The current hot-reload implementation [0] will perform a reload by
first sending OOB messages to all isolates and waiting until those OOB
messages are being handled. The handler of the OOB message will block
the thread (and unschedule isolate) and notify the thread performing
reload it's ready.
This requires that all isolates within a group can actually run & block.
This is the case for the VM implementation of isolates (as they are
run an unlimited size thread pool).
Though flutter seems to multiplex several engine isolates on the same OS
thread. Reloading can then result in one engine isolate performing
reload waiting for another to act on the OOB message (which it will not
do as it's multiplexed on the same thread as the former).
Now that we have a more flexible safepointing mechanism (introduced in
[1]) we can utilize for hot reloading by introducing a new "reloading"
safepoint level.
Reload safepoints
-----------------------
We introduce a new safepoint level (SafepointLevel::kGCAndDeoptAndReload).
Being at a "reload safepoint" implies being at a "deopt safepoint"
which implies being at a "gc safepoint".
Code has to explicitly opt-into making safepoint checks participate /
check into "reload safepoints" using [ReloadParticipationScope]. We do
that at certain well-defined places where reload is possible (e.g. event
loop boundaries, descheduling of isolates, OOM message processing, ...).
While running under [NoReloadScope] we disable checking into "reload
safepoints".
Initiator of hot-reload
-----------------------
When a mutator initiates a reload operation (e.g. as part of a
`ReloadSources` `vm-service` API call) it will use a
[ReloadSafepointOperationScope] to get all other mutators to a
safepoint.
For mutators that aren't already at a "reload safepoint", we'll
notify them via an OOB message (instead of scheduling kVMInterrupt).
While waiting for all mutators to check into a "reload safepoint", the
thread is itself at a safepoint (as other mutators may perform lower
level safepoint operations - e.g. GC, Deopt, ...)
Once all mutators are at a "reload safepoint" the thread will take
ownership of all safepoint levels.
Other mutators
-----------------------
Mutators can be at a "reload safepoint" already (e.g. isolate is not
scheduled). If they try to exit safepoint they will block until the
reload operation is finished.
Mutators that are not at a "reload safepoint" (e.g. executing Dart or VM
code) will be sent an OOB message indicating it should check into a
"reload safepoint". We assume mutators make progress until they can
process OOB message.
Mutators may run under a [NoReloadScope] when handling the OOM message.
In that case they will not check into the "reload safepoint" and simply
ignore the message. To ensure the thread will eventually check-in,
we'll make the destructor of [~NoReloadScope] check & send itself a new OOB
message indicating reload should happen. Eventually getting the mutator
to process the OOM message (which is a well-defined place where we can
check into the reload safepoint).
Non-isolate mutators such as the background compiler do not react to OOB
messages. This means that either those mutators have to be stopped (e.g.
bg compiler) before initiating a reload safepoint operation, the
threads have to explicitly opt-into participating in reload safepoints
or the threads have to deschedule themselves eventually.
Misc
----
Owning a reload safepoint operation implies also owning the deopt &
gc safepoint operation. Yet some code would like to ensure it actually
runs under a [DeoptSafepointOperatoinScope]/[GCSafepointOperationScope].
=> The `Thread::OwnsGCSafepoint()` handles that.
While performing hot-reload we may exercise common code (e.g. kernel
loader, ...) that acquires safepoint locks. Normally it's disallows to
acquire safepoint locks while holding a safepoint operation (since
mutators may be stopped at places where they hold locks, creating
deadlock scenarios).
=> We explicitly opt code into participating in reload safepointing
requests. Those well-defined places aren't holding safepoint locks.
=> The `Thread::CanAcquireSafepointLocks()` will return `true` despite
owning a reload operation. (But if one also holds deopt/gc safepoint
operation it will return false)
Example where this matters: As part of hot-reload, we load kernel which
may create new symbols. The symbol creation code may acquire the symbol
lock and `InsertNewOrGet()` a symbol. This is safe as other mutators
don't hold the symbol lock at reload safepoints. The same cannot be said
for Deopt/GC safepoint operations - as they can interrupt code at many
more places where there's no guarantee that no locks are held.
[0] https://dart-review.googlesource.com/c/sdk/+/187461
[1] https://dart-review.googlesource.com/c/sdk/+/196927
Issue https://github.com/flutter/flutter/issues/124546
TEST=Newly added Reload_* tests.
Change-Id: I6842d7d2b284d043cc047fd702b7c5c7dd1fa3c5
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/296183
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
An embedder (or the VM) can exit an isolate via `Thread::ExitIsolate()`
at a point where there's still active state (e.g. dart frames).
