summaryrefslogtreecommitdiff
path: root/bootstraptest
Commit message (Collapse)AuthorAgeFilesLines
* YJIT: Add codegen for Integer methods (#7665)Takashi Kokubun2023-04-051-0/+5
| | | | | | | * YJIT: Add codegen for Integer methods * YJIT: Update dependencies * YJIT: Fix Integer#[] for argc=2
* RJIT: Fix arguments for shift_stackTakashi Kokubun2023-04-031-11/+25
|
* YJIT: Test more kw and rest cases and change exit nameJimmy Miller2023-03-301-5/+7
|
* YJIT: Rest and keyword (non-supplying) (#7608)Jimmy Miller2023-03-291-0/+15
| | | | | | | | | * YJIT: Rest and keyword (non-supplying) * Update yjit/src/codegen.rs --------- Co-authored-by: Maxime Chevalier-Boisvert <maximechevalierb@gmail.com>
* YJIT: Rest and block_arg support (#7584)Jimmy Miller2023-03-241-0/+21
|
* YJIT: Use starting context for status === CantCompile (#7583)Jimmy Miller2023-03-231-0/+42
|
* Use shape information in YJIT's definedivar implementation (#7579)Ole Friis Østergaard2023-03-231-0/+47
| | | | | * Use shape information in YJIT's definedivar implementation * Handle complex shape for definedivar
* YJIT: Fix large ISeq rejection (#7576)Alan Wu2023-03-211-0/+13
| | | | | | | | | | | | We crashed in some edge cases due to the recent change to not compile encoded iseqs that are larger than `u16::MAX`. - Match the C signature of rb_yjit_constant_ic_update() and clamp down to `IseqIdx` size - Return failure instead of panicking with `unwrap()` in codegen when the iseq is too large Co-authored-by: Maxime Chevalier-Boisvert <maxime.chevalierboisvert@shopify.com> Co-authored-by: Noah Gibbs <noah.gibbs@shopify.com>
* YJIT: Fix incorrect exit in splat (#7575)Jimmy Miller2023-03-211-0/+15
| | | | | | | So by itself, this shouldn't have been a correctness issue, but we also pop the stack for block_args. Doing stack manipulation like that and then side-exiting causes issues. So, while this fixes the immediate failure, we have a bigger issue with block_args popping and then exiting that we need to deal with.
* Revert "YJIT: Rest and block_arg support (#7557)"Peter Zhu2023-03-211-21/+0
| | | | | | | | | | | | | | | | | | | | | | | This reverts commit 5d0a1ffafa61da04dbda38a5cb5565bcb8032a78. This commit is causing sequel in yjit-bench to raise with this stack trace: ``` sequel-5.64.0/lib/sequel/dataset/sql.rb:266:in `literal': wrong argument type Array (expected Proc) (TypeError) from sequel-5.64.0/lib/sequel/database/misc.rb:269:in `literal' from sequel-5.64.0/lib/sequel/adapters/shared/sqlite.rb:314:in `column_definition_default_sql' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:564:in `block in column_definition_sql' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:564:in `each' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:564:in `column_definition_sql' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:634:in `block in column_list_sql' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:634:in `map' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:634:in `column_list_sql' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:753:in `create_table_sql' from sequel-5.64.0/lib/sequel/adapters/shared/sqlite.rb:348:in `create_table_sql' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:702:in `create_table_from_generator' from sequel-5.64.0/lib/sequel/database/schema_methods.rb:203:in `create_table' from benchmarks/sequel/benchmark.rb:19:in `<main>' ```
* Remove a warning in bootstraptest/runner.rbTakashi Kokubun2023-03-191-1/+1
| | | | ../bootstraptest/runner.rb:121: warning: assigned but unused variable - e
* YJIT: Rest and block_arg support (#7557)Jimmy Miller2023-03-171-0/+21
| | | | | | | | | * YJIT: Rest and block_arg support * Update bootstraptest/test_yjit.rb --------- Co-authored-by: Maxime Chevalier-Boisvert <maximechevalierb@gmail.com>
* Skip a flaky test that might not workTakashi Kokubun2023-03-171-1/+1
|
* YJIT: Remove exit for rest and send combo (#7546)Jimmy Miller2023-03-161-0/+8
|
* Fix indirect counter incrementNobuyoshi Nakada2023-03-151-1/+2
| | | | | `*pcnt++` just dereferences `pcnt` then increments the local variable, but has no side effect.
