Activity
From 09/13/2025 to 09/19/2025
09/18/2025
- 09:52 PM Revision e74f5f74 (git): ZJIT: Compile sendforward with dynamic dispatch (#14501)
- * ZJIT: Compile sendforward with dynamic dispatch
* Reload locals only if it has blockiseq
* Add a test case of ...... - 08:52 PM Revision 7ff036d5 (git): Prevent GC from running during `newobj_of` for internal_event_newobj.
- If another ractor is calling for GC, we need to prevent the current one
from joining the barrier. Otherwise, our half... - 06:46 PM Revision 797a4115 (git): ZJIT: Support variadic C calls (#14575)
- * ZJIT: Support variadic C calls
This reduces the `dynamic_send_count` in `liquid-render` by ~21%
* ZJIT: Reuse gen... - 05:13 PM Revision ff5105f7 (git): [ruby/prism] Fix example marker for case statement predicate
- https://github.com/ruby/prism/commit/2ddedf650a
-
05:12 PM Revision 8073cd4f (git): Pass -w to lsof
-
02:03 PM Revision 7e6f7fbf (git): ZJIT: Annotate BasicObject#initialize (#14590)
- It's a no-op and always returns `nil`.
Fixes https://github.com/Shopify/ruby/issues/755 -
01:25 PM Revision 4e05fe54 (git): Use union in cross_ractor_require
- A cross_ractor_require is either a require or an autoload, so we can make
the fields an union. This reduces the size ... -
12:26 PM Revision 44d1feb7 (git): Remove test causing warning
- The test is no longer useful since
5c7dfe85a1dc49334e2828791f0ade42eee662db because Module#initialize_copy
is empty/n... -
09:25 AM Misc #21609: Propose Stan Lo (@st0012) as a core committer
- +1
-
07:04 AM Misc #21609: Propose Stan Lo (@st0012) as a core committer
- Thank you everyone for your support! I'd be my honor to become a committer.
- 07:00 AM Revision e3675e85 (git): Bump rexml
- Bumps the bundler group with 1 update in the /spec/bundler/realworld/fixtures/warbler directory: [rexml](https://gith...
-
06:50 AM Feature #20163: Introduce #bit_count method on Integer
- garrison (Garrison Jensen) wrote in #note-6:
> ```
> unsigned int pop_count(long value) {
> #ifdef __GNUC__
> ... -
05:04 AM Revision 176b3b51 (git): [ruby/erb] [DOC] `ERB::Version`
- https://github.com/ruby/erb/commit/7fed01c4de
-
- The documentation must be just before the definition.
https://github.com/ruby/erb/commit/62282e32d9 -
03:28 AM Bug #21610: Use ec->interrupt_mask to prevent interrupts.
- A slightly different change may be required for 3.3 and 3.2 backports: https://github.com/ruby/ruby/commit/e900e9fc9a...
-
- Fixed in https://github.com/ruby/ruby/commit/e6879401feba22e3657a231cbedc751998cb7176
-
- The following program can trigger `rb_bug` in `rb_fiber_scheduler_unblock`:
```ruby
require 'async/scheduler'
... - 03:24 AM Revision e900e9fc (git): Use `ec->interrupt_mask` to prevent interrupts. [Backport #21610]
- Disallow pending interrupts to be checked during `FiberScheduler#unblock`.
Ractors can send signals at any time, so ... -
- (https://github.com/ruby/erb/pull/78)
https://github.com/ruby/erb/commit/de0f18579e - 02:24 AM Revision e6879401 (git): Use `ec->interrupt_mask` to prevent interrupts. (#14588)
- Disallow pending interrupts to be checked during `FiberScheduler#unblock`.
Ractors can send signals at any time, so ...
09/17/2025
-
- Previously when we copied base_encoding on top of the encoding, other
threads could briefly see the name and ruby_enc... -
- 09:27 PM Revision bacb687f (git): ZJIT: Call instance allocation function directly
- 09:27 PM Revision 1b0ca8f1 (git): ZJIT: Rename optimize_direct_sends to type_specialize
- 09:27 PM Revision 88e0ac35 (git): ZJIT: Prevent custom allocator in ObjectAllocClass
- 09:27 PM Revision 7a82f1fa (git): ZJIT: Const-fold IsMethodCfunc
- 09:27 PM Revision c31a73d7 (git): ZJIT: Specialize ObjectAlloc with known class pointer
- This has fewer effects (can be elided!) and will eventually get better
codegen, too.
