diff --git a/common/bmk-driver.sh b/common/bmk-driver.sh
index a737968b9bb7eafda0e62b72535ff97346971b22..b1cafe7614d35547489bd921dc18643f984b7d2d 100755
--- a/common/bmk-driver.sh
+++ b/common/bmk-driver.sh
@@ -86,7 +86,7 @@ function check_mandatory_functions(){
 # Check that mandatory variables have been defined (default values)
 function check_mandatory_variables(){
   # Variables NCOPIES, NTHREADS, NEVENTS_THREAD have default values specific to each benchmark
-  for var in NCOPIES NTHREADS NEVENTS_THREAD; do
+  for var in NCOPIES NTHREADS NEVENTS_THREAD; do # NB: EXTRA_ARGS is not a mandatory variable (BMK-1014)
     if [ "${!var}" == "" ]; then
       echo "[$bmkDriver] ERROR! A default value of $var must be set in $bmkScript" # internal error (missing code)
       exit 1;
@@ -95,10 +95,11 @@ function check_mandatory_variables(){
   echo
 }
 
-# Variables USER_NCOPIES, USER_NTHREADS, USER_NEVENTS_THREAD are empty by default
+# Variables USER_NCOPIES, USER_NTHREADS, USER_NEVENTS_THREAD, USER_EXTRA_ARGS are empty by default
 USER_NCOPIES=
 USER_NTHREADS=
 USER_NEVENTS_THREADS=
+USER_EXTRA_ARGS=
 
 # Variable resultsDir has default value /results
 # Variables skipSubDir and DEBUG are 0 by default
@@ -108,29 +109,31 @@ DEBUG=0
 MOP="custom"
 
 function advertise_user_defined_variables(){
-  for var in NCOPIES NTHREADS NEVENTS_THREAD; do
-    echo "Default (from $bmkScript): $var=${!var}"
+  for var in NCOPIES NTHREADS NEVENTS_THREAD EXTRA_ARGS; do
+    echo "Default (from $bmkScript): $var='${!var}'"
   done
   echo
-  for var in USER_NCOPIES USER_NTHREADS USER_NEVENTS_THREAD; do
-    echo "Default (from $bmkDriver): $var=${!var}"
+  for var in USER_NCOPIES USER_NTHREADS USER_NEVENTS_THREAD USER_EXTRA_ARGS; do
+    echo "Default (from $bmkDriver): $var='${!var}'"
   done
   echo
   for var in resultsDir skipSubDir DEBUG MOP; do
-    echo "Default (from $bmkDriver): $var=${!var}"
+    echo "Default (from $bmkDriver): $var='${!var}'"
   done
 }
 
 # Usage function
 function usage(){
   echo ""
-  echo "Usage: $0 [-w | --resultsdir <resultsDir>] [-W] [-c | --copies <NCOPIES>] [-t | --threads <NTHREADS>]" \
-                 "[-e | --events <NEVENTS_PER_THREAD>] [-m | --mop <mode>] [-d | --debug] [-h | --help]"
+  echo "Usage: $0 [-w | --resultsdir <resultsDir>] [-W] [-c | --copies <NCOPIES>] [-t | --threads <NTHREADS>] "\
+                 "[-e | --events <NEVENTS_PER_THREAD>] [-a | --args <EXTRA_ARGS>] "\
+                 "[-m | --mop <mode>] [-d | --debug] [-h | --help]"
   echo "  -w --resultsdir <resultsDir> : (string) results directory (default: /results , current: $resultsDir)"
   echo "  -W                           : (bool) store results in <resultsDir> directly (default: 0 , current: $skipSubDir)"
   echo "  -c --copies <NCOPIES>        : (int) # identical copies (default $NCOPIES)"
-  echo "  -t --threads <NTHREADS>      : (ubt# threads (or processes, or threads*processes) per copy (default $NTHREADS)"
+  echo "  -t --threads <NTHREADS>      : (int) # threads (or processes, or threads*processes) per copy (default $NTHREADS)"
   echo "  -e --events <NEVENTS_THREAD> : # events per thread (default $NEVENTS_THREAD)"
+  echo "  -a --args '<EXTRA_ARGS>'     : optional workload-specific command line arguments '<arg1> <arg2>...' (default '$EXTRA_ARGS')"
   echo "  -m --mop <mode>              : clean working directory mode: none/all/custom (current: $MOP)"
   echo "  -d --debug                   : debug mode (current: $DEBUG)"
   echo "  -h --help                    : display this help and exit"
@@ -144,11 +147,13 @@ function usage(){
           none   == do not remove working files, 
           all    == remove all produced files (but summary json), 
           custom == custom implementation"
+  echo ""
   echo "Without -W (default): results are stored in a new subdirectory of <resultsDir>:"
   echo "  <resultsDir>/<uniqueid>/*.json"
   echo "  <resultsDir>/<uniqueid>/proc_1/*.log"
   echo "  <resultsDir>/<uniqueid>/proc_.../*.log"
   echo "  <resultsDir>/<uniqueid>/proc_<COPIES>/*.log"
+  echo ""
   echo "With -W (e.g. in the CI): results are stored in <resultsDir> directly:"
   echo "  <resultsDir>/*.json"
   echo "  <resultsDir>/proc_1/*.log"
@@ -157,16 +162,15 @@ function usage(){
   echo ""
   echo "Without -w (default) and without -W: <resultsDir> is /results"
   echo "Without -w (default) and with -W: <resultsDir> is a tmp directory /tmp/xxxx"
-  echo ""
   if [ "$(type -t usage_detailed)" == "function" ]; then
     echo -e "\nDetailed Usage:\n----------------\n"
     ( usage_detailed ) # as a subprocess, just in case this has a 0 exit code...
   fi
-  echo -e "DESCRIPTION\n"
+  echo -e "\nDESCRIPTION:\n----------------\n"
   if [ -e $BMKDIR/DESCRIPTION ]; then
-      cat $BMKDIR/DESCRIPTION
+    cat $BMKDIR/DESCRIPTION
   else
-      echo "Sorry there is no description included."
+    echo "Sorry there is no description included!" # NB This cannot be made mandatory as this function exits an error anyway! BMK-1027
   fi
   echo ""
   exit 2 # early termination (help or invalid arguments to benchmark script)
@@ -177,7 +181,7 @@ function usage(){
 #####################
 
 debug_args=$@
-OPTPARSE=`getopt -o c:t:e:w:Wdhm: --long help,debug,events:,threads:,copies:,mop: -n $bmkScript -- "$@"`
+OPTPARSE=`getopt -o c:t:e:w:Wdhma: --long help,debug,events:,threads:,copies:,mop:,args: -n $bmkScript -- "$@"`
 if [ $? != 0 ] ; then echo "Invalid options provided." >&2 ; usage ; fi
 eval set -- "$OPTPARSE"
 
@@ -216,6 +220,10 @@ while true; do
       fi
       shift 2
       ;;
+    -a | --args )
+      USER_EXTRA_ARGS="$2" # this may be a space-separated list of arguments (BMK-1014)
+      shift 2
+      ;;
     -w | --resultsdir )
       resultsDir=$2
       shift 2
@@ -267,12 +275,12 @@ check_mandatory_functions
 check_mandatory_variables
 
 # Dump all relevant variables after parsing the input arguments
-for var in USER_NCOPIES USER_NTHREADS USER_NEVENTS_THREAD; do
-  echo "Current value: $var=${!var}"
+for var in USER_NCOPIES USER_NTHREADS USER_NEVENTS_THREAD USER_EXTRA_ARGS; do
+  echo "Current value: $var='${!var}'"
 done
 echo
 for var in resultsDir skipSubDir DEBUG MOP; do
-  echo "Current value: $var=${!var}"
+  echo "Current value: $var='${!var}'"
 done
 echo
 
@@ -311,6 +319,7 @@ if [ "$(type -t validateInputArguments)" != "function" ]; then
   if [ "$USER_NCOPIES" != "" ]; then NCOPIES=$USER_NCOPIES; fi
   if [ "$USER_NTHREADS" != "" ]; then NTHREADS=$USER_NTHREADS; fi # already checked that USER_NTHREADS must be 1 if NTHREADS is 1
   if [ "$USER_NEVENTS_THREAD" != "" ]; then NEVENTS_THREAD=$USER_NEVENTS_THREAD; fi
+  if [ "$USER_EXTRA_ARGS" != "" ]; then EXTRA_ARGS="$USER_EXTRA_ARGS"; fi
 else
   echo -e "[$bmkDriver] function 'validateInputArguments' starting\n"
   if ! validateInputArguments; then fail=-1; fi
@@ -342,6 +351,10 @@ for var in NCOPIES NTHREADS NEVENTS_THREAD; do
   echo "Current value: $var=${!var}"
   echo "$var=${!var}" >> $baseWDir/inputs.log
 done
+for var in EXTRA_ARGS; do
+  echo "Current value: $var='${!var}'"
+  echo "$var='${!var}'" >> $baseWDir/inputs.log # this is a string not a number (BMK-1014 and BMK-1029)
+done
 echo
 
 # Keep a copy of the version.json file (in gitlab CI artifacts) for parser tests on previous logs
@@ -351,7 +364,9 @@ if [ -f $BMKDIR/version.json ]; then
 fi
 
 # Add the containement info in the version.json file
-enrich_version_json
+if ! enrich_version_json; then # add error checking to jq (BMK-1022)
+  exit 1 # early termination (cannot start processing)
+fi
 
 # Define APP before doOne (BMK-152) and parseResults
 APP=$(basename ${BMKDIR}) # or equivalently here $(basename $0 -bmk.sh)
@@ -406,7 +421,7 @@ function doOneWrapper(){
 }
 