Because of this the VM has so far conservatively retained the [Thread]
object of dart mutators throughout the isolate's lifetime. After which
is was manually `delete`ed. We'd never re-use those [Thread] objects (we
do re-use [Thread] objects of non-dart-mutator threads).
When exiting via `Thread::ExitIsolate()` with active state, the mutator
was assumed to be at-safepoint at all levels. It was removed from the
thread registry's active threads. This also means that when e.g. GC runs
it can't use the thread registry to find all active threads it may
need to scan, instead it uses [Isolate::mutator_thread_] of all isolates.
This causes a variety of subtle issues, but the main one that motivated
this change is the following:
If a thread obtains a safepoint operation it means all other mutators
are parked. The thread owning the safepoint can do whatever it likes.
When introducing reload operation safepoints, a thread may want to
ReloadSafepointOperation reload(thread);
...
// Compile sources.
{
TransitionVMToNative transition(thread);
// Will temporarily exit & re-enter current isolate.
response_port = Dart_NewNativePort();
Dart_PostCObject(kernel_isolate_port, ...);
// Wait on [response_port] for response.
}
This will cause the reloading thread to own the reload safepoint
operation but still transition states and even exit/re-enter the
isolate. Though this is currently not possible in the way enter/exit is
implemented.
So we'll refactor this fragile code in the following way:
* Move thread enter/exit logic entirely to the [Thread] object.
* Keep used threads in the thread registry's active list.
=> This allows us to keep various state on the [Thread] and thereby
avoids clearing it when suspending & re-initialing it when resuming
=> It makes nested `Thread::ExitIsolate()` faster as we mainly have
to enter safepoint (avoid acquiring threads lock, avoid releasing
storebuffers, ...)
=> It makes nested `Thread::EnterIsolate()` faster as we mainly have
to leave the safepoint (avoid acquiring threads lock, avoid acquiring
storebuffers, ...).
=> A mutator can now own a safepoint operation (e.g. reload safepoint
operation) and still `ExitSafepoint()` / `EnterSafepoint()` safely -
as those are based on the normal `EnterSafepoint()` and
`LeaveSafepoint()` APIs.
* We separate
- Suspend & Resume of a dart mutator (possibly with active stack)
- Setup & Reset of state only relevant for dart mutators
- Setup & Reset of state relevant for any mutator
* We unify how the [Thread] objects are freed between dart mutator and
non-dart mutators: [Thread] objects without state can be given back to
the [ThreadRegistry] and re-used (instead of being deleted in
`Isolate::~Isolate`)
* We have capability to free [Thread] objects if a dart mutator has an
empty stack & re-use for another isolate of the same group.
(In future we may have N Thread objects for N cores and the threads
would even maintain their TLABs when switching between isolates)
* Since we allow reusing of [Thread] objects also for dart mutators now,
we have extensive asserts to ensure they are "clean" when they get
into the free list and come out "clean" again.
TEST=ci
Change-Id: Id85e8e484efd98d28e323b33795716420e619986
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/296585
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Martin Kustermann <kustermann@google.com>
For every isolate there should be only one mutator with
a unique [Thread] object.
We change existing tests that use this functionality to instead use
`Thread::{Enter,Exit}IsolateGroupAsHelper`. It also results in a net
removal of code.
TEST=ci
Change-Id: Ic326e868a98ddedbab5b8c429252d38ea71bbf04
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/295940
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
The PC marker slots are transiently invalid during a compaction, but the profiler's stack walker no longer uses them, as they are completely absent in bare instructions mode.
TEST=ci
Change-Id: I3dba373ccccaba48e6dcc2b01d48e5e75620a4db
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/284744
Reviewed-by: Ben Konyi <bkonyi@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Our compiler shouldn't depend on current isolate, since it can
run on any isolate within an IG.
Doing this change, reveals two existing dependencies on current
isolate from compiler
- resolving native symbols in unoptimized compilations
- issuing of debug events for breakpoints
For the former we'll re-enter the currently active isolate that
triggered unoptimized compilation.
=> We may want to change that embedder API to not be based on
handles and instead give embedder a simple `const char*`.
For the ladder we'll enter the isolate corresponding to the
breakpoint debug event to be issued. We are at place where
all mutators are stopped, so that does seem okish.
=> Future could remove this by making Object Id Ring per-IG
Issue https://github.com/dart-lang/sdk/issues/48523
TEST=service_2/break_on_function_many_child_isolates_test/dds
Change-Id: Id246db5972ae505e82f637ce04bb2302bed76257
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/278901
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
- Flip flag to make strong null safety the default
- Remove code that auto detects null safety mode from source files,
it is necessary to specify --no-strong-null-safety to opt out.
- Retains sniffing of AOT/JIT snapshots and kernel files to determine
null safety mode, the opt out has to be done when generating these
file.
