To make implicit getters/setters frameless in common cases we use graph
intrinsics. We do this for boxed as well as unboxed fields. We do this
only for simple cases (e.g. not if field guards or possible boxing are
involved).
It also removes the hand-written assembly intrinsics for the getters,
which avoids bugs if boxing-state of field is not the same as
boxing-state of setter/getter.
Issue https://github.com/dart-lang/sdk/issues/40876
Change-Id: Id3bd7842ead0278fee61d4a21c8661ce18dbccee
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/155900
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Tess Strickland <sstrickl@google.com>
With the introduction of paged new-space, it became possible for the scavenger to fail to allocate if old-space could not grow and survivors did not pack after being reordered, and the scavenger would FATAL out. Long before that, the scavenger could fail to allocate the other semispace and FATAL out. Now an allocation failure causes the scavenge to abort, and partial forwarding to be reversed.
Change-Id: Idadfa4d9797aa7afce362bf70343e66b1b1dfeaf
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/155683
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
- Makes Future.wait a recognised function, and asserts its chained
future, _future is allocated at a known index in the context.
- Adds logic to locate, extract the chained future during lazy async
stack unwinding.
- Adds tests for the Future.wait async case.
- Minor consistency nits, comments.
This change is similar to a previous CL, adding Future.timeout support:
https://dart-review.googlesource.com/c/sdk/+/152328
Change-Id: I7439750968595d25d7bbac0068ad64fcc891e176
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/155420
Commit-Queue: Clement Skau <cskau@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
runGenKernel helper function now passes all null safety options to
gen_kernel tool implicitly, so callers don't bother.
runGenKernelWithoutStandardOptions helper is added in case those
options should not be passed (e.g. for null safety autodetection test).
This change fixes vm/dart/minimal_kernel_test and
vm/dart/bytecode_and_ast_mix_test tests (with null-safe package:path).
Change-Id: I772f715c1c84362f8a1c96e6f77cb8423f6743ab
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/155066
Reviewed-by: Régis Crelier <regis@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
(Assignment of libraries to loading units is already done in the kernel generation step.)
After compiling and before serializing, we walk the program and for each Code we assign its Instructions, CodeSourceMap and CompressedStackMap to the loading unit of that Code's defining library. Deduplication may cause Instructions, CodeSourceMaps and CompressedStackMaps to belong to more than one loading unit; in this case the objects are assigned to the root loading unit. Later they can be more precisely assigned to the dominating loading unit.
All objects except some Instructions, CodeSourceMaps and CompressedStackMaps belong to the root loading unit's snapshot. This snapshot is written like an unsplit snapshot, except that when serializing Code, we will write a reference to a stub or null when the Code's Instructions, CodeSourceMap or CompressedStackMap belongs to a non-root loading unit. The snapshots of non-root loading units contain these deferred objects and references to the corresponding Code objects to patch. The types of objects we defer (Instructions, CodeSourceMaps and CompressedStackMaps) usually represent 70+% of the snapshot size.
Bare instructions mode must be disabled when splitting because we cannot have PC-relative calls between loading units. Later we can re-enable this for calls within loading units.
Broken: Compactor probably crashes we can now have an unbounded number of image pages and the compactor assumes a fixed number.
Embedder's guide:
At compile-time, gen_snapshot should be passed --loading_unit_manifest with a path, which will enable splitting and output a mapping from loading unit ids to snapshot output paths.
At runtime, sometime during isolate startup, an embedder should call Dart_SetDeferredLoadHandler, probably near an existing call to Dart_SetLibraryTagHandler. The callback is given a loading unit id, and should eventually call Dart_DeferredLoadComplete[Error].
Bug: https://github.com/dart-lang/sdk/issues/41974
Change-Id: Ib597eb87c8cd634416d5ee1f00629c5550aebb00
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/152427
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
If inline bump-pointer allocation fails, allocation stubs must check whether the allocation result from the runtime is old and ensure it is in the remembered set if so to make write barrier elimination safe. CreateArrayInstr would directly call the runtime entry if inline bump-pointer allocation failed without performing the remembered check. This CL changes the slow path to call the AllocateArray stub instead, matching every other allocation slow path.
Bug: https://github.com/dart-lang/sdk/issues/42489
Change-Id: Ic502e43683607e1d0f85154f0e3a8c3ea11deefa
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/154623
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
This CL adds support for unboxing of return values of getters and
therefore completes the work on unboxing support for normal members
(closures are still missing).
As with existing unboxing support for methods and setters, we utilize
TFA information to proof a getter will always return non-nullable
int/double.
We then make such a getter return unboxed int/double. If there are
dynamic calls to the getter a dyn:get:* forwarder will be created
which performs boxing of the return value.
Overall this reduces RX by eliminating BoxInt64 instructions. It
sometimes increases metadata due to more dyn:get:* function objects.