 # Export variables to the doOne subprocesses
-for var in NCOPIES NTHREADS NEVENTS_THREAD BMKDIR DEBUG APP; do
+for var in NCOPIES NTHREADS NEVENTS_THREAD EXTRA_ARGS BMKDIR DEBUG APP; do
   export $var
 done
 
@@ -451,8 +466,7 @@ if [ $fail -eq 0 ]; then
     \mv $baseWDir/proc_1/build ${resultsDir}/build
   else
     echo "[$bmkDriver] directory $baseWDir/proc_1/build not found - no need to move it"
-    mkdir ${resultsDir}/build
-    mkdir ${resultsDir}/build/.keepme2
+    mkdir -p ${resultsDir}/build/.keepme2 # use "mkdir -p" to fix "keepme2 file exists" (BMK-1020)
   fi
   echo -e "------------------------------------------------------------------------\n"
 
diff --git a/common/json-differ.py b/common/json-differ.py
index 5151f695b14235b5c22ac097b423a9ccf71e4327..1943e8caec9068f68be04b912082485991202562 100755
--- a/common/json-differ.py
+++ b/common/json-differ.py
@@ -11,6 +11,9 @@ for ajson in sys.argv[1:3]:
     print("Reading file %s" % ajson)
     json_list.append( json.load(open(ajson)) )
 
+if "extra_arguments" not in json_list[0]["run_info"]:
+    json_list[0]["run_info"]["extra_arguments"] = "" # TEMPORARY while this is added to all workloads (BMK-1029 and BMK-1014)
+
 ###for jdict in json_list: print(jdict["app"]["containment"])
 if json_list[1]["app"]["containment"] == "unknown":
     json_list[0]["app"]["containment"] = "DO_NOT_DIFF_CONTAINMENT" # FIX BMK-1009
diff --git a/common/parser-driver.sh b/common/parser-driver.sh
index 2df9c4b1478b4b486e797248bb18b71b46dc49d4..78695d635c881ff8c15283f0d2f0add9f23c2bdd 100644
--- a/common/parser-driver.sh
+++ b/common/parser-driver.sh
@@ -14,6 +14,7 @@ The script ${parserDriver} at  ${parserDriverDIR} has been sourced
 """
 
 function enrich_version_json() {
+  status=0
   # Keep a copy of the version.json file for parser tests on previous logs
   echo "[$parserDriver: enrich_version_json]"
   if [ -f $BMKDIR/version.json ]; then
@@ -25,7 +26,10 @@ function enrich_version_json() {
     elif [ -f /singularity ]; then export flavor=singularity
     else export flavor=unknown; fi
     echo "[$parserDriver: enrich_version_json] adding to version.json containement : $flavor "
-    jq --arg flavor $flavor '. + { "containment" : $flavor }' $BMKDIR/version.json > $baseWDir/version_derived.json # AV eventually could we call this version_enriched.json instead, for consistency?
+    if ! jq --arg flavor $flavor '. + { "containment" : $flavor }' $BMKDIR/version.json > $baseWDir/version_derived.json; then # AV eventually could we call this version_enriched.json instead, for consistency?
+      echo "[$parserDriver: enrich_version_json] ERROR! lint validation failed for $BMKDIR/version.json"
+      status=1 # add error checking to jq (BMK-1022)
+    fi
     echo -e "[$parserDriver] DEBUG: $BMKDIR/version.json\n"
     cat $BMKDIR/version.json
     echo -e "[$parserDriver] DEBUG: $baseWDir/version_derived.json\n"
@@ -34,6 +38,7 @@ function enrich_version_json() {
     # AV should this not rather be an error? BMK-1006
     echo "[$parserDriver: enrich_version_json] WARNING! $BMKDIR/version.json does not exist"
   fi
+  return $status
 }
 
 # Function generateSummary
@@ -42,7 +47,7 @@ function enrich_version_json() {
 # - <fail> < 0: validateInputArguments failed
 # - <fail> > 0: doOne failed (<fail> processes failed out of $NCOPIES)
 # - <fail> = 0: OK
-# The following variables are guaranteed to be defined and exported: NCOPIES, NTHREADS, NEVENTS_THREAD, BMKDIR, DEBUG, APP
+# The following variables are guaranteed to be defined and exported: NCOPIES, NTHREADS, NEVENTS_THREAD, EXTRA_ARGS, BMKDIR, DEBUG, APP
 # The environment variable APP=<vo>-<workload> defines the name of the json file ${APP}_summary.json
 function generateSummary(){
   echo "[$parserDriver:generateSummary] (reported status from calling function: $1)"
@@ -53,18 +58,36 @@ function generateSummary(){
   # Generate the json summary
   echo -e "\n[$parserDriver:generateSummary] Final report"
   local app="{}"
-  if [ -f $baseWDir/version_derived.json ]; then app=$(jq . $baseWDir/version_derived.json); fi
+  if [ -f $baseWDir/version_derived.json ]; then
+    app=$(jq . $baseWDir/version_derived.json)
+    if [ "$?" != "0" ]; then # add error checking to jq (BMK-1022)
+      echo "[$parserDriver:generateSummary] ERROR! lint validation failed for $baseWDir/version_derived.json"
+      status=1
+    fi
+  fi
   local OUTPUT=${APP}_summary.json
   local resJSON="{\"wl-scores\":{}}"
-  if [ -f $baseWDir/parser_output.json ]; then resJSON=$(jq . $baseWDir/parser_output.json); fi
+  if [ -f $baseWDir/parser_output.json ]; then
+    resJSON=$(jq . $baseWDir/parser_output.json)
+    if [ "$?" != "0" ]; then # add error checking to jq (BMK-1022)
+      echo "[$parserDriver:generateSummary] ERROR! lint validation failed for $baseWDir/parser_output.json"
+      status=1
+    fi
+  fi
   report=$(echo $resJSON | jq --arg log "${s_msg}" '. + {"log":$log}')
+  if [ "$?" != "0" ]; then # add error checking to jq (BMK-1022)
+    echo "[$parserDriver:generateSummary] ERROR! lint validation failed for json derived from $baseWDir/parser_output.json"
+    status=1
+  fi
   echo -e "\n[$parserDriver:generateSummary] report from output parser $report"
   echo -e "{
-    \"run_info\": {\"copies\":$NCOPIES , \"threads_per_copy\":$NTHREADS , \"events_per_thread\" : $NEVENTS_THREAD },
+    \"run_info\": {\"copies\":$NCOPIES , \"threads_per_copy\":$NTHREADS , \"events_per_thread\" : $NEVENTS_THREAD, \"extra_arguments\" : \"$EXTRA_ARGS\" },
     \"report\": ${report}, 
     \"app\":${app} }" | jq '.'> $OUTPUT
-  status=$?
-
+  if [ "$?" != "0" ]; then # add error checking to jq (BMK-1022)
+    echo "[$parserDriver:generateSummary] ERROR! lint validation failed for run_info json"
+    status=1
+  fi
   echo -e "\n[$parserDriver:generateSummary] Total bmk report"
   cat $OUTPUT
   return ${status}
@@ -86,7 +109,6 @@ function parseResultsWrapper(){
     subparse=$?
     echo -e "\n[$parserDriver:parseResultsWrapper] WL parser completed (status=$subparse)"
     echo -e "\n[$parserDriver:parseResultsWrapper] calling generate summary"
-
     generateSummary $subparse "ERROR parser" #the second argument is used only if $parse is !=0
     failedSummary=${?}
   fi
diff --git a/common/parsertest.sh b/common/parsertest.sh
index a1f5ca9877b316454135e7ebaae2de3cbf62cbd1..c6c0f1e3942bc7077c8f630fdafd9bc603033c64 100755
--- a/common/parsertest.sh
+++ b/common/parsertest.sh
@@ -64,7 +64,7 @@ for job in $jobs; do
   echo -e "\n======================================================="
   echo -e "\nReference job directory: ${refdir}"
 
-  # Define NCOPIES, NTHREADS, NEVENTS_JSON from inputs.log
+  # Define NCOPIES, NTHREADS, NEVENTS_THREAD, EXTRA_ARGS from inputs.log
   if [ -f ${refdir}/inputs.log ]; then
     source ${refdir}/inputs.log
   else
@@ -89,6 +89,7 @@ for job in $jobs; do
   enrich_version_json
   parseResultsWrapper 0 # > /dev/null # comment out "> /dev/null" to keep debug printouts
   tstatus1=$?
+  tstatus=$tstatus1 # fail parsertest also if parsing or jq fails (BMK-1023): comment this out to use old behaviour
 
