[#101174] [Ruby master Bug#17359] Ractor copy mode is not Ractor-safe — marcandre-ruby-core@...

Issue #17359 has been reported by marcandre (Marc-Andre Lafortune).

17 messages 2020/12/01

[#101217] [Ruby master Feature#17363] Timeouts — marcandre-ruby-core@...

Issue #17363 has been reported by marcandre (Marc-Andre Lafortune).

19 messages 2020/12/03

[#101250] [Ruby master Bug#17369] Introduce non-blocking `Process.wait`, `Kernel.system` and related methods. — samuel@...

Issue #17369 has been reported by ioquatix (Samuel Williams).

12 messages 2020/12/05

[#101276] [Ruby master Bug#17373] Ruby 3.0 is slower at Discourse bench than Ruby 2.7 — sam.saffron@...

Issue #17373 has been reported by sam.saffron (Sam Saffron).

11 messages 2020/12/07

[#101278] [Ruby master Bug#17374] Refined methods aren't visible from a refinementRefinements that include/prepend module — marcandre-ruby-core@...

Issue #17374 has been reported by marcandre (Marc-Andre Lafortune).

17 messages 2020/12/07

[#101317] [Ruby master Feature#17378] Ractor#receive with filtering like other actor langauge — ko1@...

Issue #17378 has been reported by ko1 (Koichi Sasada).

9 messages 2020/12/08

[#101343] [Ruby master Bug#17382] Segfault in String#inspect — lionel.perrin@...

Issue #17382 has been reported by lionelperrin (Lionel Perrin).

10 messages 2020/12/09

[#101381] [Ruby master Bug#17385] Test failures on gcc 11 — jaruga@...

Issue #17385 has been reported by jaruga (Jun Aruga).

18 messages 2020/12/10

[#101458] [Ruby master Bug#17394] TCPServer is not thread safe on win32 — aladjev.andrew@...

Issue #17394 has been reported by puchuu (Andrew Aladjev).

9 messages 2020/12/14

[#101472] [Ruby master Feature#17397] shareable_literal_constant should check at runtime, not at parse time — marcandre-ruby-core@...

Issue #17397 has been reported by marcandre (Marc-Andre Lafortune).

10 messages 2020/12/16

[#101475] [Ruby master Bug#17398] SyntaxError in endless method — zverok.offline@...

Issue #17398 has been reported by zverok (Victor Shepelev).

15 messages 2020/12/16

[#101477] [Ruby master Misc#17399] Are endless methods experimental? — zverok.offline@...

Issue #17399 has been reported by zverok (Victor Shepelev).

13 messages 2020/12/16

[#101480] [Ruby master Bug#17400] Incorrect character downcase for Greek Sigma — xfalcox@...

Issue #17400 has been reported by xfalcox (Rafael Silva).

10 messages 2020/12/16

[#101513] [Ruby master Bug#17405] irb ---nomultiline gets exception when output contains some non-ascii characters — rsharman@...

Issue #17405 has been reported by rsharman (Richard Sharman).

8 messages 2020/12/18

[#101534] [Ruby master Bug#17409] Endless range of dates stuck on include? when miss — sergey.gnuskov@...

Issue #17409 has been reported by gsmetal (Sergey G).

9 messages 2020/12/19

[#101546] [Ruby master Bug#17411] Syntax error with . in pattern — zverok.offline@...

Issue #17411 has been reported by zverok (Victor Shepelev).

10 messages 2020/12/19

[#101598] [Ruby master Bug#17420] Unsafe mutation of $" when doing non-RubyGems require in Ractor — eregontp@...

Issue #17420 has been reported by Eregon (Benoit Daloze).

12 messages 2020/12/21

[#101635] [Ruby master Bug#17428] Method#inspect bad output for class methods — marcandre-ruby-core@...

Issue #17428 has been reported by marcandre (Marc-Andre Lafortune).

13 messages 2020/12/23

[#101639] [Ruby master Bug#17429] Prohibit include/prepend in refinement modules — shugo@...

Issue #17429 has been reported by shugo (Shugo Maeda).

32 messages 2020/12/23

[#101707] [Ruby master Feature#17472] HashWithIndifferentAccess like Hash extension — naruse@...

Issue #17472 has been reported by naruse (Yui NARUSE).

31 messages 2020/12/26

[#101710] [Ruby master Feature#17473] Make Pathname to embedded class of Ruby — hsbt@...

