Approximation of Worst-case Execution Time for Preemptive ...
Approximation of Worst-case Execution Time for Preemptive ... Approximation of Worst-case Execution Time for Preemptive ...
Results: Simple Tests 3000 2500 +8% 2000 ms 1500 1000 500 -5% +11% +6% +7% -5% +13% 0 Matrix Mult. Matrix Mult. FP Array Max Array Max FP Runge-Kutta Polynomial Eval Distribution Count WCET Approximation Matteo Corti, ETH Zurich, LCTES 2000 18
Results: Approximations • Worst case assumptions about caches and pipeline produce non usable durations • Example: no cache approximation (but all other included) Test Matr. Mul. Array Max. Pol. Eval. Measured value 280 ms 520 ms 1252 ms Full predictor 311 ms 555 ms 1188 ms No cache hits 1403 ms 1901 ms 3193 ms Matteo Corti, ETH Zurich, LCTES 2000 19
- Page 1 and 2: Approximation of Worst-case Executi
- Page 3 and 4: Environment: User Needs • Complex
- Page 5 and 6: Problem Description • Admission t
- Page 7 and 8: Other approaches • Longest path:
- Page 9 and 10: Longest Path ... instr op op op ins
- Page 11 and 12: Instruction Length • Preemption,
- Page 13 and 14: Performance Monitor • Not specifi
- Page 15 and 16: Cycles Per Instruction (CPI) …
- Page 17: Testing the Predictor • First pha
- Page 21 and 22: Results: Real Applications • Only
- Page 23 and 24: Comments … • Performance monito
- Page 25: Conclusions • The WCET can be app
Results: <strong>Approximation</strong>s<br />
• <strong>Worst</strong> <strong>case</strong> assumptions about caches and pipeline<br />
produce non usable durations<br />
• Example: no cache approximation (but all other<br />
included)<br />
Test<br />
Matr. Mul.<br />
Array Max.<br />
Pol. Eval.<br />
Measured value<br />
280 ms<br />
520 ms<br />
1252 ms<br />
Full predictor<br />
311 ms<br />
555 ms<br />
1188 ms<br />
No cache hits<br />
1403 ms<br />
1901 ms<br />
3193 ms<br />
Matteo Corti, ETH Zurich, LCTES 2000 19