   echo -e "\n parsertest.sh: ls of summaries available"
   ls -l *${APPprod}_summary*
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu.spec b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu.spec
index 3482b47bcc6f130c6af6dec5ddc062af40023c95..0286902f1baae837a8cf511f5bc9f14782417e5d 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu.spec
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu.spec
@@ -1,9 +1,9 @@
 HEPWL_BMKEXE=mg5amc-madgraph4gpu-bmk.sh
-HEPWL_BMKOPTS="" # No options for the moment: eventually "-avxall -flt -inl"?
+HEPWL_BMKOPTS="-c 1 --args '--both'" # Build and run both the CPP and CUDA benchmarks using a single CPU core (BMK-1018)
 HEPWL_BMKDIR=mg5amc-madgraph4gpu
 HEPWL_BMKDESCRIPTION="Madgraph5_aMCatNLO madgraph4gpu benchmark" # NB the '@' character breaks sed scripts, do not use it!
 HEPWL_BMKOS="gitlab-registry.cern.ch/linuxsupport/cc7-base:20220801-1.x86_64"
 HEPWL_BMKUSEGPU=1
 HEPWL_DOCKERIMAGENAME=mg5amc-madgraph4gpu-bmk
-HEPWL_DOCKERIMAGETAG=v0.6 # NB: use ci-vX.Y for tests (can be rebuilt) and vX.Y for production (cannot be rebuilt)
+HEPWL_DOCKERIMAGETAG=v0.7 # NB: use ci-vX.Y for tests (can be rebuilt) and vX.Y for production (cannot be rebuilt)
 HEPWL_CVMFSREPOS=sft.cern.ch
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/cishortlog.txt b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/cishortlog.txt
index 2671fd8be3a5022814c63c845cd3d0c8940b7941..9a759df9849dad7bb38d267b43043d615683895f 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/cishortlog.txt
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/cishortlog.txt
@@ -1,9 +1,9 @@
-    "copies": 4,
-{"run_info":{"copies":4,"threads_per_copy":1,"events_per_thread":12},"report":{"wl-scores":{"cuda-d-inl0":4412.73,"cuda-f-inl0":8450.237,"cuda-d-inl1":4363.146,"cuda-f-inl1":6600.3685,"cpp-d-inl0-none":4.9017,"cpp-d-inl0-sse4":9.3986,"cpp-d-inl0-avx2":16.5045,"cpp-d-inl0-512y":17.7301,"cpp-d-inl0-512z":12.3581,"cpp-f-inl0-none":4.594,"cpp-f-inl0-sse4":16.418,"cpp-f-inl0-avx2":31.1695,"cpp-f-inl0-512y":33.2729,"cpp-f-inl0-512z":26.4114,"cpp-d-inl1-none":16.9365,"cpp-d-inl1-sse4":29.3373,"cpp-d-inl1-avx2":38.2394,"cpp-d-inl1-512y":42.3531,"cpp-d-inl1-512z":29.4415,"cpp-f-inl1-none":25.7683,"cpp-f-inl1-sse4":62.7303,"cpp-f-inl1-avx2":86.377,"cpp-f-inl1-512y":89.2037,"cpp-f-inl1-512z":61.1574},"wl-stats":{"cuda-d-inl0":{"avg":1103.1825,"median":1099.0995,"min":1085.296,"max":1129.235,"count":4},"cuda-f-inl0":{"avg":2112.5592,"median":2249.7535,"min":1363.266,"max":2587.464,"count":4},"cuda-d-inl1":{"avg":1090.7865,"median":1091.33,"min":1073.924,"max":1106.562,"count":4},"cuda-f-inl1":{"avg":1650.0921,"median":1964.5985,"min":407.2425,"max":2263.929,"count":4},"cpp-d-inl0-none":{"avg":1.2254,"median":1.2268,"min":1.2097,"max":1.2384,"count":4},"cpp-d-inl0-sse4":{"avg":2.3497,"median":2.3899,"min":2.2099,"max":2.4089,"count":4},"cpp-d-inl0-avx2":{"avg":4.1261,"median":4.0816,"min":4.0197,"max":4.3217,"count":4},"cpp-d-inl0-512y":{"avg":4.4325,"median":4.4327,"min":4.1853,"max":4.6794,"count":4},"cpp-d-inl0-512z":{"avg":3.0895,"median":3.0666,"min":2.928,"max":3.2969,"count":4},"cpp-f-inl0-none":{"avg":1.1485,"median":1.1516,"min":1.1204,"max":1.1704,"count":4},"cpp-f-inl0-sse4":{"avg":4.1045,"median":4.0471,"min":3.9549,"max":4.3688,"count":4},"cpp-f-inl0-avx2":{"avg":7.7924,"median":7.7917,"min":7.7387,"max":7.8474,"count":4},"cpp-f-inl0-512y":{"avg":8.3182,"median":8.3606,"min":8.0657,"max":8.4859,"count":4},"cpp-f-inl0-512z":{"avg":6.6029,"median":6.6135,"min":6.4677,"max":6.7166,"count":4},"cpp-d-inl1-none":{"avg":4.2341,"median":4.2371,"min":4.2047,"max":4.2575,"count":4},"cpp-d-inl1-sse4":{"avg":7.3343,"median":7.3373,"min":7.2728,"max":7.39,"count":4},"cpp-d-inl1-avx2":{"avg":9.5598,"median":9.8387,"min":8.5139,"max":10.048,"count":4},"cpp-d-inl1-512y":{"avg":10.5883,"median":10.5666,"min":10.4499,"max":10.7701,"count":4},"cpp-d-inl1-512z":{"avg":7.3604,"median":7.3179,"min":7.3138,"max":7.4919,"count":4},"cpp-f-inl1-none":{"avg":6.4421,"median":6.4191,"min":6.374,"max":6.556,"count":4},"cpp-f-inl1-sse4":{"avg":15.6826,"median":15.6804,"min":15.6149,"max":15.7546,"count":4},"cpp-f-inl1-avx2":{"avg":21.5943,"median":21.6456,"min":21.2261,"max":21.8598,"count":4},"cpp-f-inl1-512y":{"avg":22.3009,"median":22.2852,"min":22.1744,"max":22.4589,"count":4},"cpp-f-inl1-512z":{"avg":15.2894,"median":15.2395,"min":15.0886,"max":15.5898,"count":4}},"log":"ok"},"app":{"version":"ci-v0.6","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"48a860dd15d851b2c6b82bfd568bfef7","bmkdata_checksum":"09778836b962d01072d52dc09c61007b","bmk_checksum":"9f37480cfd749d5fadbb610ca7235046","containment":"docker"}}
-{"run_info":{"copies":4,"threads_per_copy":1,"events_per_thread":12},"report":{"wl-scores":{"cuda-d-inl0":4412.73,"cuda-f-inl0":8450.237,"cuda-d-inl1":4363.146,"cuda-f-inl1":6600.3685,"cpp-d-inl0-none":4.9017,"cpp-d-inl0-sse4":9.3986,"cpp-d-inl0-avx2":16.5045,"cpp-d-inl0-512y":17.7301,"cpp-d-inl0-512z":12.3581,"cpp-f-inl0-none":4.594,"cpp-f-inl0-sse4":16.418,"cpp-f-inl0-avx2":31.1695,"cpp-f-inl0-512y":33.2729,"cpp-f-inl0-512z":26.4114,"cpp-d-inl1-none":16.9365,"cpp-d-inl1-sse4":29.3373,"cpp-d-inl1-avx2":38.2394,"cpp-d-inl1-512y":42.3531,"cpp-d-inl1-512z":29.4415,"cpp-f-inl1-none":25.7683,"cpp-f-inl1-sse4":62.7303,"cpp-f-inl1-avx2":86.377,"cpp-f-inl1-512y":89.2037,"cpp-f-inl1-512z":61.1574},"wl-stats":{"cuda-d-inl0":{"avg":1103.1825,"median":1099.0995,"min":1085.296,"max":1129.235,"count":4},"cuda-f-inl0":{"avg":2112.5592,"median":2249.7535,"min":1363.266,"max":2587.464,"count":4},"cuda-d-inl1":{"avg":1090.7865,"median":1091.33,"min":1073.924,"max":1106.562,"count":4},"cuda-f-inl1":{"avg":1650.0921,"median":1964.5985,"min":407.2425,"max":2263.929,"count":4},"cpp-d-inl0-none":{"avg":1.2254,"median":1.2268,"min":1.2097,"max":1.2384,"count":4},"cpp-d-inl0-sse4":{"avg":2.3497,"median":2.3899,"min":2.2099,"max":2.4089,"count":4},"cpp-d-inl0-avx2":{"avg":4.1261,"median":4.0816,"min":4.0197,"max":4.3217,"count":4},"cpp-d-inl0-512y":{"avg":4.4325,"median":4.4327,"min":4.1853,"max":4.6794,"count":4},"cpp-d-inl0-512z":{"avg":3.0895,"median":3.0666,"min":2.928,"max":3.2969,"count":4},"cpp-f-inl0-none":{"avg":1.1485,"median":1.1516,"min":1.1204,"max":1.1704,"count":4},"cpp-f-inl0-sse4":{"avg":4.1045,"median":4.0471,"min":3.9549,"max":4.3688,"count":4},"cpp-f-inl0-avx2":{"avg":7.7924,"median":7.7917,"min":7.7387,"max":7.8474,"count":4},"cpp-f-inl0-512y":{"avg":8.3182,"median":8.3606,"min":8.0657,"max":8.4859,"count":4},"cpp-f-inl0-512z":{"avg":6.6029,"median":6.6135,"min":6.4677,"max":6.7166,"count":4},"cpp-d-inl1-none":{"avg":4.2341,"median":4.2371,"min":4.2047,"max":4.2575,"count":4},"