Activity
From 10/10/2025 to 10/16/2025
Today
-
03:40 AM Revision 1d95d75c (git): ZJIT: Profile opt_succ and inline Integer#succ for Fixnum (#14846)
- This is only really called a lot in the benchmark harness, as far as I
can tell. -
02:53 AM Feature #21615: Introduce `Array#values`
- This feels like opening Pandora's box. If this is accepted, I can foresee the following discussions arising, just off the top of my head:
* Should `Array#keys` be defined as a method that returns `(0...ary.size).to_a`?
* `Array#find_... -
-
01:21 AM Revision 45c01686 (git): Use explicit memory orders in concurrent_set
- The atomic load/store operations here should mostly be using
release/acquire semantics. This may lead to better performance than what
we had under the default seq_cst.
On x86 this may make the atomic store of hash faster, as it can avoi... -
10/15/2025
-
11:29 PM Revision 5a9fac69 (git): Fix assert_equal order in test_namespace.rb
- The expected value is the first parameter and the actual value is the
second in assert_equal. -
09:24 PM Revision d272a81f (git): ZJIT: Rewrite arm64_split_with_scratch_reg for clarity
- * The while loop pattern can be rewritten to be more idiomatic, which
also allows the iterator to no longer be mutable. -
08:36 PM Revision 4c426e98 (git): ZJIT: Use rb_gc_disable() over rb_gc_disable_no_rest()
- no_rest() trips an assert inside the GC when we allocate with the GC
disabled this way:
(gc_continue) ../src/gc/default/default.c:2029
(newobj_cache_miss+0x128) [0x105040048] ../src/gc/default/default.c:2370
(rb_gc_impl_new_... -
06:55 PM Misc #21606: DevMeeting-2025-10-23
- - [Feature #21615] Introduce `Array#values`
- Seems reasonable to have an API to simplify traversal of hash/array values
- There is some precedence, since both Hashes/Arrays have #values_at.
- If approved, should it be added to ... -
03:30 PM Misc #21606: DevMeeting-2025-10-23
- * [Bug #21640] Core Pathname is missing 3 methods / is partially-defined (eregon)
* This seems a bad user experience: partially-defined class missing some methods, `require "pathname"` doesn't load the gem as expected (even with Bundl... -
06:34 PM Revision b052d706 (git): [ruby/prism] explicitly cast constants in initializers as well
- https://github.com/ruby/prism/commit/e7db2b06ab
-
- https://github.com/ruby/prism/commit/0b2710a6c9
-
- Fixes https://github.com/Shopify/ruby/issues/793
## Testing on `liquid-render`:
<details>
<summary>Before patch:</summary>
```
***ZJIT: Printing ZJIT statistics on exit***
Top-20 not inlined C methods (96.8% of total 20,222,783):
... - 05:49 PM Revision 27ff5861 (git): We can't grab the VM Lock in free functions
- This is due to the way MMTK frees objects, which is on another native thread.
Due to this, there's no `ec` so we can't grab the VM Lock.
This was causing issues in release builds of MMTK on CI like:
```
/home/runner/work/ruby/ruby/bu... -
- This fixes a reliable "ZJIT saw a dead object" repro on my machine, and should
fix the flaky ones on CI. The code for disabling the GC is the same as
the code in newobj_of().
See: https://github.com/ruby/ruby/actions/runs/18511676257/jo... -
- I'm happy to help making the changes in ruby/ruby to make Pathname "just a default gem" again like it was in 3.4, if that is the decided outcome of the dev meeting (I added this ticket to https://bugs.ruby-lang.org/issues/21606).
-
- ```
$ ruby -e 'puts Pathname.instance_methods(false).sort; puts Pathname.singleton_methods.sort' > core_pathname_methods.txt
$ ruby -rpathname -e 'puts Pathname.instance_methods(false).sort; puts Pathname.singleton_methods.sort' > requ... -
- Accidentally added in https://github.com/ruby/ruby/pull/14679
-
- 09:54 AM Revision dce202d6 (git): [rubygems/rubygems] Add checksum of gems hosted on private servers:
- - ### Problem
Running `bundle lock --add-checksums` doesn't add the checksum of
gems hosted on server that don't implement the compact index API.
This result in a lockfile which is unusable in production as
some checksums will ... -
- Unfortunately, no such guidelines exist. Anyone with access to a z/OS Linux environment will likely need to investigate the code on their own.
