Commit Graph

28 Commits

Author SHA1 Message Date
Asger Feldthaus a2e55c953d Add LibraryIndex class for finding classes and members by name.
The new class becomes a superclass of CoreTypes, which now gets its
functionality from that class, but also exposes commonly used built-in
members as fields.

This replaces a couple of ad-hoc linear searches spread throughout
the code.

The canonical names are not a convenient interface for transformers, so
the idea is that they can use LibraryIndex to address their own runtime
libraries.

To avoid name clashes and still have a somewhat convenient interface,
getters and setters are prefixed by 'get:' and 'set:' in the lookup
table (and only in the lookup table), and top-level members can be
accessed using '::' as the class name. This is the same convention
used by the VM, although it is currently not required to be the same.

R=dmitryas@google.com, karlklose@google.com, kmillikin@google.com

Review-Url: https://codereview.chromium.org/2706243011 .
2017-02-24 14:49:12 +01:00
Asger Feldthaus 8bfc4b47c0 Implement canonical name scheme in kernel.
This adds a class CanonicalName that can represent a library, class,
or member.  All references now go through a Reference object, which is
linked to both the AST node and its CanonicalName, so either can be
created first.

dartk now accepts multiple input files:
- If multiple dart files are given, they are all compiled.
- If multiple binaries are given, they are linked together.
Mixed dart and binary input is not supported by dartk.

dartk now has a flag --include-sdk which includes the entire SDK in
the output.  This is so the SDK can be compiled alone and then linked.

Example of compiling separately and then linking:
  dartk foo.dart -o foo.dill
  dartk main.dart -o main.dill
  dartk --include-sdk -o sdk.dill
  dartk main.dill foo.dill sdk.dill --target=vm --link -o program.dill

dartk still has incredibly slow cold start due to the analyzer loading
the dart sdk, so this does not actually speed things up at the moment.

BUG=
R=ahe@google.com, kmillikin@google.com, kustermann@google.com, sigmund@google.com

Review-Url: https://codereview.chromium.org/2665723002 .
2017-02-23 14:12:10 +01:00
Jens Johansen 86d788e699 [Kernel] replace function debuggable originalAsyncMarker
Previously a functions debuggable field was set to false in the async transformation (and because it wasn't done yet, not in async* and sync* transformation, though it should have been). Now instead, set the original async status and set debuggable on the c++ side equivalently.

Also, on the c++ side, set a functions modifier to async etc if they were originally async etc.

BUG=
R=asgerf@google.com, kmillikin@google.com, kustermann@google.com

Review-Url: https://codereview.chromium.org/2697193008 .
2017-02-22 12:12:54 +01:00
pq ce8dd9cabb Revert "Enable causal stacktrace in kernel"
This reverts commit b29829660b.

(In quest for build greenness.)

BUG=
R=kustermann@google.com

Review-Url: https://codereview.chromium.org/2697083002 .
2017-02-15 11:11:30 -08:00
Jens Johansen b29829660b Enable causal stacktrace in kernel
R=kustermann@google.com

Review-Url: https://codereview.chromium.org/2690873005 .
2017-02-15 14:17:29 +01:00
Paul Berry 7ea43baad1 Fix error "Couldn't infer type parameter" in kernel.
R=jensj@google.com

Review-Url: https://codereview.chromium.org/2666043003 .
2017-02-01 09:13:15 -08:00
Martin Kustermann 56e84a40be VM: [Kernel] Fix checked mode issues with await/async transformer
R=vegorov@google.com

Review-Url: https://codereview.chromium.org/2659393002 .
2017-01-31 11:03:10 +01:00
Martin Kustermann 673eb56471 VM: [Kernel] Fix remaining issues with kernel-based async/await implementation
This CL brings us on-par with the VM implementation in terms of tests.

  * Ensure we have saved-try-ctx/exception/stacktrace variables hoisted out for
    try-catch **and** try-finally.

  * Instead of closing the stream controller for 'async*' functions on return we
    do it inside an try-finally block so we get there in terms of normal and
    exceptional exit.

R=vegorov@google.com

Review-Url: https://codereview.chromium.org/2627873002 .
2017-01-11 19:35:02 +01:00
Jens Johansen 3f55b8e2b7 Non-format-changing kernel offset changes
Updates to kernel that sets more offsets (and introduces end offests
and 'debuggability') - but doesn't necessarily persist them,
i.e. the format doesn't change and no C++ changes are neccessary yet.

This is step #1 in introducing these things, next step(s) will be
persisting the new stuff and using it on the C++ side.

R=asgerf@google.com, kmillikin@google.com

Committed: https://github.com/dart-lang/sdk/commit/5edca8c4d3733f319ab2d328aa28ceb286a30904

Reopening as I reverted the change as it broke package:compiler.

Review-Url: https://codereview.chromium.org/2610133002 .
2017-01-10 09:16:19 +01:00
Jens Johansen 5215ec6ef2 Revert "Non-format-changing kernel offset changes"
The commit breaks package:compiler.

This reverts commit 5edca8c4d3.

BUG=

Review-Url: https://codereview.chromium.org/2614663007 .
2017-01-05 14:19:23 +01:00
Jens Johansen 5edca8c4d3 Non-format-changing kernel offset changes
Updates to kernel that sets more offsets (and introduces end offests
and 'debuggability') - but doesn't necessarily persist them,
i.e. the format doesn't change and no C++ changes are neccessary yet.

This is step #1 in introducing these things, next step(s) will be
persisting the new stuff and using it on the C++ side.

