A Method to Control Compensation Code during Global Scheduling
نویسنده
چکیده
| Code motion is a technique used to improve global scheduling either during the com-paction of acyclic code or the software-pipelining of loops containing conditional constructs. It can be applied in the context of high-level synthesis of ASICs or in the code generation for ASIPs. Despite its impact on the exploitation of parallelism across basic-block boundaries, code motion might have side eeects which must be compensated with extra code. We show how code compensation can be controlled in order not to increase the schedule length of the longest path. Traditional code compensation mechanisms deal with the duplication of a given operation into different paths as a consequence of non-speculative code motion. We extend the notion of code compensation to provide support for speculative code motion. Classical notions like control equivalence, domination and postdomination are generally used to keep track of code compensation. We present a new method to cast those notions into boolean form such that a BDD package can be used to implement them eeciently.
منابع مشابه
A method to control compensation codeduring global schedulingLuiz
| Code motion is a technique used to improve global scheduling either during the com-paction of acyclic code or the software-pipelining of loops containing conditional constructs. It can be applied in the context of high-level synthesis of ASICs or in the code generation for ASIPs. Despite its impact on the exploitation of parallelism across basic-block boundaries, code motion might have side e...
متن کاملComparing Tail Duplication with CompensationCode in Single Path Global
Global instruction scheduling allows operations to move across basic block boundaries to create tighter schedules. When operations move above control ow joins, some code duplication is generally necessary to preserve semantics. Tail duplication and compensation code are approaches to duplicating the necessary code, used by Superblock Scheduling and Trace Scheduling respectively. Compensation co...
متن کاملCompensation of Doppler Effect in Direct Acquisition of Global Positioning System using Segmented Zero Padding
Because of the very high chip rate of global positioning system (GPS), P-code acquisition at GPS receiver will be challenging. A variety of methods for increasing the probability of detection and reducing the average time of acquisition have been provided, among which the method of Zero Padding (ZP) is the most essential and the most widely used. The method using the Fast Fourier Transform (FFT...
متن کاملWavefront Scheduling : Path Based Data Representation andScheduling
The IA-64 architecture is rich with features that enable aggressive exploitation of instruction-level parallelism. Features such as speculation, predication, multiway branches and others provide compilers with new opportunities for the extraction of parallelism in programs. Code scheduling is a central component in any compiler for the IA-64 architecture. This paper describes the implementation...
متن کاملExploring the Performance Potential of Itanium® Processors with ILP-based Scheduling
HP and Intel’s Itanium Processor Family (IPF) is considered as one of the most challenging processor architectures to generate code for. During global instruction scheduling, the compiler must balance the use of strongly interdependent techniques like code motion, speculation and predication. A too conservative application of these features can lead to empty execution slots, contrary to the EPI...
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 1997