Re: Why does JDK 1.6 recommend creating Swing components on the EDT?
Albert wrote:
Le 01/12/2009 09:54, Roedy Green a ?crit :
On Mon, 30 Nov 2009 13:39:07 -0800, markspace<nospam@nowhere.com>
wrote, quoted or indirectly quoted someone who said :
Yeah, Sun fuggered up their earlier recommendation and had to take it
back. Everything goes on the EDT now.
The problem is threading bugs are very hard to reproduce and track
down. My putting everything onto the EDT thread, they avoid a large
number of headaches.
have you actually seen such a bug ? Something like corrupted data in a
swing model for example ? I've never triggered such a bug, even if i did
a lot of gui calls off the EDT.
In a different context, I've had to track down a hardware bug whose
problem description required a prototype server with at least 50
processors running a kernel stress test plus a special program that
caused excessive numbers of block stores. Given the right configuration
and workload, the bug happened about once a week of continuous operation.
When we finally nailed it, there was a real bug that would have caused
crashes at customer sites. Although customers would not run such a
stressful workload, they do run far more computers far longer than is
possible in system test.
Similarly, a software bug that causes failures about once every week of
run time with just the right sort of workload may be annoying to a
customer with several continuously running copies and a workload that
happens to trigger it, but practically impossible to find in system test.
Threading bugs are *very* hard to reproduce and track down.
Patricia