Hans-Juergen Boehm

Garbage collection in an uncooperative environment

We describe a technique for storage allocation and garbage collection in the absence of significant co‐operation from the code using the allocator. This limits garbage collection overhead to the time actually required for garbage collection. In particular, application programs that rarely or never make use of the collector no longer encounter a substantial performance penalty. This approach greatly simplifies the implementation of languages supporting garbage collection. It further allows conventional compilers to be used with a garbage collector, either as the primary means of storage reclamation, or as a debugging tool.

Foundations of the C++ concurrency memory model

Currently multi-threaded C or C++ programs combine a single-threaded programming language with a separate threads library. This is not entirely sound [7]. We describe an effort, currently nearing completion, to address these issues by explicitly providing semantics for threads in the next revision of the C++ standard. Our approach is similar to that recently followed by Java [25], in that, at least for a well-defined and interesting subset of the language, we give sequentially consistent semantics to programs that do not contain data races. Nonetheless, a number of our decisions are often surprising even to those familiar with the Java effort:We (mostly) insist on sequential consistency for race-free programs, in spite of implementation issues that came to light after the Java work. We give no semantics to programs with data races. There are no benign C++ data races. We use weaker semantics for trylock than existing languages or libraries, allowing us to promise sequential consistency with an intuitive race definition, even for programs with trylock. This paper describes the simple model we would like to be able to provide for C++ threads programmers, and explain how this, together with some practical, but often under-appreciated implementation constraints, drives us towards the above decisions.

Mostly parallel garbage collection

We present a method for adapting garbage collectors designed to run sequentially with the client, so that they may run concurrently with it. We rely on virtual memory hardware to provide information about pages that have been updated or “dirtied” during a given period of time. This method has been used to construct a mostly parallel trace-and-sweep collector that exhibits very short pause times. Performance measurements are given.

Threads cannot be implemented as a library

In many environments, multi-threaded code is written in a language that was originally designed without thread support (e.g. C), to which a library of threading primitives was subsequently added. There appears to be a general understanding that this is not the right approach. We provide specific arguments that a pure library approach, in which the compiler is designed independently of threading issues, cannot guarantee correctness of the resulting code.We first review why the approach almost works, and then examine some of the surprising behavior it may entail. We further illustrate that there are very simple cases in which a pure library-based approach seems incapable of expressing an efficient parallel algorithm.Our discussion takes place in the context of C with Pthreads, since it is commonly used, reasonably well specified, and does not attempt to ensure type-safety, which would entail even stronger constraints. The issues we raise are not specific to that context.

Combining generational and conservative garbage collection: framework and implementations

Two key ideas in garbage collection are generational collection and conservative pointer-finding. Generational collection and conservative pointer-finding are hard to use together, because generational collection is usually expressed in terms of copying objects, while conservative pointer-finding precludes copying. We present a new framework for defining garbage collectors. When applied to generational collection, it generalizes the notion of younger/older to a partial order. It can describe traditional generational and conservative techniques, and lends itself to combining different techniques in novel ways. We study in particular two new garbage collectors inspired by this framework. Both these collectors use conservative pointer-finding. The first one is based on a rewrite of an existing trace-and-sweep collector to use one level of generation. The second one has a single parameter, which controls how objects are partitioned into generations: the value of this parameter can be changed dynamically with no overhead. We have implemented both collectors and present measurements of their performance in practice.

Exact real arithmetic: a case study in higher order programming

Two methods for implementing ezact real arithmetic are explored One method is based on formulating real numbers as functions that map rational tolerances to rational approximations. This approach, which was developed by constructive mathematicians as a concrete formalization of the real numbers, has lead to a surprisingly successful implementation. The second method formulates real numbers as potentially infinite sequences of digits, evaluated on demand. This approach has frequently been advocated by proponents of lazy functional languages in the computer science community Ironically, it leads to much less satisfactory implementations. We discuss the theoretical problems involved m both methods, give algortthms for the basic arithmetic operations, and give an empirical comparison of the two techniques. We conclude wtth some general observations about the lazy evaluation paradigm and its implementation.

Conflict exceptions: simplifying concurrent language semantics with precise hardware exceptions for data-races

We argue in this paper that concurrency errors should be treated as exceptions, i.e., have fail-stop behavior and precise semantics. We propose an exception model based on conflict of synchronization free regions, which precisely detects a broad class of data-races. We show that our exceptions provide enough guarantees to simplify high-level programming language semantics and debugging, but are significantly cheaper to enforce than traditional data-race detection. To make the performance cost of enforcement negligible, we propose architecture support for accurately detecting and precisely delivering these exceptions. We evaluate the suitability of our model as well as the behavior of our architectural mechanisms using the PARSEC benchmark suite and commercial applications. Our results show that the exception model largely reflects how programmers are already writing code and that the main memory, traffic and performance overheads of the enforcement mechanisms we propose are very low.

Atlas: leveraging locks for non-volatile memory consistency

Non-volatile main memory, such as memristors or phase change memory, can revolutionize the way programs persist data. In-memory objects can themselves be persistent without the need for a separate persistent data storage format. However, the challenge is to ensure that such data remains consistent if a failure occurs during execution. In this paper, we present our system, called Atlas, which adds durability semantics to lock-based code, typically allowing us to automatically maintain a globally consistent state even in the presence of failures. We identify failure-atomic sections of code based on existing critical sections and describe a log-based implementation that can be used to recover a consistent state after a failure. We discuss several subtle semantic issues and implementation tradeoffs. We confirm the ability to rapidly flush CPU caches as a core implementation bottleneck and suggest partial solutions. Experimental results confirm the practicality of our approach and provide insight into the overheads of such a system.

IFRit: interference-free regions for dynamic data-race detection

We propose a new algorithm for dynamic data-race detection. Our algorithm reports no false positives and runs on arbitrary C and C++ code. Unlike previous algorithms, we do not have to instrument every memory access or track a full happens-before relation. Our data-race detector, which we call IFRit, is based on a run-time abstraction called an interference-free region (IFR). An IFR is an interval of one threads execution during which any write to a specific variable by a different thread is a data race. We insert instrumentation at compile time to monitor active IFRs at run-time. If the runtime observes overlapping IFRs for conflicting accesses to the same variable in two different threads, it reports a race. The static analysis aggregates information for multiple accesses to the same variable, avoiding the expense of having to instrument every memory access in the program. We directly compare IFRit to FastTrack and ThreadSanitizer, two state-of-the-art fully-precise data-race detectors. We show that IFRit imposes a fraction of the overhead of these detectors. We show that for the PARSEC benchmarks, and several real-world applications, IFRit finds many of the races detected by a fully-precise detector. We also demonstrate that sampling can further reduce IFRits performance overhead without completely forfeiting precision.

Reducing garbage collector cache misses

Cache misses are currently a major factor in the cost of garbage collection, and we expect them to dominate in the future. Traditional garbage collection algorithms exhibit relatively little temporal locality; each live object in the heap is likely to be touched exactly once during each garbage collection. We measure two techniques for dealing with this issue: prefetch-on-grey, and lazy sweeping. The first of these is new in this context. Lazy sweeping has been in common use for a decade. It was introduced as a mechanism for reducing paging and pause times; we argue that it is also crucial for eliminating cache misses during the sweep phase. Our measurements are obtained in the context of a non-moving garbage collector. Fully copying garbage collection inherently requires more traffic through the cache, and thus probably also stands to benefit substantially from something like the prefetch-on-grey technique. Generational garbage collection may reduce the benefit of these techniques for some applications, but experiments with a non-moving generational collector suggest that they remain quite useful.