On 06/22/11 04:47 PM, Joe Darcy wrote:
David Holmes wrote:
Hi Joe,
Joe Darcy said the following on 06/18/11 11:38:
Please review this (somewhat tedious) change to make the behavior of
the Number subtypes in the JDK more explicit:
6253144: Long narrowing conversion should describe the algorithm used
and implied "risks"
http://cr.openjdk.java.net/~darcy/6253144.0/
David, how are changes to AtomicInteger and AtomicLong managed?
Normally they would go into Doug Lea's CVS for jsr166, we (Chris
Hegarty) would pull them over and then push to OpenJDK. It can work
the other way but the sync's can get messier.
Chris,
Off-list, Mike approved this set of changes and I'd like to get them
pushed once the matching ccc is approved. (The ccc request is needed
since the long-standing behavior of the non-abstract methods on number
is being specified.)
How would you like to handle updates to the Atomic classes?
I think in this case it should be fine to make the changes in OpenJDK
first. Then I can create a patch for the Atomic changes based on Dougs CVS.
I'll double check this with Doug, but unless you hear otherwise let's
assume we can do this. In fact, if Doug is watching he may be able to
move faster than us!
-Chris.
Thanks,
-Joe