TEST=ci
Change-Id: If2c9608eedb7c46d9c3cd85e261ee9640e0d28eb
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/261140
Reviewed-by: Alexander Thomas <athom@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Commit-Queue: Siva Annamalai <asiva@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
We have many tests that explicitly use --optimization-counter-threshold=
flag. We also have a CI builder that uses this flag for all tests.
This is an issue for cases where `kernel-isolate` is not AppJIT'ed,
which is the case for simulators and ia32. Especially in debug builds
this causes a very-very slow time-to main due to JIT'ing the CFE code in
`kernel-isolate` and running flow graph checker etc (in debug mode).
This causes various tests to sporadically hit the timeout limit, become
flaky and require gardening attention.
As workarounds: the actual threshold was modified in tests, status files
were updated to mark tests as Pass,Slow etc.
=> This Cl will make `kernel-isolate` no longer be affected by the
`--optimization-counter-threshold`
=> This should make the cycle times faster on those modes and avoid
flaky timeouts that gardeners constantly have to pay attention to.
TEST=ci
Change-Id: Ia58e807b22f69f924315a43c6764427afe398ee6
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/266683
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Martin Kustermann <kustermann@google.com>
* Merge ClassTable and SharedClassTable back together;
* Simplify handling of multiple arrays growing in sync;
* Refactor how reload deals with ClassTable.
The last change is the most important because it makes it
much easier to reason about the code. We move away from
copying bits and pieces of the class table and shared
class table into reload contexts.
Having two class table fields in the isolate group makes
it easier to reason about. One field contains program
class table (one modified by kernel loader and accessed
by various program structure cid lookups) and heap
walk class table (used by GC visitors). Normally these
two fields point to the same class table, but during
hot reload we temporary split them apart: original
class table is kept as a heap walk class table, while
program class table is replaced by a clone and updated
by reload.
If reload succeeds we drop original class table and
set program class table as heap walk one.
If reload fails we drop the program class table and
restore original one from heap walk table.
TEST=ci
Cq-Include-Trybots: luci.dart.try:vm-kernel-reload-linux-release-x64-try,vm-kernel-reload-linux-debug-x64-try,vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-reload-rollback-linux-release-x64-try,vm-kernel-linux-debug-x64-try,vm-kernel-precomp-tsan-linux-release-x64-try,vm-kernel-tsan-linux-release-x64-try,vm-kernel-precomp-asan-linux-release-x64-try,vm-kernel-asan-linux-release-x64-try
Change-Id: I8b66259fcc474dea7dd2af063e4772df99be06c4
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/258361
Commit-Queue: Slava Egorov <vegorov@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
``` before
$ dart isolate_exit_pause_time_test.dart
.Min(RunTimeRaw): 0.001 ms.
.Avg(RunTimeRaw): 212.399592 ms.
.Percentile50(RunTimeRaw): 49.055 ms.
.Percentile90(RunTimeRaw): 809.049 ms.
.Percentile95(RunTimeRaw): 1009.049 ms.
.Percentile99(RunTimeRaw): 1169.049 ms.
.Max(RunTimeRaw): 1208.049 ms.
```
``` after
$ dart isolate_exit_pause_time_test.dart
.Min(RunTimeRaw): 0.001 ms.
.Avg(RunTimeRaw): 53.87510125 ms.
.Percentile50(RunTimeRaw): 8.05 ms.
.Percentile90(RunTimeRaw): 174.005 ms.
.Percentile95(RunTimeRaw): 217.975 ms.
.Percentile99(RunTimeRaw): 275.033 ms.
.Max(RunTimeRaw): 314.033 ms.
```
Remaining latency after the change is due to large root set formed by pointers_to_verify_at_exit_ that have to be visited during GC.
Fixes https://github.com/dart-lang/sdk/issues/49050
TEST=ci
Change-Id: I1ed7af017f80890900b137a1514e5ffdeb63f53f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/246482
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Alexander Aprelev <aam@google.com>
The new implementation is based on suspend/resume stubs and doesn't
use desugaring of async functions on kernel AST.
Previously, new implementation of async/async* was only supported in
AOT mode. This change adds all necessary bits for the JIT mode:
* Suspending variable-length frames (for unoptimized code).
* Handling of Code and pool pointers in Dart stack frames.
* OSR.
* Deoptimization.
* Hot reload.
* Debugger.
The new implementation is not enabled in JIT mode yet.
Design doc: go/compact-async-await.
