Name
Abstract | ||
|---|---|---|
GPUs are becoming first-class compute citizens and increasingly support programmability-enhancing features such as shared virtual memory and hardware cache coherence. This enables them to run a wider variety of programs. However, a key aspect of general-purpose programming where GPUs still have room for improvement is the ability to invoke system calls.
We explore how to directly invoke system calls from GPUs. We examine how system calls can be integrated with GPGPU programming models, where thousands of threads are organized in a hierarchy of execution groups. To answer questions on GPU system call usage and efficiency, we implement Genesys, a generic GPU system call interface for Linux. Numerous architectural and OS issues are considered and subtle changes to Linux are necessary, as the existing kernel assumes that only CPUs invoke system calls. We assess the performance of Genesys using micro-benchmarks and applications that exercise system calls for signals, memory management, filesystems, and networking.
|
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/ISCA.2018.00075 | ISCA |
Keywords | Field | DocType |
accelerators and domain-specific architecture, graphics oriented architecture, multicore and parallel architectures, virtualization and OS, GPUs | Programming paradigm,Computer science,Instruction set,Parallel computing,Thread (computing),Memory management,System call,General-purpose computing on graphics processing units,Hierarchy,Operating system,Cache coherence | Conference |
ISSN | ISBN | Citations |
1063-6897 | 978-1-5386-5984-7 | 3 |
PageRank | References | Authors |
0.37 | 16 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ján Veselý | 1 | 44 | 2.59 |
| Arkaprava Basu | 2 | 1145 | 40.24 |
| Abhishek Bhattacharjee | 3 | 538 | 21.22 |
| Gabriel H. Loh | 4 | 12 | 1.85 |
| Mark Oskin | 5 | 906 | 76.63 |
| Steven K. Reinhardt | 6 | 3885 | 226.69 |