cpp-d-inl1-sse4":{"avg":7.3343,"median":7.3373,"min":7.2728,"max":7.39,"count":4},"cpp-d-inl1-avx2":{"avg":9.5598,"median":9.8387,"min":8.5139,"max":10.048,"count":4},"cpp-d-inl1-512y":{"avg":10.5883,"median":10.5666,"min":10.4499,"max":10.7701,"count":4},"cpp-d-inl1-512z":{"avg":7.3604,"median":7.3179,"min":7.3138,"max":7.4919,"count":4},"cpp-f-inl1-none":{"avg":6.4421,"median":6.4191,"min":6.374,"max":6.556,"count":4},"cpp-f-inl1-sse4":{"avg":15.6826,"median":15.6804,"min":15.6149,"max":15.7546,"count":4},"cpp-f-inl1-avx2":{"avg":21.5943,"median":21.6456,"min":21.2261,"max":21.8598,"count":4},"cpp-f-inl1-512y":{"avg":22.3009,"median":22.2852,"min":22.1744,"max":22.4589,"count":4},"cpp-f-inl1-512z":{"avg":15.2894,"median":15.2395,"min":15.0886,"max":15.5898,"count":4}},"log":"ok"},"app":{"version":"ci-v0.6","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"48a860dd15d851b2c6b82bfd568bfef7","bmkdata_checksum":"09778836b962d01072d52dc09c61007b","bmk_checksum":"9f37480cfd749d5fadbb610ca7235046","containment":"docker"}}
-    "copies": 4,
-{"run_info":{"copies":4,"threads_per_copy":1,"events_per_thread":12},"report":{"wl-scores":{"cuda-d-inl0":4546.582,"cuda-f-inl0":6980.786,"cuda-d-inl1":4209.8032,"cuda-f-inl1":11498.399,"cpp-d-inl0-none":4.9951,"cpp-d-inl0-sse4":10.0095,"cpp-d-inl0-avx2":17.3854,"cpp-d-inl0-512y":18.8182,"cpp-d-inl0-512z":12.5551,"cpp-f-inl0-none":4.6394,"cpp-f-inl0-sse4":17.5725,"cpp-f-inl0-avx2":30.7403,"cpp-f-inl0-512y":31.1316,"cpp-f-inl0-512z":25.4174,"cpp-d-inl1-none":16.5604,"cpp-d-inl1-sse4":28.2918,"cpp-d-inl1-avx2":37.2652,"cpp-d-inl1-512y":41.4713,"cpp-d-inl1-512z":29.4372,"cpp-f-inl1-none":25.1617,"cpp-f-inl1-sse4":61.078,"cpp-f-inl1-avx2":81.8479,"cpp-f-inl1-512y":87.2765,"cpp-f-inl1-512z":60.4356},"wl-stats":{"cuda-d-inl0":{"avg":1136.6455,"median":1137.867,"min":1099.445,"max":1171.403,"count":4},"cuda-f-inl0":{"avg":1745.1965,"median":1928.028,"min":1031.338,"max":2093.392,"count":4},"cuda-d-inl1":{"avg":1052.4508,"median":1101.201,"min":686.2372,"max":1321.164,"count":4},"cuda-f-inl1":{"avg":2874.5998,"median":2816.86,"min":2725.214,"max":3139.465,"count":4},"cpp-d-inl0-none":{"avg":1.2488,"median":1.2435,"min":1.2175,"max":1.2906,"count":4},"cpp-d-inl0-sse4":{"avg":2.5024,"median":2.5021,"min":2.4956,"max":2.5096,"count":4},"cpp-d-inl0-avx2":{"avg":4.3464,"median":4.3842,"min":4.1884,"max":4.4286,"count":4},"cpp-d-inl0-512y":{"avg":4.7046,"median":4.7239,"min":4.6115,"max":4.759,"count":4},"cpp-d-inl0-512z":{"avg":3.1388,"median":3.1754,"min":2.9104,"max":3.2938,"count":4},"cpp-f-inl0-none":{"avg":1.1598,"median":1.1602,"min":1.1423,"max":1.1766,"count":4},"cpp-f-inl0-sse4":{"avg":4.3931,"median":4.4124,"min":4.2593,"max":4.4884,"count":4},"cpp-f-inl0-avx2":{"avg":7.6851,"median":7.6443,"min":7.5496,"max":7.902,"count":4},"cpp-f-inl0-512y":{"avg":7.7829,"median":8.0084,"min":6.9233,"max":8.1914,"count":4},"cpp-f-inl0-512z":{"avg":6.3544,"median":6.3962,"min":5.8989,"max":6.7262,"count":4},"cpp-d-inl1-none":{"avg":4.1401,"median":4.1677,"min":3.9664,"max":4.2587,"count":4},"cpp-d-inl1-sse4":{"avg":7.0729,"median":7.071,"min":6.9938,"max":7.156,"count":4},"cpp-d-inl1-avx2":{"avg":9.3163,"median":9.3614,"min":8.6676,"max":9.8749,"count":4},"cpp-d-inl1-512y":{"avg":10.3678,"median":10.4715,"min":9.9459,"max":10.5824,"count":4},"cpp-d-inl1-512z":{"avg":7.3593,"median":7.3619,"min":7.3035,"max":7.4099,"count":4},"cpp-f-inl1-none":{"avg":6.2904,"median":6.4205,"min":5.8145,"max":6.5061,"count":4},"cpp-f-inl1-sse4":{"avg":15.2695,"median":15.203,"min":15.0594,"max":15.6125,"count":4},"cpp-f-inl1-avx2":{"avg":20.462,"median":20.3871,"min":19.7117,"max":21.362,"count":4},"cpp-f-inl1-512y":{"avg":21.8191,"median":21.7672,"min":21.1139,"max":22.6281,"count":4},"cpp-f-inl1-512z":{"avg":15.1089,"median":15.0218,"min":14.683,"max":15.709,"count":4}},"log":"ok"},"app":{"version":"ci-v0.6","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"cbc70b62ff7971cfa1e4d45109dbedc2","bmkdata_checksum":"09778836b962d01072d52dc09c61007b","bmk_checksum":"1ede317ade9476b2cf025418c42f45e8","containment":"docker"}}
-{"run_info":{"copies":4,"threads_per_copy":1,"events_per_thread":12},"report":{"wl-scores":{"cuda-d-inl0":4412.73,"cuda-f-inl0":8450.237,"cuda-d-inl1":4363.146,"cuda-f-inl1":6600.3685,"cpp-d-inl0-none":4.9017,"cpp-d-inl0-sse4":9.3986,"cpp-d-inl0-avx2":16.5045,"cpp-d-inl0-512y":17.7301,"cpp-d-inl0-512z":12.3581,"cpp-f-inl0-none":4.594,"cpp-f-inl0-sse4":16.418,"cpp-f-inl0-avx2":31.1695,"cpp-f-inl0-512y":33.2729,"cpp-f-inl0-512z":26.4114,"cpp-d-inl1-none":16.9365,"cpp-d-inl1-sse4":29.3373,"cpp-d-inl1-avx2":38.2394,"cpp-d-inl1-512y":42.3531,"cpp-d-inl1-512z":29.4415,"cpp-f-inl1-none":25.7683,"cpp-f-inl1-sse4":62.7303,"cpp-f-inl1-avx2":86.377,"cpp-f-inl1-512y":89.2037,"cpp-f-inl1-512z":61.1574},"wl-stats":{"cuda-d-inl0":{"avg":1103.1825,"median":1099.0995,"min":1085.296,"max":1129.235,"count":4},"cuda-f-inl0":{"avg":2112.5592,"median":2249.7535,"min":1363.266,"max":2587.464,"count":4},"cuda-d-inl1":{"avg":1090.7865,"median":1091.33,"min":1073.924,"max":1106.562,"count":4},"cuda-f-inl1":{"avg":1650.0921,"median":1964.5985,"min":407.2425,"max":2263.929,"count":4},"cpp-d-inl0-none":{"avg":1.2254,"median":1.2268,"min":1.2097,"max":1.2384,"count":4},"cpp-d-inl0-sse4":{"avg":2.3497,"median":2.3899,"min":2.2099,"max":2.4089,"count":4},"cpp-d-inl0-avx2":{"avg":4.1261,"median":4.0816,"min":4.0197,"max":4.3217,"count":4},"cpp-d-inl0-512y":{"avg":4.4325,"median":4.4327,"min":4.1853,"max":4.6794,"count":4},"cpp-d-inl0-512z":{"avg":3.0895,"median":3.0666,"min":2.928,"max":3.2969,"count":4},"cpp-f-inl0-none":{"avg":1.1485,"median":1.1516,"min":1.1204,"max":1.1704,"count":4},"cpp-f-inl0-sse4":{"avg":4.1045,"median":4.0471,"min":3.9549,"max":4.3688,"count":4},"cpp-f-inl0-avx2":{"avg":7.7924,"median":7.7917,"min":7.7387,"max":7.8474,"count":4},"cpp-f-inl0-512y":{"avg":8.3182,"median":8.3606,"min":8.0657,"max":8.4859,"count":4},"cpp-f-inl0-512z":{"avg":6.6029,"median":6.6135,"min":6.4677,"max":6.7166,"count":4},"cpp-d-inl1-none":{"avg":4.2341,"median":4.2371,"min":4.2047,"max":4.2575,"count":4},"cpp-d-inl1-sse4":{"avg":7.3343,"median":7.3373,"min":7.2728,"max":7.39,"count":4},"cpp-d-inl1-avx2":{"avg":9.5598,"median":9.8387,"min":8.5139,"max":10.048,"count":4},"cpp-d-inl1-512y":{"avg":10.5883,"median":10.5666,"min":10.4499,"max":10.7701,"count":4},"cpp-d-inl1-512z":{"avg":7.3604,"median":7.3179,"min":7.3138,"max":7.4919,"count":4},"cpp-f-inl1-none":{"avg":6.4421,"median":6.4191,"min":6.374,"max":6.556,"count":4},"cpp-f-inl1-sse4":{"avg":15.6826,"median":15.6804,"min":15.6149,"max":15.7546,"count":4},"cpp-f-inl1-avx2":{"avg":21.5943,"median":21.6456,"min":21.2261,"max":21.8598,"count":4},"cpp-f-inl1-512y":{"avg":22.3009,"median":22.2852,"min":22.1744,"max":22.4589,"count":4},"cpp-f-inl1-512z":{"avg":15.2894,"median":15.2395,"min":15.0886,"max":15.5898,"count":4}},"log":"ok"},"app":{"version":"ci-v0.6","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"48a860dd15d851b2c6b82bfd568bfef7","bmkdata_checksum":"09778836b962d01072d52dc09c61007b","bmk_checksum":"9f37480cfd749d5fadbb610ca7235046","containment":"docker"}}
