Object-oriented programming languages confer many benefits, including abstraction, which lets the programmer hide the details of an object’s implementation from the object’s clients. Unfortunately, crossing abstraction boundaries often incurs a substantial run-time overhead in the form of frequent procedure calls. Thus, pervasive use of abstraction, while desirable from a design standpoint, may be impractical when it leads to inefficient programs. Aggressive compiler optimizations can reduce the overhead of abstraction. However, the long compilation times introduced by optimizing compilers delay the programming environment‘s responses to changes in the program. Furthermore, optimization also conflicts with source-level debugging. Thus, programmers are caught on the horns of two dilemmas: they have to choose between abstraction and efficiency, and between responsive programming environments and efficiency. This dissertation shows how to reconcile these seemingly contradictory goals by performing optimizations lazily. Four new techniques work together to achieve high performance and high responsiveness: • Type feedback achieves high performance by allowing the compiler to inline message sends based on information extracted from the runtime system. On average, programs run 1.5 times faster than the previous SELF system; compared to a commercial Smalltalk implementation, two medium-sized benchmarks run about three times faster. This level of performance is obtained with a compiler that is both simpler and faster than previous SELF compilers. • Adaptive optimization achieves high responsiveness without sacrificing performance by using a fast nonoptimizing compiler to generate initial code while automatically recompiling heavily used parts of the program with an optimizing compiler. On a previous-generation workstation like the SPARCstation-2, fewer than 200 pauses exceeded 200 ms during a 50-minute interaction, and 21 pauses exceeded one second. On a currentgeneration workstation, only 13 pauses exceed 400 ms. • Dynamic deoptimization shields the programmer from the complexity of debugging optimized code by transparently recreating non-optimized code as needed. No matter whether a program is optimized or not, it can always be stopped, inspected, and single-stepped. Compared to previous approaches, deoptimization allows more debugging while placing fewer restrictions on the optimizations that can be performed. • Polymorphic inline caching generates type-case sequences on-the-fly to speed up messages sent from the same call site to several different types of object. More significantly, they collect concrete type information for the optimizing compiler. With better performance yet good interactive behavior, these techniques make exploratory programming possible both for pure object-oriented languages and for application domains requiring higher ultimate performance, reconciling exploratory programming, ubiquitous abstraction, and high performance.

Linear logic has been proposed as one solution to the problem of garbage collection and providing efficient "updatein -place" capabilities within a more functional language. Linear logic conserves accessibility, and hence provides a mechanical metaphor which is more appropriate for a distributed-memory parallel processor in which copying is explicit. However, linear logic's lack of sharing may introduce significant inefficiencies of its own. We show an efficient implementation of linear logic called Linear Lisp that runs within a constant factor of non-linear logic. This Linear Lisp allows RPLACX operations, and manages storage as safely as a non-linear Lisp, but does not need a garbage collector. Since it offers assignments but no sharing, it occupies a twilight zone between functional languages and imperative languages. Our Linear Lisp Machine offers many of the same capabilities as combinator/graph reduction machines, but without their copying and garbage collection problems. Intr...

this paper. associated with a relocating collector is saved; other costs remain, however, such as the costs of updating all pointers and foregoing some compiler optimizations.

Higher order functional programs constantly allocate objects dynamically. These objects are typically cons cells, closures, and records and are generally allocated in the heap and reclaimed later by some garbage collection process. This paper describes a compile time analysis, called escape analysis, for determining the lifetime of dynamically created objects in higher order functional programs, and describes optimizations that can be performed, based on the analysis, to improve storage allocation and reclamation of such objects. In particular, our analysis can be applied to programs manipulating lists, in which case optimizations can be performed to allow whole cons cells in spines of lists to be either reclaimed at once or reused without incurring any garbage collection overhead. In a previous paper on escape analysis [10], we had left open the problem of performing escape analysis on lists. Escape analysis simply determines when the argument (or some part of the argument) to a func...

We describe a new method for determining when an object can be garbage collected. The method does not require marking live objects. Instead, each object X is dynamically associated with a stack frame , such that is collectable when pops. Because could havebeendead earlier, our method is conservative. Our results demonstrate that the method nonetheless identifies a large percentage of collectable objects. The method has been implemented in Sun's Java tm Virtual Machine interpreter, and results are presented based on this implementation.

Lazy allocation is a model for allocating objects on the execution stack of a high-level language which does not create dangling references. Our model provides safe transportation into the heap for objects that may survive the deallocation of the surrounding stack frame. Space for objects that do not survive the deallocation of the surrounding stack frame is reclaimed without additional effort when the stack is popped. Lazy allocation thus performs a first-level garbage collection, and if the language supports garbage collection of the heap, then our model can reduce the amortized cost of allocation in such a heap by filtering out the short-lived objects that can be more efficiently managed in LIFO order. A run-time mechanism called result expectation further filters out unneeded results from functions called only for their effects. In a shared-memory multi-processor environment, this filtering reduces contention for the allocation and management of global memory. Our model performs s...

The need to reverse a computation arises in many contexts---debugging, editor undoing, optimistic concurrency undoing, speculative computation undoing, trace scheduling, exception handling undoing, database recovery, optimistic discrete event simulations, subjunctive computing, etc. The need to analyze a reversed computation arises in the context of static analysis---liveness analysis, strictness analysis, type inference, etc. Traditional means for restoring a computation to a previous state involve checkpoints; checkpoints require time to copy, as well as space to store, the copied material. Traditional reverse abstract interpretation produces relatively poor information due to its inability to guess the previous values of assigned-to variables. We propose an abstract computer model and a programming language---Y-Lisp---whose primitive operations are injective and hence reversible, thus allowing arbitrary undoing without the overheads of checkpointing. Such a computer can be built from reversible conservative logic circuits, with the serendipitous advantage of dissipating far less heat than traditional Boolean AND/OR/NOT circuits. Unlike functional languages, which have one "state " for all times, Y-Lisp has at all times one "state", with unique predecessor and successor states. Compiling into a reversible pseudocode can have benefits even when targeting a traditional computer. Certain optimizations, e.g., update-in-place, and compile-time garbage collection may be more easily performed, because the

One of the major overheads in implementing functional languages is the storage management overhead due to dynamic allocation and automatic reclamation of indefinite-extent storage. This dissertation investigates the problems of statically inferring lifetime information about dynamically-allocated objects in higher-order polymorphic functional languages, both strict and non-strict, and of applying that information to reduce the storage management overhead. We have developed a set of compile-time semantic analyses for a higher-order, monomorphic, strict functional language based on denotational semantics and abstract interpretation. They are 1) escape analysis, which provides information about the relative lifetimes of objects such as arguments and local objects defin...

SELF's debugging system provides complete source-level debugging (expected behavior) with globally optimized code. It shields the debugger from optimizations performed by the compiler by dynamically deoptimizing code on demand. Deoptimization only affects the procedure activations that are actively being debugged; all other code runs at full speed. Deoptimization requires the compiler to supply debugging information at discrete interrupt points; the compiler can still perform extensive optimizations between interrupt points without affecting debuggability. At the same time, the inability to interrupt between interrupt points is invisible to the user. Our debugging system also handles programming changes during debugging. Again, the system provides expected behavior: it is possible to change a running program and immediately observe the effects of the change. Dynamic deoptimization transforms old compiled code (which may contain inlined copies of the old version of the changed procedure) into new versions reflecting the current source-level state. To the best of our knowledge, SELF is the first practical system providing full expected behavior with globally optimized code.