Id: I74dd581ecb573ff4069a2fe19f9d04aaa352538e
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/417801
Reviewed-by: Tess Strickland <sstrickl@google.com>
Commit-Queue: Ivan Inozemtsev <iinozemtsev@google.com>
Reviewed-by: Ben Konyi <bkonyi@google.com>
Before this change, Slot::Equals would hit the default UNREACHABLE()
case if used to compare two equal native slots. That's surprising and
also unavoidable if code comparing arbitrary slots wants to check Slot
equality, since there's no existing predicate to determine whether an
arbitrary slot is native.
TEST=ci
Cq-Include-Trybots: luci.dart.try:vm-aot-linux-debug-x64-try,vm-aot-mac-release-arm64-try,vm-aot-mac-product-arm64-try,vm-aot-linux-product-x64-try,vm-aot-dwarf-linux-product-x64-try,vm-linux-debug-x64-try,vm-mac-debug-arm64-try,vm-mac-release-arm64-try
Change-Id: Ia794a33fc607013bf35435b0427ceb02bdae4b0d
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/419560
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
This change adds pragmas which can be used in dart:* libraries to
specify classes and members which are used by the compiler
to implement certain language features. Compiler can insert
references to those classes and members when desugaring language
constructs, so dynamic module code can use these classes and members
implicitly even if they are private.
Previously, we required to specify these members in the
dynamic_interface.yaml before dynamic modules could use them.
However, this is cumbersome and error-prone.
Now, the following pragmas are recognized:
@pragma('dyn-module:language-impl:extendable')
@pragma('dyn-module:language-impl:can-be-overridden')
@pragma('dyn-module:language-impl:callable')
If dynamic interface yaml file is specified during compilation,
then classes/members annotated with these pragmas are treated as if
they are used in corresponding sections of dynamic_interface.yaml.
If dynamic interface yaml file is not specified, these pragmas
are ignored.
Bug: b/395992622
TEST=existing
CoreLibraryReviewExempt: no API changes, only adding pragmas.
Change-Id: I6606467fc5e5264a2565d7eb8e8baccc2eebea84
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/418663
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
For out-of-memory events vm reserves (and if used, then replenishes) small chunk of memory for handling those, so preallocation of oom and stacktrace is not needed.
Unwind error can be created once as part of vm launch, kept in vm isolate.
This should reduce overhead of starting new isolate.
TEST=ci
Change-Id: Ibe2c85f250a521a8e838cc5690984fe9fa287a98
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/419261
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Alexander Aprelev <aam@google.com>
There were `pubspec.yaml` files in `tools/` that were not
included in the global package config.
That means that their imports were irrelevant, which could
be misleading. And if any tool would look at the
`pubspec.yaml` file, it might be inconsistent with
the actual package resolution.
Makes every `pubspec.yaml` file either be included by the
`tools/generate_package_config.dart` script, or deletes them
if they seem to be stale and unused.
(Compare vs. `git ls-files '**/pubspec.yaml'`.)
Excepted `tools/dart2js/sourceMapViewer/pubspec.yaml`.
The entire directory might be stale. The dependencies of
that pubspec are not SDK dependencies otherwise,
the pubspec has no SDK min-version, which is now a requirement,
and the README refers to a *packages directory*.
Keeping as-is and filing issue to have owners take a look.
Added `lib/` directory to `tools/` to avoid `tools/bots/`
being inside the package URI root of the `tools/` package,
which would cause its `../../pkg/...` import to fail.
Makes every `pubspec.yaml` file use a `^...` SDK constraint instead
of the longer `>= ... < ...` format.
Tested: No new tests, goal is to keep running the same way
Change-Id: I688e463fe985fc4de43550a1f4c7ff350536cffc
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/414020
Commit-Queue: Lasse Nielsen <lrn@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Sigurd Meldgaard <sigurdm@google.com>
Reviewed-by: Brian Quinlan <bquinlan@google.com>
Intern the following fields:
* Category names
* Event labels
* Debug annotation keys and values
This significantly reduces the size of the timeline
(e.g. a timeline containing 60k slices goes from 15Mb
to 5Mb timeline)
This relands commit f2614d24f8
with a fix for Android build.
TEST=ci and manually
Cq-Include-Trybots: luci.dart.try:vm-ffi-android-release-arm64c-try
Change-Id: I88d4c5e1142ff66b270a22b82bacd1e9313fa953
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/418220
Reviewed-by: Derek Xu <derekx@google.com>
Intern the following fields:
* Category names
* Event labels
* Debug annotation keys and values
This significantly reduces the size of the timeline
(e.g. a timeline containing 60k slices goes from 15Mb
to 5Mb timeline)
TEST=ci and manually
Change-Id: I59e850279b6714b8b75c5b8e79e6a8b0a981a261
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/417202
Reviewed-by: Derek Xu <derekx@google.com>
Consolidate code populating TracePacket in a single
function instead of multiple helpers. Most of the
duplication in the previous code was caused by
how Duration events were handled - this CL
avoids this duplication by making the core helper
takes event_type, timestamp and timeline event.
This allows to perform Duration split simply by
doing:
PopulateAndWritePacket(TimelineEvent::kBegin, event.TimeOrigin(), event);
PopulateAndWritePacket(TimelineEvent::kEnd, event.TimeEnd(), event);
Instead of duplicating large chunks of code.
TEST=ci
Change-Id: Ica5d00264f3e872b939059ea90fcbfbf6c8768e4
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/417201
Reviewed-by: Derek Xu <derekx@google.com>
Derek Xu