Transactional memory (TM) simplifies parallel programming by guaranteeing that transactions appear to execute atomically and in isolation. Implementing these properties includes providing data version management for the simultaneous storage of both new (visible if the transaction commits) and old (retained if the transaction aborts) values. Most (hardware) TM systems leave old values “in place” (the target memory address) and buffer new values elsewhere until commit. This makes aborts fast, but penalizes (the much more frequent) commits. In this paper, we present a new implementation of transactional memory, Log-based Transactional Memory (LogTM), that makes commits fast by storing old values to a per-thread log in cacheable virtual memory and storing new values in place. LogTM makes two additional contributions. First, LogTM extends a MOESI directory protocol to enable both fast conflict detection on evicted blocks and fast commit (using lazy cleanup). Second, LogTM handles aborts in (library) software with little performance penalty. Evaluations running micro- and SPLASH-2 benchmarks on a 32way multiprocessor support our decision to optimize for commit by showing that only 1-2 % of transactions abort. 1.

This paper proposes a hardware transactional memory (HTM) system called LogTM Signature Edition (LogTM-SE). LogTM-SE uses signatures to summarize a transaction’s readand write-sets and detects conflicts on coherence requests (eager conflict detection). Transactions update memory “in place ” after saving the old value in a per-thread memory log (eager version management). Finally, a transaction commits locally by clearing its signature, resetting the log pointer, etc., while aborts must undo the log. LogTM-SE achieves two key benefits. First, signatures and logs can be implemented without changes to highly-optimized cache arrays because LogTM-SE never moves cached data, changes a block’s cache state, or flash clears bits in the cache. Second, transactions are more easily virtualized because signatures and logs are software accessible, allowing the operating system and runtime to save and restore this state. In particular, LogTM-SE allows cache victimization, unbounded nesting (both open and closed), thread context switching and migration, and paging. 1

Early implementations of software transactional memory (STM) assumed that sharable data would be accessed only within transactions. Memory may appear inconsistent in programs that violate this assumption, even when program logic would seem to make extra-transactional accesses safe. Designing STM systems that avoid such inconsistency has been dubbed the privatization problem. We argue that privatization comprises a pair of symmetric subproblems: private operations may fail to see updates made by transactions that have committed but not yet completed; conversely, transactions that are doomed but have not yet aborted may see updates made by private code, causing them to perform erroneous, externally visible operations. We explain how these problems arise in different styles of STM, present strategies to address them, and discuss their implementation tradeoffs. We also propose a taxonomy of contracts between the system and the user, analogous to programmer-centric memory consistency models, which allow us to classify programs based on their privatization requirements. Finally, we present empirical comparisons of several privatization strategies. Our results suggest that the best strategy may depend on application characteristics.

Software transactions have received significant attention as a way to simplify shared-memory concurrent programming, but insufficient focus has been given to the precise meaning of software transactions or their interaction with other language features. This work begins to rectify that situation by presenting a family of formal languages that model a wide variety of behaviors for software transactions. These languages abstract away implementation details of transactional memory, providing high-level definitions suitable for programming languages. We use small-step semantics in order to represent explicitly the interleaved execution of threads that is necessary to investigate pertinent issues. We demonstrate the value of our core approach to modeling transactions by investigating two issues in depth. First, we consider parallel nesting, in which parallelism and transactions can nest arbitrarily. Second, we present multiple models for weak isolation, in which nontransactional code can violate the isolation of a transaction. For both, type-and-effect systems let us soundly and statically restrict what computation can occur inside or outside a transaction. We prove some key language-equivalence theorems to confirm that under sufficient static restrictions, in particular that each mutable memory location is used outside transactions or inside transactions (but not both), no program can determine whether the language implementation uses weak isolation or strong isolation.

Transactional memory provides a new concurrency control mechanism that avoids many of the pitfalls of lock-based synchronization. High-performance software transactional memory (STM) implementations thus far provide weak atomicity: Accessing shared data both inside and outside a transaction can result in unexpected, implementation-dependent behavior. To guarantee isolation and consistent ordering in such a system, programmers are expected to enclose all shared-memory accesses inside transactions. A system that provides strong atomicity guarantees isolation even in the presence of threads that access shared data outside transactions. A strongly-atomic system also orders transactions with conflicting non-transactional memory operations in a consistent manner. In this paper, we discuss some surprising pitfalls of weak atomicity, and we present an STM system that avoids these problems

