<div dir="ltr">Yes (I wrote this privately to Vincent too).<div><br></div><div>Also, it has the advantage of being clear that you're not sending any meaningful values (for future maintainers of this code).</div><div><br> </div><div>Robby</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, May 3, 2013 at 11:06 AM, Vincent St-Amour <span dir="ltr"><<a href="mailto:stamourv@ccs.neu.edu" target="_blank">stamourv@ccs.neu.edu</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Sounds good, I'll use a semaphore.<br> <br> Vincent<br> <br> <br> At Fri, 3 May 2013 09:58:25 -0600,<br> <div class="HOEnZb"><div class="h5">Matthew Flatt wrote:<br> ><br> > Thanks for tracking this down!<br> ><br> > In case you mean a channel in the sense of `make-channel', I recommend<br> > a semaphore, instead, since a semaphore is the lightest-weight<br> > synchronization construct.<br> ><br> > At Fri, 03 May 2013 11:45:38 -0400, Vincent St-Amour wrote:<br> > > Sam, Asumu and I found and fixed the bug. Here's the problem in a<br> > > nutshell.<br> > ><br> > > This is the implementation of `delay/thread' from racket/promise.<br> > ><br> > > (define (delay/thread thunk)<br> > > (let ()<br> > > (define (run)<br> > > (call-with-exception-handler<br> > > (lambda (e) (pset! p (make-reraise e)) (kill-thread<br> > > (current-thread)))<br> > > (lambda () (pset! p (call-with-values thunk list)))))<br> > > (define p<br> > > (make-promise/thread<br> > > (make-running-thread<br> > > (object-name thunk)<br> > > (thread run))))<br> > > p))<br> > ><br> > > `run' depends on `p', and vice versa. The `run' thread may start<br> > > executing *before* `p' is actually defined. This causes `pset!' (which<br> > > is just promise mutation) to mutate `#<undefined>', which raises an<br> > > exception[1]. The exception handler then calls `pset!' again, which<br> > > raises an exception, and kills the thread without setting the promise's<br> > > value.<br> > ><br> > > Our solution is to have `run' block on a channel until `p' is fully<br> > > initialized. I'll push the fix.<br> > ><br> > > Vincent<br> > ><br> > ><br> > > [1] Actually, since `pset!' uses `unsafe-struct-set!', there is no<br> > > actual exception. The thread gets killed some other way.<br> > ><br> > ><br> > > At Thu, 2 May 2013 22:27:14 -0400,<br> > > Asumu Takikawa wrote:<br> > > ><br> > > > On 2013-05-02 22:14:44 -0400, Asumu Takikawa wrote:<br> > > > > This produces a promise error on every third or fifth run or so for me.<br> > > > > Commenting out the #:when line makes it work, very oddly.<br> > > ><br> > > > Tried it on another machine, the #:when line didn't matter there. Also,<br> > > > I can reproduce this more reliably on a machine with fewer cores.<br> > > ><br> > > > Cheers,<br> > > > Asumu<br> > > > _________________________<br> > > > Racket Developers list:<br> > > > <a href="http://lists.racket-lang.org/dev" target="_blank">http://lists.racket-lang.org/dev</a><br> > > _________________________<br> > > Racket Developers list:<br> > > <a href="http://lists.racket-lang.org/dev" target="_blank">http://lists.racket-lang.org/dev</a><br> _________________________<br> Racket Developers list:<br> <a href="http://lists.racket-lang.org/dev" target="_blank">http://lists.racket-lang.org/dev</a><br> </div></div></blockquote></div><br></div>