I don't seem to have much time for hacking on the test runner these
days, unfortunately, but I'm still happy to review changes and keep an
eye on it.
Change-Id: I273f720e7bbd0b6658d1554898cfc8ff2f815111
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/235480
Auto-Submit: Bob Nystrom <rnystrom@google.com>
Reviewed-by: Alexander Thomas <athom@google.com>
* Add team "groups" in tools/OWNERS_<group name>.
* Add top-level OWNERS as a fallback.
* Add OWNERS for all top-level directories.
* Add OWNERS to all packages.
For additional background information see go/dart-sdk-owners.
TEST=No op until code-owners is enabled.
Bug: b/200915407
Change-Id: I7fe6116cc599c749cd50ca16151d6d6a801d99d7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/229147
Reviewed-by: Jonas Termansen <sortie@google.com>
Use `hasScheme` in place of comparing against the empty string, and
`isScheme` to compare against all other schemes.
TEST=No behavior changes.
Change-Id: Ifc9fd13c6cf37933ebd4a754c4b500dedbcb291b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/231185
Reviewed-by: Kevin Moore <kevmoo@google.com>
Commit-Queue: Nate Bosch <nbosch@google.com>
These scripts will be removed in a future version of Dart,
so we switch to run ddc from either sources or from the snapshot instead.
Change-Id: I1c8e2cf6014d484b5951ec66ae9c4f667470b435
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/229303
Reviewed-by: Alexander Thomas <athom@google.com>
Reviewed-by: Nicholas Shahan <nshahan@google.com>
Changes are:
* remove case in createOutput, this is unreachable and
now handled by the subclass
* remove "|| true" - it was meant for debugging and I forgot
to delete it.
Change-Id: I32dbc34ec51c02c9f311fa1b3a246b3cb73db194
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/229380
Reviewed-by: Alexander Thomas <athom@google.com>
We will be deprecating the dart2js script shipped with the sdk under
dart-sdk/bin/. This change starts using 'dart compile js' when
running dart2js in `useSdk` mode.
For host-asserts, we continue to use a helper script that invokes
dart2js from sources. We will continue to support this helper
script, but it will become an internal only tool. It is a bit
confusing that such script lives under sdk/bin/dart2js and
sdk/bin/dart2js_developer. We may consider renaming this in a later
CL or moving it to a new location (e.g. under tools).
Change-Id: I4f426936f5e0e0a1a7854f4f555198d5ae382079
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/229060
Reviewed-by: Alexander Thomas <athom@google.com>
Dart2xCompilerConfiguration is no longer needed now that the only
use of it is to be subclassed by Dart2jsCompilerConfiguration.
Change-Id: I6c52ee477d9a3aceecbd8f65b41bc14bc2e1b58b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/229040
Reviewed-by: Alexander Thomas <athom@google.com>
Implements a backend targeting RV32GC and RV64GC, based on Linux standardizing around GC. The assembler is written to make it easy to disable usage of C, but because the sizes of some instruction sequences are compile-time constants, an additional build configuration would need to be defined to make use of it.
The assembler and disassembler cover every RV32/64GC instruction. The simulator covers all instructions except accessing CSRs and the floating point state accessible through such, include accrued exceptions and dynamic rounding mode.
Quirks:
- RISC-V is a compare-and-branch architecture, but some existing "architecture-independent" parts of the Dart compiler assume a condition code architecture. To avoid rewriting these parts, we use a peephole in the assembler to map to compare-and-branch. See Assembler::BranchIf. Luckily nothing depended on taking multiple branches on the same condition code set.
- There are no hardware overflow checks, so we must use Hacker's Delight style software checks. Often these are very cheap: if the sign of one operand is known, a single branch is needed.
- The ranges of RISC-V branches and jumps are such that we use 3 levels of generation for forward branches, instead of the 2 levels of near and far branches used on ARM[64]. Nearly all code is handled by the first two levels with 20-bits of range, with enormous regex matchers triggering the third level that uses aupic+jalr to get 32-bits of range.
