Call sites targeting a procedure annotated with `external-effect` will
not produce any code, including the argument which will not be
evaluated.
However, the single parameter will be treated as 'live' for the purposes
of any global analysis the backends do. This is useful for things like
protobuf shaking where a user may want to retain certain protobuf
messages without actually emitting the code that retains those messages.
Today this functionality is available internally in the vm and wasm SDK
libraries. dart2js has similar functionality represented via the
opaqueTrue and opaqueFalse booleans (which will cause conditional
branches to get shaken after analysis). This will replace dart2js's
opaque(True/False).
This also adds validation to the frontend to ensure a method annotated
with 'external-effect' is well-formed.
Change-Id: If1c4096673e655c58fe7638840a16125003e7809
Tested: Backend tests for codegen were added. A frontend test was added for the validation. A language test was added to confirm the behavior.
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/476020
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Nate Biggs <natebiggs@google.com>
To support checking for dynamically-callable targets in dart dynamic
modules, we need to use a bit in the Function header in AOT. Currently
all 32 bits of `kind_tags_` are in use. To make space for that new bit,
we need to evict one of the current properties that is not needed by AOT.
Among them, `IsRedirectingFactory` made the top of the list. It has only
one use in non-AOT logic.
TEST=existing
Bug: b/448095881
Change-Id: I3112d8865523696ed8e906a6c59f53c23db3090a
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/498281
Auto-Submit: Sigmund Cherem <sigmund@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
The 2nd argument of _simpleInstanceOf call is always a
Constant (type). However, a Redefinition can be inserted in the middle.
So, instead of ArgumentAt(1)->AsConstant()->value() it is
more safe to use ArgumentValueAt(1)->BoundConstant() as
BindsToConstant/BoundConstant unwraps Redefinition(s) via
OriginalDefinition().
TEST=runtime/tests/vm/dart/regress_63211_test.dart
Fixes https://github.com/dart-lang/sdk/issues/63211
Change-Id: Ie4a473ebe2deee8562e6634a792f02b0dcefc918
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/497761
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
The five binary operators on Int32x4 (+, -, |, &, ^) were never added
to the recognized-method list as graph intrinsics, so calls to them
were left as runtime calls through external-name bodies. In JIT the
call specializer picked kInt32x4Cid from IC feedback and emitted a
native SimdOpInstr, but AOT has no IC feedback and therefore fell back
to boxed calls, making Int32x4List inner loops 10-70x slower than both
the JIT version and a hand-written scalar equivalent.
This CL wires the same specialization paths that already exist for
Float32x4.+,-,*,/:
- recognize the five operators as graph intrinsics and mark them
with `@pragma("vm:recognized", "graph-intrinsic")` plus an
exact-result-type pragma;
- add Build_Int32x4{Add,Sub,BitAnd,BitOr,BitXor} helpers that
delegate to the existing BuildSimdOp;
- extend SimdOpInstr::KindForOperator and CreateFromCall;
- extend CallSpecializer::InlineSimdOp and TryInlineRecognizedMethod
so the non-speculative null-check path used for Float32x4 operators
in AOT also applies here.
Measured on macOS arm64 (M-series), `dart compile exe`:
Issue 63217 orSimd : 12.58 -> 0.32 us/iter (39x)
Issue 63217 andNotSimd : 23.51 -> 0.34 us/iter (69x)
Issue 53662 mandelbrot : 4038.5 -> 55.5 ms (72x)
A new benchmark benchmarks/SimdInt32x4 exercises all five operators
with a scalar and a SIMD variant so the specialization stays covered
by the benchmark bots; it is registered in Omnibus and OmnibusDeferred.
Existing tests/lib/typed_data/simd_*_test.dart still pass in JIT and
AOT.
TEST=tests/lib/typed_data/int32x4_arithmetic_test; benchmarks/SimdInt32x4
Bug: https://github.com/dart-lang/sdk/issues/53662
Bug: https://github.com/dart-lang/sdk/issues/63217
Change-Id: I9b76ab4fff228ff1a5e3d3c86f4bfc059e66a49a
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/497000
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Slava Egorov <vegorov@google.com>
Auto-Submit: Modestas Valauskas <valauskasmodestas@gmail.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
Some objects have their canonical bit set but don't have the deeply immutable bit set. It's simpler if they have this bit clear so there is only cluster need, and these objects are instances of VM-internal classes that don't really do canonicalization.
Also give the oddballs their header hash while doing the rest of their initialization. This makes it easier to use the as snapshot base objects when there is no finalize VM isolate step.