R=asgerf@google.com, kmillikin@google.com

Review-Url: https://codereview.chromium.org/2610133002 .
2017-01-05 13:25:46 +01:00
Asger Feldthaus 440813c67b Check that invocations have well-formed targets in kernel verifier.
This also fixes some issues in the frontend and transformers that
generated calls without the correct number of type arguments.

BUG=
R=kmillikin@google.com

Review URL: https://codereview.chromium.org/2533793005 .
2016-11-30 08:24:03 +01:00
Asger Feldthaus d056236180 Enable strong-mode for pkg/kernel and fix some strong-mode warnings.
BUG=
R=kmillikin@google.com

Review URL: https://codereview.chromium.org/2523673005 .
2016-11-23 16:16:37 +01:00
Asger Feldthaus 7ca4400f5b [kernel] Fix parent pointer issue in async rewriter.
BUG=
R=kmillikin@google.com

Review URL: https://codereview.chromium.org/2502943002 .
2016-11-15 14:27:22 +01:00
Kevin Millikin fb4123b566 [kernel] Kernel: Fix a bug with await in finally.
The translation for async and async* functions translates returns into
two statements, where the second one is a return.  The VM's compilation
strategy is to inline the code for all finally blocks immediately before
a return statement.

This causes future completion/stream cancellation to occur before
finally blocks, instead of after as intended.

The solution here is to translate returns into a break from a labeled
block and put a single async/async* return sequence, outside any finally
blocks, after the labeled block.

BUG=
R=asgerf@google.com

Review URL: https://codereview.chromium.org/2476513003 .
2016-11-03 10:44:24 +01:00
Kevin Millikin 5f93106653 [kernel] Support both older and newer implementations of dart:async.
The VM target's continuation transformer uses a class that was originally
named _StreamIteratorImpl and then renamed to _StreamIterator.  If looking
up the old name fails, try the new name.

BUG=
R=jensj@google.com

Review URL: https://codereview.chromium.org/2465163004 .
2016-11-02 09:32:25 +01:00
Kevin Millikin a6bba9520b [kernel] Remove BlockExpression from the Kernel language.
BlockExpression allowed statements to appear in an expression context.
It is unsafe for its original use case, allowing 'yield' to appear as an
expression, because the VM's yield implementation does not work when
there are live expression intermediate values.

Eliminate it by changing the async ExpressionLifter to return a
transformed expression and emit statements into a given list as a side
effect.  This requires implementing the transformation for all
statements to account for how they may contain a nested expression.

BUG=
R=asgerf@google.com

Review URL: https://codereview.chromium.org/2460373002 .
2016-11-01 14:13:30 +01:00
Vyacheslav Egorov b33181adbd [kernel] New VM libraries renamed _StreamIteratorImpl into _StreamInterator.
R=kmillikin@google.com
BUG=

Review URL: https://chromereviews.googleplex.com/527977013 .
2016-10-24 13:19:03 +02:00
Kevin Millikin 79743bee75 [kernel] A one-pass version of the await transformation.
By transforming subexpressions from right-to-left, the transformation knows
whether an expression is possibly live across a await when the expression is
translated.

R=asgerf@google.com

Review URL: https://chromereviews.googleplex.com/508487014 .
2016-09-27 10:57:49 +02:00
Asger Feldthaus 1d2bced4e8 [kernel] Allow async transformer to run on already-transformed code.
BUG=
R=kmillikin@google.com

Review URL: https://chromereviews.googleplex.com/504977013 .
2016-09-09 14:51:02 +02:00
Asger Feldthaus 1316073b81 [kernel] Apply async transformation when --target=vm.
BUG=
R=kustermann@google.com

Review URL: https://chromereviews.googleplex.com/504897014 .
2016-09-08 15:25:20 +02:00
Harry Terkelsen a74bb4e438 [kernel] Some small fixes, typos, etc.
R=vegorov@google.com

Review URL: https://chromereviews.googleplex.com/493937013 .
2016-08-22 09:56:05 -07:00
Vyacheslav Egorov d512d0b51e [kernel] Async rewriter: check if statements list is empty before trying to access the last statement.
BUG=
R=kustermann@google.com

Review URL: https://chromereviews.googleplex.com/488077013 .
2016-08-16 20:57:55 +02:00
Martin Kustermann 86e914b8a8 [kernel] Implement support for "await for (...)"
R=vegorov@google.com

Review URL: https://chromereviews.googleplex.com/463347013 .
2016-08-02 12:17:12 +02:00
Vyacheslav Egorov f26a9269d3 [kernel] Support async exceptions
Transformer lifts try/catch auxiliary variables that need to survive across
yield points into the context.

Drive by fix: we were calling asyncThenWrapper to create an :async_op_error
callback by mistake.

R=kustermann@google.com
BUG=

Review URL: https://chromereviews.googleplex.com/466247013 .
2016-08-02 12:17:07 +02:00
Vyacheslav Egorov 01f1fe8252 [kernel] Various fixes to make async/await work on the VM.
- Rewrite all expressions in functions, not only expression-statements.
- Inject :await_ctx_var and :await_jump_var into the wrapper.
- Don't use transformed expression directly as part of completer.complete call.

BUG=
R=kustermann@google.com

Review URL: https://chromereviews.googleplex.com/468717014 .
2016-08-02 12:15:21 +02:00
Martin Kustermann 439447c41a [kernel] Add AsyncMarker.SyncCoroutine
R=vegorov@google.com

Review URL: https://chromereviews.googleplex.com/464127013 .
2016-08-02 12:14:50 +02:00
Martin Kustermann 55fb22d91a [kernel] Implement proof-of-concept for async/async*/sync* rewriting
R=vegorov@google.com

Review URL: https://chromereviews.googleplex.com/461327013 .
2016-08-02 12:14:50 +02:00