- For PC-relative calls in AOT, we always generate auipc+jalr pairs with 32-bits of range, so we never generate trampolines.
- Only a subset of registers are available in some compressed instructions, so we assign the most popular uses to these registers. In particular, THR, TMP[2], CODE and PP. This has the effect of assigning CODE and PP to volatile registers in the C calling convention, whereas they are assigned preserved registers on the other architectures. As on ARM64, PP is untagged; this is so short indices can be accessed with a compressed instruction.
- There are no push or pop instructions, so combining pushes and pops is preferred so we can update SP once.
- The C calling convention has a strongly aligned stack, but unlike on ARM64 we don't need to use an alternate stack pointer. The author ensured language was added to the RISC-V psABI making the OS responsible for realigning the stack pointer for signal handlers, allowing Dart to leave the stack pointer misaligned from the C calling convention's point of view until a foreign call.
- We don't bother with the link register tracking done on ARM[64]. Instead we make use of an alternate link register to avoid inline spilling in the write barrier.
Unimplemented:
- non-trivial FFI cases
- Compressed pointers - No intention to implement.
- Unboxed SIMD - We might make use of the V extension registers when the V extension is ratified.
- BigInt intrinsics
TEST=existing tests for IL level, new tests for assembler/disassembler/simulator
Bug: https://github.com/dart-lang/sdk/issues/38587
Bug: https://github.com/dart-lang/sdk/issues/48164
Change-Id: I991d1df4be5bf55efec5371b767b332d37dfa3e0
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/217289
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Slava Egorov <vegorov@google.com>
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Providing `--log_service_response_sizes` will cause the VM to log VM service
response sizes to the provided file in CSV format.
Also added `--service_response_sizes_directory` to the service test
runner to allow for collecting response size data for the entire service
test suite.
TEST=Local
Change-Id: I7aaf4ba936e2593e67d46ff9052e2130374ef461
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/226805
Reviewed-by: Siva Annamalai <asiva@google.com>
Commit-Queue: Ben Konyi <bkonyi@google.com>
* Add support for dumping flow graphs in JSON (to make it machine
readable);
* Add support for controlling which functions are dumped through
a pragma annotation:
`@pragma('vm:testing:print-flow-graph'[, "passes filter"])`
* Replace simple matching DSL with programmatic matching
* Support obfuscated builds
Fixes https://github.com/dart-lang/sdk/issues/47340
TEST=vm/dart{,_2}/aot_prefer_equality_comparison_il_test
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-product-x64-try,vm-kernel-precomp-linux-release-x64-try
Change-Id: Ie067ba451d311e6019a8c3a88c012231b0c50eb5
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/219240
Commit-Queue: Slava Egorov <vegorov@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
* Remove all uses of `--dart2js-batch`.
* Enable batch mode for dart2js by default.
* Rename `--noBatch` to `--no-batch`.
* Make `--no-batch` affect dart2js configurations.
Change-Id: I33bdb5cfe8a82ccfc6ce9e43843f1b7a34ff1fbf
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/217006
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Commit-Queue: Alexander Thomas <athom@google.com>
Our current unit testing infrastructure does not make it possible to
test AOT compilation pipeline end-to-end, because it does not run TFA
when generating Kernel.
This makes it challenging to write regression tests for certain issues.
Instead we extend test runner with a capability to perform IL matching
when running AOT tests.
runtime/docs/infa/il_tests.md provides details on how to write such tests.
TEST=manually
Change-Id: I6f5220b814f4a5d8c053efacd3711df495dea404
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/214961
Commit-Queue: Slava Egorov <vegorov@google.com>
Reviewed-by: Alexander Markov <alexmarkov@google.com>
* Move compare_results.dart to pkg/test_runner.
* Clean up analyzer issues in compare_results.dart.
* Add a test for compare_results.dart.
