On 10/14/07, Brian Goetz <[EMAIL PROTECTED]> wrote: > > > > I'm not aware of any work on improving the performance/correctness > > tradeoff of integral arithmetic on the Java platform. If you're > > generating byte-code for the Java VM, you have to choose between good > > performance but limited precision with silent overflow or poor > > performance but flexible precision. The general problem was "solved" > > long ago, including in the StrongTalk system that was a predecessor to > > HotSpot. Unfortunately those results have not been applied to the Java > > platform. > > Neal, could you sketch out the correct general solution, for those of us > not familiar with Strongtalk?
There are many possible solutions, depending on your hardware architecture. One solution is to compile to arithmetic instructions that trap on overflow; on overflow JIT-rewrite the code to use extended precision. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "JVM Languages" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/jvm-languages?hl=en -~----------~----~----~----~------~----~------~--~---
