|
|
|
|
|
|
by computerphage
3389 days ago
|
|
|
No, even in theory there's an advantage to having information only available at runtime. Some optimizations depend on the probability distribution of data coming in (eg. an optimization for a CSV parser that makes it faster when dealing with well-formed CSVs that would slow the program down if it starts getting a lot of mis-formatted CSVs). The compiler can choose one or the other, but not both. The JVM can effectively choose both by deciding during execution. If the compiler is allowed to embed a JIT or other runtime system, then all bets are off. |
|
|