3 ms·

> that _doesn't_ do this kind of thing Why does that matter to anyone? > you'd need a source language that offers much better static guarantees than the typic

> that _doesn't_ do this kind of thing Why does that matter to anyone? > you'd need a source language that offers much better static guarantees than the typical language that a JIT compiler is written for Java and C# are both statically, strongly typed languages where JITs are the dominant implementation. Almost by definition, it's impossible to write a JIT compiler that outperforms AOT compilation without looking at runtime data, because AOT compilers have a lot more time to look for difficult static optimizations. The reason JITs can keep up is because they have access to information that an AOT compiler does not. > And there's no need to use statistics to optimize anything at runtime when your ahead-of-time compilation step already emits optimal target machine code. This is simply untrue. For example, it's not possible to statically determine whether a function should be inlined or not. However, a JIT can see that it's used in a hot loop and dynamically inline. For any language, no matter the type system, runtime information will always be a superset of compile-time information. There will always exist optimizations in a JIT that aren't possible in an AOT compiler. > Java is as dynamically typed as it gets: `instanceof`, downcasts and reflection, all conspire to reduce the usefulness of static type information to zero. Idiomatic Java code doesn't use these features heavily. Just because it's possible to wipe out type information doesn't mean that the vast majority of code that an AOT or JIT compiler sees won't be strongly typed.

catnaroek 10y ago
> For example, it's not possible to statically determine whether a function should be inlined or not. MLton has absolutely no problems inlining functions, even higher-order functions, at compile time. This is only difficult in languages with virtual methods, because they can be overridden anywhere. If anything, that's an indictment of virtual methods, not AOT compilers. > However, a JIT can see that it's used in a hot loop and dynamically inline. What if it's a virtual method call that's known to be overridden in several places? You can't inline it, even if it's in the middle of a hot loop. > For any language, no matter the type system, runtime information will always be a superset of compile-time information. Runtime information is always anecdotal, specific to one particular run of a program, so... > There will always exist optimizations in a JIT that aren't possible in an AOT compiler. ... for every “optimization” a JIT can perform, there will always exist a program for which the “optimization” will have to be rolled back after it has already been performed, because it turned out to be unsound. > Idiomatic Java code doesn't use these features heavily. Language implementations must work correctly whether you write idiomatic or unidiomatic code. > Just because it's possible to wipe out type information doesn't mean that the vast majority of code that an AOT or JIT compiler sees won't be strongly typed. Most code I write in Python could be given static types too. That doesn't make Python a statically typed language. And “strongly typed” doesn't really mean anything.
- smallnamespace 10y ago
  > MLton has absolutely no problems inlining functions, even higher-order functions, at compile time Of course, but how does it know which functions to inline? If you inline everything, then you will blow through your cache. > What if it's a virtual method call that's known to be overridden in several places? You can't inline it, even if it's in the middle of a hot loop. That's not true--a JIT could optimistically replace with a concrete realization. > Runtime information is always anecdotal, specific to one particular run of a program, so... That's a benefit. No matter what AOT compiled code you have, it is possible to speed up execution if you know what code paths you will take. > ... for every “optimization” a JIT can perform, there will always exist a program for which the “optimization” will have to be rolled back after it has already been performed, because it turned out to be unsound. Yes, but so what? As long as it improves performance in the average case, and the worst case is bounded, then that is a net win. You can equally well write deliberately obfuscated code that an AOT compiler has trouble with. > Language implementations must work correctly whether you write idiomatic or unidiomatic code. Implementation is correct. Only reflection is slow. If you don't want that, don't write reflection.
  - catnaroek 10y ago
    > Of course, but how does it know which functions to inline? Small functions and higher-order functions are the most natural candidates. (The two categories greatly overlap in most cases.) > That's not true--a JIT could optimistically replace with a concrete realization. You'd have to roll back an unsound optimization in the middle of a hot loop. I'm pretty sure that's not what you want. > it is possible to speed up execution if you know what code paths you will take. That knowledge can be encoded statically in many cases, if only you used the right languages. > As long as it improves performance in the average case, and the worst case is bounded, then that is a net win. This is only the case when your program wasn't close to optimal to begin with. > You can equally well write deliberately obfuscated code that an AOT compiler has trouble with. Yes, but languages amenable to static analysis will actively get in your way if you try to write such obfuscated code. The static analysis either tells you that your program is gibberish, or outputs nonsensical gibberish of its own. So the path of least resistance is to write code that the static analysis knows how to optimize. Which is not the case in dynamic languages (including pseudo-static ones like Java). > Only reflection is slow. If you don't want that, don't write reflection. So. basically, you're telling me to ditch Java's entire library and framework ecosystem?