Fix https://github.com/Shopify/... -
08:49 PM Revision 96272ba1 (git): [ruby/erb] [DOC] Doc for #new_toplevel
- https://github.com/ruby/erb/commit/dee5dd4da8
-
- https://github.com/ruby/erb/commit/d5f6f65a89
-
- https://github.com/ruby/erb/commit/77830cb205
-
07:54 PM Bug #21607: require 'concurrent-ruby' causes segfault with Ruby 3.4.6 on linux/i686 built with GCC 15
- FWIW concurrent-ruby creates and resumes a Fiber while loading, this has found several issues like https://bugs.ruby-...
-
- +1 with that amount of contributions I thought he was already a committer!
-
- +1. Stan has been doing great work and really passionate about making the Ruby ecosystem better.
-
04:32 PM Misc #21609: Propose Stan Lo (@st0012) as a core committer
- Stan has also been instrumental in helping prism be useful for IRB. +1
-
04:23 PM Misc #21609: Propose Stan Lo (@st0012) as a core committer
- +1. Stan has been doing an amazing job. We need him to become a committer to see as many contributions from him as po...
-
03:59 PM Misc #21609: Propose Stan Lo (@st0012) as a core committer
- +1
-
- ## Proposal
I'd like to propose Stan Lo (@st0012) as a committer. He's been doing excellent work on ZJIT for the p... - 07:40 PM Revision bb335576 (git): ZJIT: Split Insn::Const from Insn::GetBlockParamProxy (#14583)
- * ZJIT: Split Insn::Const from Insn::GetBlockParamProxy
* Print [BlockParamProxy]
* Link a TODO comment to a Shopif... - 07:17 PM Revision c995faa8 (git): ZJIT: Add stat for `def_type` of send fallbacks (#14533)
- I thought about creating a new HIR like `SendWithoutBlockFailedToOptimize` that can carry very specific reasons later...
-
05:35 PM Revision 4d003eab (git): GitHub CI: JITs: Dump RUBY_CRASH_REPORT on failure
- We sometimes see silent crashes on CI such as
<https://github.com/ruby/ruby/actions/runs/17661854762/job/50196469793>... -
-
- (https://github.com/ruby/erb/pull/75)
https://github.com/ruby/erb/commit/8dc0eacaad -
- I'm not sure if you ran with my patch enabled, since the patch above explicitly adds column information. When I run i...
-
- I agree, but there's precedence here of some things only being detectable at run-time, like illegal `break` usage. I ...
-
04:39 PM Revision 3c7a8972 (git): [ruby/prism] Reject `1 if foo = bar baz`
- and also `1 and foo = bar baz`
This is a partial fix for https://github.com/ruby/prism/issues/3106
It still accepts ... -
04:29 PM Revision c22d84a1 (git): ZJIT: Revert documentation indent (#14580)
-
- It crashes frequently on CI but I am not able to reproduce locally:
https://ci.rvm.jp/results/trunk-random1@ruby-sp2... -
-
- `Dir.[]` returns list of strings, `map(&:to_s)` does nothing.
-
- Print a concatenated string instead of `print`ing multiple strings,
since `$\` is set to `"\n"` because of `-l` comma... -
-
-
-
-
-
-
-
-
-
-
- Setting v1, v2, v3 when we allocate an object assumes that we always
allocate 40 byte objects. By removing v1, v2, v3... -
- For prism https://github.com/ruby/prism/pull/3652. I implemented the rules as how I understood they should be, please...
- 12:33 PM Revision 73b08ff4 (git): [ruby/openssl] Fix test_ssl.rb in FIPS.
