Activity
From 10/16/2025 to 10/22/2025
Today
10/21/2025
-
10:52 PM Revision a2a107c2 (git): ZJIT: Handle when SetIvar raises FrozenError
-
10:42 PM Revision 4a23b6a8 (git): Fix memory leak in RCLASS_SET_NAMESPACE_CLASSEXT
- The st_insert in RCLASS_SET_NAMESPACE_CLASSEXT may overwrite an existing
rb_classext_t, causing it to leak memory. This commit changes it to use
st_update to free the existing one before overwriting it. -
-
10:03 PM Revision 0cc4819f (git): Improve test_fixnum_{and,or}* by checking all results and using more interesting inputs
-
- * Handled in cruby_methods.rs because there is no basic operation for Fixnum#^.
-
09:11 PM Bug #21645: Can't `require "resolve"` on Windows under Bundler without warnings
- Would this problem be avoided if mini_portile2 declared an explicit dependency on net-http? If it would be helpful I'm happy to cut a new release of mini_portile with that change.
-
11:57 AM Bug #21645: Can't `require "resolve"` on Windows under Bundler without warnings
- Here is a standalone workflow that demonstrates the issue closer to how I actually encountered it:
```yml
name: Test
on: push
jobs:
make:
runs-on: windows-latest
steps:
- run: echo "source 'https://rubygems.or... -
- This helps ZJIT optimize ~300,000 more sends in ruby-bench's lobsters
Top-6 not optimized method types for send_without_block
Before After
iseq: 713,899 (48.0%) iseq: 725,668 (62.4... -
- The type name bindgen picks for anonymous enums creates desync issues on
the bindgen CI checks. -
- ```
/src/jit.c:19:5: error: ISO C restricts enumerator values to range of 'int' (4294967295 is too large) [-Werror,-Wpedantic]
19 | RB_INVALID_SHAPE_ID = INVALID_SHAPE_ID,
| ^ ~~~~~~~~~~~~~~~~
``` -
08:48 PM Revision 193b299b (git): [DOC] Tweaks for String#setbyte
-
-
-
-
-
08:14 PM Misc #21606: DevMeeting-2025-10-23
- * [Misc #21646] Propose Luke Gruber as a Ruby committer (jhawthorn)
* Luke has been fixing bugs in Ractors and elsewhere. Would be helpful if he could merge his own PRs.
-
06:46 PM Misc #21646: Propose Luke Gruber as a Ruby committer
- I would be honored to join as a committer if you'll have me. Thank you everyone!
-
06:41 PM Misc #21646: Propose Luke Gruber as a Ruby committer
- +1 Luke is very helpful with Ractor development, and I think he would make a great addition to the team!
-
03:45 PM Feature #21572: Make illegal variable in alternation pattern a syntax error
- I agree with you, @kddnewton.
Your proposal makes sense. -
08:02 AM Feature #21636 (Closed): Proposal to Introduce a Dedicated Warning Category for Regular Expressions
- Thank you, I'll close this for now. If there's anything else you'd like to discuss, please let me know.
-
- I talked with @akr-san. He said it should be first confirmed whether you have any real use cases that actually want to handle situations like "can't read but can write" (or vice versa).
Pipes are typically unidirectional, so when a wr... -
-
10/20/2025
-
- +1 Luke has been doing great work and is really knowledgeable in Ractors
-
- I'd like to propose adding Luke Gruber (@luke-gru) as a Ruby committer.
He's been contributing to CRuby [since 2019](https://github.com/ruby/ruby/commit/ec8e5f5aa64e2a54cf1e303f2b012c98e8d521ba) fixing a variety of bugs and being acti... -
09:30 PM Revision 17368234 (git): ZJIT: Implement codegen for FixnumMod (#14857)
- This is mostly to see what happens to the loops-times benchmark.
- 08:55 PM Revision 6e9f7974 (git): [DOC] Create doc/contributing/concurrency_guide.md
- This guide is for those that want to contribute to ruby but don't
understand where they need to use locks or other concurrency mechanisms.
It teaches them how to use these locks safely and what is prohibited in
certain circumstances. - 08:10 PM Revision e047cea2 (git): ZJIT: Optimize send with block into CCallWithFrame (#14863)
- Since `Send` has a block iseq, I updated `CCallWithFrame` to take an optional `blockiseq` as well, and then generate `CCallWithFrame` for `Send` when the condition is right.