TEST=ci
Issue: https://github.com/dart-lang/sdk/issues/48378
Change-Id: I477d6684bdce7cbc1edb179ae2271ff598b7dcc5
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/246081
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
Isolate may be entering or leaving the group while a scavenge is running and using the active mutator count to decide the size of the next to-space. A stale value seen by the scavenger only has performance implications rather than correctness implications, which will get fixed at the next scavenge.
TEST=iso-stress-linux
Bug: https://github.com/dart-lang/sdk/issues/48921
Change-Id: I171e5a460eadba3db6659dccda29c863549d27ea
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/242925
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Rounding usage up to the TLAB causes excessive growth because it gives the growth policy a very inflated view of the survivor ratio.
Ensure new-space grows specifically to provide enough TLABs for active mutators.
TEST=golem
Change-Id: Iab01df18c6a8ac996e0f0b2707c9da7e14c7f4a4
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/241689
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
1. Run native finalizers before sending isolate exit message. This way
users can rely on the isolate exit message for both (1) no Dart code
executing anymore on that isolate _and_ (2) all native finalizers
having run. Alternatively, we could opt to not provide this second
guarantee and document this.
2. Fix UBSAN error by making SetArgumentTo42's argument a `void*`.
3. Fix bug in debug mode, stop reading tags from forwarding addresses.
4. Small optimization: Don't promote 0 bytes from new to old space.
5. Skip finalizer_isolate_groups_run_gc_test in optcounter mode.
In hot reload mode, GC is guaranteed to run. On all other bots GC
does not run (nothing happens in the other isolate group). However,
in optcounter mode enough happens in the other isolate group to
trigger GCs. Alternatively, we could loosen the test to allow either
a GC to happen or not.
TEST=runtime/tests/vm/dart(_2)/finalizer/*
TEST=runtime/tests/vm/dart(_2)/isolates/fast_object_copy_test.dart
TEST=runtime/vm/object_test.cc
TEST=tests/ffi(_2)/vmspecific_native_finalizer_*
Closes: https://github.com/dart-lang/sdk/issues/48740
Closes: https://github.com/dart-lang/sdk/issues/48715
Closes: https://github.com/dart-lang/sdk/issues/48674
Change-Id: I5e260e087aef48524f2214f5b332caeda18f2e37
Cq-Include-Trybots: luci.dart.try:vm-kernel-ubsan-linux-release-x64-try,vm-kernel-precomp-ubsan-linux-release-x64-try,vm-canary-linux-debug-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-optcounter-threshold-linux-release-ia32-try,app-kernel-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/240043
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Daco Harkes <dacoharkes@google.com>
This CL implements `NativeFinalizer` in the GC.
`FinalizerEntry`s are extended to track `external_size` and in which
`Heap::Space` the finalizable value is.
On attaching a native finalizer, the external size is added to the
relevant heap. When the finalizable value is promoted from new to old
space, the external size is promoted as well. And when a native
finalizer is run or is detached, the external size is removed from the
relevant heap again.
In contrast to Dart `Finalizer`s, `NativeFinalizer`s are run on isolate
shutdown.
When the `NativeFinalizer`s themselves are collected, the finalizers are
not run. Users should stick the native finalizer in a global variable to
ensure finalization. We will revisit this design when we add send and
exit support, because there is a design space to explore what to do in
that case. This current solution promises the least to users.
In this implementation native finalizers have a Dart entry to clean up
the entries from the `all_entries` field of the finalizer. We should
consider using another data structure that avoids the need for this Dart
entry. See the TODO left in the code.
Bug: https://github.com/dart-lang/sdk/issues/47777
TEST=runtime/tests/vm/dart(_2)/isolates/fast_object_copy_test.dart
TEST=runtime/vm/object_test.cc
TEST=tests/ffi(_2)/vmspecific_native_finalizer_*
Change-Id: I8f594c80c3c344ad83e1f2de10de028eb8456121
Cq-Include-Trybots: luci.dart.try:vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-reload-linux-debug-x64-try,vm-ffi-android-debug-arm64c-try,dart-sdk-mac-arm64-try,vm-kernel-mac-release-arm64-try,pkg-mac-release-arm64-try,vm-kernel-precomp-nnbd-mac-release-arm64-try,vm-kernel-win-debug-x64c-try,vm-kernel-win-debug-x64-try,vm-kernel-precomp-win-debug-x64c-try,vm-kernel-nnbd-win-release-ia32-try,vm-ffi-android-debug-arm-try,vm-precomp-ffi-qemu-linux-release-arm-try,vm-kernel-mac-debug-x64-try,vm-kernel-nnbd-mac-debug-x64-try,vm-kernel-nnbd-linux-debug-ia32-try,benchmark-linux-try,flutter-frontend-try,pkg-linux-debug-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-gcc-linux-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64c-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/236320
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Daco Harkes <dacoharkes@google.com>
Original CL in patchset 1.