=> On our main size benchmark targets show no significant change.
Issue https://github.com/dart-lang/sdk/issues/40876
Change-Id: If7450ef7e5e3fc9c2e0eaa6b86ffa817699a7e17
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/154329
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Tess Strickland <sstrickl@google.com>
CFE adds signature members to classes based on the NNBD mode of
an enclosing library, so identical anonymous mixin applications
in libraries with different NNBD modes could end up being different.
This change fixes mixin deduplication transformation to take
NNBD mode of anonymous mixin applications into account.
Fixes https://github.com/dart-lang/sdk/issues/42656
Change-Id: Ib83ae55f8193177012fa1a592fecf381ca8fdbff
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/154164
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Note: This is a reland of https://dart-review.googlesource.com/c/sdk/+/150343
Adds a new SDK hash to kernels and the VM which is optionally checked
to verify kernels are built for the same SDK as the VM.
This helps catch incompatibilities that are currently causing
subtle bugs and (not so subtle) crashes.
The SDK hash is encoded in kernels as a new field in components.
The hash is derived from the 10 byte git short hash.
This new check can be disabled via:
tools/gn.py ... --no-verify-sdk-hash
This CL bumps the min. (and max.) supported kernel format version,
making the VM backwards incompatible from this point back.
This also bumps the min. and current ABI version.
Bug: https://github.com/dart-lang/sdk/issues/41802
Change-Id: I2f85945045a603eb9dcfd1f2c0d0d024bd84a956
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/152802
Commit-Queue: Clement Skau <cskau@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
This reverts commit edde575dcd.
Reason for revert: Breaks the Dart to Flutter roll and golem
Original change's description:
> [SDK] Adds an SDK hash to kernels and the VM.
>
> Adds a new SDK hash to kernels and the VM which is optionally checked
> to verify kernels are built for the same SDK as the VM.
> This helps catch incompatibilities that are currently causing
> subtle bugs and (not so subtle) crashes.
>
> The SDK hash is encoded in kernels as a new field in components.
> The hash is derived from the 10 byte git short hash.
>
> This new check can be disabled via:
> tools/gn.py ... --no-verify-sdk-hash
>
> This CL bumps the min. (and max.) supported kernel format version,
> making the VM backwards incompatible from this point back.
>
> Bug: https://github.com/dart-lang/sdk/issues/41802
> Change-Id: I3cbb2d481239ee64dafdaa0e4aac36c80281931b
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/150343
> Commit-Queue: Clement Skau <cskau@google.com>
> Reviewed-by: Jens Johansen <jensj@google.com>
> Reviewed-by: Martin Kustermann <kustermann@google.com>
TBR=kustermann@google.com,jensj@google.com,cskau@google.com
Change-Id: I34cc7d378e2babdaaca4d932d19c19d0f35422fc
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: https://github.com/dart-lang/sdk/issues/41802
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/152703
Reviewed-by: Siva Annamalai <asiva@google.com>
Commit-Queue: Siva Annamalai <asiva@google.com>
Adds a new SDK hash to kernels and the VM which is optionally checked
to verify kernels are built for the same SDK as the VM.
This helps catch incompatibilities that are currently causing
subtle bugs and (not so subtle) crashes.
The SDK hash is encoded in kernels as a new field in components.
The hash is derived from the 10 byte git short hash.
This new check can be disabled via:
tools/gn.py ... --no-verify-sdk-hash
This CL bumps the min. (and max.) supported kernel format version,
making the VM backwards incompatible from this point back.
Bug: https://github.com/dart-lang/sdk/issues/41802
Change-Id: I3cbb2d481239ee64dafdaa0e4aac36c80281931b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/150343
Commit-Queue: Clement Skau <cskau@google.com>
Reviewed-by: Jens Johansen <jensj@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
We reconstruct ProgramInfo tree structure from the snapshot
by applying a simple graph algorithm, which takes parts of the
snapshot with clear ownership (e.g. library, class, function objects)
and then attempts to attribute the rest of the snapshot to these
clearly owned objects based on reachability.
Issue https://github.com/dart-lang/sdk/issues/41249
Cq-Include-Trybots: luci.dart.try:pkg-linux-debug-try,pkg-linux-release-try,pkg-win-release-try,pkg-mac-release-try
Change-Id: I17c0f8323ee9092a2214b18bd948ff51fa2ccc49
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/151384
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Changes:
Doing this always in precompiled mode meant increased data segment sizes
when CodeSourceMaps are stored, since encoded line/column information is
larger in the LEB-like encoding used. Now we only store column
information when we produce non-symbolic stacks, since the increased
space needed to store the columns is instead in DWARF sections and can
be stripped or elided.
Original description:
Previously, we passed line number information to the stack trace printer
and to DWARF by changing the non-special positions in the CodeSourceMap
to line numbers in precompiled mode. However, doing this lost column
information.