## Stats
`liquid-render` Benchmark
| Metric ... -
04:24 PM Revision 5d2a244b (git): [Bug #21644] compile.c: fix `newrange` INSN peephole optimization for chilled string
- ref: https://bugs.ruby-lang.org/issues/21644
```shell
$ ruby -v -e '("a" || "b").."c"'
ruby 3.4.7 (2025-10-08 revision 7a5688e2a2) +PRISM [x86_64-linux]
-e:1: warning: possibly useless use of .. in void context
-e:1: [BUG] Stack consist... -
- It turns out that we don't use it anywhere.
-
03:14 PM Bug #19017: Net::HTTP may block when attempting to reuse a persistent connection
- Hi, I am not sure we should solve this in IO, the error lies more in calling eof? in the first place I would say as eof? is meant to be blocking - and currently I only see solutions that involve non-blocking calls.
The solution propos... -
- Only support the simple case: no splat or rest.
lobsters before:
<details>
```
***ZJIT: Printing ZJIT statistics on exit***
Top-20 not inlined C methods (60.5% of total 11,039,954):
Kernel#is_a?: 1,030,7... -
- `prism` had disabled some workflows for ruby-head on windows because of the extraction from fiddle to being a bundled gem.
I attempted to renable them now, but it is still an issue: https://github.com/ruby/prism/pull/3687
During `bun... -
12:09 PM Revision ac27bec2 (git): merge revision(s) 957c832db137e67289e93dfd9fd9e915b1f2fc87:
- [PATCH] Fix memory leak in rb_const_remove when using namespace
We need to free the rb_const_entry_t we remove from the RCLASS_WRITABLE_CONST_TBL
otherwise it will leak memory. -
09:12 AM Bug #21643 (Closed): 'make' or 'make up' results in: uncommon.mk:1988: *** missing separator. Stop.
- @nobu: Many thanks! This is solved, most probably due to removing uncommon.mk.
[After that, I had to upgrade from Ubuntu 22.04 to Ubuntu 24.04 to make sure BASERUBY wasn't too old, so there may have been other reasons, too.] -
04:43 AM Bug #21643: 'make' or 'make up' results in: uncommon.mk:1988: *** missing separator. Stop.
- duerst (Martin Dürst) wrote:
> Line 1988 and 1989 (the last two lines) in upcommon.mk read:
> ...
The `prism/srcs.mk` and `depend` file contents should follow it.
After removing that file and running `./config.status --recheck`, what ... -
- I tried to get my ruby checkout up to date using 'make up', but after a lot of output, it stopped with the message "uncommon.mk:1988: *** missing separator. Stop.". Now it just gives this message when trying 'make up' or 'make'.
Line... - 07:04 AM Revision 6eb75f6c (git): [ruby/rubygems] update magnus version in rust extension gem template
- https://github.com/ruby/rubygems/commit/1ba8eb4ab3
-
06:52 AM Revision b6f1c4ed (git): [ruby/rubygems] Use ruby/rubygems instead of rubygems/rubygems at document, tool and configurations
- https://github.com/ruby/rubygems/commit/749b498822
-
- https://github.com/ruby/rubygems/commit/c637007e91
-
- Applied in changeset commit:git|7587e92910e7604a4c66f2b804bfa2076339c6ff.
----------
[Bug #21644] compile.c: fix `newrange` INSN peephole optimization for chilled string
ref: https://bugs.ruby-lang.org/issues/21644
```shell
$ ruby -v ... -
- PR:https://github.com/ruby/ruby/pull/14879
Confirmed on every version since 3.4.
``` shell
$ ruby -v -e '("a" || "b").."c"'
ruby 3.4.7 (2025-10-08 revision 7a5688e2a2) +PRISM [x86_64-linux]
-e:1: warning: possibly useless use of... -
- ref: https://bugs.ruby-lang.org/issues/21644
```shell
$ ruby -v -e '("a" || "b").."c"'
ruby 3.4.7 (2025-10-08 revision 7a5688e2a2) +PRISM [x86_64-linux]
-e:1: warning: possibly useless use of .. in void context
-e:1: [BUG] Stack consist... -
- ioquatix (Samuel Williams) wrote in #note-4:
> - "Connection Reset" occurs during read, and indicates that the remote end has dropped the connection, and it's likely that we are missing data (as opposed to reaching end of stream).
I ... -
03:50 AM Feature #21642: Introduce `IO::ConnectionResetError` and `IO::BrokenPipeError` as standardized IO-level exceptions.
- Thanks for your feedback. I understand your point and I think it makes sense.
