While some tools exist to detect data races in GPU pro- grams, they are often prohibitively slow or focused only on a small class of data races in shared memory. These con- currency bugs can be extremely subtle and di cult to debug due to the massive numbers of threads running concurrently on a modern GPU. GPU programming requires a combination of extreme levels of parallelism and low-level programming, making it easy for concurrency bugs such as data races to arise. Therefore, we believe that inaddition to being a debugging tool, the DataRaceException may be a viable mechanism to enforce the safety of executions of multithreaded Java programs.Īs GPUs have become an integral part of nearly every pro- cessor, GPU programming has become increasingly popular. Our experiments indicate that our implementation has reasonable overhead. We have evaluated our system on a variety of publicly available Java benchmarks and a few microbenchmarks that combine lock-based and transaction-based synchronization. We have implemented our algorithm and the DataRaceException in the Kaffe Java Virtual Machine. This algorithm generalizes the Goldilocks algorithm for data-race detectionby handling transactions and providing the ability to distinguish between read and write accesses.
#Iguard in georgia free
We present a precise and efficient algorithm for dynamically verifying that an execution is free of data races. We generalize this definition to executions that use transactions in addition to locks and volatile variables for synchronization. We start with the definition of data-race-free executions in the Java memory model. Alternatively, the DataRaceException can serve as a conflict-detection mechanism inoptimistic uses of concurrency. When a DataRaceException is caught, the operation, thread, or program causing it can be terminated gracefully. This strong guarantee helps to rule out many concurrency-related possibilities as the cause of erroneous behavior.
Second, if no DataRaceException is thrown in an execution, it is guaranteed to be sequentially consistent. First, accesses causing race conditions are interruptedand handled before they cause errors that may be difficult to diagnose later. Analogous to other runtime exceptions, the DataRaceException provides two key benefits. Motivated by these observations, we have designed and implemented a Java runtime system that monitors program executions and throws a DataRaceException when a data race is about to occur.
In addition to causing data corruption and leading programs to crash, the presence of data races complicates the semantics of an execution which might no longer be sequentially consistent. Our evaluation with several microbenchmarks and benchmarks shows that PreDataR has better data race coverage compared to prior techniques at practical run-time overheads.ĭata races often result in unexpected and erroneous behavior.
PreDataR includes three optimizations and a novel vector clock compression scheme that are readily applicable to other partial order based analyses. We implement a tool called PreDataR that tracks the GWCP relation using binary instrumentation. GWCP takes into account the GPU thread hierarchy and different synchronization semantics such as barrier synchronization and scoped atomics and locks. GWCP extends a sound and precise relation called weak-causally-precedes (WCP) proposed in the context of multithreaded shared memory CPU programs to GPU kernels. This work proposes GWCP, a predictive partial order for data race detection of GPU kernels. Existing predictive partial orders cannot be na\"ively applied to analyze GPU kernels because of the differences in programming models. Predictive partial order relations for CPU programs aim to expose data races that can be hidden during a dynamic execution.
Data races pose one such concurrency correctness challenge, and therefore, effective methods of detecting as many data races as possible are required. The high degree of parallelism and relatively complicated synchronization mechanisms in GPUs make writing correct kernels difficult.
0 kommentar(er)
