<div dir="ltr"><div dir="ltr"><div class="gmail_default" style="font-size:small">There are many reasons that may lead to the variance of the execution time of a software component (such as the overhead value you measured). </div><div class="gmail_default" style="font-size:small">One reason is that different runs simply execute different instructions because the input is slightly different. Another reason is that all types of interference caused by other workloads (e.g., interrupts) in the systems can delay the execution of the software component you want to measure. If you search for research related to "Worst-Case Execution Time" in real-time systems, you will get more sophisticated answers.</div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">Without knowing what you have done to figure out the question and what you want to achieve, it is hard to provide insightful answers. </div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">BTW, Dr. Brandenburg discusses the design, implementation, and theory of LITMUS^RT in detail in his awesome dissertation (<a href="https://www.cs.unc.edu/~bbb/diss/brandenburg-diss.pdf">https://www.cs.unc.edu/~bbb/diss/brandenburg-diss.pdf</a>). </div><div class="gmail_default" style="font-size:small">It is a very helpful and insightful material if you are working on LITMUS^RT. </div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">Meng</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Jan 6, 2020 at 12:50 AM sasna ms <<a href="mailto:sasnams.mec@gmail.com">sasnams.mec@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>Hi,</div><div> I am trying to plot graph of release overhead v/s release overhead of GSN-EDF scheduler . I had created real time tasks using the command " ./mktasks.py -m 8 -t 4 -u 0.5 " and the resulting overhead was traced using feature-trace tool. The result was</div><div> SCHED2 0</div><div> RELEASE 7546</div><div> CXS 0</div><div> SCHED 0</div><div> RELEASE LATENCY 23996</div><div> SEND-RESCHED 62376</div><div>Each time when i run this i am getting a different value. What is the reason for that? Should that value remain constant for a particular scheduler?<br></div></div>
_______________________________________________<br>
litmus-dev mailing list<br>
<a href="mailto:litmus-dev@lists.litmus-rt.org" target="_blank">litmus-dev@lists.litmus-rt.org</a><br>
<a href="https://lists.litmus-rt.org/listinfo/litmus-dev" rel="noreferrer" target="_blank">https://lists.litmus-rt.org/listinfo/litmus-dev</a><br>
</blockquote></div></div>