TEST=ci
Change-Id: I7bdf7dc59b04ae30f49c5c67c0dc4f69c2d2524f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/497121
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
When enabled, indirect calls and jumps must set X7 to a label value that matches the label in a so-called landing pad at the call target. This check does not apply if the indirect call is made through X1/RA, X5/RA2 or X7/T2, with the expectation that the compiler will limit use of these registers for logically direct calls like an AUPIC+JALR pair, or for software guarded branches like a bounds-checked jump table. We use FAR_TMP for logically direct jumps that exceed J-type range, and blocking it from register allocation limits the availablity of gadgets that set X7.
Compare x64's IBT and arm64's BTI.
TEST=ci
Change-Id: Ic00195eb3c04cc10379f9965e05ef84052e02933
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/494861
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Make VM always return empty array in the response.
Current implementation for this field comes with a bunch of complexity
because it locks message handler and then invokes Dart code which
makes it difficult to reason about various invariants. This code is
furthermore demonstrated to cause deadlocks. Given that nobody uses
it - it is simpler to remove this code altogether.
Fixes https://github.com/flutter/flutter/issues/185156
TEST=ci
Change-Id: I497210e0f1542860caa0d765d634f8ec6a6a6964
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/496340
Reviewed-by: Ben Konyi <bkonyi@google.com>
Commit-Queue: Slava Egorov <vegorov@google.com>
Currently, the failure message for EXPECT_STREQ just prints the
(escaped) expected and actual strings on failure, but if the
expected and actual strings are mostly the same, it can be difficult
to determine what the difference actually is.
Instead, do a bit of work to find the common prefix and suffix of
the expected and actual strings, so that the actual difference is
more apparent. If the lengths of the common prefix and suffix
combined is considered too short to separately print them, then
the output is the same as before.
TEST=ci (failure output change only, so tested with
vm/cc/SourceReport_Coverage output on vm-dyn-linux-debug-x64)
Change-Id: Iecaacde6ebf6628643bdf1440c6d77d4ff6599af
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/495980
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
Also allows the --print-classes flag to be used in the AOT runtime if
dynamic modules are enabled and crashes with an appropriate error
message if class finalization fails when loading a member from bytecode.
TEST=co19/LibTest/math/Rectangle co19/LibTest/math/MutableRectangle
Cq-Include-Trybots: luci.dart.try:vm-dyn-linux-debug-x64-try,vm-aot-dyn-linux-debug-x64-try,vm-aot-dyn-linux-product-x64-try
Change-Id: I1921aa0189eb587cd4658c592a779be190af092e
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/495724
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
We can do a "native" stack walk starting from the interrupt context, a Dart stack walk starting from the exit frame, or a Dart stack walk starting from the interrupt context / simulator state / interpreter state.
Decide which kind of stack walk to do once instead both inside and outside CollectSample. Remove CollectSample, as we should not need SEH guards now that we have accurate stack bounds.
TEST=ci
Bug: https://github.com/dart-lang/sdk/issues/63105
Change-Id: I8f46f2ab860220e0be373f02aa882ac096bb2fb1
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/493969
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Dynamic modules may make dynamic calls to compiled methods that had no
dynamic calls in the original whole world compilation. If this happens,
then dynamically create an interpreted dynamic invocation forwarder that
then appropiately checks and delegates to the compiled function.
TEST=co19/LibTest
Cq-Include-Trybots: luci.dart.try:vm-dyn-linux-debug-x64-try,vm-aot-dyn-linux-debug-x64-try,vm-aot-dyn-linux-product-x64-try
Change-Id: I283ca501f50118606650cf434b8debe30a43f676
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/494520
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
This class is structurally equivalent to _HashVMBase and the only reason
for it to exist for inability of normal Dart classes to extend
_HashVMBase - but this can worked around by properly configuring
_HashVMBase class in bootstrapping.
TEST=ci
Change-Id: I5ca401e274920d2b4739424c6b6595306a6a6964
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/494140
Commit-Queue: Slava Egorov <vegorov@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
The DartRuntimeService based VM service implementation now has support
for launching DDS instances and responding to _yieldControlToDDS RPC
invocations from DDS instances.
package:vm_service test suite is ~97% passing with this change.
TEST=Local testing.