Split-off https://dart-review.googlesource.com/c/sdk/+/238341
And pulled in fix https://dart-review.googlesource.com/c/sdk/+/238582
(Should merge cleanly when this lands later.)
This CL implements the `Finalizer` in the GC.
The GC is specially aware of two types of objects for the purposes of
running finalizers.
1) `FinalizerEntry`
2) `Finalizer` (`FinalizerBase`, `_FinalizerImpl`)
A `FinalizerEntry` contains the `value`, the optional `detach` key, and
the `token`, and a reference to the `finalizer`.
An entry only holds on weakly to the value, detach key, and finalizer.
(Similar to how `WeakReference` only holds on weakly to target).
A `Finalizer` contains all entries, a list of entries of which the value
is collected, and a reference to the isolate.
When a the value of an entry is GCed, the enry is added over to the
collected list.
If any entry is moved to the collected list, a message is sent that
invokes the finalizer to call the callback on all entries in that list.
When a finalizer is detached by the user, the entry token is set to the
entry itself and is removed from the all entries set.
This ensures that if the entry was already moved to the collected list,
the finalizer is not executed.
To speed up detaching, we use a weak map from detach keys to list of
entries. This ensures entries can be GCed.
Both the scavenger and marker tasks process finalizer entries in
parallel.
Parallel tasks use an atomic exchange on the head of the collected
entries list, ensuring no entries get lost.
The mutator thread is guaranteed to be stopped when processing entries.
This ensures that we do not need barriers for moving entries into the
finalizers collected list.
Dart reads and replaces the collected entries list also with an atomic
exchange, ensuring the GC doesn't run in between a load/store.
When a finalizer gets posted a message to process finalized objects, it
is being kept alive by the message.
An alternative design would be to pre-allocate a `WeakReference` in the
finalizer pointing to the finalizer, and send that itself.
This would be at the cost of an extra object.
Send and exit is not supported in this CL, support will be added in a
follow up CL. Trying to send will throw.
Bug: https://github.com/dart-lang/sdk/issues/47777
TEST=runtime/tests/vm/dart/finalizer/*
TEST=runtime/tests/vm/dart_2/isolates/fast_object_copy_test.dart
TEST=runtime/vm/object_test.cc
Change-Id: Ibdfeadc16d5d69ade50aae5b9f794284c4c4dbab
Cq-Include-Trybots: luci.dart.try:vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-reload-linux-debug-x64-try,vm-ffi-android-debug-arm64c-try,dart-sdk-mac-arm64-try,vm-kernel-mac-release-arm64-try,pkg-mac-release-arm64-try,vm-kernel-precomp-nnbd-mac-release-arm64-try,vm-kernel-win-debug-x64c-try,vm-kernel-win-debug-x64-try,vm-kernel-precomp-win-debug-x64c-try,vm-kernel-nnbd-win-release-ia32-try,vm-ffi-android-debug-arm-try,vm-precomp-ffi-qemu-linux-release-arm-try,vm-kernel-mac-debug-x64-try,vm-kernel-nnbd-mac-debug-x64-try,vm-kernel-nnbd-linux-debug-ia32-try,benchmark-linux-try,flutter-analyze-try,flutter-frontend-try,pkg-linux-debug-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-gcc-linux-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64c-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/238086
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Daco Harkes <dacoharkes@google.com>
This reverts commit 7dca34c235.
Reason for revert: b/226085355 dart_vm_test crashing. Unclear what
the cause is. Reverting so we can triage the issue.
TEST=This is a revert.
Original change's description:
> [vm] Implement `Finalizer`
>
> This CL implements the `Finalizer` in the GC.
>
> (This CL does not yet implement `NativeFinalizer`.)
>
> The GC is specially aware of two types of objects for the purposes of
> running finalizers.
>
> 1) `FinalizerEntry`
> 2) `Finalizer` (`FinalizerBase`, `_FinalizerImpl`)
>
> A `FinalizerEntry` contains the `value`, the optional `detach` key, and
> the `token`, and a reference to the `finalizer`.
> An entry only holds on weakly to the value, detach key, and finalizer.
> (Similar to how `WeakReference` only holds on weakly to target).
>
> A `Finalizer` contains all entries, a list of entries of which the value
> is collected, and a reference to the isolate.
>
> When a the value of an entry is GCed, the enry is added over to the
> collected list.
> If any entry is moved to the collected list, a message is sent that
> invokes the finalizer to call the callback on all entries in that list.
>
> When a finalizer is detached by the user, the entry token is set to the
> entry itself and is removed from the all entries set.
> This ensures that if the entry was already moved to the collected list,
> the finalizer is not executed.