-    "copies": 4,
-{"run_info":{"copies":4,"threads_per_copy":1,"events_per_thread":12},"report":{"wl-scores":{"cuda-d-inl0":4137.625,"cuda-f-inl0":10500.875,"cuda-d-inl1":4664.8816,"cuda-f-inl1":13775.954,"cpp-d-inl0-none":5.2035,"cpp-d-inl0-sse4":9.9695,"cpp-d-inl0-avx2":17.6938,"cpp-d-inl0-512y":18.0838,"cpp-d-inl0-512z":12.4559,"cpp-f-inl0-none":4.7863,"cpp-f-inl0-sse4":17.8144,"cpp-f-inl0-avx2":31.9468,"cpp-f-inl0-512y":34.466,"cpp-f-inl0-512z":25.9076,"cpp-d-inl1-none":17.0965,"cpp-d-inl1-sse4":28.783,"cpp-d-inl1-avx2":38.2948,"cpp-d-inl1-512y":40.3101,"cpp-d-inl1-512z":28.294,"cpp-f-inl1-none":26.0602,"cpp-f-inl1-sse4":61.9017,"cpp-f-inl1-avx2":82.3965,"cpp-f-inl1-512y":83.6453,"cpp-f-inl1-512z":59.778},"wl-stats":{"cuda-d-inl0":{"avg":1034.4062,"median":1019.2905,"min":1003.895,"max":1095.149,"count":4},"cuda-f-inl0":{"avg":2625.2188,"median":2424.7025,"min":2058.619,"max":3592.851,"count":4},"cuda-d-inl1":{"avg":1166.2204,"median":1116.7095,"min":984.3946,"max":1447.068,"count":4},"cuda-f-inl1":{"avg":3443.9885,"median":3419.1935,"min":2961.966,"max":3975.601,"count":4},"cpp-d-inl0-none":{"avg":1.3009,"median":1.3017,"min":1.2931,"max":1.3071,"count":4},"cpp-d-inl0-sse4":{"avg":2.4924,"median":2.4928,"min":2.4667,"max":2.5173,"count":4},"cpp-d-inl0-avx2":{"avg":4.4234,"median":4.4406,"min":4.3243,"max":4.4882,"count":4},"cpp-d-inl0-512y":{"avg":4.521,"median":4.5765,"min":4.2348,"max":4.696,"count":4},"cpp-d-inl0-512z":{"avg":3.114,"median":3.1546,"min":2.9011,"max":3.2456,"count":4},"cpp-f-inl0-none":{"avg":1.1966,"median":1.1961,"min":1.195,"max":1.1992,"count":4},"cpp-f-inl0-sse4":{"avg":4.4536,"median":4.4514,"min":4.4359,"max":4.4757,"count":4},"cpp-f-inl0-avx2":{"avg":7.9867,"median":7.9982,"min":7.9395,"max":8.0108,"count":4},"cpp-f-inl0-512y":{"avg":8.6165,"median":8.6126,"min":8.5912,"max":8.6496,"count":4},"cpp-f-inl0-512z":{"avg":6.4769,"median":6.5943,"min":6.0632,"max":6.6558,"count":4},"cpp-d-inl1-none":{"avg":4.2741,"median":4.2691,"min":4.2644,"max":4.294,"count":4},"cpp-d-inl1-sse4":{"avg":7.1958,"median":7.1968,"min":7.1848,"max":7.2047,"count":4},"cpp-d-inl1-avx2":{"avg":9.5737,"median":9.5756,"min":9.4725,"max":9.6711,"count":4},"cpp-d-inl1-512y":{"avg":10.0775,"median":10.0963,"min":9.9714,"max":10.1461,"count":4},"cpp-d-inl1-512z":{"avg":7.0735,"median":7.0426,"min":6.9533,"max":7.2554,"count":4},"cpp-f-inl1-none":{"avg":6.5151,"median":6.5176,"min":6.4896,"max":6.5355,"count":4},"cpp-f-inl1-sse4":{"avg":15.4754,"median":15.4715,"min":15.406,"max":15.5527,"count":4},"cpp-f-inl1-avx2":{"avg":20.5991,"median":20.5902,"min":20.3878,"max":20.8283,"count":4},"cpp-f-inl1-512y":{"avg":20.9113,"median":21.1249,"min":19.6155,"max":21.78,"count":4},"cpp-f-inl1-512z":{"avg":14.9445,"median":14.9337,"min":14.7676,"max":15.143,"count":4}},"log":"ok"},"app":{"version":"ci-v0.6","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"cbc70b62ff7971cfa1e4d45109dbedc2","bmkdata_checksum":"09778836b962d01072d52dc09c61007b","bmk_checksum":"1ede317ade9476b2cf025418c42f45e8","containment":"singularity"}}
-{"run_info":{"copies":4,"threads_per_copy":1,"events_per_thread":12},"report":{"wl-scores":{"cuda-d-inl0":4412.73,"cuda-f-inl0":8450.237,"cuda-d-inl1":4363.146,"cuda-f-inl1":6600.3685,"cpp-d-inl0-none":4.9017,"cpp-d-inl0-sse4":9.3986,"cpp-d-inl0-avx2":16.5045,"cpp-d-inl0-512y":17.7301,"cpp-d-inl0-512z":12.3581,"cpp-f-inl0-none":4.594,"cpp-f-inl0-sse4":16.418,"cpp-f-inl0-avx2":31.1695,"cpp-f-inl0-512y":33.2729,"cpp-f-inl0-512z":26.4114,"cpp-d-inl1-none":16.9365,"cpp-d-inl1-sse4":29.3373,"cpp-d-inl1-avx2":38.2394,"cpp-d-inl1-512y":42.3531,"cpp-d-inl1-512z":29.4415,"cpp-f-inl1-none":25.7683,"cpp-f-inl1-sse4":62.7303,"cpp-f-inl1-avx2":86.377,"cpp-f-inl1-512y":89.2037,"cpp-f-inl1-512z":61.1574},"wl-stats":{"cuda-d-inl0":{"avg":1103.1825,"median":1099.0995,"min":1085.296,"max":1129.235,"count":4},"cuda-f-inl0":{"avg":2112.5592,"median":2249.7535,"min":1363.266,"max":2587.464,"count":4},"cuda-d-inl1":{"avg":1090.7865,"median":1091.33,"min":1073.924,"max":1106.562,"count":4},"cuda-f-inl1":{"avg":1650.0921,"median":1964.5985,"min":407.2425,"max":2263.929,"count":4},"cpp-d-inl0-none":{"avg":1.2254,"median":1.2268,"min":1.2097,"max":1.2384,"count":4},"cpp-d-inl0-sse4":{"avg":2.3497,"median":2.3899,"min":2.2099,"max":2.4089,"count":4},"cpp-d-inl0-avx2":{"avg":4.1261,"median":4.0816,"min":4.0197,"max":4.3217,"count":4},"cpp-d-inl0-512y":{"avg":4.4325,"median":4.4327,"min":4.1853,"max":4.6794,"count":4},"cpp-d-inl0-512z":{"avg":3.0895,"median":3.0666,"min":2.928,"max":3.2969,"count":4},"cpp-f-inl0-none":{"avg":1.1485,"median":1.1516,"min":1.1204,"max":1.1704,"count":4},"cpp-f-inl0-sse4":{"avg":4.1045,"median":4.0471,"min":3.9549,"max":4.3688,"count":4},"cpp-f-inl0-avx2":{"avg":7.7924,"median":7.7917,"min":7.7387,"max":7.8474,"count":4},"cpp-f-inl0-512y":{"avg":8.3182,"median":8.3606,"min":8.0657,"max":8.4859,"count":4},"cpp-f-inl0-512z":{"avg":6.6029,"median":6.6135,"min":6.4677,"max":6.7166,"count":4},"cpp-d-inl1-none":{"avg":4.2341,"median":4.2371,"min":4.2047,"max":4.2575,"count":4},"cpp-d-inl1-sse4":{"avg":7.3343,"median":7.3373,"min":7.2728,"max":7.39,"count":4},"cpp-d-inl1-avx2":{"avg":9.5598,"median":9.8387,"min":8.5139,"max":10.048,"count":4},"cpp-d-inl1-512y":{"avg":10.5883,"median":10.5666,"min":10.4499,"max":10.7701,"count":4},"cpp-d-inl1-512z":{"avg":7.3604,"median":7.3179,"min":7.3138,"max":7.4919,"count":4},"cpp-f-inl1-none":{"avg":6.4421,"median":6.4191,"min":6.374,"max":6.556,"count":4},"cpp-f-inl1-sse4":{"avg":15.6826,"median":15.6804,"min":15.6149,"max":15.7546,"count":4},"cpp-f-inl1-avx2":{"avg":21.5943,"median":21.6456,"min":21.2261,"max":21.8598,"count":4},"cpp-f-inl1-512y":{"avg":22.3009,"median":22.2852,"min":22.1744,"max":22.4589,"count":4},"cpp-f-inl1-512z":{"avg":15.2894,"median":15.2395,"min":15.0886,"max":15.5898,"count":4}},"log":"ok"},"app":{"version":"ci-v0.6","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"48a860dd15d851b2c6b82bfd568bfef7","bmkdata_checksum":"09778836b962d01072d52dc09c61007b","bmk_checksum":"9f37480cfd749d5fadbb610ca7235046","containment":"docker"}}
+    "copies": 1,
