Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #4877

Re: Making one or more threads wait for another to produce a value or fail

Show all headers | View raw

On 6/1/11 2:07 PM, Tom Anderson wrote:
> [...]
>> In particular, as long as you ensure that the writer doesn't get to
>> arrive at its "synchronized" statement in which it will call notify()
>> until after all the readers have arrived at their wait() statement,
>> you're assured all the readers will see the new value.
>
> That is true, but that's a more stringent requirement than is needed:
> the happens-before relationship is between the writer's lock release,
> and the reader's lock acquisition. The reader's prospective wait must be
> inside a synchronized block, so even if a reader arrives at its wait
> statement long after the writer has called notify() (and unlocked) its
> lock acquisition will mean the writer's actions will have
> happened-before that moment.

Actually, the scenario that is problematic (or at least the one that I 
was thinking of and addressing above) is the reader arriving early, not 
late.

Depending on how the rest of the code is structured, this may or may not 
even be possible anyway.  But if it is, one needs to protect against it.

> I imagine you knew that, of course, and we're talking slightly at
> cross-purposes.
>
>> Alternatively, maintain a separate "do I really need to wait?" flag
>> that is cleared when the delivered value is set, so that readers don't
>> bother waiting if the value's already been updated.
>
> Yes. Which you need anyway, to deal with the spurious wakeups!

Yes, but I think it still depends on the specific implementation.  You 
need some way of making sure that flag is cleared before any of the 
readers can potentially check it.

As long as the specific implementation naturally protects against that 
(e.g. doesn't even start the readers until the flag has been cleared), 
it's not an issue at all.  But as with all concurrent code, one should 
think carefully to make sure it's not an issue.  :)

Pete

Back to comp.lang.java.programmer | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

Making one or more threads wait for another to produce a value or fail Tom Anderson <twic@urchin.earth.li> - 2011-05-31 15:00 +0100
  Re: Making one or more threads wait for another to produce a value or fail Peter Duniho <NpOeStPeAdM@NnOwSlPiAnMk.com> - 2011-05-31 07:14 -0700
    Re: Making one or more threads wait for another to produce a value or fail Tom Anderson <twic@urchin.earth.li> - 2011-05-31 16:46 +0100
      Re: Making one or more threads wait for another to produce a value or fail Deeyana <d.awlberg@hotmail.invalid> - 2011-05-31 19:23 +0000
        Re: Making one or more threads wait for another to produce a value or fail Tom Anderson <twic@urchin.earth.li> - 2011-06-01 22:12 +0100
      Re: Making one or more threads wait for another to produce a value or fail Peter Duniho <NpOeStPeAdM@NnOwSlPiAnMk.com> - 2011-05-31 20:45 -0700
        Re: Making one or more threads wait for another to produce a value or fail Patricia Shanahan <pats@acm.org> - 2011-05-31 21:37 -0700
          Re: Making one or more threads wait for another to produce a value or fail Peter Duniho <NpOeStPeAdM@NnOwSlPiAnMk.com> - 2011-06-01 07:18 -0700
            Re: Making one or more threads wait for another to produce a value or fail Patricia Shanahan <pats@acm.org> - 2011-06-01 09:01 -0700
        Re: Making one or more threads wait for another to produce a value or fail Lawrence D'Oliveiro <ldo@geek-central.gen.new_zealand> - 2011-06-01 16:40 +1200
          Re: Making one or more threads wait for another to produce a value or fail "John B. Matthews" <nospam@nospam.invalid> - 2011-06-01 11:22 -0400
            Re: Making one or more threads wait for another to produce a value or fail Lawrence D'Oliveiro <ldo@geek-central.gen.new_zealand> - 2011-06-02 10:15 +1200
              Re: Making one or more threads wait for another to produce a value or fail "John B. Matthews" <nospam@nospam.invalid> - 2011-06-01 21:38 -0400
        Re: Making one or more threads wait for another to produce a value or fail Tom Anderson <twic@urchin.earth.li> - 2011-06-01 22:07 +0100
          Re: Making one or more threads wait for another to produce a value or fail Peter Duniho <NpOeStPeAdM@NnOwSlPiAnMk.com> - 2011-06-01 22:46 -0700
  Re: Making one or more threads wait for another to produce a value or fail markspace <-@.> - 2011-05-31 08:26 -0700
    Re: Making one or more threads wait for another to produce a value or fail Tom Anderson <twic@urchin.earth.li> - 2011-05-31 19:13 +0100
  Re: Making one or more threads wait for another to produce a value or fail Patricia Shanahan <pats@acm.org> - 2011-05-31 09:06 -0700
    Re: Making one or more threads wait for another to produce a value or fail Paul Cager <paul.cager@googlemail.com> - 2011-05-31 10:08 -0700
      Re: Making one or more threads wait for another to produce a value or fail Patricia Shanahan <pats@acm.org> - 2011-05-31 10:39 -0700
        Re: Making one or more threads wait for another to produce a value or fail Paul Cager <paul.cager@googlemail.com> - 2011-05-31 16:24 -0700
  Re: Making one or more threads wait for another to produce a value or fail Lawrence D'Oliveiro <ldo@geek-central.gen.new_zealand> - 2011-06-01 12:33 +1200
    Re: Making one or more threads wait for another to produce a value or fail "John B. Matthews" <nospam@nospam.invalid> - 2011-06-01 00:38 -0400
  Re: Making one or more threads wait for another to produce a value or fail markspace <-@.> - 2011-06-02 14:10 -0700
    Re: Making one or more threads wait for another to produce a value or fail markspace <-@.> - 2011-06-02 14:25 -0700
    Re: Making one or more threads wait for another to produce a value or fail Patricia Shanahan <pats@acm.org> - 2011-06-02 14:43 -0700

csiph-web