By the way, for a long time, @ReiOdaira had been maintaining the CI for Ruby on z/OS Linux,... -
07:05 AM Misc #21639 (Feedback): Guidelines or documentation for porting Ruby to a new platform (z/OS)
- Hi Ruby team,
I work on the IBM z/OS platform and am currently working on porting Ruby to it.
I wanted to ask whether there are any existing porting guidelines, documentation, or recommended steps for bringing Ruby up on a new plat... -
07:15 AM Revision 51b2c5a4 (git): [rubygems/rubygems] Removed obsoleted option from bundle-exec manpages
- https://github.com/rubygems/rubygems/commit/6a3342541a
-
- https://github.com/rubygems/rubygems/commit/cd1493eec4
-
- https://github.com/rubygems/rubygems/commit/59b909fa74
- 07:15 AM Revision a60b56c3 (git): [rubygems/rubygems] Use `default_cache_path` helper for brevity
- https://github.com/rubygems/rubygems/commit/29a12c3d46
- 07:15 AM Revision 92cbd7ec (git): [rubygems/rubygems] Test current clean after bundle update behavior
- https://github.com/rubygems/rubygems/commit/c43e35c3ea
-
- https://github.com/rubygems/rubygems/commit/06508374aa
-
- https://github.com/rubygems/rubygems/commit/89bcdfc941
-
- https://github.com/rubygems/rubygems/commit/15be905c44
- 04:58 AM Revision 1142abb1 (git): [DOC] Update making_changes_to_stdlibs.md mirror Example
- CSV is no more part of the standard lib, but the documentation was not
updated (the example link was broken for the master branch)
Selected ERB that has the dedicated directory, like CSV. -
- https://github.com/rubygems/rubygems/commit/b1b963b34a
Co-authored-by: David Rodríguez <[email protected]> -
- Strictly more info than just the builtin_type from `assert_ne!`.
Old:
assertion `left != right` failed: ZJIT should only see live objects
left: 0
right: 0
New:
ZJIT saw a dead object. T_type=0, out-of-heap:0x000000... -
02:05 AM Revision 8d438678 (git): [rubygems/rubygems] remove some memoization
- I don't think these methods are hotspots, and since gem specifications
are sometimes serialized to yaml / marshal, I think we should remove as
many instance variables as possible
https://github.com/rubygems/rubygems/commit/40490d918b
10/14/2025
-
-
- 08:36 PM Revision ed94e543 (git): ZJIT: Centralize the allocation of scratch registers (#14815)
-
- These bring `send_without_block_no_profiles` numbers down more.
On lobsters:
Before: send_without_block_no_profiles: 1,293,375
After: send_without_block_no_profiles: 998,724
all stats before:
```
***ZJIT: Printing ZJIT statisti... -
07:48 PM Feature #21615: Introduce `Array#values`
- matheusrich (Matheus Richard) wrote in #note-6:
> @Dan0042 I considered that. I thought it would be too much for one PR. I'll wait for more feedback, and if people are positive about this, I'll propose `Set#values` too.
One principle... -
- These bring `send_without_block_no_profiles` numbers down dramatically.
On lobsters:
Before: send_without_block_no_profiles: 3,466,375
After: send_without_block_no_profiles: 1,293,375
all stats before:
```
***ZJIT: Printing ZJIT ... -
- Gets rid of one transient vec copy/allocation.
-
- Fix https://github.com/Shopify/ruby/issues/795
-
- Add the ability to create a Rust function that "open-codes" HIR
implementations of specific well-known C functions, starting with
`String#to_s`. It supports emitting multiple instructions into a
temporary block, but does not support emit... -
02:45 PM Revision 55e76b4c (git): [ruby/erb] Add `changelog_uri` to spec metadata
- (https://github.com/ruby/erb/pull/89)
This project's `NEWS.md` file appears to be the closest thing to a
changelog file that I could find. The change here links to the file in
the released version's branch. RubyGems.org will use this me... - 01:59 PM Revision cad692de (git): Remove useless comments
- Namespace frame exists, but it is used only for Namespace#eval now.
- 01:59 PM Revision 29adf0bb (git): Split gvar space between root and main namespaces
- 01:59 PM Revision d60ee6fb (git): Remove a comment - we cannot remove this method now probably
- 01:59 PM Revision 9743b518 (git): Define main.to_s even in namespaces
- It just shows "main" just like the main object without namespace.