Hardware transactional memory has great potential to simplify the creation of correct and efficient multithreaded programs, allowing programmers to exploit more effectively the soon-to-be-ubiquitous multi-core designs. Several recent proposals have extended the original bounded transactional memory to unbounded transactional memory, a crucial step toward transactions becoming a generalpurpose primitive. Unfortunately, supporting the concurrent execution of an unbounded number of unbounded transactions is challenging, and as a result, many proposed implementations are complex. This paper explores a different approach. First, we introduce the permissions-only cache to extend the bound at which transactions overflow to allow the fast, bounded case to be used as frequently as possible. Second, we propose ONETM to simplify the implementation of unbounded transactional memory by bounding the concurrency

pa•thol•o•gy any deviation from a healthy, normal, or efficient condition. Hardware Transactional Memory (HTM) systems reflect choices from three key design dimensions: conflict detection, version management, and conflict resolution. Previously proposed HTMs represent three points in this design space: lazy conflict detection, lazy version management, committer wins (LL); eager conflict detection, lazy version management, requester wins (EL); and eager conflict detection, eager version management, and requester stalls with conservative deadlock avoidance (EE). To isolate the effects of these high-level design decisions, we develop a common framework that abstracts away differences in cache write policies, interconnects, and ISA to compare these three design points. Not surprisingly, the relative performance of these systems depends on the workload. Under light transactional loads they perform similarly, but under heavy loads they differ by up to 80%. None of the systems performs best on all of our benchmarks. We identify seven performance pathologies—interactions between workload and system that degrade performance—as the root cause of many performance differences: FRIENDLYFIRE, STARVINGWRITER, SERIALIZEDCOMMIT, FUTILESTALL, STARVIN-GELDER, RESTARTCONVOY, and DUELINGUPGRADES. We discuss when and on which systems these pathologies can occur and show that they actually manifest within TM workloads. The insight provided by these pathologies motivated four enhanced systems that often significantly reduce transactional memory overhead. Importantly, by avoiding transaction pathologies, each enhanced system performs well across our suite of benchmarks. Categories and Subject Descriptors C.4 [Performance of Systems]: performance attributes, design

TxLinux is a variant of Linux that is the first operating system to use hardware transactional memory (HTM) as a synchronization primitive, and the first to manage HTM in the scheduler. This paper describes and measures TxLinux and discusses two innovations in detail: cooperation between locks and transactions, and the integration of transactions with the OS scheduler. Mixing locks and transactions requires a new primitive, cooperative transactional spinlocks (cxspinlocks) that allow locks and transactions to protect the same data while maintaining the advantages of both synchronization primitives. Cxspinlocks allow the system to attempt execution of critical regions with transactions and automatically roll back to use locking if the region performs I/O. Integrating the scheduler with HTM eliminates priority inversion. On a series of real-world benchmarks TxLinux has similar performance to Linux, exposing concurrency with as many as 32 concurrent threads on 32 CPUs in the same critical region.

There has been considerable recent interest in both hardware and software transactional memory (TM). We present an intermediate approach, in which hardware serves to accelerate a TM implementation controlled fundamentally by software. Specifically, we describe an alert on update mechanism (AOU) that allows a thread to receive fast, asynchronous notification when previously-identified lines are written by other threads, and a programmable data isolation mechanism (PDI) that allows a thread to hide its speculative writes from other threads, ignoring conflicts, until software decides to make them visible. These mechanisms reduce bookkeeping, validation, and copying overheads without constraining software policy on a host of design decisions. We have used AOU and PDI to implement a hardwareaccelerated software transactional memory system we call RTM.

Transactional memory proposes an alternative synchronization primitive to traditional locks. Its promise is to simplify the software development of multi-threaded applications while at the same time delivering the performance of parallel applications using (complex and error prone) fine grain locking. This study reports our experience implementing a realistic application using transactional memory (TM). The application is Lee’s routing algorithm and was selected for its abundance of parallelism but difficulty of expressing it with locks. Each route between a source and a destination point in a grid can be considered a unit of parallelism. Starting from this simple approach, we evaluate the exploitable parallelism of a transactional parallel implementation and explore how it can be adapted to deliver better performance. The adaptations do not introduce locks nor alter the essence of the implemented algorithm, but deliver up to 20 times more parallelism. The adaptations are derived from understanding the application itself and TM. The evaluation simulates an abstracted TM system and, thus, the results are independent of specific software or hardware TM implemented, and describe properties of the application. 1.