* RJIT: Skip a flaky test_thread test for nowTakashi Kokubun2023-03-102-3/+2
| | | | and unskip a ractor test that was actually running
* YJIT: Handle special case of splat and rest lining up (#7422)Jimmy Miller2023-03-071-0/+22
| | | | | | | | | | | | | | | | If you have a method that takes rest arguments and a splat call that happens to line up perfectly with that rest, you can just dupe the array rather than move anything around. We still have to dupe, because people could have a custom to_a method or something like that which means it is hard to guarantee we have exclusive access to that array. Example: ```ruby def foo(a, b, *rest) end foo(1, 2, *[3, 4]) ```
* s/mjit/rjit/Takashi Kokubun2023-03-062-3/+3
|
* s/MJIT/RJIT/Takashi Kokubun2023-03-063-5/+5
|
* Rename MJIT filenames to RJITTakashi Kokubun2023-03-061-0/+0
|
* Add more GC guardsTakashi Kokubun2023-03-051-1/+1
|
* Deal with too slow testTakashi Kokubun2023-03-051-1/+1
|
* Support SP motion in all insnsTakashi Kokubun2023-03-051-0/+6
|
* Implement initial opt_ltTakashi Kokubun2023-03-051-0/+6
|
* Put nil on an appropriate indexTakashi Kokubun2023-03-051-0/+4
|
* add a test for RactorKoichi Sasada2023-03-061-0/+28
| | | | | Ractor should take care method cache invalidation. Added test will miss method cache on each method call.
* Another attempt to skip test_ractor on ci.rvm.jpTakashi Kokubun2023-03-031-1/+1
| | | | This reverts commit 8d31a60f47fb053bcfe0c744a89bd666dae48539.
* Fix a YJIT enablement checkTakashi Kokubun2023-03-031-1/+1
| | | | | This should be enough for `make test` and `make btest-ruby` while it doesn't work for `make btest`.
* YJIT: fix CI issue reported by Koichi caused by small stack patch (#7442)Maxime Chevalier-Boisvert2023-03-031-0/+27
| | | | | Includes small reproduction produced by Kokubun. http://ci.rvm.jp/results/trunk-yjit@ruby-sp2-docker
* Re-skip an unstable Ractor testTakashi Kokubun2023-03-021-1/+2
| | | | | https://github.com/ruby/ruby/actions/runs/4316423442/jobs/7532190115 http://ci.rvm.jp/results/trunk-yjit@ruby-sp2-docker/4466770
* Revert "Revert "Re-enable test_ractor for YJIT""Takashi Kokubun2023-03-021-5/+3
| | | | | | This reverts commit 9792d9e40f790e6deb18ead56a8befc9d5c4bc51. Ractor implementation has been rewritten. Let's see if it works now.
* YJIT: Fix cfunc splatJimmy Miller2023-03-021-1/+12
| | | Follow-up for cb8a040b7906c09d9d3ac3d3fe853f633005024f.
* `Ractor::Selector#empty?`Koichi Sasada2023-03-031-0/+96
| | | | | | It returns the waiting set is empty or not. Also add Ractor::Selector's tests.
* Enable flaky ractor testKoichi Sasada2023-03-021-2/+1
| | | | I hope a4421bd73c286253311c2cdf8c78ed258f8cff44 will solve the issue...
* Rewrite Ractor synchronization mechanismKoichi Sasada2023-03-021-4/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch rewrites Ractor synchronization mechanism, send/receive and take/yield. * API * Ractor::Selector is introduced for lightweight waiting for many ractors. * Data structure * remove `struct rb_ractor_waiting_list` and use `struct rb_ractor_queue takers_queue` to manage takers. * remove `rb_ractor_t::yield_atexit` and use `rb_ractor_t::sync::will_basket::type` to check the will. * add `rb_ractor_basket::p.take` to represent a taking ractor. * Synchronization protocol * For the Ractor local GC, `take` can not make a copy object directly so ask to generate the copy from the yielding ractor. * The following steps shows what `r1.take` does on `r0`. * step1: (r0) register `r0` into `r1`'s takers. * step2: (r0) check `r1`'s status and wakeup r0 if `r1` is waiting for yielding a value. * step3: (r0) sleep until `r1` wakes up `r0`. * The following steps shows what `Ractor.yield(v)` on `r1`. * step1: (r1) check first takers of `r1` and if there is (`r0`), make a copy object of `v` and pass it to `r0` and wakes up `r0`. * step2: (r1) if there is no taker ractors, sleep until another ractor try to take.