We get the column information back in the majority of cases by encoding
the line number and column information when neither is too large to pack
together into 30 bits. (Here, 20 bits for line and 10 bits for column.)
Otherwise, we just store the line information as before, though due to
using a bit to encode whether column info exists, it's reduced to 30
bits. If the line info is too big for that, we just return kNoSourcePos.
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-nnbd-linux-release-x64-try,vm-kernel-precomp-linux-release-x64-try,vm-kernel-precomp-linux-product-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-release-simarm_x64-try
Change-Id: Ia8baee71468da6100a170fa305d03059ffd17f78
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/151822
Commit-Queue: Tess Strickland <sstrickl@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Disassembling on every run makes it difficult to get an idea of which
specific configuration failed. Instead, just pick a configuration that
already succeeded and test that configuration with disassembly
separately so that it's clear if the disassembly was what caused the
failure.
Also add reasons to Expect.isTrue/isFalse uses in
package:vm/v8_snapshot_profile.dart. Now we don't have to track down the
specific line to find out what failed when reading logs. Another benefit
is that the error message now includes the values involved in the
failure and not just "Expect.isTrue(false)" or "Expect.isFalse(true)".
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-nnbd-linux-release-x64-try,vm-kernel-precomp-linux-release-x64-try,vm-kernel-precomp-linux-product-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-release-simarm_x64-try
Change-Id: I3d4b533f4df041dfcced4eaeb8bc5729dfad0b82
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/151502
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
AOT transformations (mixin deduplication and TFA tree shaker) were
using CanonicalName.unbind() in order to make sure that kernel
writer will not be able to write dangling references to the deleted
classes or members (to detect such dangling references earlier).
This conflicts with lazy reading from a kernel file if --from-dill
option is used: lazy reader may read a reference to such class after
unbind() and it creates a new Reference object for such canonical name.
This causes crash later when the fresh Reference is used as it doesn't
have Class node filled in.
The solution is to avoid calling CanonicalName.unbind().
Instead, it is enough to nullify node.reference.canonicalName, which
won't break CanonicalName.reference used by lazy reader.
This crash can be reproduced on tests/language_2/vm/regress_flutter_55345_test.dart
if it is compiled in AOT mode using 2-step kernel generation.
Bug: b/158853113
Change-Id: Ib2004dbbbda85d9f56adc56b48882f4ef08869a7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/151120
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
The newly added --worker-thread-priority can be used by users as an
opt-in to set the thread priority / nice value.
It is the responsibility of the user to ensure that the process has the
right to change the priority to the given value. Failure to set the
priority will be fatal.
See b/154918152
Change-Id: I3b3791e88b6ddf4fa6e39e4587f7fe1ab4d2312f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/150560
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
This option can be used to split AOT kernel generation into two rounds:
* The 1st round uses CFE to parse source files and generate intermediate
dill file in *non-AOT* mode.
* The 2nd round loads intermediate dill file and applies AOT global
transformations, then generates final dill file.
Each round is faster than one-shot generation of a dill file in AOT mode.
In order to use this 2-round compilation, command
gen_kernel <other-options> --aot -o output.dill main.dart
should be replaced with the following 2 commands:
gen_kernel <other-options> --no-aot -o intermediate.dill main.dart
gen_kernel <other-options> --aot --from-dill intermediate.dill -o output.dill main.dart
Bug: b/158101962
Change-Id: I2d4deef45b50eab11df47f4a372598e6cf1d0edf
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/150272
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
This change actually migrates vm/dart/* tests to NNBD by fixing
compile-time errors and adjusting tests where needed.
vm/dart/null_float32x4_simd_ops_test and vm/dart/null_float64x2_simd_ops_test
are deleted as they are superseded by static type checks.
vm/dart/regress_40462_test.dart is a huge source code auto-generated
by fuzzer and migrating and maintaining this source doesn't have much
value, so it is deleted.
There are still failures in strong mode due to dependencies on
the packages which are not migrated yet (https://github.com/dart-lang/sdk/issues/42146).
Migrated vm/dart/null_checks_with_dwarf_stack_traces_test fails both in
weak and strong mode due to https://github.com/dart-lang/sdk/issues/42149.
Issue: https://github.com/dart-lang/sdk/issues/41314
Change-Id: I5561f1c8705ec16def0c4e0efa495d15f4ea7259
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/149493
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Régis Crelier <regis@google.com>
Reviewed-by: Siva Annamalai <asiva@google.com>
Load*Field and Init*Field IL instructions are fused into load
instructions with 'calls_initializer' flag.
A new, more powerful elimination of lazy initializer calls uses
data flow analysis and load forwarding (place numbering).
In addition to improved elimination of initializer calls,
fusing instructions avoids extra load after field is already
loaded for lazy initialization check.
