[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: Re[6]: [aspectj-users] Once again aboutn non-this access
|
Konstantin,
Thanks for digging in and fixing my test. So apparently the VM doesn't optimize for this case. It certainly *could* by performing an analysis to determine that the results of the if pointcut tests are constant and then simplify the resulting code. So there's no requirement that the JIT perform these calls. But obviously I was too optimistic (and sorry for the typo!).
You've convinced me that there is a good performance reason to implement the optimization. I'd suggest you enter a request enhancement into bugzilla. To get the feature implemented most quickly, I'd suggest you implement this optimization yourself in the AspectJ compiler. To contribute like this, I'd post to aspectj-dev to coordinate and get help on the design strategy and then submit a patch.
Ron
Ron Bodkin
Chief Technology Officer
New Aspects of Software
m: (415) 509-2895