- test_post_connect_check_with_anon_ciphers:
test_tmp_dh_callback:
test_tmp_dh:
DH missing the q value on unknown name... - 12:32 PM Revision a1f39b4b (git): [ruby/openssl] Add AuthTagError exception for AEAD authentication
- failures
(https://github.com/ruby/openssl/pull/939)
* Add AuthTagError exception for AEAD authentication failures
-... - 12:25 PM Revision 0f3c6ca4 (git): [ruby/openssl] c_rehash: fix hash_name output for small hashes
- The hash lookup is done by 8-character hash.
https://github.com/ruby/openssl/commit/fedb57255c -
- This reverts commit fbc7e935761d892ba6c031256ccbb914963e4ce1.
Failures with "GCC 8" and "clang 12".
```
/github/work... -
12:18 PM Revision 08e7b5f2 (git): [ruby/openssl] Revert "pkey: stop retrying after non-retryable error from OSSL_DECODER"
- This reverts commit https://github.com/ruby/openssl/commit/5347880c6eb0 and
https://github.com/ruby/openssl/commit/98... -
-
06:42 AM Revision 6094cb51 (git): Revert "Handle `uninitialized constant JSON::GenericObject` at ruby/ruby."
- This reverts commit 0dc1cd407e7775610f2bcaef6c1282369867f91c.
1213adfe5526d65cce81a9fb127074130c8faea7 is fixed this... -
06:42 AM Revision e4fc0d4b (git): [rubygems/rubygems] Stop trying to remove every file on extraction
- When we're extracting a gem, it should be extracted to an empty
directory. Trying to remove every file before extrac... - 06:42 AM Revision 5a07417c (git): [rubygems/rubygems] Finalize `bundle install --binstubs` removal
- https://github.com/rubygems/rubygems/commit/ccb65ce0ea
- 06:42 AM Revision a16c88ec (git): [rubygems/rubygems] Fix grammar in `bundle install --binstubs` removal message
- https://github.com/rubygems/rubygems/commit/feb258c712
- 06:42 AM Revision 7c5613a6 (git): [rubygems/rubygems] Move `bundle install --binstubs` deprecation to cli.rb
- For consistency with the other deprecations.
https://github.com/rubygems/rubygems/commit/28e300cee1 - 06:42 AM Revision 19a36368 (git): [rubygems/rubygems] Make `bundle remove --install` raise an error
- https://github.com/rubygems/rubygems/commit/444022cfd3
-
- No one uses `sync_default_gems_with` on other platforms; it is used by
the dedicated workflow and a few developers only. -
-
- To elaborate, if you replace `break` with a method call, it was already a syntax error. I suspect this was changed wi...
-
- In ruby code, operations with arguments without parentheses have quite lower precedence.
That code is inverting the ... -
- Thank you for the report, I will get this fixed.
-
12:25 AM Bug #21608 (Rejected): 三項演算子と break を組み合わせると SyntaxError
- ```
>> RUBY_VERSION
=> "3.4.5"
>> [1, 2].find {|i| i.even? ? break i : nil }
```
```
>> RUBY_VERSION
=> "3.4... -
-
-
- Follow-up on https://github.com/ruby/ruby/pull/14565
I meant to push this this as well but it was left un-committed ... -
- (https://github.com/ruby/erb/pull/74)
https://github.com/ruby/erb/commit/125ce1f897
09/16/2025
-
11:38 PM Revision 52b22f81 (git): ZJIT: Debug mechanism to verify leafness for `gen_prepare_call_with_gc` (#14553)
- * functional debug mechanism; small refactor
* make all tests pass
* clean up implementation of debug mechanism for... -
- We've published https://speed.ruby-lang.org/. Thank you @hsbt!
-
10:46 PM Bug #21607: require 'concurrent-ruby' causes segfault with Ruby 3.4.6 on linux/i686 built with GCC 15
- I can confirm that building with GCC 14 resolves the issue. So this seems like a problem with building ruby for i686 ...
-
- Hmm. Ok, I will try rebuilding with GCC 14, in case this is possibly a regression in GCC.