* YJIT: Properly deal with cfunc splat when no args needed (#7413)Jimmy Miller2023-03-011-0/+10
| | | | | | | | | | Related to: https://github.com/ruby/ruby/pull/7377 Previously it was believed that there was a problem with a combination of cfuncs + splat + send, but it turns out the same issue happened without send. For example `Integer.sqrt(1, *[])`. The issue was happened not because of send, but because of setting the wrong argc when we don't need to splat any args.
* YJIT: Reject __send__ with splat to cfunc for nowAlan Wu2023-02-271-0/+5
| | | | | | | | | | `make test-spec` revealed this issue after applying an unrelated bug fix. A crashing case is included, though I suspect there are other scenarios where it misbehaves. Don't compile for now. Note that this is *not* an issue on the 3.2.x series; it has `send_args_splat_non_iseq` which already rejects all splats to cfuncs, including sends with splats.
* YJIT: `Kernel#{is_a?,instance_of?}` fast paths (GH-7297)Jimmy Miller2023-02-151-0/+27
| | | | Co-authored-by: Maxime Chevalier-Boisvert <maxime.chevalierboisvert@shopify.com> Co-authored-by: Alan Wu <XrXr@users.noreply.github.com>
* YJIT: Check correct BOP on gen_fixnum_cmp (#7303)Takashi Kokubun2023-02-141-0/+39
|
* YJIT: Support invokesuper in a block (#7264)Maple Ong2023-02-091-0/+32
| | | | | | | Support invokesuper in a block on YJIT invokesuper previously side exited when it is in a block. To make sure we're compiling the correct method in super, we now use the local environment pointer (LEP) to get the method, which will work in a block. Co-authored-by: John Hawthorn <john@hawthorn.email>
* Implement splat for cfuncs. Split exit exit cases to better capture where we ↵Jimmy Miller2023-01-191-0/+25
| | | | | | | | | | | | | | are exiting (#6929) YJIT: Implement splat for cfuncs. Split exit cases This also implements a new check for ruby2keywords as the last argument of a splat. This does mean that we generate more code, but in actual benchmarks where we gained speed from this (binarytrees) I don't see any significant slow down. I did have to struggle here with the register allocator to find code that didn't allocate too many registers. It's a bit hard when everything is implicit. But I think I got to the minimal amount of copying and stuff given our current allocation strategy.
* Remove Encoding#replicateBenoit Daloze2023-01-111-15/+0
|
* Revert "Re-enable test_ractor for YJIT"Takashi Kokubun2023-01-091-3/+5
| | | | | | | This reverts commit 650a20a3e1205f47224a987676cdbad7d826d597. Now that 3.2.0 is released, let's disable flaky tests. Koichi said he'll rework Ractor implementation for this, and it has not been done yet.
* MJIT: Cancel all on disastrous situations (#7019)Takashi Kokubun2022-12-241-3/+3
| | | | | | | | | | I noticed this while running test_yjit with --mjit-call-threshold=1, which redefines `Integer#<`. When Ruby is monkey-patched, MJIT itself could be broken. Similarly, Ruby scripts could break MJIT in many different ways. I prepared the same set of hooks as YJIT so that we could possibly override it and disable it on those moments. Every constant under RubyVM::MJIT is private and thus it's an unsupported behavior though.
* Skip a flaky Ractor test for mswinTakashi Kokubun2022-12-211-1/+1
|
* Re-enable test_ractor for YJITTakashi Kokubun2022-12-191-4/+3
| | | | | | This would be still flaky, but we want to make sure there's no YJIT-specific issue when Ruby 3.2 is released. We might skip it again after the release.
* fixed encoding tableKoichi Sasada2022-12-161-2/+2
| | | | | This reduces global lock acquiring for reading. https://bugs.ruby-lang.org/issues/18949
* Skip yet another flaky Ractor testTakashi Kokubun2022-12-131-1/+1
|
* Skip another flaky Ractor test for YJITTakashi Kokubun2022-12-021-1/+1
|
* Skip a couple of Ractor testsTakashi Kokubun2022-12-021-2/+4
| | | | | Koichi plans to rework Ractor implementation to address these failures. He agreed to skip flaky Ractor tests for now.
View Lorry Controller Status