All main objects in namespaces will show "main" and it is impossible
to determine a main from main objects if it returns "main".
But it's not a problem because we don't de... - 01:59 PM Revision 7e07a8d8 (git): Remove a debug method that is useless now
-
12:12 PM Revision 25a42035 (git): [rubygems/rubygems] Fix typo
- https://github.com/rubygems/rubygems/commit/e4f1772d80
- 11:05 AM Revision 34ee5cbf (git): [DOC] Fix minor typos in YJIT comments (#14829)
- [DOC] Fix typos in YJIT core
-
- https://github.com/rubygems/rubygems/commit/376e4ec8c7
Co-authored-by: David Rodríguez <[email protected]> -
10:29 AM Feature #21636: Proposal to Introduce a Dedicated Warning Category for Regular Expressions
- Thank you so much for your feedback! 🙏
You were absolutely right: I simplified the PR and made it explicit now to run only on the tested types, making it a far more predictable outcome.
Thus I'm not so keen about my proposal right no... -
- Could you please elaborate on your use case?
I'm not entirely clear on which specific warnings you want to suppress. Looking at the patch's comments, it seems you want to suppress warnings about "character class overlaps." This warnin... -
- https://github.com/rubygems/rubygems/commit/9d70887185
-
- https://github.com/rubygems/rubygems/commit/90130c0648
-
- https://github.com/rubygems/rubygems/commit/7b0da18764
-
- https://github.com/rubygems/rubygems/commit/7d910dd94c
Co-authored-by: David Rodríguez <[email protected]> -
04:59 AM Bug #21632 (Feedback): Backport REXML CVE-2025-58767 fix
- > I'm not sure what to do for Ruby 3.2. It's a security fix so it qualifies for a backport, but there's other changes included in a version bump.
I am a maintainer of REXML.
Ruby 3.2 is subject to security fixes, so I believe an upda... -
04:30 AM Bug #21638 (Closed): Ractor-local `$DEBUG` is not marked
- Applied in changeset commit:git|e94a2f691d67ad98be9036e76c765fcfa7d22552.
----------
[Bug #21638] Mark ractor-local `$VERBOSE` and `$DEBUG`
https://github.com/sampersand/blog/blob/master/the%20-s%20flag.md#the-segfault -
- https://github.com/sampersand/blog/blob/master/the%20-s%20flag.md#the-segfault
https://github.com/ruby/ruby/pull/14828 -
- https://github.com/sampersand/blog/blob/master/the%20-s%20flag.md#the-segfault
-
- https://github.com/ruby/strscan/commit/29ad49f89d
10/13/2025
-
10:14 PM Revision 25821f34 (git): [DOC] Tweaks for String#rjust
-
-
09:13 PM Feature #21637: Tracing global variable assignment
- alanwu (Alan Wu) wrote in #note-2:
> > Currently, there is no straightforward way to track global variable assignments at runtime.
> ...
I'm aware of `Kernel#trace_var`, but it does not cover cases where the global variables aren't kno... -
- > Currently, there is no straightforward way to track global variable assignments at runtime.
There is [Kernel#trace_var](https://docs.ruby-lang.org/en/3.4/Kernel.html#method-i-trace_var). -
- ## Motivation
Currently, there is no straightforward way to track global variable assignments at runtime.
The only workaround I am aware of involves using the `line` TracePoint event, combined with manipulation of the source code's... -
- Inspired by 42ba82424d908c290a4a34ced8853f0a403b734b, I looked for other
occurrences of "the the". -
- Hi folks,
while working on adding regex support for the marcel gem (see PR: https://github.com/rails/marcel/pull/132), I encountered regex warnings triggered by some of Tika’s regular expressions. Typically, these warnings are valid b... -
02:48 PM Revision d11df417 (git): [ruby/erb] html_escape: refactor redundant if
- (https://github.com/ruby/erb/pull/88)
https://github.com/ruby/erb/commit/c231ced3f4 -
02:12 PM Bug #21634: Combining read(1) with eof? causes dropout of results unexpectedly on Windows.
- That is interesting behavior I hadn't considered.
My understanding is that with 'rt' uses universal newline conversion and
0x1A is treated as a regular character, on both Windows and other platforms.