-
- It does not seem to be related to prism (or ruby at all?):
```sh
$ ruby -e "Fiber.new {}.resume"
Segmentation fa... -
- It doesn't appear to get to a backtrace:
```
chronos@buppie-i686:2.23 M58 /usr/local/lib/crew/packages (master <>|SP... -
07:17 PM Bug #21607: require 'concurrent-ruby' causes segfault with Ruby 3.4.6 on linux/i686 built with GCC 15
- It would help us a lot if you could get the output from the crashing process. It should report a backtrace to `stderr...
-
- Hello all, a build from earlier this summer's 3.4 branch using the current bundled gems also gave the same error when...
-
- @kddnewton I'm not sure it is related to ractors because maybe those tests always failed on this platform. I'm also s...
-
- Thanks @luke-gru I was struggling to figure out what that had to do with prism at all, seeing as all the failures wer...
-
- There appear to be issues with Ractors. I'm going to reassign this to the ractor team.
-
- Our build run for this build, including make check is here:
https://github.com/chromebrew/chromebrew/actions/runs/... -
- Our downstream PR/Issue is here: https://github.com/chromebrew/chromebrew/pull/12806
-
- As mentioned at https://github.com/ruby/ruby/pull/14532#issuecomment-3288983469
chronos@buppie-i686:2.23 M58 /usr/... -
- ruby_3_4 commit:6882012473.
-
09:17 PM Revision 68820124 (git): [3.4] compile.c: Handle anonymous variables in `outer_variable_cmp` (#13493)
-
- ruby_3_4 commit:1f6d845011cf8900c0e9c6fe5e4243a63a5ecae3.
-
08:35 PM Bug #21313: `it` in rescue/ensure on prism
- I created a backport PR with this fix to 3.4 as https://github.com/ruby/ruby/pull/14573
-
08:36 PM Revision b5346f2c (git): Add a test case for `it` in a regex.
- Co-authored-by: Alan Wu <XrXr@users.noreply.github.com>
-
- The following is crashing for me:
```shell
ruby --yjit --yjit-call-threshold=1 -e '1.tap { raise rescue p it }'
rub... -
-
- (https://github.com/ruby/erb/pull/73)
https://github.com/ruby/erb/commit/04bb746fc7 -
- (https://github.com/ruby/erb/pull/71)
https://github.com/ruby/erb/commit/f4abab7195 -
- (https://github.com/ruby/erb/pull/72)
https://github.com/ruby/erb/commit/df7bdcd5cb -
- https://github.com/ruby/erb/commit/aae3a5be34
-
-
- Nothing needs rb_imemo_new exported anymore.
-
11:50 AM Feature #21555: Add support for predicate attribute reader names
- I think Matz's position is clear and will not change, so what about closing the related tickets that are still open?
-
- `log.showSignature` configuration and `--no-show-signature` option was
added at git 2.10.0. -
-
- This option is available since git 1.8.5 that was released in 2013.
- 10:37 AM Revision 97117405 (git): Add a macro to manage the condition of no-inline version rb_current_ec
- Add the macro `RB_THREAD_CURRENT_EC_NOINLINE` to manage the condition to use
no-inline version rb_current_ec for a be... -
- Thanks for opening this.
It seems just like a normal bug though, no? I understand that this doesn't deviate from wha... -
- In the Ruby test suite, this test class is causing trouble because
ostruct is not available. Having an autoload for J... - 08:17 AM Revision db027afe (git): [rubygems/rubygems] Remove aggregate source mentions
- It's a term from times with multiple remote sources, let's move on :)
https://github.com/rubygems/rubygems/commit/64... - 08:17 AM Revision 6adcc559 (git): [rubygems/rubygems] Completely remove multisources support
- https://github.com/rubygems/rubygems/commit/8f9d6c54a1
- 08:17 AM Revision 98780601 (git): [rubygems/rubygems] Simplify an edge case of not adding lower bound requirements
- https://github.com/rubygems/rubygems/commit/1bc5e74281
- 08:17 AM Revision 26f9911c (git): [rubygems/rubygems] Multisource checks are only relevant when there's a lockfile
- https://github.com/rubygems/rubygems/commit/4c110d3289
- 08:17 AM Revision 0a5a0eea (git): [rubygems/rubygems] Handle locked sources more simillarly to locked specs
- https://github.com/rubygems/rubygems/commit/0a2f5ed717
-
- Before this patch we would use `IO.copy_stream` with the tar entry
object rather than just straight to the IO. That ... - 08:17 AM Revision 6b0af313 (git): [rubygems/rubygems] Remove `allow_offline_install` setting
- And let the feature always be enabled, so I'm not sure why we'd need
this configurable.
