Sciweavers

110 search results - page 6 / 22
» Exploiting Vector Parallelism in Software Pipelined Loops
Sort
View
IEEEPACT
2007
IEEE
14 years 1 months ago
Speculative Decoupled Software Pipelining
In recent years, microprocessor manufacturers have shifted their focus from single-core to multi-core processors. To avoid burdening programmers with the responsibility of paralle...
Neil Vachharajani, Ram Rangan, Easwaran Raman, Mat...
HPCA
2004
IEEE
14 years 8 months ago
Creating Converged Trace Schedules Using String Matching
This paper focuses on generating efficient software pipelined schedules for in-order machines, which we call Converged Trace Schedules. For a candidate loop, we form a string of t...
Satish Narayanasamy, Yuanfang Hu, Suleyman Sair, B...
CGO
2008
IEEE
14 years 2 months ago
Compiling for vector-thread architectures
Vector-thread (VT) architectures exploit multiple forms of parallelism simultaneously. This paper describes a compiler for the Scale VT architecture, which takes advantage of the ...
Mark Hampton, Krste Asanovic
IPPS
1998
IEEE
13 years 11 months ago
Register-Sensitive Software Pipelining
In this paper, we propose an integrated approach for register-sensitive software pipelining. In this approach, the heuristics proposed in the stage scheduling method of Eichenberg...
Amod K. Dani, V. Janaki Ramanan, Ramaswamy Govinda...
CODES
2005
IEEE
14 years 1 months ago
Improving superword level parallelism support in modern compilers
Multimedia vector instruction sets are becoming ubiquitous in most of the embedded systems used for multimedia, networking and communications. However, current compiler technology...
Christian Tenllado, Luis Piñuel, Manuel Pri...