runBashTestMultiple.sh

#!/usr/bin/env bash

# Il faut :
# Un fichier chain.in pour chaque chaine de tâche à benchmarker.
# Pour chaque fichier, lance ./MoCoAgent.out - 20min [chaine] [chaineBenchOutput]

# De là, on a :
# La répartition du temps d'execution des chaines en isolation,
# et donc le WCET des tâches de la chaine + WCET de la chaîne.

# on fixe donc ces temps dans le fichier général d'informations.
# ---------------------------------------- #

# Ensuite vient la phase de benchmarks :
calc(){ awk "BEGIN { print "$*" }"; }
commentaire="//"

Infile="input_chaine.txt"
duration=100
load=100
schedPolicy=1

while [ "$1" != "" ]; do
    case $1 in
        -i | -f | --input )  shift
                             Infile=$1
        ;;
        -l | --load )        shift
                             load=$1
        ;;
        -d | --duration )    shift
                             duration=$1
        ;;
        -s | --sched )     shift
                           schedPolicy="-s $1"
                           if test $1 = 1
                              then schedName="FIFO"
                           elif test $1 = 2
                              then schedName="RR"
                           elif test $1 = 7
                              then schedName="RM"
                           else schedName=$1
                           fi
        ;;
        -FIFO | -RR | -RM | -EDF ) schedName="${1:1}"
                                   schedPolicy=$1
        ;;
        -h | --help )        echo "usage: runBashTest.sh [[[-f file ] [-d duration] [-l load]] | [-h]]"
                             exit 0
        ;;
        * )                  echo "usage: runBashTest.sh [[[-f file ] [-d duration] [-l load]] | [-h]]"
                             exit 1
    esac
    shift
done

echo "Duration : $duration | Load : $load | Scheduling : $schedPolicy (${schedName}) | Input : $Infile | Output : $dirName"
expeResume=${duration}s_${load}_${schedName}
dirName=./Exps/`date +%d-%m-%Hh`_$expeResume
exit 0

if test -f $Infile
then
    mkdir -p $dirName
    for toExecute in {2..6} # exécuter toutes les chaines
    do
        #echo $toExecute
        while read -r line
        numLigne=1
        do
            if test $numLigne -eq $toExecute
            then
                echo $line
                name=`echo $line | awk '{print $5}'`
            else
                echo "${commentaire}$line"
            fi
            numLigne=`expr $numLigne + 1`
        done < $Infile > _inputFile.tmp.in
        #echo "Name is "$name
        sudo sar -o ${dirName}/IODatas${name}_1 -P 0-3 1 $duration > /dev/null 2>&1 &
        ./MoCoAgent.out -e 2  -d $duration -l $load $schedPolicy -i ./_inputFile.tmp.in -o ${dirName}/${name}_1 #_${expeResume}
        expe1Out=$?
        sudo rm -f ./bench/output/*

        sudo sar -o ${dirName}/IODatas${name}_0 -P 0-3 1 $duration > /dev/null 2>&1 &
        ./MoCoAgent.out -e 0 -d $duration -l $load $schedPolicy -i ./_inputFile.tmp.in -o ${dirName}/${name}_0 #_${expeResume}
        expe0Out=$?
        sudo rm -f ./bench/output/*
        rm -f ./_inputFile.tmp.in
        echo "expe1out : $expe1Out"
        echo "expe0out : $expe0Out"
    done

exit 0

    # Convertir les fichiers SAR en .csv... la totale..!
cd ${dirName}
for file in IODatas*
do
	sudo chmod 666 ${file}
    sadf -dh -- -P 0-3 ${file} > tmp1
    sadf -dh -- -P 0-3 -rB -m CPU,TEMP ${file} | cut -d\; -f1-3 --complement > tmp2
    headerBefore="# hostname;interval;timestamp; CPU0;CPU0%user;CPU0%nice;CPU0%system;CPU0%iowait;CPU0%steal;CPU0%idle; CPU1;CPU1%user;CPU1%nice;CPU1%system;CPU1%iowait;CPU1%steal;CPU1%idle; CPU2;CPU2%user;CPU2%nice;CPU2%system;CPU2%iowait;CPU2%steal;CPU2%idle; CPU3;CPU3%user;CPU3%nice;CPU3%system;CPU3%iowait;CPU3%steal;CPU3%idle; pgpgin-s;pgpgout-s;fault-s;majflt-s;pgfree-s;pgscank-s;pgscand-s;pgsteal-s; %vmeff;kbmemfree;kbavail;kbmemused;%memused;kbbuffers;kbcached;kbcommit;%commit;kbactive;kbinact;kbdirty; CPU0;CPU0MHz;CPU1;CPU1MHz;CPU2;CPU2MHz;CPU3;CPU3MHz; CPU0TEMP;CPU0DEVICE;CPU0degC;CPU0%temp; CPU1TEMP;CPU1DEVICE;CPU1degC;CPU1%temp; CPU2TEMP;CPU2DEVICE;CPU2degC;CPU2%temp; CPU3TEMP;CPU3DEVICE;CPU3degC;CPU3%temp; CPU4TEMP;CPU4DEVICE;CPU4degC;CPU4%temp; CPU5TEMP;CPU5DEVICE;CPU5degC;CPU5%temp"
    # On dégage toutes les colonnes inutiles
    linesToCut="2,4,6,9,11,13,16,18,20,23,25,27,30,35,37-40,52,54,56,58,60,61,63-65,67-69,71-73,75-83"
    newfileName=`echo $file | cut -c 8-`
    newfileName=IOD_$newfileName
    paste -d ';' tmp1 tmp2 | cut -d\; -f${linesToCut} --complement > ${newfileName}.csv
    header=`echo $headerBefore | cut -d\; -f${linesToCut} --complement`
#"# hostname;timestamp; CPU0%user;CPU0%system;CPU0%iowait;CPU0%idle; CPU1%user;CPU1%system;CPU1%iowait;CPU1%idle; CPU2%user;CPU2%system;CPU2%iowait;CPU2%idle; CPU3%user;CPU3%system;CPU3%iowait;CPU3%idle; pgpgin/s;pgpgout/s;fault/s;pgfree/s;kbmemfree;kbavail;kbmemused;%memused;kbbuffers;kbcached;kbcommit;%commit;kbactive;kbinact;kbdirty;CPU0MHz;CPU1MHz;CPU2MHz;CPU3MHz;CPU0degC;CPU1degC;CPU2degC;CPU3degC"
    # Remplace le header selon les colonnes dégagées
    sed -i "1s/.*/$header/" ${newfileName}.csv
    rm tmp1
    rm tmp2
done
    exit 0

else
    exit 1
fi