>
> To speed up detaching, we use a weak map from detach keys to list of
> entries. This ensures entries can be GCed.
>
> Both the scavenger and marker tasks process finalizer entries in
> parallel.
> Parallel tasks use an atomic exchange on the head of the collected
> entries list, ensuring no entries get lost.
> The mutator thread is guaranteed to be stopped when processing entries.
> This ensures that we do not need barriers for moving entries into the
> finalizers collected list.
> Dart reads and replaces the collected entries list also with an atomic
> exchange, ensuring the GC doesn't run in between a load/store.
>
> When a finalizer gets posted a message to process finalized objects, it
> is being kept alive by the message.
> An alternative design would be to pre-allocate a `WeakReference` in the
> finalizer pointing to the finalizer, and send that itself.
> This would be at the cost of an extra object.
>
> Send and exit is not supported in this CL, support will be added in a
> follow up CL. Trying to send will throw.
>
> Bug: https://github.com/dart-lang/sdk/issues/47777
>
> TEST=runtime/tests/vm/dart/finalizer/*
> TEST=runtime/tests/vm/dart_2/isolates/fast_object_copy_test.dart
> TEST=runtime/vm/object_test.cc
>
> Change-Id: I03e6b4a46212316254bf46ba3f2df333abaa686c
> Cq-Include-Trybots: luci.dart.try:vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-reload-linux-debug-x64-try,vm-ffi-android-debug-arm64c-try,dart-sdk-mac-arm64-try,vm-kernel-mac-release-arm64-try,pkg-mac-release-arm64-try,vm-kernel-precomp-nnbd-mac-release-arm64-try,vm-kernel-win-debug-x64c-try,vm-kernel-win-debug-x64-try,vm-kernel-precomp-win-debug-x64c-try,vm-kernel-nnbd-win-release-ia32-try,vm-ffi-android-debug-arm-try,vm-precomp-ffi-qemu-linux-release-arm-try,vm-kernel-mac-debug-x64-try,vm-kernel-nnbd-mac-debug-x64-try,vm-kernel-nnbd-linux-debug-ia32-try,benchmark-linux-try,flutter-analyze-try,flutter-frontend-try,pkg-linux-debug-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-gcc-linux-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64c-try
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/229544
> Reviewed-by: Lasse Nielsen <lrn@google.com>
> Reviewed-by: Slava Egorov <vegorov@google.com>
> Reviewed-by: Martin Kustermann <kustermann@google.com>
> Reviewed-by: Ryan Macnak <rmacnak@google.com>
> Commit-Queue: Daco Harkes <dacoharkes@google.com>
TBR=lrn@google.com,vegorov@google.com,kustermann@google.com,rmacnak@google.com,dacoharkes@google.com
Change-Id: I991f6e49896d18a8d70210cf315d858b462d66c9
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: https://github.com/dart-lang/sdk/issues/47777
Cq-Include-Trybots: luci.dart.try:vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-reload-linux-debug-x64-try,vm-ffi-android-debug-arm64c-try,dart-sdk-mac-arm64-try,vm-kernel-mac-release-arm64-try,pkg-mac-release-arm64-try,vm-kernel-precomp-nnbd-mac-release-arm64-try,vm-kernel-win-debug-x64c-try,vm-kernel-win-debug-x64-try,vm-kernel-precomp-win-debug-x64c-try,vm-kernel-nnbd-win-release-ia32-try,vm-ffi-android-debug-arm-try,vm-precomp-ffi-qemu-linux-release-arm-try,vm-kernel-mac-debug-x64-try,vm-kernel-nnbd-mac-debug-x64-try,vm-kernel-nnbd-linux-debug-ia32-try,benchmark-linux-try,flutter-analyze-try,flutter-frontend-try,pkg-linux-debug-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-gcc-linux-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64c-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/238080
Reviewed-by: Slava Egorov <vegorov@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Commit-Queue: Daco Harkes <dacoharkes@google.com>
This CL implements the `Finalizer` in the GC.
(This CL does not yet implement `NativeFinalizer`.)
The GC is specially aware of two types of objects for the purposes of
running finalizers.
1) `FinalizerEntry`
2) `Finalizer` (`FinalizerBase`, `_FinalizerImpl`)
A `FinalizerEntry` contains the `value`, the optional `detach` key, and
the `token`, and a reference to the `finalizer`.
An entry only holds on weakly to the value, detach key, and finalizer.
(Similar to how `WeakReference` only holds on weakly to target).
A `Finalizer` contains all entries, a list of entries of which the value
is collected, and a reference to the isolate.
When a the value of an entry is GCed, the enry is added over to the
collected list.
