[#99856] [Ruby master Feature#17143] Improve support for warning categories — merch-redmine@...

Issue #17143 has been reported by jeremyevans0 (Jeremy Evans).

16 messages 2020/09/03

[#99868] [Ruby master Bug#17144] Tempfile.open { ... } does not unlink the file — eregontp@...

Issue #17144 has been reported by Eregon (Benoit Daloze).

15 messages 2020/09/03

[#99885] [Ruby master Feature#17145] Ractor-aware `Object#deep_freeze` — marcandre-ruby-core@...

Issue #17145 has been reported by marcandre (Marc-Andre Lafortune).

32 messages 2020/09/03

[#99903] [Ruby master Bug#17146] Queue operations are allowed after it is frozen — eregontp@...

Issue #17146 has been reported by Eregon (Benoit Daloze).

16 messages 2020/09/03

[#100016] [Ruby master Feature#17171] Why is the visibility of constants not affected by `private`? — marcandre-ruby-core@...

Issue #17171 has been reported by marcandre (Marc-Andre Lafortune).

10 messages 2020/09/15

[#100024] [Ruby master Bug#17175] Ruby 2.5: OpenSSL related test failures — jaruga@...

Issue #17175 has been reported by jaruga (Jun Aruga).

10 messages 2020/09/16

[#100025] [Ruby master Feature#17176] GC.enable_autocompact / GC.disable_autocompact — tenderlove@...

Issue #17176 has been reported by tenderlovemaking (Aaron Patterson).

11 messages 2020/09/16

[#100099] [Ruby master Bug#17184] No stdlib function to perform simple string replacement — sheerun@...

Issue #17184 has been reported by sheerun (Adam Stankiewicz).

18 messages 2020/09/24

[#100192] [Ruby master Bug#17197] Some Hash methods still have arity 2 instead of 1 — marcandre-ruby-core@...

Issue #17197 has been reported by marcandre (Marc-Andre Lafortune).

14 messages 2020/09/28

[#100200] [Ruby master Misc#17199] id outputed by inspect and to_s output does not allow to find actual object_id and vice-versa — baptiste.courtois@...

Issue #17199 has been reported by Annih (Baptiste Courtois).

7 messages 2020/09/28

[#100206] [Ruby master Misc#17200] DevelopersMeeting20201026Japan — mame@...

Issue #17200 has been reported by mame (Yusuke Endoh).

18 messages 2020/09/28

[#100239] [Ruby master Feature#17206] Introduce new Regexp option to avoid MatchData allocation — fatkodima123@...

Issue #17206 has been reported by fatkodima (Dima Fatko).

8 messages 2020/09/30

[ruby-core:100190] [Ruby master Bug#15661] Disallow concurrent Dir.chdir with block

From: sam.saffron@...
Date: 2020-09-28 00:07:46 UTC
List: ruby-core #100190
Issue #15661 has been updated by sam.saffron (Sam Saffron).


I guess my bigger point here is that even with this fix the block form remains unsafe under concurrent use. At best this catches a few multithreading bugs. The construct is incompatible with multithreaded programming cause state leaks.

I do not object to making this "a little less terrible". But ... it remains terrible. 

This fix also does nothing really for single threaded programs which are not in scope. 

----------------------------------------
Bug #15661: Disallow concurrent Dir.chdir with block
https://bugs.ruby-lang.org/issues/15661#change-87761

* Author: headius (Charles Nutter)
* Status: Open
* Priority: Normal
* ruby -v: all
* Backport: 2.4: UNKNOWN, 2.5: UNKNOWN, 2.6: UNKNOWN
----------------------------------------
`Dir.chdir` with a block should disallow concurrent use, since it will almost never produce the results a user expects.

In https://bugs.ruby-lang.org/issues/9785 calls for `Dir.chdir` to be made thread-safe were rejected because the underlying native call is process-global. This is reasonable because there's no way to easily make the native chdir be thread-local (at least not without larger changes to CRuby itself).

However the block form of `chdir` is explicitly bounded, and clearly indicates that the dir should be changed only for the given block. I believe `Dir.chdir` should prevent multiple threads from attempting to do this bounded `chdir` at the same time.

Currently, if two threads attempt to do a `Dir.chdir` with a block, one of them will see a warning: "conflicting chdir during another chdir block". This warning is presumably intended to inform the user that they may see unpredictable results, but I can think of no cases where you would ever see predictable results.

In other words, there's no reason to allow a user to do concurrent `Dir.chdir` with a block because they will always be at risk of unpredictable results, and I believe this only leads to hard-to-diagnose bugs.

The warning should be a hard error.

The warning should also be turned into an error if a non-block `Dir.chdir` call happens while a block Dir.chdir is in operation. The important thing is to protect the integrity of the current directory during the lifetime of that block invocation.

In CRuby terms, the patch would be to check for `chdir_blocking > 0` and then simply error out, unless it's happening on the same thread.

Concurrent non-block `Dir.chdir` would be unaffected. This only protects the block form from having the current dir change while it is executing.

See https://github.com/jruby/jruby/issues/5649



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

Unsubscribe: <mailto:[email protected]?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>

In This Thread

Prev Next