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":12,"extra_arguments":"--both"},"report":{"wl-scores":{"cuda-d-inl0":1336.788,"cuda-f-inl0":3259.709,"cuda-d-inl1":1355.664,"cuda-f-inl1":3260.011,"cpp-d-inl0-none":1.3055,"cpp-d-inl0-sse4":2.5348,"cpp-d-inl0-avx2":4.5044,"cpp-d-inl0-512y":4.6305,"cpp-d-inl0-512z":3.4765,"cpp-f-inl0-none":1.2016,"cpp-f-inl0-sse4":4.5242,"cpp-f-inl0-avx2":8.2438,"cpp-f-inl0-512y":8.8604,"cpp-f-inl0-512z":7.117,"cpp-d-inl1-none":4.3259,"cpp-d-inl1-sse4":7.6645,"cpp-d-inl1-avx2":10.7028,"cpp-d-inl1-512y":11.6234,"cpp-d-inl1-512z":8.0197,"cpp-f-inl1-none":6.5605,"cpp-f-inl1-sse4":16.5001,"cpp-f-inl1-avx2":23.8756,"cpp-f-inl1-512y":24.8866,"cpp-f-inl1-512z":16.8995},"wl-stats":{"cuda-d-inl0":{"avg":1336.788,"median":1336.788,"min":1336.788,"max":1336.788,"count":1},"cuda-f-inl0":{"avg":3259.709,"median":3259.709,"min":3259.709,"max":3259.709,"count":1},"cuda-d-inl1":{"avg":1355.664,"median":1355.664,"min":1355.664,"max":1355.664,"count":1},"cuda-f-inl1":{"avg":3260.011,"median":3260.011,"min":3260.011,"max":3260.011,"count":1},"cpp-d-inl0-none":{"avg":1.3055,"median":1.3055,"min":1.3055,"max":1.3055,"count":1},"cpp-d-inl0-sse4":{"avg":2.5348,"median":2.5348,"min":2.5348,"max":2.5348,"count":1},"cpp-d-inl0-avx2":{"avg":4.5044,"median":4.5044,"min":4.5044,"max":4.5044,"count":1},"cpp-d-inl0-512y":{"avg":4.6305,"median":4.6305,"min":4.6305,"max":4.6305,"count":1},"cpp-d-inl0-512z":{"avg":3.4765,"median":3.4765,"min":3.4765,"max":3.4765,"count":1},"cpp-f-inl0-none":{"avg":1.2016,"median":1.2016,"min":1.2016,"max":1.2016,"count":1},"cpp-f-inl0-sse4":{"avg":4.5242,"median":4.5242,"min":4.5242,"max":4.5242,"count":1},"cpp-f-inl0-avx2":{"avg":8.2438,"median":8.2438,"min":8.2438,"max":8.2438,"count":1},"cpp-f-inl0-512y":{"avg":8.8604,"median":8.8604,"min":8.8604,"max":8.8604,"count":1},"cpp-f-inl0-512z":{"avg":7.117,"median":7.117,"min":7.117,"max":7.117,"count":1},"cpp-d-inl1-none":{"avg":4.3259,"median":4.3259,"min":4.3259,"max":4.3259,"count":1},"cpp-d-inl1-sse4":{"avg":7.6645,"median":7.6645,"min":7.6645,"max":7.6645,"count":1},"cpp-d-inl1-avx2":{"avg":10.7028,"median":10.7028,"min":10.7028,"max":10.7028,"count":1},"cpp-d-inl1-512y":{"avg":11.6234,"median":11.6234,"min":11.6234,"max":11.6234,"count":1},"cpp-d-inl1-512z":{"avg":8.0197,"median":8.0197,"min":8.0197,"max":8.0197,"count":1},"cpp-f-inl1-none":{"avg":6.5605,"median":6.5605,"min":6.5605,"max":6.5605,"count":1},"cpp-f-inl1-sse4":{"avg":16.5001,"median":16.5001,"min":16.5001,"max":16.5001,"count":1},"cpp-f-inl1-avx2":{"avg":23.8756,"median":23.8756,"min":23.8756,"max":23.8756,"count":1},"cpp-f-inl1-512y":{"avg":24.8866,"median":24.8866,"min":24.8866,"max":24.8866,"count":1},"cpp-f-inl1-512z":{"avg":16.8995,"median":16.8995,"min":16.8995,"max":16.8995,"count":1}},"log":"ok"},"app":{"version":"ci-v0.7","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"e62fa439bdb4a8e6d4888a0789613b3f","bmkdata_checksum":"4eb8273563af22b829ddbf985ae0b0a9","bmk_checksum":"bebb6ecc27bce4bdc1f79862424d3faf","containment":"docker"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":12,"extra_arguments":"--both"},"report":{"wl-scores":{"cuda-d-inl0":1336.788,"cuda-f-inl0":3259.709,"cuda-d-inl1":1355.664,"cuda-f-inl1":3260.011,"cpp-d-inl0-none":1.3055,"cpp-d-inl0-sse4":2.5348,"cpp-d-inl0-avx2":4.5044,"cpp-d-inl0-512y":4.6305,"cpp-d-inl0-512z":3.4765,"cpp-f-inl0-none":1.2016,"cpp-f-inl0-sse4":4.5242,"cpp-f-inl0-avx2":8.2438,"cpp-f-inl0-512y":8.8604,"cpp-f-inl0-512z":7.117,"cpp-d-inl1-none":4.3259,"cpp-d-inl1-sse4":7.6645,"cpp-d-inl1-avx2":10.7028,"cpp-d-inl1-512y":11.6234,"cpp-d-inl1-512z":8.0197,"cpp-f-inl1-none":6.5605,"cpp-f-inl1-sse4":16.5001,"cpp-f-inl1-avx2":23.8756,"cpp-f-inl1-512y":24.8866,"cpp-f-inl1-512z":16.8995},"wl-stats":{"cuda-d-inl0":{"avg":1336.788,"median":1336.788,"min":1336.788,"max":1336.788,"count":1},"cuda-f-inl0":{"avg":3259.709,"median":3259.709,"min":3259.709,"max":3259.709,"count":1},"cuda-d-inl1":{"avg":1355.664,"median":1355.664,"min":1355.664,"max":1355.664,"count":1},"cuda-f-inl1":{"avg":3260.011,"median":3260.011,"min":3260.011,"max":3260.011,"count":1},"cpp-d-inl0-none":{"avg":1.3055,"median":1.3055,"min":1.3055,"max":1.3055,"count":1},"cpp-d-inl0-sse4":{"avg":2.5348,"median":2.5348,"min":2.5348,"max":2.5348,"count":1},"cpp-d-inl0-avx2":{"avg":4.5044,"median":4.5044,"min":4.5044,"max":4.5044,"count":1},"cpp-d-inl0-512y":{"avg":4.6305,"median":4.6305,"min":4.6305,"max":4.6305,"count":1},"cpp-d-inl0-512z":{"avg":3.4765,"median":3.4765,"min":3.4765,"max":3.4765,"count":1},"cpp-f-inl0-none":{"avg":1.2016,"median":1.2016,"min":1.2016,"max":1.2016,"count":1},"cpp-f-inl0-sse4":{"avg":4.5242,"median":4.5242,"min":4.5242,"max":4.5242,"count":1},"cpp-f-inl0-avx2":{"avg":8.2438,"median":8.2438,"min":8.2438,"max":8.2438,"count":1},"cpp-f-inl0-512y":{"avg":8.8604,"median":8.8604,"min":8.8604,"max":8.8604,"count":1},"cpp-f-inl0-512z":{"avg":7.117,"median":7.117,"min":7.117,"max":7.117,"count":1},"cpp-d-inl1-none":{"avg":4.3259,"median":4.3259,"min":4.3259,"max":4.3259,"count":1},"cpp-d-inl1-sse4":{"avg":7.6645,"median":7.6645,"min":7.6645,"max":7.6645,"count":1},"cpp-d-inl1-avx2":{"avg":10.7028,"median":10.7028,"min":10.7028,"max":10.7028,"count":1},"cpp-d-inl1-512y":{"avg":11.6234,"median":11.6234,"min":11.6234,"max":11.6234,"count":1},"cpp-d-inl1-512z":{"avg":8.0197,"median":8.0197,"min":8.0197,"max":8.0197,"count":1},"cpp-f-inl1-none":{"avg":6.5605,"median":6.5605,"min":6.5605,"max":6.5605,"count":1},"cpp-f-inl1-sse4":{"avg":16.5001,"median":16.5001,"min":16.5001,"max":16.5001,"count":1},"cpp-f-inl1-avx2":{"avg":23.8756,"median":23.8756,"min":23.8756,"max":23.8756,"count":1},"cpp-f-inl1-512y":{"avg":24.8866,"median":24.8866,"min":24.8866,"max":24.8866,"count":1},"cpp-f-inl1-512z":{"avg":16.8995,"median":16.8995,"min":16.8995,"max":16.8995,"count":1}},"log":"ok"},"app":{"version":"ci-v0.7","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"e62fa439bdb4a8e6d4888a0789613b3f","bmkdata_checksum":"4eb8273563af22b829ddbf985ae0b0a9","bmk_checksum":"bebb6ecc27bce4bdc1f79862424d3faf","containment":"docker"}}
+    "copies": 1,