If any entry is moved to the collected list, a message is sent that
invokes the finalizer to call the callback on all entries in that list.
When a finalizer is detached by the user, the entry token is set to the
entry itself and is removed from the all entries set.
This ensures that if the entry was already moved to the collected list,
the finalizer is not executed.
To speed up detaching, we use a weak map from detach keys to list of
entries. This ensures entries can be GCed.
Both the scavenger and marker tasks process finalizer entries in
parallel.
Parallel tasks use an atomic exchange on the head of the collected
entries list, ensuring no entries get lost.
The mutator thread is guaranteed to be stopped when processing entries.
This ensures that we do not need barriers for moving entries into the
finalizers collected list.
Dart reads and replaces the collected entries list also with an atomic
exchange, ensuring the GC doesn't run in between a load/store.
When a finalizer gets posted a message to process finalized objects, it
is being kept alive by the message.
An alternative design would be to pre-allocate a `WeakReference` in the
finalizer pointing to the finalizer, and send that itself.
This would be at the cost of an extra object.
Send and exit is not supported in this CL, support will be added in a
follow up CL. Trying to send will throw.
Bug: https://github.com/dart-lang/sdk/issues/47777
TEST=runtime/tests/vm/dart/finalizer/*
TEST=runtime/tests/vm/dart_2/isolates/fast_object_copy_test.dart
TEST=runtime/vm/object_test.cc
Change-Id: I03e6b4a46212316254bf46ba3f2df333abaa686c
Cq-Include-Trybots: luci.dart.try:vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-reload-linux-debug-x64-try,vm-ffi-android-debug-arm64c-try,dart-sdk-mac-arm64-try,vm-kernel-mac-release-arm64-try,pkg-mac-release-arm64-try,vm-kernel-precomp-nnbd-mac-release-arm64-try,vm-kernel-win-debug-x64c-try,vm-kernel-win-debug-x64-try,vm-kernel-precomp-win-debug-x64c-try,vm-kernel-nnbd-win-release-ia32-try,vm-ffi-android-debug-arm-try,vm-precomp-ffi-qemu-linux-release-arm-try,vm-kernel-mac-debug-x64-try,vm-kernel-nnbd-mac-debug-x64-try,vm-kernel-nnbd-linux-debug-ia32-try,benchmark-linux-try,flutter-analyze-try,flutter-frontend-try,pkg-linux-debug-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-gcc-linux-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64c-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/229544
Reviewed-by: Lasse Nielsen <lrn@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Daco Harkes <dacoharkes@google.com>
These GCs usually did not free very much memory but did consume a lot of CPU. They would on low-powered, low-memory devices often take ~1s, during which time the OS might decide the OOM signal wasn't working and kill us before the compaction can complete and free pages. Instead, only release pooled memory.
Also use more appropriate GCReasons in calls of CollectMost/AllGarbage.
TEST=ci
Bug: b/216333343
Change-Id: Ia56b9ca409410f17d40508c69fb1bc9df0ce4028
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/235300
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Branch coverage can't be enabled by a simple global flag, because this
causes problems when functions in an app-jit snapshot that didn't have
the flag enabled are compiled by a VM that has the flag enabled. In
particular, the coverage array will see a different set of token
positions, causing some important asserts to fail.
Change-Id: I35227252b5d271f20b01de99466c06708ec83ed8
TEST=CI
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/223360
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Liam Appelbe <liama@google.com>
System isolates should be able to register service extensions with the
exception of a few special isolates:
- Kernel isolate
- Service isolate
- VM isolate
Required to help resolve b/203694081
TEST=CQ
Change-Id: I1f11acd89132636039a817a5dfddb6189817a7fe
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/222032
Reviewed-by: Siva Annamalai <asiva@google.com>
Commit-Queue: Ben Konyi <bkonyi@google.com>
This region is mostly unused after loading, but it may be accessed again if
- Isolate.spawn is invoke with isolate groups disabled
- A secondary snapshot produced by splitting is loaded
- An external typed data in the snapshot is accessed (usually a kernel file)
- Likely other cases
Even if these cases did not exist, the region is often part of a shared library and so unable to be released independently.
madvise(DONT_NEED) on this region will cause the OS to release the memory in this region but keep the address space reservation and mapping. If it is touched again, it will be brought back in from the file. If it is not backed by a file, such as malloc memory, it will be brought back in as zeros and the program will likely fail.
TEST=ci
Bug: https://github.com/dart-lang/sdk/issues/44019
Bug: https://github.com/flutter/flutter/issues/92120
Change-Id: I315a049b0f7d440e181d0a5e87fa6770a2fd4f79
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/216580
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Siva Annamalai <asiva@google.com>
Now that support for native extensions (aka dart-ext:*) has been
removed, the kernel loader is guaranteed not to call out to embedder
anymore (it did so before for loading shared libraries).
