M. Corti, R. Brega, and T. Gross. Approximation of WorstCase Execution Time for Preemptive Multitasking Systems. In Proc. Workshop on Languages, Compilers, and Tools for Embedded Systems. Springer, 2000.  Home/Search   Document Details and Download   Summary   Related Articles

....However, this method may be faster than other approaches since it can effectively find good input data. 3.5. Measuring Execution Times There are several possible approaches to measuring execution time, each with its own advantages and disadvantages. Here we review some of the possibilities [3, 17, 14]. Using Clock Cycle Counters: A clock cycle counter counts clock cycles on the bus or from the CPU core. The benefits are that this approach is exact, time can be measured with a fine granularity, and it is non intrusive, i.e. it does not disturb the execution of software in any way. Using ....

....for measuring the worst case execution time of programs is presented. The method relies on the actual hardware for measurements, and tackles the problem of too many paths by dividing the program into so called measurement blocks . Compared to our work this method does not produce safe values. In [3] on chip performance monitors are used to retrieve information on number of pipeline stalls, cache misses etc. An approximation of the execution time of basic blocks is derived from this. Our work does not require any performance monitors, and hence it can be applied to a wider range of hardware ....

M. Corti, R. Brega, and T. Gross. Approximation of WorstCase Execution Time for Preemptive Multitasking Systems. In Proc. Workshop on Languages, Compilers, and Tools for Embedded Systems. Springer, 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC