This way requests to compile won't end prematurely after reading
bytecode if interpreter is enabled.
Change-Id: Ic06cae98ea65bc4656ef491fbd306d34b0e1ef4c
Reviewed-on: https://dart-review.googlesource.com/75628
Reviewed-by: Zach Anderson <zra@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Before this CL, debugging code like this:
```
import "dart:developer";
main() {
Foo foo = new Foo();
foo.bar<dynamic, dynamic>(1, 2);
}
class Foo {
bar<A, B>(A a, B b) {
debugger();
print("a: $a; b: $b");
}
}
```
and trying to print for instance `a` or `b` when the debugger is stopped
in `bar` (e.g. `p a`) would yield a wrong result. I.e., this happends
when debugging methods taking type arguments, but where the type
arguments are all `dynamic`.
This also surfaces when hovering variables in e.g. VSCode with the
Dart Code plugin.
Fixes: #34353.
Change-Id: Iddf1ed119700fea61a0e29eb6ad763a8d2b657a0
Reviewed-on: https://dart-review.googlesource.com/74583
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Jens Johansen <jensj@google.com>
Previously we tried to rely on the assumption that all variables would be
boxed - so the machinery for setting correct catch-entry state only
supported tagged values and constants. However this both leads to worse code
and is not entirely correct assumption.
This also:
- renames various confusingly named classes: we move away from talking
about "catch entry state" to "catch entry moves" - because we only
record a subset of moves that needs to be performed and that does
not describe the whole state;
- refactors a bunch of associated code to be more readable and maintainable;
- adds documentation about catch implementation in optimized code
to runtime/docs/compiler;
Fixes https://github.com/flutter/flutter/issues/21685.
Change-Id: I03ae361a1bb7710acbd9f661ae014e663a163c59
Reviewed-on: https://dart-review.googlesource.com/74860
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Demote newly introduced VM class NativeEntryData from a VM instance class
to a simple TypedData wrapper class.
Change-Id: I44aacee33500c93eb283d2d349cd7a24dd94de65
Reviewed-on: https://dart-review.googlesource.com/74323
Commit-Queue: Régis Crelier <regis@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
We need to block interrupts while evaluating pragmas to prevert reentrant class finalization.
Original revision is in patchset 0.
Change-Id: I872cec4eaf4ca85567c9657c458ed39c8b2e30de
Cq-Include-Trybots: luci.dart.try:vm-kernel-win-release-x64-try, vm-kernel-optcounter-threshold-linux-release-x64-try, vm-kernel-precomp-linux-debug-x64-try, vm-kernel-precomp-linux-release-simarm-try, vm-kernel-precomp-linux-release-simarm64-try, vm-kernel-precomp-linux-release-x64-try, vm-kernel-precomp-win-release-x64-try
Reviewed-on: https://dart-review.googlesource.com/73160
Commit-Queue: Samir Jindel <sjindel@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
After this CL, the interpreter is included by default in the
JIT VM under the flag --enable-interpreter.
Reland with fix to NativeArgument setup in simulator_arm.cc
Change-Id: Ib9b4df6eb4d997dfbe361188b8a127828c1d9c6f
Reviewed-on: https://dart-review.googlesource.com/74003
Reviewed-by: Siva Annamalai <asiva@google.com>
Reviewed-by: Régis Crelier <regis@google.com>
Commit-Queue: Zach Anderson <zra@google.com>
This CL adds support for _dynamic_ calls (which are routed through
dynamic invocation forwarders) to bytecode pipeline.
This allows to skip argument type checks for non-covariant and
non-generic-convarint-impl parameters in bytecode when calling instance
members through known interface targets. In case of dynamic call, dynamic
invocation forwarders perform type checks which are otherwise skipped.
This CL also corrects argument type checks in forwarding stubs.
Forwarding stubs are synthetic methods, introduced by front-end to ensure
type safety if some of the argument type checks are skipped.
The specifics of forwarding stubs is that they should use parameter types
of a *target* of a stub for the argument type checks.
This change unifies how argument type checks are performed in bytecode
and in the VM in JIT mode. This allows to mix bytecode methods with
methods JIT-compiled from kernel AST in more cases.
