On Mon, Jan 24, 2011 at 11:13 AM, Joshua Ballanco <jballanc@gmail.com> wrote:

Regarding the potential bug...well...where to begin? So, yes, MacRuby blocks have the same semantics as regular Ruby blocks, *except* when they are dispatched through GCD. In that case, ivars are shared, but local variable get copied. I'm sure Laurent can explain why better than I, but it has to do with the semantics of libdispatch and the uncertainty inherent in when a dispatched block or function will execute (i.e. if local variables were not copied during dispatch, they might go out of scope and be collected before GCD ever gets around to running the code).

Wow, that's very surprising. I'm not sure I agree with bending Ruby semantics so drastically, even to help concurrency. Or at least, I'd expect other threaded scenarios to be consistent: ~ ➔ ruby -ve "a = 0; Thread.new { a += 1 }.join; p a" MacRuby 0.8 (ruby 1.9.2) [universal-darwin10.0, x86_64] 1 ~ ➔ ruby -ve "a = 0; q = Dispatch::Queue.concurrent; q.sync {a += 1}; p a" MacRuby 0.8 (ruby 1.9.2) [universal-darwin10.0, x86_64] 0 The implicitness in being able to mutate the surrounding scope is certainly problematic. This is one reason Java's anonymous inner classes require that referenced variables be declared final...to indicate they can't be mutated by the body of the anonymous inner class's methods. The result is that people end up using one-element arrays and the like, but people find ways around anything. So I suppose this applies to anything in the surrounding scope, including visibility, $~, and so on?

To illustrate how this impacts async semantics wrt ruby, here's a sample done with MacRuby/GCD and canonical Ruby:

I'll see how this might be implemented in JRuby.

In other words, not a bug in MacRuby...also, this code is scheduled for a rewrite that will turn it into a bit more of a continuation-passing style of handling the connections. I don't think tap is really the solution we are looking for here, since after dispatching, the main #accept loop doesn't really care about that connection (other than that, at some point, it is responsibly closed).

tap would be a clean way to explicitly isolate each loop's environment, if you weren't already doing that. - Charlie

On Wed, Jan 26, 2011 at 4:50 PM, Laurent Sansonetti <lsansonetti@apple.com> wrote:

No, only locals and dynamic (block) variables. To be honest I always disliked this semantic change too. I think it was a mistake to add it. It will probably be reverted for 1.0.

I started to try to implement it, and it's not easy. I would need to cascade all the way up the containing scopes, cloning each one. It's very messy. I would have no objections to this behavior going away. What if instead of e.g. async, you had the following form: queue.async(a, b) {|a, b| ... } Because 1.9 forces variables in the block's argument list to be local to the block, this would avoid the sharing of variables completely. This would work with anything proc-like as well: job = ->(a, b) { ... } # block-local a and b queue.async(a, b, &job) # outer scope's a and b Of course you might want to rename them in the block, to avoid confusion, but this neatly does what "tap" did in my suggested fix (and I don't know why I didn't think of it sooner). - Charlie

On Saturday, October 22, 2011 at 1:35 AM, Michael Johnston wrote:

When I need to get a queue-protected result into a local in code that is concurrent (so I can't use an ivar) is a pointer the best (only) way to get the result of a sync block? Assuming I don't want to factor out a method object.

ex:

result_p = Pointer.new(:id) some_queue.sync do result_p.assign(value_protected_by_queue) end result = result_p[0]

it's not very ruby-ish...

There's no restriction on not using ivars in block. Indeed, even locals are in scope in a block (and with a #sync dispatched block such as the one you provided, there aren't even any threading issues): result = nil Dispatch::Queue.concurrent.sync do result = true end p result #=> true

Ah, indeed, there are subtle differences with lvars at the top-level in IRB vs in a method scope. Not sure if I'd consider this a bug or not, but to illustrate:

...

result = nil => nil Dispatch::Queue.concurrent.sync do result = true end => nil result => true def test result = nil Dispatch::Queue.concurrent.sync do result = true end result end => nil test => nil

Anyway, we're getting away from the original question. If you replace your lvars with ivars (s/result/@result/ above), then things will work. Just be careful of the possibility of concurrency bugs when assigning the same ivar from multiple queued blocks (and also have a look at the dispatch gem which has some convenience methods to make this sort of thing easier). On Saturday, October 22, 2011 at 6:30 AM, Michael Johnston wrote:

Very strange, your example works for me in irb too. Irb must do something odd with locals. Have you tried this outside of irb, though? If you wrap with a method def'n in irb, it behaves differently, and the fact that locals are copied has been discussed lots on the list. But Laurent (I think) said this will likely change, has it already? I'm on 0.10

$ macirb --noprompt def check print "can I change a local scalar? " maybe = "nope" $q.sync do maybe = "yep" end puts maybe end => nil def check_p print "can I assign to a pointer? " maybe_p = Pointer.new(:id) maybe_p.assign "nope" $q.sync do maybe_p.assign "yep" end puts maybe_p[0] end => nil def check_w print "can I assign to a wrapper attr? " maybe = ResultWrapper.new("nope") $q.sync do maybe.value = "yep" end puts maybe.value end => nil ResultWrapper = Struct.new(:value) => ResultWrapper $q= Dispatch::Queue.new('q') => q check can I change a local scalar? nope => nil # but using a pointer works: => nil check_p can I assign to a pointer? yep => nil # or a wrapper (but more expensive for tight loops): => nil check_w can I assign to a wrapper attr? yep => nil $q= Dispatch::Queue.concurrent => com.apple.root.default-priority # double-checking it isn't different for the parallel queues: => nil check can I change a local scalar? nope => nil check_p can I assign to a pointer? yep => nil check_w can I assign to a wrapper attr? yep => nil

Cheerio,

Michael Johnston lastobelus@mac.com (mailto:lastobelus@mac.com)

On 2011-10-22, at 12:32 AM, Joshua Ballanco wrote:

...

On Saturday, October 22, 2011 at 1:35 AM, Michael Johnston wrote:

...

When I need to get a queue-protected result into a local in code that is concurrent (so I can't use an ivar) is a pointer the best (only) way to get the result of a sync block? Assuming I don't want to factor out a method object.

ex:

result_p = Pointer.new(:id) some_queue.sync do result_p.assign(value_protected_by_queue) end result = result_p[0]

it's not very ruby-ish...

There's no restriction on not using ivars in block. Indeed, even locals are in scope in a block (and with a #sync dispatched block such as the one you provided, there aren't even any threading issues):

result = nil Dispatch::Queue.concurrent.sync do result = true end p result #=> true _______________________________________________ MacRuby-devel mailing list MacRuby-devel@lists.macosforge.org (mailto:MacRuby-devel@lists.macosforge.org) http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel

_______________________________________________ MacRuby-devel mailing list MacRuby-devel@lists.macosforge.org (mailto:MacRuby-devel@lists.macosforge.org) http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel