tag:blogger.com,1999:blog-6373963829340632529.post5347288083182843910..comments2024-01-30T05:40:30.415-03:00Comments on Algorithmically challenged: Scala 2.9 optimizes for comprehensions way better!Danielhttp://www.blogger.com/profile/07505997833685327219noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-6373963829340632529.post-59358506091665225242011-05-22T15:07:54.555-03:002011-05-22T15:07:54.555-03:00@v6ak
Actually, this is a particular situation wh...@v6ak<br /><br />Actually, this is a particular situation where inlining doesn't work well: megamorphic call sites.<br /><br />See http://www.azulsystems.com/blog/cliff/2011-04-04-fixing-the-inlining-problem for a recent discussion about this problem.Danielhttps://www.blogger.com/profile/07505997833685327219noreply@blogger.comtag:blogger.com,1999:blog-6373963829340632529.post-83978132168907321262011-05-22T09:43:18.154-03:002011-05-22T09:43:18.154-03:00Hmm ... probably versus definately .. I'll tak...Hmm ... probably versus definately .. I'll take the definite one thank you :-)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6373963829340632529.post-34689178480078949142011-05-22T04:05:43.336-03:002011-05-22T04:05:43.336-03:00The method foreach and the lambda seem to be inlin...The method foreach and the lambda seem to be inlined. In this case, usage of IntRef does not IMHO make sense.<br /><br />However, I don't believe there will be any notable performance benefit in long-running app if you use a modern JVM. Modern JVMs probably can inline these calls and dynamic allocation can be probably also inhibited.v6aknoreply@blogger.com