This allos us to clean up some code as well as extend the scope of our
NoActiveIsolateScope.
TEST=ci
Change-Id: I5c8c9ee47298fdd316583ba3aed74772c1c7783f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/212296
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Martin Kustermann <kustermann@google.com>
Fix labels for sentinel, transition_sentinel, ImmutableArray, LinkedHashMap and LinkedHashSet.
Display unlimited children when look at successors, as the worst case is the size of the field table. For predecessors and dominators, the worst case is the whole heap.
TEST=ci
Change-Id: I6fcfdfb0833d58c9ac4f586b823244d817aeba27
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/209841
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
This CL
* Removes conditional Skip markers in status files
* Removes some outdated TODOs in the code
* Deletes tests/lib_2/isolate/isolate_stress_test
=> It's skipped everywhere, we no longer support isolates on web
* Fixes standalone{,_2}/io/directory_list_sync_test:
=> Instead of relying on the Platform.script which could be an ELF
file lying somewhere else, use Platform.executable instead
* Fixes standalone{,_2}/io/unix_socket_test:
=> The server used `sub.cancel()` to close it's reading end while
the client tried to write to it. This is racy.
If the server goes first, the write on client will throw (since
no error listener was installed on socket.done)
Avoid the race by coordinating client/server.
Issue https://github.com/dart-lang/sdk/issues/36097
TEST=Enables more tests.
Change-Id: I0464f81f540f2eda23ef4199cc536e53e6a809b9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/207120
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Martin Kustermann <kustermann@google.com>
Storing a persistent handle in [Message] objects instead of the
[Bequest] (which is just a wrapper around a persistent handle) will
allow us to use the same mechanism for sending copies of transitive
object graphs.
Since messages can have maps inside them that need rehashing, we'll
make the persistent handle point to an array of length 2 of the format:
array = [<message>, <array-of-objects-in-message-to-rehash>]
The sendAndExit doesn't use the second part atm (since it preserves
identities and therefore avoids the need for rehashing).
Though sending transitive copies of object graphs will require this
functionality (a follow-up CL which will land with this one).
For ease of reviewing this CL was split out.
Issue https://github.com/dart-lang/sdk/issues/36097
TEST=Refactoring, relying on existing test coverage.
Change-Id: I2afa78b42ef82d46477579623fd54f027136333f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/203769
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
Attempts to acquire safepoint lock while in SafepointOperationScope could result
in deadlocks if somebody else was holding that lock when they were forced to a safepoint.
Clean up few places where locks were acquired in SafepointOperationScope.
Introduce StoppedMutatorsScope and GroupDebugger::RunUnderReadLockIfNeeded that acquires
a lock only if it runs outside of StoppedMutatorsScope - to prevent such deadlocks.
Also fix tsan warning about data race around message_notify_callback by making it atomic.
TEST=tsan runs of debugger CI tests
Fixes https://github.com/dart-lang/sdk/issues/45527
Fixes https://github.com/dart-lang/sdk/issues/45549
Issue https://github.com/dart-lang/sdk/issues/36097
Change-Id: Ibb53d9ce760d869e044e17075aeebf20fc0016a5
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/193582
Commit-Queue: Alexander Aprelev <aam@google.com>
Reviewed-by: Ben Konyi <bkonyi@google.com>
CodeBreakpoint class tracks patching of the call targets for the purpose of routing the call to the debugger runtime.
With this CL CodeBreakpoint instance is managed by GroupDebugger, debugger class owned by IsolateGroup, and can be associated with multiple BreakpointLocation, one location per isolate that set a breakpoint at particular code location.
TEST=observatory/tests/service/break_on_function_child_isolate_test
Issue https://github.com/dart-lang/sdk/issues/36097
Change-Id: I3b3d0974a0fb9668742e5c1e2a2ce13b84b1d630
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/190732
Commit-Queue: Alexander Aprelev <aam@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
To avoid a O(n) traversal of isolates in order to find pending deopts,
we'll move the pending deopt state to the Thread. It's really a property
of a thread: when the thread's stack gets unwound - via return or
exceptional flow - we remove pending deopts.
This CL also encapsulates the pending deopt logic into a [PendingDeopts]
class - thereby avoiding logic around manipulating this array being
spread around. It simplifies the already very big [Isolate] class.
TEST=Refactoring of existing code.
Issue https://github.com/dart-lang/sdk/issues/36097
Closes https://github.com/dart-lang/sdk/issues/45168
Change-Id: I5a9b18c577f21a0f2a88332b1bc26641ec39d2af
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/188521
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
Martin Kustermann