[ruby-core:77372] [Ruby trunk Bug#12688] Thread unsafety in autoload
From:
headius@...
Date:
2016-09-23 13:00:30 UTC
List:
ruby-core #77372
Issue #12688 has been updated by Charles Nutter.
> It successfully runs for less than 50 dots, and then one of the threads errors out. I don't think it should.
Ok, I can't count...it does run through one full iteration and then probably fails because the autoload doesn't actually load anything. But then it still fails a hundred iterations later, so something's still not right.
----------------------------------------
Bug #12688: Thread unsafety in autoload
https://bugs.ruby-lang.org/issues/12688#change-60618
* Author: Charles Nutter
* Status: Open
* Priority: Normal
* Assignee:
* ruby -v: 2.3.0
* Backport: 2.1: UNKNOWN, 2.2: UNKNOWN, 2.3: UNKNOWN
----------------------------------------
I need clarification here. I expected, based on Ruby's assertion that autoloads are thread-safe, that the following code would never error. Instead, it gets a couple iterations in and raises NameError:
```ruby
loop do
class Foo
autoload :Bar, 'bar.rb'
end
go = false
threads = (1..50).map {Thread.new { 1 until go; print '.'; Foo.const_get(:Bar) }}
go = true
threads.each(&:join)
puts
self.class.send :remove_const, :Foo
end
```
And the output with Ruby 2.3.0:
```
$ ruby23 -I. autoload_breaker.rb
..................................................
..................................................autoload_breaker.rb:7:in `const_get': uninitialized constant Foo::Bar (NameError)
Did you mean? Foo::Bar
from autoload_breaker.rb:7:in `block (3 levels) in <main>'
```
Is there something wrong with my script? Is my expectation incorrect?
--
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>