For example:
```ruby
>./mini... -
- 08:41 AM Revision c78895b1 (git): Add "Namespace detection information" section in bug reports
- * To show environments stack when the current namespace is unexpected or
namespace detection is broken
* It is displayed only when RUBY_BUGREPORT_NAMESPACE_ENV=1 is specified
10/12/2025
- 08:00 AM Revision c938c11f (git): Update default gems list at 226312ecd45304698f5ac74f941305 [ci skip]
-
07:58 AM Revision 226312ec (git): [ruby/erb] Version 5.1.1
- https://github.com/ruby/erb/commit/3dc0bb09bf
-
- Fix https://github.com/ruby/erb/pull/87
https://github.com/ruby/erb/commit/75764f022b -
-
- Soft fails like warnings from rustc. The `rustdoc` warnings tend to be
dead links in the markup. -
10/11/2025
- 10:39 PM Revision a6f92ddd (git): Update default gems list at 02d8a001eed707f8cd130f5c644f5c [ci skip]
-
- https://github.com/ruby/erb/commit/25fdde41d6
-
- escapable character
(https://github.com/ruby/erb/pull/87)
This change improves reduces allocations and makes `html_escape` ~35% faster in
a benchmark with escaped strings taken from the `test_html_escape` test in
`test/test_erb.rb`.
- ... -
- When a process with multiple threads is forked, LSAN outputs warning
messages to stderr like:
==276855==Running thread 276851 was not suspended. False leaks are possible.
We should ignore messages like this in tests. -
07:08 PM Bug #21635 (Closed): Reading code from STDIN unexpectedly sets $_ in 3.4.6+
- Applied in changeset commit:git|d036dc0a794d87309f912e7585749c681c2438f5.
----------
For prism parser, do not update $_ from STDIN
Fixes [Bug #21635] -
- Fixes [Bug #21635]
- 05:41 PM Revision 10c0d7a8 (git): ZJIT: Count unoptimized `Send` (#14801)
- * ZJIT: Count unoptimized `Send`
This includes `Send` in `send fallback reasons` to guide future
optimizations.
* ZJIT: Create dedicated def_type counter for Send -
09:28 AM Revision a4ca3de3 (git): merge revision(s) 07b59eee6aa120537d7d72422327cc7b855e9400:
- [PATCH] Fix memory leak when load_from_binary raises
ibf_load_code will leak memory allocated for the code if an exception is
raised. The following script reproduces the leak:
bin = RubyVM::InstructionSequen... -
09:27 AM Revision 7e31d3c0 (git): Backport WASI setjmp handler memory leak fixes to Ruby 3.3
-
08:38 AM Feature #21616: date ライブラリを deprecated させたい
- jinroq (Jinroq SAITOH) wrote in #note-2:
> nobu (Nobuyoshi Nakada) wrote in #note-1:
> ...
実装をどうするかよりも、Time.parseがRubyに必要かという議論をするのがよいと思います。
今のままのものがそのまま欲しいかというとNoよりだと思います。
じゃあ全くいらないのかというと悩ましいところで、まずここでいう Time.parse とは何かから考える必要があって、
... -
- ruby_3_3 commit:27a1df7506ba566d08a536e93910fff574cd6d56 merged revision(s) commit:7ae67e8f6ad6e7fd0677b28a7a10961f79d55495.
-
- YO4 (Yoshinao Muramatsu) wrote in #note-1:
> The IO that has mode_enc "rt" will read with O_BINARY but opend with O_TEXT.
> ...
Thank you for the patch.
`IO#eof?` behavior seems changing.
With "txt" file that its content is "abcd... -
- [PATCH] load.c: Fix dest and src of MEMMOVE
When multiple files with the same name are required, the features_index
hash stores the indexes in `$LOADED_FEATURES` array into a darray.
The dest and src arguments fo... -
- Merged into ruby_3_3 branch at commit:ce7aa23f97273fa181be26aec33d3c6998e203c5. Thanks for your contributions!
- 07:53 AM Revision ce7aa23f (git): Update openssl gem to 3.2.2
- 06:00 AM Revision e8f0e142 (git): [DOC] Fix typo in Regexp Optimization section
-
03:40 AM Bug #21446: StackOverflow when changing visibility in reopened refinement
- I found that you don't even need to change the visibility twice. This issue affects all refinement visibility change methods where the method whose visibility is changed by the refinement is in an ancestor of the refined class. Here's a ...