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":12,"extra_arguments":"--both"},"report":{"wl-scores":{"cuda-d-inl0":1341.419,"cuda-f-inl0":3248.914,"cuda-d-inl1":1357.328,"cuda-f-inl1":3253.001,"cpp-d-inl0-none":1.3067,"cpp-d-inl0-sse4":2.5344,"cpp-d-inl0-avx2":4.5401,"cpp-d-inl0-512y":4.8484,"cpp-d-inl0-512z":3.4523,"cpp-f-inl0-none":1.2086,"cpp-f-inl0-sse4":4.5408,"cpp-f-inl0-avx2":8.201,"cpp-f-inl0-512y":8.8378,"cpp-f-inl0-512z":6.781,"cpp-d-inl1-none":4.3149,"cpp-d-inl1-sse4":7.4918,"cpp-d-inl1-avx2":10.2609,"cpp-d-inl1-512y":10.8104,"cpp-d-inl1-512z":7.6855,"cpp-f-inl1-none":6.5177,"cpp-f-inl1-sse4":16.0085,"cpp-f-inl1-avx2":22.5616,"cpp-f-inl1-512y":24.117,"cpp-f-inl1-512z":16.268},"wl-stats":{"cuda-d-inl0":{"avg":1341.419,"median":1341.419,"min":1341.419,"max":1341.419,"count":1},"cuda-f-inl0":{"avg":3248.914,"median":3248.914,"min":3248.914,"max":3248.914,"count":1},"cuda-d-inl1":{"avg":1357.328,"median":1357.328,"min":1357.328,"max":1357.328,"count":1},"cuda-f-inl1":{"avg":3253.001,"median":3253.001,"min":3253.001,"max":3253.001,"count":1},"cpp-d-inl0-none":{"avg":1.3067,"median":1.3067,"min":1.3067,"max":1.3067,"count":1},"cpp-d-inl0-sse4":{"avg":2.5344,"median":2.5344,"min":2.5344,"max":2.5344,"count":1},"cpp-d-inl0-avx2":{"avg":4.5401,"median":4.5401,"min":4.5401,"max":4.5401,"count":1},"cpp-d-inl0-512y":{"avg":4.8484,"median":4.8484,"min":4.8484,"max":4.8484,"count":1},"cpp-d-inl0-512z":{"avg":3.4523,"median":3.4523,"min":3.4523,"max":3.4523,"count":1},"cpp-f-inl0-none":{"avg":1.2086,"median":1.2086,"min":1.2086,"max":1.2086,"count":1},"cpp-f-inl0-sse4":{"avg":4.5408,"median":4.5408,"min":4.5408,"max":4.5408,"count":1},"cpp-f-inl0-avx2":{"avg":8.201,"median":8.201,"min":8.201,"max":8.201,"count":1},"cpp-f-inl0-512y":{"avg":8.8378,"median":8.8378,"min":8.8378,"max":8.8378,"count":1},"cpp-f-inl0-512z":{"avg":6.781,"median":6.781,"min":6.781,"max":6.781,"count":1},"cpp-d-inl1-none":{"avg":4.3149,"median":4.3149,"min":4.3149,"max":4.3149,"count":1},"cpp-d-inl1-sse4":{"avg":7.4918,"median":7.4918,"min":7.4918,"max":7.4918,"count":1},"cpp-d-inl1-avx2":{"avg":10.2609,"median":10.2609,"min":10.2609,"max":10.2609,"count":1},"cpp-d-inl1-512y":{"avg":10.8104,"median":10.8104,"min":10.8104,"max":10.8104,"count":1},"cpp-d-inl1-512z":{"avg":7.6855,"median":7.6855,"min":7.6855,"max":7.6855,"count":1},"cpp-f-inl1-none":{"avg":6.5177,"median":6.5177,"min":6.5177,"max":6.5177,"count":1},"cpp-f-inl1-sse4":{"avg":16.0085,"median":16.0085,"min":16.0085,"max":16.0085,"count":1},"cpp-f-inl1-avx2":{"avg":22.5616,"median":22.5616,"min":22.5616,"max":22.5616,"count":1},"cpp-f-inl1-512y":{"avg":24.117,"median":24.117,"min":24.117,"max":24.117,"count":1},"cpp-f-inl1-512z":{"avg":16.268,"median":16.268,"min":16.268,"max":16.268,"count":1}},"log":"ok"},"app":{"version":"ci-v0.7","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"d7711b68f507160e48c70d8cc49c27ab","bmkdata_checksum":"4eb8273563af22b829ddbf985ae0b0a9","bmk_checksum":"cf5ca117fdf0774ad6d4452ea7dfe8e9","containment":"docker"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":12,"extra_arguments":"--both"},"report":{"wl-scores":{"cuda-d-inl0":1336.788,"cuda-f-inl0":3259.709,"cuda-d-inl1":1355.664,"cuda-f-inl1":3260.011,"cpp-d-inl0-none":1.3055,"cpp-d-inl0-sse4":2.5348,"cpp-d-inl0-avx2":4.5044,"cpp-d-inl0-512y":4.6305,"cpp-d-inl0-512z":3.4765,"cpp-f-inl0-none":1.2016,"cpp-f-inl0-sse4":4.5242,"cpp-f-inl0-avx2":8.2438,"cpp-f-inl0-512y":8.8604,"cpp-f-inl0-512z":7.117,"cpp-d-inl1-none":4.3259,"cpp-d-inl1-sse4":7.6645,"cpp-d-inl1-avx2":10.7028,"cpp-d-inl1-512y":11.6234,"cpp-d-inl1-512z":8.0197,"cpp-f-inl1-none":6.5605,"cpp-f-inl1-sse4":16.5001,"cpp-f-inl1-avx2":23.8756,"cpp-f-inl1-512y":24.8866,"cpp-f-inl1-512z":16.8995},"wl-stats":{"cuda-d-inl0":{"avg":1336.788,"median":1336.788,"min":1336.788,"max":1336.788,"count":1},"cuda-f-inl0":{"avg":3259.709,"median":3259.709,"min":3259.709,"max":3259.709,"count":1},"cuda-d-inl1":{"avg":1355.664,"median":1355.664,"min":1355.664,"max":1355.664,"count":1},"cuda-f-inl1":{"avg":3260.011,"median":3260.011,"min":3260.011,"max":3260.011,"count":1},"cpp-d-inl0-none":{"avg":1.3055,"median":1.3055,"min":1.3055,"max":1.3055,"count":1},"cpp-d-inl0-sse4":{"avg":2.5348,"median":2.5348,"min":2.5348,"max":2.5348,"count":1},"cpp-d-inl0-avx2":{"avg":4.5044,"median":4.5044,"min":4.5044,"max":4.5044,"count":1},"cpp-d-inl0-512y":{"avg":4.6305,"median":4.6305,"min":4.6305,"max":4.6305,"count":1},"cpp-d-inl0-512z":{"avg":3.4765,"median":3.4765,"min":3.4765,"max":3.4765,"count":1},"cpp-f-inl0-none":{"avg":1.2016,"median":1.2016,"min":1.2016,"max":1.2016,"count":1},"cpp-f-inl0-sse4":{"avg":4.5242,"median":4.5242,"min":4.5242,"max":4.5242,"count":1},"cpp-f-inl0-avx2":{"avg":8.2438,"median":8.2438,"min":8.2438,"max":8.2438,"count":1},"cpp-f-inl0-512y":{"avg":8.8604,"median":8.8604,"min":8.8604,"max":8.8604,"count":1},"cpp-f-inl0-512z":{"avg":7.117,"median":7.117,"min":7.117,"max":7.117,"count":1},"cpp-d-inl1-none":{"avg":4.3259,"median":4.3259,"min":4.3259,"max":4.3259,"count":1},"cpp-d-inl1-sse4":{"avg":7.6645,"median":7.6645,"min":7.6645,"max":7.6645,"count":1},"cpp-d-inl1-avx2":{"avg":10.7028,"median":10.7028,"min":10.7028,"max":10.7028,"count":1},"cpp-d-inl1-512y":{"avg":11.6234,"median":11.6234,"min":11.6234,"max":11.6234,"count":1},"cpp-d-inl1-512z":{"avg":8.0197,"median":8.0197,"min":8.0197,"max":8.0197,"count":1},"cpp-f-inl1-none":{"avg":6.5605,"median":6.5605,"min":6.5605,"max":6.5605,"count":1},"cpp-f-inl1-sse4":{"avg":16.5001,"median":16.5001,"min":16.5001,"max":16.5001,"count":1},"cpp-f-inl1-avx2":{"avg":23.8756,"median":23.8756,"min":23.8756,"max":23.8756,"count":1},"cpp-f-inl1-512y":{"avg":24.8866,"median":24.8866,"min":24.8866,"max":24.8866,"count":1},"cpp-f-inl1-512z":{"avg":16.8995,"median":16.8995,"min":16.8995,"max":16.8995,"count":1}},"log":"ok"},"app":{"version":"ci-v0.7","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"e62fa439bdb4a8e6d4888a0789613b3f","bmkdata_checksum":"4eb8273563af22b829ddbf985ae0b0a9","bmk_checksum":"bebb6ecc27bce4bdc1f79862424d3faf","containment":"docker"}}
+    "copies": 1,
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":12,"extra_arguments":"--both"},"report":{"wl-scores":{"cuda-d-inl0":1374.997,"cuda-f-inl0":3227.002,"cuda-d-inl1":1327.673,"cuda-f-inl1":3248.295,"cpp-d-inl0-none":1.3166,"cpp-d-inl0-sse4":2.5513,"cpp-d-inl0-avx2":4.555,"cpp-d-inl0-512y":4.8244,"cpp-d-inl0-512z":3.4486,"cpp-f-inl0-none":1.2049,"cpp-f-inl0-sse4":4.5669,"cpp-f-inl0-avx2":8.2615,"cpp-f-inl0-512y":8.7872,"cpp-f-inl0-512z":7.0312,"cpp-d-inl1-none":4.331,"cpp-d-inl1-sse4":7.501,"cpp-d-inl1-avx2":10.2228,"cpp-d-inl1-512y":10.9143,"cpp-d-inl1-512z":7.6985,"cpp-f-inl1-none":6.5926,"cpp-f-inl1-sse4":16.0978,"cpp-f-inl1-avx2":22.6422,"cpp-f-inl1-512y":23.3967,"cpp-f-inl1-512z":16.8209},"wl-stats":{"cuda-d-inl0":{"avg":1374.997,"median":1374.997,"min":1374.997,"max":1374.997,"count":1},"cuda-f-inl0":{"avg":3227.002,"median":3227.002,"min":3227.002,"max":3227.002,"count":1},"cuda-d-inl1":{"avg":1327.673,"median":1327.673,"min":1327.673,"max":1327.673,"count":1},"cuda-f-inl1":{"avg":3248.295,"median":3248.295,"min":3248.295,"max":3248.295,"count":1},"cpp-d-inl0-none":{"avg":1.3166,"median":1.3166,"min":1.3166,"max":1.3166,"count":1},"cpp-d-inl0-sse4":{"avg":2.5513,"median":2.5513,"min":2.5513,"max":2.5513,"count":1},"cpp-d-inl0-avx2":{"avg":4.555,"median