On Reducing Overhead in Loops
نویسنده
چکیده
In this paper we discuss several techniques for reducing the overhead in Fortran-like DO loops. In particular, we describe two topics: 1. Simplifying loop bounds. 2. Removing a ne IF-statements from loops. The objective is to isolate the part of the iteration space of the original loop in which the bounds take simpler forms, or in which the condition is satis ed. The main technical tool for determining this subspace is Fourier-Motzkin elimination. We show an easy way to generate code for scanning the di erent parts of an iteration space that have been isolated in this way. We present our techniques as a number of elementary transformation steps. SInce these steps only partition the iteration space, the steps can always legally be employed. The full e ect of these steps is obtained by iterating them until no more changes occur. The result will be a loop structure in which all bounds are simple a ne expressions and from which all IF-statements with a ne conditions have been eliminated.
منابع مشابه
Innuence of Overhead on Processor Allocation for Multiple Loops
We consider two consecutive and independent forall loops and the strategy to allocate processors for their execution. One strategy is to execute each of the two loops consecutively, each time with all the available processors. Another strategy is to execute both loops simultaneously, each with a fraction of the available processors. We verify that the presence of overhead can innuence this stra...
متن کاملReducing Logging Overhead for Deterministic Execution
Deterministic execution of parallel applications can be enforced by logging speculative memory accesses and restoring saved state in the event of a conflict. However, logging all accesses has a prohibitive overhead. We propose three techniques to reduce logging in an always-on transactional memory system. The first, Threaded Section Analysis, statically determines when only a single thread is r...
متن کاملNonscan design-for-testability techniques using RT-level design information
This paper presents non-scan design-for-testability techniques applicable to register-transfer (RT) level data path circuits. Knowledge of high-level design information, in the form of the RT-level structure, as well as the functions of the RT-level components, is utilized to develop effective non-scan DFT techniques. Instead of conventional techniques of selecting flip-flops (FF) to make contr...
متن کاملMitigation of Magnetic Field under Double-Circuit Overhead Transmission Line
This paper studies the effect of various mitigation loop configurations namely: single horizontal mitigation loop (two conductors), two separate horizontal mitigation loops (four conductors) and single vertical mitigation loop (two conductors), on the magnetic field produced by 220 kV double-circuit overhead transmission line. Also, in this paper a further reduction in the magnetic field is ach...
متن کاملTHESIS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Techniques to Reduce Thread-Level Speculation Overhead
The traditional single-core processors are being replaced by chip multiprocessors (CMPs) where several processor cores are integrated on a single chip. While this is beneficial for multithreaded applications and multiprogrammed workloads, CMPs do not provide performance improvements for single-threaded applications. Threadlevel speculation (TLS) has been proposed as a way to improve single-thre...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1996