[concurrency-interest] TLR Puzzle - UI threading

Oleksandr Otenko oleksandr.otenko at oracle.com
Mon Feb 3 07:36:37 EST 2014


Deterministically visible?! I wonder what method would you, or any of 
those suggesting it is ok, propose for detecting such a bug in 
application code.


Alex



On 31/01/2014 22:36, Jeff Hain wrote:
>
> Tim Halloran wrote:
> >SWT does this [exception] when event thread calls are detected outside of
> >the SWT thread -- sadly, Swing doesn't.
>
> I'm doing a lightweight GUI framework on top of AWT, but
> something much more simple than AWT and Swing (which were
> slowing me down to a halt, not counting their glitches and hostility
> to huge data models).
> As a side effect of the simplicity, I found it was easy to parallelize
> painting, so I'm glad Graphics2D class lets me paint stuffs from
> different threads (I do check EDT at some places, but through an
> interface, so that it can also work from JUnit thread).
>
>
> As for the TLR puzzle, I agree with those that don't feel the need
> to touch anything except maybe add more warning in the spec.
> I even think that Java 8 behavior is preferable to Java 7 one,
> since it makes the coding error deterministically visible
> instead of hiding it behind some randomness.
>
>
> -Jeff
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.oswego.edu/pipermail/concurrency-interest/attachments/20140203/a571d1c2/attachment.html>


More information about the Concurrency-interest mailing list