Hi Dmitry,
I like the idea of the fix. Note that the proposed implementation still
allows for timeout to become larger than the constant, and it gets reset
back to the value of the constant on the next iteration only. Perhaps it
would be simpler to just do the following: if (t<T) { t <<= 1; } - i.e.
we don't really need the : branch of the ?: operator. IMO that would
simplify the logic. But I'm OK with the current version as well. See for
yourself if you want to update the fix.
In any case, we need at least one more reviewer for this patch.
--
best regards,
Anthony
On 05/27/2013 04:28 PM, dmitry markov wrote:
Hello,
Could you review the fix, please?
bug: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=8015375
webrev: http://cr.openjdk.java.net/~vkarnauk/8015375/webrev.00/
The function XNextSecondaryLoopEvent() should use a progressive time out
value instead of hardcoded 250 msces during secondary loop execution.
Thanks,
Dmitry