Issue #17473 has been reported by hsbt (Hiroshi SHIBATA).

28 messages 2020/12/26

[#101719] [Ruby master Feature#17474] Interpreting constants at compile time — jzakiya@...

Issue #17474 has been reported by jzakiya (Jabari Zakiya).

23 messages 2020/12/26

[#101735] [Ruby master Misc#17480] DevelopersMeeting20210113Japan — mame@...

Issue #17480 has been reported by mame (Yusuke Endoh).

12 messages 2020/12/27

[#101790] [Ruby master Bug#17486] Build fails on darwin due to libtool being removed — dark.panda@...

Issue #17486 has been reported by dark.panda (J Smith).

11 messages 2020/12/28

[#101794] [Ruby master Bug#17488] Regression in Ruby 3: Hash#key? is non-deterministic when argument uses DelegateClass — myron.marston@...

Issue #17488 has been reported by myronmarston (Myron Marston).

11 messages 2020/12/28

[#101809] [Ruby master Feature#17490] Rename RubyVM::MJIT to RubyVM::JIT — takashikkbn@...

Issue #17490 has been reported by k0kubun (Takashi Kokubun).

15 messages 2020/12/30

[#101838] [Ruby master Feature#17496] Add constant Math::TAU — jzakiya@...

Issue #17496 has been reported by jzakiya (Jabari Zakiya).

32 messages 2020/12/31

[#101840] [Ruby master Bug#17497] Ractor performance issue — marcandre-ruby-core@...

Issue #17497 has been reported by marcandre (Marc-Andre Lafortune).

21 messages 2020/12/31

[ruby-core:101798] [Ruby master Bug#17420] Unsafe mutation of $" when doing non-RubyGems require in Ractor

From: shatrov@...
Date: 2020-12-28 21:27:38 UTC
List: ruby-core #101798
Issue #17420 has been updated by kirs (Kir Shatrov).


(coming to this bug from https://bugs.ruby-lang.org/issues/17477)

I think it's fine to force `require` to be called only from the main thread/ractor, at least for now. I can imagine that would simplify a lot of things.

I'd like to make that error very clear to developers, and not `can not access non-shareable objects in constant Kernel::RUBYGEMS_ACTIVATION_MONITOR by non-main ractor` that they see right now.

----------------------------------------
Bug #17420: Unsafe mutation of $" when doing non-RubyGems require in Ractor
https://bugs.ruby-lang.org/issues/17420#change-89623

* Author: Eregon (Benoit Daloze)
* Status: Open
* Priority: Normal
* Assignee: ko1 (Koichi Sasada)
* ruby -v: ruby 3.0.0dev (2020-12-16T10:12:48Z master a9a7f4d8b8) [x86_64-linux]
* Backport: 2.5: UNKNOWN, 2.6: UNKNOWN, 2.7: UNKNOWN
----------------------------------------
With an empty file `a.rb`:
```
$ ruby --disable-gems -e 'Ractor.new { puts $" }.take'
-e:1:in `block in <main>': can not access global variables $" from non-main Ractors (RuntimeError)
```
That is expected, given the rules for global variables.

```
ruby --disable-gems -e 'Ractor.new { require "./a.rb"; }.take; p $"'
[... , "/home/eregon/a.rb"]
```
Is it OK that the Ractor can do `require`, which does modify `$"`?

I think it's not, and it might lead to segfaults if e.g. the main Ractor mutates `$"` in parallel to some other Ractor doing `require`.

Probably `require` needs to be forbidden in non-main Ractors (it does mutate `$"`, so it's logical), or there needs to be always VM-global synchronization on any access to `$"` (otherwise, segfaults are possible).
The latter doesn't seem reasonable, especially when considering the user might do `$".each { ... }`.

---

Note that RubyGems' `require` does not work on non-main Ractors (pretty much expected given it depends on a lot of global state):
```
$ ruby -e 'Ractor.new { require "./a.rb"; }.take'
<internal:/home/eregon/prefix/ruby-master/lib/ruby/3.0.0/rubygems/core_ext/kernel_require.rb>:37:in `require': can not access non-shareable objects in constant Kernel::RUBYGEMS_ACTIVATION_MONITOR by non-main ractor. (NameError)
```

This probably also has consequences for `autoload`.
Maybe the `zeitwerk` gem can help with the mode to resolve all autoload at once.



-- 
https://bugs.ruby-lang.org/

Unsubscribe: <mailto:ruby-core-request@ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>

In This Thread