https://github.com/rubygems/... - 08:17 AM Revision 190a2354 (git): [rubygems/rubygems] Complete rubocop flags and settings removal
- If the CLI flags are used, we abort early as usual.
As per the settings, I decided to ignore them. We've been migrat... - 08:17 AM Revision 1c7fd141 (git): [rubygems/rubygems] Fix `--no-rubocop` deprecation message
- https://github.com/rubygems/rubygems/commit/2c16b0e11e
- 08:17 AM Revision 9b45a25c (git): [rubygems/rubygems] Fix `rubocop` config removal message
- In Bundler 4, configuration will no longer be updated.
https://github.com/rubygems/rubygems/commit/33a4718d7a -
- When extracting tar files, the tar header actually knows the exact size
of the file we need to extract. Before this ... -
- We already have the open file descriptor, so we can avoid the overhead
of resolving the filepath (as well as the over... -
-
-
02:46 AM Feature #21573: Simpler syntax errors
- Before hastily adding modes, we should clarify the problem.
In the proposed patch, the error message will be as fo... -
- # The next dev meeting
**Date: 2025/10/23 13:00-17:00** (JST)
Log: *TBD*
- Dev meeting *IS NOT* a decision-mak... -
-
- Discussed at the dev meeting.
Matz said that he wanted to permit passing a single command (a method call without p... -
- 01:17 AM Revision 20d4e27d (git): ZJIT: Revert VM_CALL_ARGS_SPLAT and VM_CALL_KWARG support (#14565)
-
- ubuntu-24.04-arm jobs look working as of 2025-09-16.
- 12:43 AM Revision 6c5960ae (git): ZJIT: Support compiling block args (#14537)
09/15/2025
-
- These used to be used by the parser
-
- (https://github.com/ruby/erb/pull/69)
* [DOC] More on class ERB
* [DOC] More on class ERB
* More
* More
* More... -
-
- ruby_3_4 commit:52b3f006cb31672c45b26f6a44a609fd1b8e2ee5.
-
- ruby_3_4 commit:1f421d092bb8c290df5cfc9e95a01e279ebadc42.
-
- ruby_3_4 commit:52b3f006cb.
-
- imemo_tmpbuf is not write-barrier protected and uses mark maybe to mark
the buffer it holds. The normal rb_imemo_new ... -
-
-
- Symbol#name is only a thing since Ruby 3.0
https://github.com/ruby/prism/commit/2de82b15fc -
- If a character literal was followed by a string concatenation, then
the forced encoding of the string concatenation c... -
- Right now both the prism and parse.y parsers add some context to syntax errors to make it easier to understand where ...
-
-
-
- The performances are: block > proc > method object.
https://github.com/ruby/optparse/commit/9ec5d1d582 -
-
- An unspecified uplevel is not the same as an uplevel of 1:
```
$ irb
irb(main):001> warn("foo")
foo
=> nil
irb(main)... -
-
- These tests are run in the dedicated workflows for JITs.
-
- https://github.com/ruby/date/commit/afaa4a997b
-
-
- It is needed before updating the timestamp for prism/srcs.mk.
-
- Get rid of `File.identical?` on a WebDAV-mounted drive.
09/14/2025
-
12:30 PM Bug #21569: [armv7, musl] SIGBUS in ibf_load_object_float due to unaligned VFP double load when reading IBF
- Successfully compiled on ARMv7 with this patch (right now); no crash occurred during the project build.
ruby 3.3.8 (... -
- Simply it can be:
```c
memcpy(&d, IBF_OBJBODY(double, offset), sizeof(d));
```
amacxz: Would you send PR at git... -
-
-
- Now it is always compiled (and will be optimized away).