Change-Id: I2f2f1b0926845134578f08d073ed7606f1fc4173
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/490320
Reviewed-by: Jessy Yameogo <yjessy@google.com>
Reviewed-by: Nicholas Shahan <nshahan@google.com>
This is a reland of commit 88496ba1c3
Fixes on top of the original change:
* Closure layout is changed to avoid gap between fixed fields and
variable-length elements on compressed pointers architecture.
This gap was causing crashes in the GC when scanning closure
objects.
* pkg/vm_snapshot_analysis/test/instruction_sizes_test is fixed
on arm64 by decreasing threshold for detecting size changes.
Original change's description:
> [vm,dart2bytecode,modular_aot] Variable-length closure objects
>
> Extend closure objects with variable number of elements to capture.
> This is needed to support capturing multiple independent contexts
> after capturing is computed in the front-end.
>
> The following fixed Closure fields are moved into variable-length
> elements:
> - delayed type arguments;
> - instantiator type arguments;
> - function type arguments;
> - context.
>
> Number of elements and presence/indices of various type arguments
> are encoded into the new length_and_flags field in the Closure.
>
> Most closure objects don't need any of the type arguments so this
> change will reduce average Closure object size.
TEST=ci
Issue: https://github.com/dart-lang/sdk/issues/61572
Issue: https://github.com/dart-lang/sdk/issues/61635
Change-Id: I8685e632e2d0832766ecdc470f3cf9a6b880de48
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/494243
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Tess Strickland <sstrickl@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
The handling of entry frame in PrintStackFrames was broken, now it's
fixed.
In addition, the stack printer now print various info about the static
slots in stack frames, as well as printing out the objects in slots.
Now that entry frames are handled properly, printing with no limit
prints _all_ stack frames on the interpreter stack, noting call
boundaries as they are passed, instead of stopping at the closest
entry frame.
Move the separator between frame from the stack pointer for a frame
to the frame pointer for a frame, meaning that the metadata before
the frame pointer is printed as part of the previous frame, not the
current one.
TEST=ci (manual testing while debugging)
Cq-Include-Trybots: luci.dart.try:vm-dyn-linux-debug-x64-try,vm-aot-dyn-linux-product-x64-try
Change-Id: Ia2ad07c832b791f0c2ce2bbdfbfc32d5d8968476
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/493402
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Tess Strickland <sstrickl@google.com>
Extend closure objects with variable number of elements to capture.
This is needed to support capturing multiple independent contexts
after capturing is computed in the front-end.
The following fixed Closure fields are moved into variable-length
elements:
- delayed type arguments;
- instantiator type arguments;
- function type arguments;
- context.
Number of elements and presence/indices of various type arguments
are encoded into the new length_and_flags field in the Closure.
Most closure objects don't need any of the type arguments so this
change will reduce average Closure object size.
TEST=ci
Issue: https://github.com/dart-lang/sdk/issues/61572
Issue: https://github.com/dart-lang/sdk/issues/61635
Change-Id: I7ca5cec0fd8725c432a01d51781fb14e803997dd
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/489482
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
Reviewed-by: Tess Strickland <sstrickl@google.com>
Fix the following cases of incorrect optimizations:
* In TFA, when anlyzing a dynamic call with unknown receiver type,
do not assume that all possible targets can be computed at compile
time (if there can be dynamically loaded classes).
* In the AOT, disable optimizations for dynamic calls with unique
selectors (if there can be dynamically loaded classes).
TEST=pkg/dynamic_modules/test/data/dyn_call_from_host
Change-Id: I39d620aae3c116de03d4a2a3fd61864d88c48c8e
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/493960
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
Remove special case for null in message snapshots; snapshots are sometimes read or written with no current isolate group.
Currently still all copies pointing into the VM isolate.
TEST=ci
Change-Id: I4d2e35a01880885d4e92e1c623c0f39a35e06065
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/493866
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Alexander Aprelev <aam@google.com>
Remove special case for null in message snapshots; snapshots are sometimes read or written with no current isolate group.
Currently still all copies pointing into the VM isolate.
TEST=ci
Change-Id: I6fa47b2c04b0abaae74210c86f9fdf17ec932db7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/493222
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Instead of using the Function String-creating methods, use the
Function CString-creating methods which avoids allocation that
causes issues when called from within the interpreter (especially
in debug mode, where asserts are triggered).
TEST=ci
Cq-Include-Trybots: luci.dart.try:vm-dyn-linux-debug-x64-try
Change-Id: Ic43a831a6ec1ee4327e1fdf4dfc64bd09af5c04a
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/493801
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Auto-Submit: Tess Strickland <sstrickl@google.com>
Nate Biggs