-
- * assigned but unused variable
* literal string will be frozen in the future -
- `env` on macOS resets DYLD environment variables that are needed to
run ruby configured with `--enable-load-relative`. Use simple
commands instead, and `/bin/sh` that is specified in POSIX is more
portable than `/usr/bin/env`. -
- When setting `log.showSignature=true`, `git log` and `git show`
include messages gpg verfied the commits, in addition to the message
specified by `--pretty`.
10/10/2025
-
- ibf_load_code will leak memory allocated for the code if an exception is
raised. The following script reproduces the leak:
bin = RubyVM::InstructionSequence.of(1.method(:abs)).to_binary
10.times do
100_000.times do
... -
- fetched from the repository
-
-
- Previously we just had a comment stating that the code required a
barrier. Turns out it's not too difficult to properly assert that.
Co-authored-by: Luke Gruber <[email protected]> -
- https://github.com/ruby/ruby/pull/14717
-
- * ZJIT: Add Insn:: ArrayArefFixnum to accelerate Array#[]
* ZJIT: Use result from GuardType in ArrayArefFixnum
* ZJIT: Unbox index for aref_fixnum
* ZJIT: Change condition and add ArrayArefFixnum test
* ZJIT: Fix ArrayArefFixnum disp... -
- This isn't (yet?) safe to do because it concurrently modifies GC
structures and dfree functions are not necessarily safe to do without
stopping all Ractors.
If it was safe to do this we should also do it for
gc_enter_event_continue. I d... - 04:41 PM Revision f8c90e45 (git): [DOC] Update bundled gems list at 0ba6379acadf00ee0c4c92cb60ae37
-
04:39 PM Revision 0ba6379a (git): Update bundled bigdecimal version (#14809)
- * Update bigdecimal spec
* Update bundled bigdecimal to 3.3.1 -
- The IO that has mode_enc "rt" will read with O_BINARY but opend with O_TEXT.
This leads fill_cbuf using O_TEXT at rb_io_eof unexpectedly.
I made [PR #18410](https://github.com/ruby/ruby/pull/14810). -
- This is happening because of prism's use of IO#gets on STDIN. We will fix this.
-
10:56 AM Bug #21635 (Closed): Reading code from STDIN unexpectedly sets $_ in 3.4.6+
- It seems from Ruby 3.4.6+ if you pass in code that doesn't end with a newline to the ruby interpreter it sets `$_`:
```
# ruby -v
ruby 3.4.6 (2025-09-16 revision dbd83256b1) +PRISM [arm64-darwin23]
# printf 'p $_' | ruby
"p $_"
`... -
- https://github.com/rubygems/rubygems/commit/631a55be91
-
- https://github.com/rubygems/rubygems/commit/b2472e7b82
Co-authored-by: David Rodríguez <[email protected]> -
- https://github.com/rubygems/rubygems/commit/31d67ecc05
-
06:53 AM Bug #21631: Backport openssl gem bugfix releases
- `OpenSSL::X509::V_FLAG_CRL_CHECK` can make sense on cert stores with `OpenSSL::X509::Store#set_default_paths` but only if you have also did `OpenSSL::X509::Store#add_crl` as there is no such thing as a default CRL file. Calling that on `...
-
- Note: https://github.com/ruby/ruby/commit/7863389ad0e31bee853ba5c7399637edbead31a1#commitcomment-167573285
-
- https://github.com/rubygems/rubygems/commit/bfe15a4712
Co-authored-by: David Rodríguez <[email protected]> - 06:13 AM Revision 77e32902 (git): [rubygems/rubygems] Consolidate removal of `Bundler.rubygems.all_specs`
- https://github.com/rubygems/rubygems/commit/73779331ce
- 04:45 AM Revision 1399134f (git): [rubygems/rubygems] Consolidate removal of `Bundler::SpecSet#-` and `Bundler::SpecSet#<<`
- https://github.com/rubygems/rubygems/commit/aee50b31db
-
- At the moment, we're not sure which gem relies on past renames. So we
try this, and revert it if it turns out to be necessary.
Given that it slows down every single sync, however, I'm thinking of
making tool/sync_default_gems.rb respons... -
- At the moment, nothing in the notification tells you whether it was ibm
jobs or not. `matrix.os` seems like the easiest way to include that
information, so I added it in the label. -
- https://github.com/rubygems/rubygems/commit/ced8ef3a12
Co-authored-by: David Rodríguez <[email protected]> -
-
-
-