[LITMUS^RT] st-job-stats
Martinez Garcia Jorge Luis (PS-EC/ESB2)
JorgeLuis.MartinezGarcia at de.bosch.com
Thu Sep 20 16:19:52 CEST 2018
Hi Björn,
I rerun the experiment for two seconds and I get the following statistics:
Task, Job, Period, Response, DL Miss?, Lateness, Tardiness, Forced?, Exec. Time, Preemptions, Migrations, Release time, Completion time
# task NAME=rtspin PID=1985 COST=200000 PERIOD=1000000 CPU=101
1985, 1, 1000000, 522095, 0, -477905, 0, 0, 425312, 0, 0, 0, 522095
1985, 2, 1000000, 274904, 0, -725096, 0, 0, 214948, 0, 0, 1000000, 1274904
1985, 3, 1000000, 255836, 0, -744164, 0, 0, 224115, 0, 0, 2000000, 2255836
1985, 4, 1000000, 243852, 0, -756148, 0, 0, 227500, 0, 0, 3000000, 3243852
1985, 5, 1000000, 245097, 0, -754903, 0, 0, 210729, 0, 0, 4000000, 4245097
1985, 6, 1000000, 249363, 0, -750637, 0, 0, 212187, 0, 0, 5000000, 5249363
1985, 7, 1000000, 270451, 0, -729549, 0, 0, 220937, 0, 0, 6000000, 6270451
1985, 8, 1000000, 233884, 0, -766116, 0, 0, 221667, 0, 0, 7000000, 7233884
1985, 9, 1000000, 254191, 0, -745809, 0, 0, 216666, 0, 0, 8000000, 8254191
1985, 10, 1000000, 236790, 0, -763210, 0, 0, 234844, 0, 0, 9000000, 9236790
1985, 11, 1000000, 292254, 0, -707746, 0, 0, 222709, 0, 0, 10, 10292254
1985, 12, 1000000, 246572, 0, -753428, 0, 0, 221406, 0, 0, 11000000, 11246572
1985, 13, 1000000, 255889, 0, -744111, 0, 0, 217708, 0, 0, 12000000, 12255889
1985, 14, 1000000, 246093, 0, -753907, 0, 0, 229427, 0, 0, 13000000, 13246093
1985, 15, 1000000, 266817, 0, -733183, 0, 0, 218073, 0, 0, 14000000, 14266817
1985, 16, 1000000, 244312, 0, -755688, 0, 0, 218750, 0, 0, 15000000, 15244312
1985, 17, 1000000, 249984, 0, -750016, 0, 0, 211458, 0, 0, 16000000, 16249984
1985, 18, 1000000, 233520, 0, -766480, 0, 0, 211875, 0, 0, 17000000, 17233520
1985, 19, 1000000, 297994, 0, -702006, 0, 0, 234219, 0, 0, 18000000, 18297994
1985, 20, 1000000, 282937, 0, -717063, 0, 0, 237395, 0, 0, 19000000, 19282937
1985, 21, 1000000, 332306, 0, -667694, 0, 0, 239427, 0, 0, 20, 20332306
1985, 22, 1000000, 251625, 0, -748375, 0, 0, 222291, 0, 0, 21000000, 21251625
1985, 23, 1000000, 251620, 0, -748380, 0, 0, 208021, 0, 0, 22000000, 22251620
1985, 24, 1000000, 241198, 0, -758802, 0, 0, 195364, 0, 0, 23000000, 23241198
1985, 25, 1000000, 254786, 0, -745214, 0, 0, 292969, 0, 0, 24000000, 24254786
1985, 26, 1000000, 249625, 0, -750375, 0, 0, 225052, 0, 0, 25000000, 25249625
1985, 27, 1000000, 250297, 0, -749703, 0, 0, 212136, 0, 0, 26000000, 26250297
1985, 59, 1000000, 303154, 0, -696846, 0, 0, 252084, 0, 0, 58000000, 58303154
1985, 60, 1000000, 256743, 0, -743257, 0, 0, 222917, 0, 0, 59000000, 59256743
...
1985, 80, 1000000, 281589, 0, -718411, 0, 0, 243073, 0, 0, 79000000, 79281589
1985, 81, 1000000, 315073, 0, -684927, 0, 0, 237135, 0, 0, 80, 80315073
1985, 111, 1000000, 291014, 0, -708986, 0, 0, 259272, 0, 0, 110, 110291014
...
The second column shows me the job being executed. I see that for instance after job 27, job 60 is executed. After job 81, job 111 and so on. I was expecting around 2000 jobs but the statistics shows me around 1000.
The same thing happens with the other tasks.
What happens, when "rtspin" tasks are mapped onto "Unconstrained" reservations? Are they treated like real-time tasks?
Maybe the problem lies in my understanding of the "Unconstrained Server". Can you please shed light on this matter?
Best,
Jorge
> On 20. Sep 2018, at 13:28, Martinez Garcia Jorge Luis (PS-EC/ESB2) <JorgeLuis.MartinezGarcia at de.bosch.com> wrote:
>
> I am using a modern reservation-based plugin and invoking the tasks in the following fashion:
>
> resctl -n 101 -c 0 -t unconstrained-server -q 1 rtspin -w -m 0 -s 1
> -p 0 -r 101 -q 1 0.200000 1.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
>
> resctl -n 102 -c 0 -t unconstrained-server -q 2 rtspin -w -m 0 -s 1
> -p 0 -r 102 -q 2 0.150000 2.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
>
> resctl -n 103 -c 0 -t unconstrained-server -q 3 rtspin -w -m 0 -s 1
> -p 0 -r 103 -q 5 0.250000 5.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
>
> resctl -n 104 -c 0 -t unconstrained-server -q 4 rtspin -w -m 0 -s 1
> -p 0 -r 104 -q 10 2.700000 10.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
>
> resctl -n 105 -c 0 -t unconstrained-server -q 5 rtspin -w -m 0 -s 1
> -p 0 -r 105 -q 20 2.350000 20.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
>
> resctl -n 106 -c 0 -t unconstrained-server -q 6 rtspin -w -m 0 -s 1
> -p 0 -r 106 -q 50 0.600000 50.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
>
> resctl -n 107 -c 0 -t unconstrained-server -q 7 rtspin -w -m 0 -s 1
> -p 0 -r 107 -q 100 2.150000 100.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
>
> resctl -n 108 -c 0 -t unconstrained-server -q 8 rtspin -w -m 0 -s 1
> -p 0 -r 108 -q 200 0.100000 200.000000 2 & PID="$!"
> PIDS="$PIDS $PID"
> START_PIDS="$START_PIDS $PID"
Thanks for the context. One more question regarding your first email:
> when I check the results by means of “st-job-stats *.bin”, I see that
> there are many jobs of the 1 ms task that are skipped even though it
> has the highest priority in the system
All jobs seem to be present and have reasonable release times, no tardiness, and more-or-less plausible response times. What exactly do you mean by “skipped”?
- Björn
_______________________________________________
litmus-dev mailing list
litmus-dev at lists.litmus-rt.org
https://lists.litmus-rt.org/listinfo/litmus-dev
More information about the litmus-dev
mailing list