-
- Make this macro condition as compile-time constant instead of a
preprocess-time constant, and compile the body always. -
04:48 AM Bug #21567: Crash when $LOADED_FEATURES is modified during require
- ruby_3_3 commit:01b0beb818e74c290fdf8fcc824946fd8c93b373 merged revision(s) commit:928fea3bfa86053c0bc6f7a5bf7559b115...
-
- Fix crash when $LOADED_FEATURES is modified during require
[Bug #21567]
When we require an object t... -
- ruby_3_3 commit:210643c51d706a486ad460111eac51d4d34639cf merged revision(s) commit:7c28fb2fb2544e5fed75ef216c4dd08084...
-
- [Bug #21546] Make the generated pc file relocatable
-
- The backslashs in these tests are parsed as escapes in string literals
by the parser, but not passed to the `Regexp` ... -
- ruby_3_3 commit:e39a0a19c03fbb37d1c9d1e050bb551dccfc11c6 merged revision(s) commit:ce849d565bf6aae8e0179fffb04eb1f665...
- 03:53 AM Revision 9c65da01 (git): Bump Prism version to 1.5.1 (#14549)
-
- (https://github.com/ruby/erb/pull/68)
https://github.com/ruby/erb/commit/9591b5d23b -
- ruby2_keywords warnings: Quote non-UTF8 method names fully
It used to quote only part of the method name bec... -
- Phalado (Raphael Cordeiro) wrote in #note-6:
> The idea here came when working with JSON from requests and, especial... -
- Phalado (Raphael Cordeiro) wrote in #note-5:
> I also think this is something very 'Ruby' to do, like the method `.z...
09/13/2025
-
- This commit fills the space that is not initialized with garbage for
newly allocated objects on debug mode. -
- matz (Yukihiro Matsumoto) wrote in #note-5:
> The behavior of Prism in 3.5 is close to my intention.
>
> Matz.
... -
- Should this now be assigned to prism since there is an incompatibility? Sorry I am not clear on the conclusion.
-
- Make it clear that it parses with the most recent version of Ruby
syntax.
https://github.com/ruby/prism/commit/7285d... -
- Currently when you have an illegal variable in an alternation pattern, you get a compile error, but no syntax error.
... - 02:01 PM Revision 6c9408d1 (git): Update default gems list at f4ce5e90b2b9a4ccc7b4a0a25416c5 [ci skip]
-
- https://github.com/ruby/prism/commit/cac5118884
-
- I think this might not be fixed. I'm reopening to keep investigating.
-
- Fixed by https://github.com/ruby/ruby/commit/d781d69a06e7d4eef3334e44a25b02d05bad1e2d
-
- This reverts commit https://github.com/ruby/prism/commit/4052d93cf852, reversing
changes made to https://github.com/r... -
- When a line is going to be displayed in an error message that
contains multibyte characters, we need to respect the e... -
- Applied in changeset commit:git|4f4b4e3b37c884093a165801b5028c14a7de1604.
----------
Fill in lead num for blocks wit... -
- Fixes [Bug #21256]
Co-Authored-By: Earlopain <14981592+Earlopain@users.noreply.github.com> -
-
-
-
- When `-F` option is given to strip comments in bundled_gems file, `$;`
is set to the regexp.
On the other hand, old ... -
-
-
-
- It depends on ostruct gem that is no longer a part of the default
gems, and the all tests are just skipped with a war... -
-
-
-
- Applied in changeset commit:git|618224adf620bd72645b3fada4f51e805196f5ca.
----------
[Feature #20925] Support leadin... -
-
-
- On OpenBSD:
```
ld: warning: namespace.c:731(namespace.o:(rb_namespace_local_extension)): warning: strcpy() is almos... -
-
-
-
02:58 AM Feature #17398 (Closed): SyntaxError in endless method
- Applied in changeset commit:git|a8c829e7f0dc20c3213d23fb9722fabc74fe1a9d.
----------
[Bug #17398] Allow `private def... -
-
- > Do you think this is possible to implement?
This is possible to implement like [this](https://github.com/yui-knk...