* Fix compare and extend results to detect expectation changes.
Change-Id: I451ddc2d0b01a7e5fd7c9d791b000c786c2f2c93
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/214805
Commit-Queue: Alexander Thomas <athom@google.com>
Reviewed-by: William Hesse <whesse@google.com>
* Fail test.py if an invalid named configuration is used.
Cq-Include-Trybots: luci.dart.try:vm-ffi-android-product-arm64c-try
Change-Id: Ib700511a639ee39fd58441dd6fee8bd6dad8c2e5
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/214040
Reviewed-by: William Hesse <whesse@google.com>
macOS arm64 builds only create the x64 gen_snapshot by default because
they are cross-compiled in Rosetta. Adding the gen_snapshot target
ensures that we have a gen_snapshot for arm64 as well.
These bots require clang which currently isn't provisioned on shards, so
they are running on the main bot.
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-nnbd-mac-release-arm64-try
Change-Id: I842f37019a65dfb30e522d2d9bead5e62bb3ff79
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/204505
Reviewed-by: Siva Annamalai <asiva@google.com>
Reviewed-by: Jonas Termansen <sortie@google.com>
No qemu-using CI bot runs vm/cc tests currently, but this allows
developers to run vm/cc tests locally using qemu. This way, tests
like those from assembler_arm_test.cc can be tested both with and
without our own ARM7/ARM8 simulators.
Change-Id: I7fd97e411936954b59d9f0fc65e16c518e9ee332
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/204781
Commit-Queue: Tess Strickland <sstrickl@google.com>
Reviewed-by: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Clement Skau <cskau@google.com>
Error tests with explicit line/column markers, like
[error line 123, column 456]
currently require both "line" and "column".
All existing occurrences of `[error line 123, column 456]` always
refer to the most recent non-comment line *just like* the `// ^^` lines,
so omitting the line, and having the most recent non-comment line
as the *default* line, seems reasonable.
This CL makes the line optional, the new format without a "line" entry
will be just `[error column 456]`.
This allows files to change the number of lines in the code without
invalidating any later expectations.
(Or, say, remove a comment line at the top of the file!)
Omits the line entry by default when writing/updating expectations.
Change-Id: Ia0c95bf804a1c6b11c0c1621bfd481e43f8fa0c0
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/200429
Commit-Queue: Lasse R.H. Nielsen <lrn@google.com>
Reviewed-by: Bob Nystrom <rnystrom@google.com>
There are two ways the test runner might want to parse and show errors
from analyzer:
1. When a static error test's output doesn't match the expected errors.
2. When a normal runtime test produces unexpected errors.
The former were correctly migrated over to parse and read the analyzer's
JSON output. The latter wasn't. It was still trying to parse the old
"|"-separated machine format.
Worse, when migrating to the new JSON output, a try-catch-block was
added to handle any unexpected errors. That had the unintended side
effect of silently catching this parse error and then continuing. The
end result is that if an analyzer test failed on a normal runtime test,
the test runner didn't print any of the analyzer errors. :(
This fixes that. It uses a single code path to parse the JSON output to
AnalyzerError. Then, for static error tests, it converts those to
StaticErrors.
Change-Id: I51c1277f0ed04faead82f579d668f8897f379a49
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/201382
Auto-Submit: Bob Nystrom <rnystrom@google.com>
Reviewed-by: Paul Berry <paulberry@google.com>
Commit-Queue: Bob Nystrom <rnystrom@google.com>
There was a race between the subprocess listening on the socket and
the test trying to connect.
Also the Android bits of the test harness needed to be taught about
the abstract_socket_test program, similar to how process_test is
handled.
TEST=Fixes tests
Change-Id: I5e1c5d2e23e4bff2a01cd5c491a45e1c58d2578e
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/201861
Reviewed-by: Alexander Aprelev <aam@google.com>
Commit-Queue: Zach Anderson <zra@google.com>
Ryan Macnak