Fixes https://github.com/dart-lang/sdk/issues/41417
Change-Id: Ibb2a63f84b9b0c970db67b3d6684ec15384a64e7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/148283
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Before this change we would replace unboxed definitions (those with
representation not equal to kTagged) which is too loose and causes issues
in subsequent passes, which assume that replacements have normalised
representation.
After this change we make sure that replacement has a matching
representation.
Closes https://github.com/dart-lang/sdk/issues/35371
Fixes https://github.com/dart-lang/sdk/issues/41971
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-release-x64-try,vm-kernel-precomp-linux-debug-x64-try
Change-Id: I48543893562acf4f8516651196eba7c0c0769a22
Fixed: 35371,41971
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/103139
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Previous variant of the code did not handle FutureOr<T> and Comparable<U>
correctly - variables of these types can contain a smi value if T can be a
subclass of num or if U can be num.
Rename HasNonSmiAssignableInterface to CanReceiverBeSmiBasedOnInterfaceTarget
to better conway its meaning.
Fixes https://github.com/flutter/flutter/issues/57398
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-release-x64-try,vm-kernel-precomp-linux-product-x64-try,vm-kernel-precomp-linux-debug-x64-try
Change-Id: I67b29908e057632e62c46fb327a5c805fcc78d74
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/149063
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
The GC was sometimes incorrectly calculating the "is-unboxed" bit due to
shifter >= 64. The change in [0] fixed that issue and this CL is adding a
regresion test.
Our tests in `runtime/tests/vm/dart/*` have access to `dart:_internal`.
To make the newly added test reliable we add a `collectAllGarbage()` native
and use it in the test.
[0] https://dart-review.googlesource.com/c/sdk/+/143000
Change-Id: Ifb3f16469a9dc58623acdb029b78807ed7fcc857
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/148402
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
If a custom embedder does not use the VM's thread pool implementation
(1) but instead installs a custom message handler and invokes that
message handler on possibly unlimted number of threads, those mutator
threads will be constantly fighting to obtain TLABs.
To avoid this, the VM will ensure that only a limited number of mutators
can execute at the same time. The limit is the same as we use for
determining the VM's mutator thread pool.
In order to test this, we simulate such an embedder by disabling the
mutator pool limit, thereby allowing unlimited number of thread pool
workers.
=> The added test would timeout without without the new limit imposed
by this CL.
(1) Which has a maximum size adjusted based on TLAB size and new space size.
Issue https://github.com/dart-lang/sdk/issues/36097
Change-Id: I6d459f85fbeb3e4b1d756b752853157e40097271
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/148121
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
To avoid deadlock scenarios where an application runs N isolates on a
thread pool with T threads (T < N) where all scheduled T isolates will
do FFI call and block. That prevents any other isolates from executing.
To avoid such a scenario we add support for the isolate-group specific
thread pool to dynamically increase the maximum size if a worker is
running a isolate mutator thread, which calls to C which exits the
isolate.
=> While such an isolate is descheduled but still occupies a thread
pool worker we temporarily incrase the maximum size of the thread pool.
Issue https://github.com/dart-lang/sdk/issues/36097
Change-Id: Id61c39b06766da11f76d607ac7cbe8a8e623f250
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/148120
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Instance hash calculation on host and on target should produce same
results(like when you use simarm_x64/gen_snapshot to produce arm snapshot).
It was not producing same results because of how unboxed fields could be
taking more than one word(on 32-bit arm they are taking two words).
Hash calculation was done in terms of words so simarm_x64/gen_snapshot
saw one word unboxed field, but same calculation saw two words.
The fix is to pay special attention when iterating over unboxed field
during hash calculation and accommodate host and target word size
difference.
Fixes https://github.com/dart-lang/sdk/issues/41824.
Change-Id: Ib280736fa458a3954ca1f9994734ad04ee8db80a
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/148040
Commit-Queue: Alexander Aprelev <aam@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Since the virtual addresses in ELF snapshots are the same as in
separately saved debugging information, print the virtual address in
non-symbolic frames again when running from a snapshot compiled directly
to ELF."
Storing the relocated address as an extra field in the Image header,
which requires increasing the Image header size on 64-bit platforms,
means Image pages cannot be used reliably as HeapPages as objects no
longer start after kMaxObjectAlignment bytes.
Instead, we return to an older design that just uses the lowest bit
in the BSS offset to store whether the instructions in an Image were
compiled directly to ELF.
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-release-x64-try,vm-kernel-precomp-linux-product-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-win-release-x64-try,vm-kernel-precomp-linux-release-simarm_x64-try,vm-precomp-ffi-qemu-linux-release-arm-try
Change-Id: I3819b0dc2719d69f5e8764ca8be8c6ae7171a7bc
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/146560
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
Martin Kustermann