Change-Id: I3a2720a488c8bbb91de1b189eb36ee8c51684a78
Reviewed-on: https://dart-review.googlesource.com/72543
Reviewed-by: Samir Jindel <sjindel@google.com>
Reviewed-by: Régis Crelier <regis@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Also move deopt id computation logic into a separate class and add a comment
explaining while deopt ids are incremented by 2.
Change-Id: Ife489be7d10c7198a8e7adf9e97e0c516d78ea55
Reviewed-on: https://dart-review.googlesource.com/72685
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
This class replaces a TypedData array used to describe native calls in the
constant pool of kernel bytecode.
Change-Id: Ic19c592fabde8c24f903d9c94a4d282da3e7b383
Reviewed-on: https://dart-review.googlesource.com/72242
Commit-Queue: Régis Crelier <regis@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
There should be no GCs between Instructions::New and
all write into the instructions object, because GC indiscriminately
write protects all code objects at the end of the GC.
In the previous version of the code assembler->GetCodeComments would
potentially trigger a GC and the code would crash trying to write
pointers into the code.
Also move FlushICache to the point after everything is written
into the code object.
Fixes#34030
Change-Id: I83742715f825f897a8612dda5ef7352c572517ee
Reviewed-on: https://dart-review.googlesource.com/72103
Reviewed-by: Siva Annamalai <asiva@google.com>
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
- Don't add dependency on fields that are either unitialized or already marked
as non guarded (guarded cid is kIllegalCid in the first case and kDynamicCid
in the second case).
- Exit early from FieldStore if guarded_cid is kDynamicCid (or we are storing
null into a nullable field).
- Don't track exactness of fields which have function type type.
Also move logic for starting to track fields into class_finalizer because we
need resolved field types to make our decisions.
Change-Id: I164b42abe3108a70b0769f4c78d92459b4e78883
Reviewed-on: https://dart-review.googlesource.com/70620
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
We say that field's static type G<T0, ..., Tn> is exact if for
any value that can be loaded from this field, its runtime type
T is such that T at G has type arguments exactly equal to
<T0, ..., Tn>.
Know if field's static type is exact allows us to apply optimizations
that require knowing type arguments e.g.
- we can fold o.f.:type_arguments
to a constant value if we know that o.f is trivially exact;
- for method invocations o.f.m(...) we can skip argument type checks
on the callee if we know that o.f is invariant (this optimization will
be enabled once multiple entry points CLs will land).
Bug: https://github.com/dart-lang/sdk/issues/31798
Change-Id: Id565046d45a842625d41feb002b65db48451034c
Reviewed-on: https://dart-review.googlesource.com/69969
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
The check which disables entrypoints in precompiled mode currently only triggers for
regular functions. Here we make it trigger for implicit closures as well.
Change-Id: I68749a2109bb24ee5d299e4b139e3b63e3be5f72
Reviewed-on: https://dart-review.googlesource.com/70340
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Samir Jindel <sjindel@google.com>
Test Plan:
Will be tested by following revisions which introduce optimizations that invoke the
new entrypoint.
Change-Id: I56d714bceb92c44a13c46d52457154b82577aa8c
Cq-Include-Trybots: luci.dart.try:vm-kernel-win-release-x64-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-release-simarm-try,vm-kernel-precomp-linux-release-simarm64-try,vm-kernel-precomp-linux-release-x64-try,vm-kernel-precomp-win-release-x64-try
Reviewed-on: https://dart-review.googlesource.com/67345
Commit-Queue: Samir Jindel <sjindel@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
Instantiator type arguments are sometimes reused instead of doing
instantiation at run time. VM has a bunch of assertions to verify
that this optimization happened. In case of bytecode pipeline, decision
to reuse instantiator type arguments is done while generating bytecode.
For this optimization to correctly happen, bytecode generator
should be able to construct instantiator type arguments the same way
as VM does at run time. Besides flattening of type arguments of
superclasses, VM shrinks them by overlapping type parameters with
type arguments of a superclass (if they match). This CL implements
such overlapping in bytecode generator.
Also, --overlap_type_arguments VM option is removed, as disabling
of type arguments overlapping would break bytecode pipeline.
