<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;"><br><div><div>On Oct 7, 2014, at 1:21 PM, Mikyung Kang <<a href="mailto:mkkang01@gmail.com">mkkang01@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">r="ltr">Hello all,<div><br></div><div style="">I'm trying to check schedulability for simple HRT/SRT task sets using Litmus-rt 2014.2.</div><div style="">* 8 Cores, no hyper-threading, 16G memory</div><div style="">* Bare-metal testing, Ubuntu 12.04 (Linux 3.10.41)</div><div style="">* rt_launch program, 20-30 seconds</div><div style=""><br></div><div style="">On top of clean OS and litmus-rt, only rt_launch is running and then st-* trace files are recorded.</div><div style="">I expected that all cases should be schedulable and c-edf result is better than g-edf result. </div><div style="">(At least, if Utilization <= 70% then c-edf schedulability >= 0.8)</div><div style=""><br></div><div style="">But, as shown in the first following table and the second graph (bare-metal litmus-rt), </div><div style="">even w/ 45% CPUs utilization (6*(120,200)=3.6), the schedulability is dropped so much and c-edf result isn't better than g-edf case.<br></div><div style="">The schedulability for c-edf and g-edf (6th and 7th column) is an averaged value (20 times).</div><div style=""><br></div><div style="">Based on other related papers, I expected that the schedulability (c-edf) will be larger than 0.7~0.8 (at least) up to 74.4% utilization (5.95).</div><div style="">But, HRT and SRT, both cases produced around 0.5 schedulability. </div><div style=""><br></div><div style="">Even though I tried 1*(180,200) ~ 8*(180,200) as shown in the third graph, it's little bit better but still not good and fluctuation is happened.</div><div style="">And, even though the utillization is similar, the schedulability is very different if the number of task sets is different.</div><div style=""><br></div><div style="">Is this correct? How can I increase the schedulability? How can I see the difference between c-edf and g-edf?</div><div style="">Is there anything that I should check for schedulability? Any tips to get best schedulability?</div><div><br></div><div>(1)</div><div>
<table border="0" cellpadding="0" cellspacing="0" width="467" style="border-collapse:collapse;width:467pt">
<colgroup><col width="65" span="4" style="width:65pt">
<col width="77" style="width:77pt">
<col width="65" span="2" style="width:65pt">
</colgroup><tbody><tr height="15" style="height:15pt">
<td height="15" class="" width="65" style="height:15pt;width:65pt">num-tasks</td>
<td class="" width="65" style="width:65pt">period</td>
<td class="" width="65" style="width:65pt">wcet</td>
<td class="" width="65" style="width:65pt">utilization</td>
<td class="" width="77" style="width:77pt">ratio</td>
<td class="" width="65" style="width:65pt">bm-c-edf</td>
<td class="" width="65" style="width:65pt">bm-g-edf</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">4.00</td>
<td class="">200.00</td>
<td class="">60.00</td>
<td class="">1.20</td>
<td class="">15.0%</td>
<td class="">0.99</td>
<td class="">1.00</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">4.00</td>
<td class="">200.00</td>
<td class="">100.00</td>
<td class="">2.00</td>
<td class="">25.0%</td>
<td class="">1.00</td>
<td class="">0.99</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">5.00</td>
<td class="">200.00</td>
<td class="">110.00</td>
<td class="">2.75</td>
<td class="">34.4%</td>
<td class="">0.87</td>
<td class="">0.92</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">6.00</td>
<td class="">200.00</td>
<td class="">120.00</td>
<td class="">3.60</td>
<td class="">45.0%</td>
<td class="">0.64</td>
<td class="">0.79</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">6.00</td>
<td class="">200.00</td>
<td class="">145.00</td>
<td class="">4.35</td>
<td class="">54.4%</td>
<td class="">0.70</td>
<td class="">0.70</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">7.00</td>
<td class="">200.00</td>
<td class="">150.00</td>
<td class="">5.25</td>
<td class="">65.6%</td>
<td class="">0.49</td>
<td class="">0.51</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">7.00</td>
<td class="">200.00</td>
<td class="">170.00</td>
<td class="">5.95</td>
<td class="">74.4%</td>
<td class="">0.50</td>
<td class="">0.59</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">7.00</td>
<td class="">200.00</td>
<td class="">180.00</td>
<td class="">6.30</td>
<td class="">78.8%</td>
<td class="">0.57</td>
<td class="">0.50</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">7.00</td>
<td class="">200.00</td>
<td class="">190.00</td>
<td class="">6.65</td>
<td class="">83.1%</td>
<td class="">0.49</td>
<td class="">0.51</td>
</tr>
<tr height="15" style="height:15pt">
<td height="15" class="" style="height:15pt">8.00</td>
<td class="">200.00</td>
<td class="">180.00</td>
<td class="">7.20</td>
<td class="">90.0%</td>
<td class="">0.41</td>
<td class="">0.41</td>
</tr>
</tbody></table><br></div><div>(2)</div><div><span><bm.png></span><br></div><div><br></div><div>(3) 1*(180,200) ~ 8*(180,200)</div><div><span><bm2.png></span><br><br></div><div><br></div><div style="">Thanks,</div><div style="">Mikyung</div></blockquote><br></div><div><br></div><div>Hi Mikyung,</div><div><br></div><div>What is your criteria for determining schedulability? That is, what makes a task set unschedulable? One deadline miss (HRT)? A deadline miss of more than X time units (SRT)? From your email, I get the feeling that you’re observing runtime behavior to determine schedulability, rather than theoretical analysis-based schedulability. Is this correct? If this is the case, what is the collective utilization of your tasks as reported by the *nix utility ‘top’? Can you copy/paste the line that starts with “%Cpu(s)”? The “%us” should be relatively close to what you provision with rt_launch.</div><div><br></div><div>Did you manage to resolve your problems you reported on last September? The issues you reported made it sound like your system CPUs were experiencing much heavier load than you had expected (resulting in incomplete traces).</div><div><br></div><div>-Glenn</div><div><br></div><br></body></html>