"Connection Reset" and "Broken Pipe" have well defined meanings. On the face of it, I don't know what "CloseUnrecoverable" means. Is there a concept like this... -
- I sympathize with this proposal. Having to rescue a wide variety of exceptions is a common pain point, especially when writing applications like chatbots.
I believe the best approach is to classify exceptions based on the required use... -
12:38 AM Revision 4353187d (git): Fix extension file permissions on Cygwin in namespace feature
10/19/2025
-
- Currently, different IO implementations in Ruby raise inconsistent exception types when a connection is reset or broken.
For example:
```ruby
# Plain TCP socket:
socket.read_nonblock(1024)
# => Errno::ECONNRESET
# SSL socket:... -
- We need to free the rb_const_entry_t we remove from the RCLASS_WRITABLE_CONST_TBL
otherwise it will leak memory. - 11:35 AM Revision 2f20dc5d (git): [DOC] Improve loop code example documentation
-
-
- https://github.com/ruby/English/commit/5e60c1068a
-
- https://github.com/ruby/English/commit/70b46b58cc
10/18/2025
-
-
-
-
01:34 PM Feature #19326: Please add a better API for passing a Proc to a Ractor
- byroot (Jean Boussier) wrote in #note-13:
> I think we can close now?
```ruby
require 'pp'
class Worker
def initialize(*args, &block)
@args = args
@block = Ractor.shareable_proc(&block)
end
def run(count)
... -
-
- ```
/github/workspace/src/ext/zlib/zlib.c:2608:25: warning: default initialization of an object of type 'struct read_raw_arg' with const member leaves the object uninitialized [-Wdefault-const-init-field-unsafe]
2608 | struct re... -
-
- [PATCH] Win32: OpenSSL 1.1 DLLs are no longer used
We are already using OpenSSL 3.3 and have no possibility to use system
provided DLLs. -
- ruby_3_3 commit:e526fbe770909974f610b7a2a00bd8d5dd045f1f merged revision(s) commit:e94a2f691d67ad98be9036e76c765fcfa7d22552.
-
- [PATCH] [Bug #21638] Mark ractor-local `$VERBOSE` and `$DEBUG`
https://github.com/sampersand/blog/blob/master/the%20-s%20flag.md#the-segfault -
- We know the exact capacity for the constant table created in
rb_singleton_class_clone_and_attach so we can preallocate it. -
10/17/2025
-
10:37 PM Revision a0bf6d34 (git): ZJIT: Add inlining for Kernel#respond_to? (#14873)
- lobsters before:
<details>
```
***ZJIT: Printing ZJIT statistics on exit***
Top-20 not inlined C methods (61.1% of total 10,568,718):
Kernel#is_a?: 1,030,925 ( 9.8%)
Stri... -
- We duplicate the method table in rb_class_duplicate_classext, so we should
set RCLASSEXT_ICLASS_IS_ORIGIN and unset RCLASSEXT_ICLASS_ORIGIN_SHARED_MTBL
to signal that the iclass owns the method table and it should be freed. - 08:29 PM Revision 1c119f02 (git): Revert "ZJIT: Mark commonly-edited files as merge=union (#14865)"
- This reverts commit 23287c45806cac060ed63e87176d1a87968b9267.
This looks like a mixed bag... -
- This helps the default merge driver make reasonable decisions (and
therefore avoid conflicts) when multiple people are e.g. adding tests
to src/hir.rs at the same time. -
07:33 PM Revision 0594646c (git): ZJIT: Don't push frame for Hash#size (#14871)
- `Hash#size` was not in "Top-20 not annotated C methods" on lobsters so our before / after benchmarks are not very helpful for this change.
<details>
<summary>Before</summary>
<br>
```
***ZJIT: Printing ZJIT statistics on exit***
Top-20 ... -
- lobsters before:
```
***ZJIT: Printing ZJIT statistics on exit***
Top-20 not inlined C methods (71.9% of total 15,247,103):
Hash#[]: 4,516,006 (29.6%)
Class#current: 1,15... -
- https://github.com/rubygems/rubygems/commit/9b3a5a8ae9
-
-
- 09:37 AM Revision 943b5f66 (git): CI: Launchable: Fix errors at actions/setup-python on ppc64le/s390x
- The following errors happened at the actions/setup-python step.
https://github.com/ruby/ruby/actions/runs/18229870239
> The version '3.x' with architecture 's390x' was not found for Ubuntu 24.04.
> ...
