| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
And use this as the target of the default methods or
statically resolved super or $init calls.
The call-site change is predicated on `-Yuse-trait-statics`
as a stepping stone for experimentation / bootstrapping.
I have performed this transformation in the backend,
rather than trying to reflect this in the view from
Scala symbols + ASTs.
We also need to add an restriction related to invokespecial to Java
parents: to support a super call to one of these to implement a
super accessor, the interface must be listed as a direct parent
of the class.
The static method names has a trailing $ added to avoid duplicate
name and signature errors in classfiles.
|
| |
|
|
| |
Keep -Yopt-inline-heuristics and -Yopt-trace unchanged
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
These manual mixins were forwarding to the impl classes have
just been removed. We can now rely on default methods instead.
Update Tests:
- Fix test/files/pos/t1237.scala, we can't have an outer field
in an interface, always use the outer method.
- Don't crash on meaningless trait early init fields
test/files/neg/t2796.scala
- Remove impl class relate parts of inner class test
- Remove impl class relate parts of elidable test
- Remove impl class related reflection test.
- Remove test solely about trait impl classes renaming
- Update check file with additional stub symbol error
- Disable unstable parts of serialization test.
- TODO explain, and reset the expectation
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The previous encodings created static fields in the enclosing class
to host caches. However, this isn't an option once emit code in default
methods of interfaces, as Java interfaces don't allow private static
fields.
We could continue to emit fields, and make them public when added to
traits.
Or, as chosen in this commit, we can emulate a call-site specific
static field by using invokedynamic: when the call site is linked,
our bootstrap methid can perform one-time computation, and we can
capture the result in the CallSite.
To implement this, I've allowed encoding of arbitrary invokedynamic
calls in ApplyDynamic.
The encoding is:
ApplyDynamic(
NoSymbol.newTermSymbol(TermName("methodName")).setInfo(invokedType)
Literal(Constant(bootstrapMethodSymbol)) :: (
Literal(Constant(staticArg0)) :: Literal(Constant(staticArgN)) :: Nil
) :::
(dynArg0 :: dynArgN :: Nil)
)
So far, static args may be `MethodType`, numeric or string literals, or
method symbols, all of which can be converted to constant pool entries.
`MethodTypes` are transformed to the erased JVM type and are converted
to descriptors as String constants.
I've taken advantage of this for symbol literal caching and
for the structural call site cache.
I've also included a test case that shows how a macro could target this
(albeit using private APIs) to cache compiled regexes.
I haven't managed to use this for LambdaMetafactory yet, not sure
if the facility is general enough.
|
| |
|
|
|
|
|
|
| |
With GenBCode being the default and only supported backend for Java 8,
we can get rid of GenASM.
This commit also fixes/migrates/moves to pending/deletes tests which
depended on GenASM before.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- private-inline, t8601-closure-elim, inline-in-constructors
- test closure inlining / elimination, which is not yet implemented
in GenBCode. noted in https://github.com/scala-opt/scala/issues/14.
- constant-optimization, t7006
- no constant folding in GenBCode yet.
noted in https://github.com/scala-opt/scala/issues/29.
- patmat_opt_ignore_underscore, patmat_opt_no_nullcheck, patmat_opt_primitive_typetest
- not all optimizations in GenBCode yet.
noted in https://github.com/scala-opt/scala/issues/30.
- t3234
- tests a warning of trait inlining - trait inlining works in
GenBCode
- synchronized
- ignore inliner warnings (they changed a bit)
- t6102
- account for the changed outputo of -Ydebug has under GenBCode
|
| |
|
|
|
|
|
|
|
|
| |
Instrumentation is (currently) not supported on Avian.
This causes
- tests instrumented/InstrumentationTest.scala and
- instrumented/inline-in-constructors
to fail.
Let's skip these tests on Avian.
|
| |
|
|
|
|
| |
The differences when running with method based delambdafication aren't
important enough yet to create specialized versions that use method
based delambdafication.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
1) Deprecates much of Predef and scala.Console, especially:
- the read* methods (see below)
- the set{Out,Err,In} methods (see SI-4793)
2) Removed long-deprecated:
- Predef#exit
- Predef#error should have gone, but could not due to sbt
At least the whole source base has now been future-proofed
against the eventual removal of Predef#error.
The low justification for the read* methods should be readily
apparent: they are little used and have no call to be in global
namespace, especially given their weird ad hoc semantics and
unreasonably tempting names such as readBoolean().
3) Segregated the deprecated elements in Predef from the part
which still thrives.
4) Converted all the standard Predef implicits into implicit
classes, value classes where possible:
- ArrowAssoc, Ensuring, StringFormat, StringAdd, RichException (value)
- SeqCharSequence, ArrayCharSequence (non-value)
Non-implicit deprecated stubs prop up the names of the
formerly converting methods.
|
| |
|
|
|
|
|
| |
Expands an existing optimization for reference arrays to
apply to primitives, as well.
Fixes one aspect of SI-6247.
|
| |
|
|
| |
This reverts commit 8265175ecc42293997d59049f430396c77a2b891.
|
| | |
|
| |
|
|
|
| |
The net was cast too wide and was unsafely optimizing away array
copies.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Inlining in constructors has been disabled a long
time ago due to some VerifyErrors. Unfortunately,
@dragos cannot recall what exactly was the problem.
I tried to enable inlining in constructors and I
didn't see any problem.
`Predef.assert` calls in class constructors are one
of the biggest contributors to closure allocation
in a compiler so we better off get rid of it.
Added a test-case that checks if inlining in
constructors works properly.
Review by @magarciaEPFL and @paulp.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Extend instrumenting infrastructure to instrument
classes in `instrumented` package. This is useful
because very often you need to put your classes
into non-empty package. E.g. inliner doesn't work
properly with empty package at the moment so in
order to test any behaviour we need to put classes
in some other package that is instrumented.
Added testing code for that to instrumentation
test-case.
Review by @phaller.
|
|
|
---
(taken from README)
Tests in `instrumented/` directory are executed the same way as in `run/` but
they have additional byte-code instrumentation performed for profiling. You
should put your tests in `instrumented/` directory if you are interested in
method call counts. Examples include tests for specialization (you want to
count boxing and unboxing method calls) or high-level tests for optimizer
where you are interested if methods are successfuly inlined (so they should
not be called at runtime) or closures are eliminated (so no constructors
of closures are called).
Check `scala.tools.partest.instrumented.Instrumentation` to learn how to
use the instrumentation infrastructure.
The instrumentation itself is achieved by attaching a Java agent to the
forked VM process that injects calls to profiler. Check
`scala.tools.partest.javaagent.ProfilingAgent` for details.
---
A few notes on low-level details of this change:
* Partest now depends on asm library for byte-code instrumentation
(`build.xml`)
* Build additional jar called `scala-partest-javaagent.jar` that is
used with `-javaagent:` option. (`build.xml`)
* Set `-javaagent:` option for all tests in `instrumented/` directory.
(`RunnerManger.scala`)
* Introduce a new category of tests called `instrumented`.
* Add one instrumented test to demonstrate usage and test new
infrastructure itself. (`InstrumentationTest.scala`)
Review by @phaller.
|