catnaroek 10y ago

> For example, it's not possible to statically determine whether a function should be inlined or not. MLton has absolutely no problems inlining functions, even higher-order functions, at compile time. This is only difficult in languages with virtual methods, because they can be overridden anywhere. If anything, that's an indictment of virtual methods, not AOT compilers. > However, a JIT can see that it's used in a hot loop and dynamically inline. What if it's a virtual method call that's known to be overridden in several places? You can't inline it, even if it's in the middle of a hot loop. > For any language, no matter the type system, runtime information will always be a superset of compile-time information. Runtime information is always anecdotal, specific to one particular run of a program, so... > There will always exist optimizations in a JIT that aren't possible in an AOT compiler. ... for every “optimization” a JIT can perform, there will always exist a program for which the “optimization” will have to be rolled back after it has already been performed, because it turned out to be unsound. > Idiomatic Java code doesn't use these features heavily. Language implementations must work correctly whether you write idiomatic or unidiomatic code. > Just because it's possible to wipe out type information doesn't mean that the vast majority of code that an AOT or JIT compiler sees won't be strongly typed. Most code I write in Python could be given static types too. That doesn't make Python a statically typed language. And “strongly typed” doesn't really mean anything.

smallnamespace 10y ago
> MLton has absolutely no problems inlining functions, even higher-order functions, at compile time Of course, but how does it know which functions to inline? If you inline everything, then you will blow through your cache. > What if it's a virtual method call that's known to be overridden in several places? You can't inline it, even if it's in the middle of a hot loop. That's not true--a JIT could optimistically replace with a concrete realization. > Runtime information is always anecdotal, specific to one particular run of a program, so... That's a benefit. No matter what AOT compiled code you have, it is possible to speed up execution if you know what code paths you will take. > ... for every “optimization” a JIT can perform, there will always exist a program for which the “optimization” will have to be rolled back after it has already been performed, because it turned out to be unsound. Yes, but so what? As long as it improves performance in the average case, and the worst case is bounded, then that is a net win. You can equally well write deliberately obfuscated code that an AOT compiler has trouble with. > Language implementations must work correctly whether you write idiomatic or unidiomatic code. Implementation is correct. Only reflection is slow. If you don't want that, don't write reflection.
- catnaroek 10y ago
  > Of course, but how does it know which functions to inline? Small functions and higher-order functions are the most natural candidates. (The two categories greatly overlap in most cases.) > That's not true--a JIT could optimistically replace with a concrete realization. You'd have to roll back an unsound optimization in the middle of a hot loop. I'm pretty sure that's not what you want. > it is possible to speed up execution if you know what code paths you will take. That knowledge can be encoded statically in many cases, if only you used the right languages. > As long as it improves performance in the average case, and the worst case is bounded, then that is a net win. This is only the case when your program wasn't close to optimal to begin with. > You can equally well write deliberately obfuscated code that an AOT compiler has trouble with. Yes, but languages amenable to static analysis will actively get in your way if you try to write such obfuscated code. The static analysis either tells you that your program is gibberish, or outputs nonsensical gibberish of its own. So the path of least resistance is to write code that the static analysis knows how to optimize. Which is not the case in dynamic languages (including pseudo-static ones like Java). > Only reflection is slow. If you don't want that, don't write reflection. So. basically, you're telling me to ditch Java's entire library and framework ecosystem?