After skipping the actions/setup-... -
08:24 AM Revision 5ced99dd (git): [rubygems/rubygems] :Revamp CmakeBuilder to fix the issues described in #8572. Specifically:
- * Correctly pass command line arguments to CMake
* Call CMake twice - once to configure a project and a second time to build (which is the standard way to use CMake). This fixes the previously incorrect assumption that CMake generates a ... -
-
- Disable the shorten-64-to-32 warning for now, because it currently
generates a lot of warnings on platforms where `sizeof(void*)` is
larger than `sizeof(long)`.
TODO: Replace `long` with `ptrdiff_t` or something in the all sources. -
-
- Get rid of conflict with inline versions provided in time.h.
-
-
-
-
-
- Add `CLOCK_PROCESS_CPUTIME_ID` and `CLOCK_THREAD_CPUTIME_ID`.
- 05:11 AM Revision ff6bd7fb (git): [ruby/ipaddr] Fix InvalidAddressError message
- https://github.com/ruby/ipaddr/commit/c96dbadee3
-
03:01 AM Bug #21632 (Closed): Backport REXML CVE-2025-58767 fix
- Merged.
-
-
- This update addresses CVE-2025-58767 (GHSA-c2f4-jgmc-q2r5).
-
12:53 AM Bug #21641 (Closed): Hash#to_s behaves different in 3.4
- closed by request of OP
10/16/2025
-
- The remote fetcher only works with certain schemes (`http`, `https`,
`s3`, and `file`). It's possible for other schemes to show up in this
code and it can cause bugs.
Before this patch, doing `gem install path:///hello` would result in... -
- remove_class_from_subclasses calls st_insert, which mallocs. Malloc is not
allowed in GC. This commit replaces the st_insert with an st_update since
we know that ns_id exists in the st_table.
The following script reproduces the crash:
... - 08:47 PM Revision 3fa84846 (git): ZJIT: Add to counters when FnProperties.inline inlining succeeds
- This counts methods that can be folded away to nothing *in addition* to
the already-counted `CCall`. - 08:47 PM Revision 9598b444 (git): ZJIT: Fix singleton class qualified method names in stats
- Now methods on singleton classes (for example, `new`) get split up into
`String*#new`, `Array*#new`, ... (where the `*` indicates a singleton
class) instead of all looking like `Class#new`.
before:
```
***ZJIT: Printing ZJIT statistics... -
07:24 PM Bug #21641: Hash#to_s behaves different in 3.4
- Please feel free to join us on [Ruby Discord](https://discord.gg/ad2acQFtkh) or [#ruby IRC](https://web.libera.chat/#ruby) to discuss all the options! Refining Hash is one way:
``` ruby
module LegacyHashToS
refine Hash do
def t... -
07:12 PM Bug #21641: Hash#to_s behaves different in 3.4
- It's mentioned in https://www.ruby-lang.org/en/news/2024/12/25/ruby-3-4-0-released/ under Hash#inspect rendering have been changed. [Bug #20433]
-
06:03 PM Bug #21641 (Closed): Hash#to_s behaves different in 3.4
- I couldn't find anywhere documenting this change from Ruby 3.3 to Ruby 3.4:
3.3.9:
``` ruby
> irb
3.3.9 :001 > {x:1}.to_s
=> "{:x=>1}"
3.3.9 :002 >
```
3.4.0:
```ruby
3.4.0 :001 > {x:1}.to_s
=> "{x: 1}"
3.4.0 :002 >
... -
- We have a lot of patchpoint exits on some applications and this helps
pin down why. -
-
- It just returns nil.
-
02:38 PM Revision 68f45a6d (git): [ruby/prism] Do not rely on RUBY_VERSION being consistent
- https://github.com/ruby/prism/commit/34428946db
-
- https://github.com/ruby/prism/commit/dda0dc81df
-
- https://github.com/ruby/prism/commit/44c4306247
-
- https://github.com/ruby/prism/commit/ebf4425d49
-
- https://github.com/ruby/prism/commit/2466940e49
- 12:45 PM Revision ea4c3169 (git): Update default gems list at aa2d3cd5133e6a5deb24fa047e5a66 [ci skip]
-
- https://github.com/ruby/prism/commit/b72fcc6183
-
- The docs currently say to use `Prism.parse(foo, version: RUBY_VERSION)` for this.
By specifying "current" instead, we can have prism raise a more specifc error.
Note: Does not use `ruby_version` from `ruby/version.h` because writing a t... -
- This is only really called a lot in the benchmark harness, as far as I
can tell. -
- 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_... -
-
- 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... -