Change-Id: Ib276c0e688e0e86b44278d88a5c0af09dd342798
Reviewed-on: https://dart-review.googlesource.com/70060
Reviewed-by: Régis Crelier <regis@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Until now the expression evaluation implementation was using normal kernel loader
functionality, which registered a new class with the isolate. Then the cid of that
class was set to kIllegalCid.
This caused the direct_subclasses/direct_implementors CHA information to contain a
class with kIllegalCid.
This CL fixes this by ensuring we never register the libraries/classes
created for expression evaluation (which are not even used, they are an artifact
of how the expressions are encoded in kernel).
Issue https://github.com/flutter/flutter/issues/20255
Issue https://github.com/flutter/flutter/issues/20307
Change-Id: Ie6dd76c7ff696cd8adf4f27e9a072274afd90136
Reviewed-on: https://dart-review.googlesource.com/68681
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
This reverts commit 08f59e5de3.
There are no changes to take 3 except fixing merge conflicts.
Change-Id: I45d5a16274c80f469827c19f4e4f4ca724e85a67
Cq-Include-Trybots: luci.dart.try:vm-kernel-win-release-x64-try,vm-kernel-optcounter-threshold-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-release-simarm-try,vm-kernel-precomp-linux-release-simarm64-try,vm-kernel-precomp-linux-release-x64-try,vm-kernel-precomp-win-release-x64-try
Reviewed-on: https://dart-review.googlesource.com/68363
Commit-Queue: Samir Jindel <sjindel@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
After this change, ICData map is always unpacked in CompileParsedFunctionHelper::Compile
regardless of the optimizing mode. The unpacked ICData array is passed
both to FlowGraphBuilder and to FlowGraphCompiler.
FlowGraphCompiler::FlowGraphCompiler no longer unpacks ICData array, and
the duplicated code for unpacking of ICData map is replaced with debug asserts.
Change-Id: Ib6864acdb6bcd02fe7acd1cb5bcf70d6c69808af
Reviewed-on: https://dart-review.googlesource.com/67505
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
Our IL graphs can contain portions that can not be reached due to execution
and all optimizations need to take that into account.
In this particular case AssertAssignable::Canonicalize should have taken
into account that constant instantiator type arguments might not necessarily
correspond to the correct class - because AssertAssignable is on the
path that will never be reached during execution.
Fixes https://github.com/dart-lang/sdk/issues/33999
Bug: http://dartbug.com/33999
Change-Id: Ia151a3902b2bc7f0f8d4aa6dcdb302fe6b1d626d
Reviewed-on: https://dart-review.googlesource.com/67321
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
This CL introduces, similar to RawClass::direct_subclasses_, a
RawClass::direct_implementors field which is kept up-to-date.
The generation of type testing stubs uses this cached
reverse-relationship information for faster cid-range calculation (time
spent in type testing stub generation is decreased by an order of
magnitude).
This CL also enable type testing stubs for any instantiated type.
This seems to improve
* analysis-server-cold-analysis by 6%
* analysis-server-warm-analysis by 11%
as well as other non-analysis benchmarks in JIT mode.
Issue https://github.com/dart-lang/sdk/issues/33257
Change-Id: If01ee06ac08251b2ed27f79d6b82ad7354c8336e
Reviewed-on: https://dart-review.googlesource.com/66576
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
Rationale:
While writing tests for the ongoing native 64-bit
MOD/TRUNCDIV support in AOT, I noticed a floating-point
crash in our VM due to evaluating the constant mod case:
Expect.equals(0, mod(minInt64, -1));
Expect.equals(minInt64, truncdiv(minInt64, -1));
This fixes the constant evaluation part.
https://github.com/dart-lang/sdk/issues/33967
Change-Id: I9a4e6b3cd4d0d0dee39c690d2b981b5812501be4
Reviewed-on: https://dart-review.googlesource.com/67281
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Aart Bik <ajcbik@google.com>
which was blocking the Dart SDK roll.
Revert "Take 3 for "[vm/kernel/precomp] Remove procedures from entry points files.""
This reverts commit 567109df7f.
Revert "[vm/precomp] Extend @pragma entry-points to classes."
This reverts commit 232698047c.
Change-Id: Ib63d1afb8a1c978be7ddf282af0e7d5547111cc3
Reviewed-on: https://dart-review.googlesource.com/67300
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Alexander Markov