":4.555,"min":4.555,"max":4.555,"count":1},"cpp-d-inl0-512y":{"avg":4.8244,"median":4.8244,"min":4.8244,"max":4.8244,"count":1},"cpp-d-inl0-512z":{"avg":3.4486,"median":3.4486,"min":3.4486,"max":3.4486,"count":1},"cpp-f-inl0-none":{"avg":1.2049,"median":1.2049,"min":1.2049,"max":1.2049,"count":1},"cpp-f-inl0-sse4":{"avg":4.5669,"median":4.5669,"min":4.5669,"max":4.5669,"count":1},"cpp-f-inl0-avx2":{"avg":8.2615,"median":8.2615,"min":8.2615,"max":8.2615,"count":1},"cpp-f-inl0-512y":{"avg":8.7872,"median":8.7872,"min":8.7872,"max":8.7872,"count":1},"cpp-f-inl0-512z":{"avg":7.0312,"median":7.0312,"min":7.0312,"max":7.0312,"count":1},"cpp-d-inl1-none":{"avg":4.331,"median":4.331,"min":4.331,"max":4.331,"count":1},"cpp-d-inl1-sse4":{"avg":7.501,"median":7.501,"min":7.501,"max":7.501,"count":1},"cpp-d-inl1-avx2":{"avg":10.2228,"median":10.2228,"min":10.2228,"max":10.2228,"count":1},"cpp-d-inl1-512y":{"avg":10.9143,"median":10.9143,"min":10.9143,"max":10.9143,"count":1},"cpp-d-inl1-512z":{"avg":7.6985,"median":7.6985,"min":7.6985,"max":7.6985,"count":1},"cpp-f-inl1-none":{"avg":6.5926,"median":6.5926,"min":6.5926,"max":6.5926,"count":1},"cpp-f-inl1-sse4":{"avg":16.0978,"median":16.0978,"min":16.0978,"max":16.0978,"count":1},"cpp-f-inl1-avx2":{"avg":22.6422,"median":22.6422,"min":22.6422,"max":22.6422,"count":1},"cpp-f-inl1-512y":{"avg":23.3967,"median":23.3967,"min":23.3967,"max":23.3967,"count":1},"cpp-f-inl1-512z":{"avg":16.8209,"median":16.8209,"min":16.8209,"max":16.8209,"count":1}},"log":"ok"},"app":{"version":"ci-v0.7","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"d7711b68f507160e48c70d8cc49c27ab","bmkdata_checksum":"4eb8273563af22b829ddbf985ae0b0a9","bmk_checksum":"cf5ca117fdf0774ad6d4452ea7dfe8e9","containment":"singularity"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":12,"extra_arguments":"--both"},"report":{"wl-scores":{"cuda-d-inl0":1336.788,"cuda-f-inl0":3259.709,"cuda-d-inl1":1355.664,"cuda-f-inl1":3260.011,"cpp-d-inl0-none":1.3055,"cpp-d-inl0-sse4":2.5348,"cpp-d-inl0-avx2":4.5044,"cpp-d-inl0-512y":4.6305,"cpp-d-inl0-512z":3.4765,"cpp-f-inl0-none":1.2016,"cpp-f-inl0-sse4":4.5242,"cpp-f-inl0-avx2":8.2438,"cpp-f-inl0-512y":8.8604,"cpp-f-inl0-512z":7.117,"cpp-d-inl1-none":4.3259,"cpp-d-inl1-sse4":7.6645,"cpp-d-inl1-avx2":10.7028,"cpp-d-inl1-512y":11.6234,"cpp-d-inl1-512z":8.0197,"cpp-f-inl1-none":6.5605,"cpp-f-inl1-sse4":16.5001,"cpp-f-inl1-avx2":23.8756,"cpp-f-inl1-512y":24.8866,"cpp-f-inl1-512z":16.8995},"wl-stats":{"cuda-d-inl0":{"avg":1336.788,"median":1336.788,"min":1336.788,"max":1336.788,"count":1},"cuda-f-inl0":{"avg":3259.709,"median":3259.709,"min":3259.709,"max":3259.709,"count":1},"cuda-d-inl1":{"avg":1355.664,"median":1355.664,"min":1355.664,"max":1355.664,"count":1},"cuda-f-inl1":{"avg":3260.011,"median":3260.011,"min":3260.011,"max":3260.011,"count":1},"cpp-d-inl0-none":{"avg":1.3055,"median":1.3055,"min":1.3055,"max":1.3055,"count":1},"cpp-d-inl0-sse4":{"avg":2.5348,"median":2.5348,"min":2.5348,"max":2.5348,"count":1},"cpp-d-inl0-avx2":{"avg":4.5044,"median":4.5044,"min":4.5044,"max":4.5044,"count":1},"cpp-d-inl0-512y":{"avg":4.6305,"median":4.6305,"min":4.6305,"max":4.6305,"count":1},"cpp-d-inl0-512z":{"avg":3.4765,"median":3.4765,"min":3.4765,"max":3.4765,"count":1},"cpp-f-inl0-none":{"avg":1.2016,"median":1.2016,"min":1.2016,"max":1.2016,"count":1},"cpp-f-inl0-sse4":{"avg":4.5242,"median":4.5242,"min":4.5242,"max":4.5242,"count":1},"cpp-f-inl0-avx2":{"avg":8.2438,"median":8.2438,"min":8.2438,"max":8.2438,"count":1},"cpp-f-inl0-512y":{"avg":8.8604,"median":8.8604,"min":8.8604,"max":8.8604,"count":1},"cpp-f-inl0-512z":{"avg":7.117,"median":7.117,"min":7.117,"max":7.117,"count":1},"cpp-d-inl1-none":{"avg":4.3259,"median":4.3259,"min":4.3259,"max":4.3259,"count":1},"cpp-d-inl1-sse4":{"avg":7.6645,"median":7.6645,"min":7.6645,"max":7.6645,"count":1},"cpp-d-inl1-avx2":{"avg":10.7028,"median":10.7028,"min":10.7028,"max":10.7028,"count":1},"cpp-d-inl1-512y":{"avg":11.6234,"median":11.6234,"min":11.6234,"max":11.6234,"count":1},"cpp-d-inl1-512z":{"avg":8.0197,"median":8.0197,"min":8.0197,"max":8.0197,"count":1},"cpp-f-inl1-none":{"avg":6.5605,"median":6.5605,"min":6.5605,"max":6.5605,"count":1},"cpp-f-inl1-sse4":{"avg":16.5001,"median":16.5001,"min":16.5001,"max":16.5001,"count":1},"cpp-f-inl1-avx2":{"avg":23.8756,"median":23.8756,"min":23.8756,"max":23.8756,"count":1},"cpp-f-inl1-512y":{"avg":24.8866,"median":24.8866,"min":24.8866,"max":24.8866,"count":1},"cpp-f-inl1-512z":{"avg":16.8995,"median":16.8995,"min":16.8995,"max":16.8995,"count":1}},"log":"ok"},"app":{"version":"ci-v0.7","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"e62fa439bdb4a8e6d4888a0789613b3f","bmkdata_checksum":"4eb8273563af22b829ddbf985ae0b0a9","bmk_checksum":"bebb6ecc27bce4bdc1f79862424d3faf","containment":"docker"}}
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/inputs.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/inputs.log
index fa6f5d726bd557facfff8a0382b36e86d67bc912..50c8d3455af2e74f31496c4446149c21c1ebffdf 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/inputs.log
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/inputs.log
@@ -1,3 +1,4 @@
-NCOPIES=4
+NCOPIES=1
 NTHREADS=1
 NEVENTS_THREAD=12
+EXTRA_ARGS='--both'
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/joburl.txt b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/joburl.txt
index 4ba57c5224e9db4faae720cfe052b62215065f50..9f56f68f3fd64370af45310585451e2ed10e10df 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/joburl.txt
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/joburl.txt
@@ -1 +1 @@
-https://gitlab.cern.ch/valassi/hep-workloads/-/jobs/23945119
+https://gitlab.cern.ch/valassi/hep-workloads/-/jobs/24061392
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/mg5amc-madgraph4gpu_summary.json b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/mg5amc-madgraph4gpu_summary.json
index 8cfe75abad31d4e40722e76292b6764de32a82cb..23b5c519853f53923ae3e680f09ab4b0e5cc8f2e 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/mg5amc-madgraph4gpu_summary.json
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/mg5amc-madgraph4gpu_summary.json
@@ -1,214 +1,215 @@
 {
   "run_info": {
-    "copies": 4,
+    "copies": 1,
     "threads_per_copy": 1,
-    "events_per_thread": 12
+    "events_per_thread": 12,
+    "extra_arguments": "--both"
   },
   "report": {
     "wl-scores": {
-      "cuda-d-inl0": 4412.73,
-      "cuda-f-inl0": 8450.237,
-      "cuda-d-inl1": 4363.146,
-      "cuda-f-inl1": 6600.3685,
-      "cpp-d-inl0-none": 4.9017,
-      "cpp-d-inl0-sse4": 9.3986,
-      "cpp-d-inl0-avx2": 16.5045,
-      "cpp-d-inl0-512y": 17.7301,
-      "cpp-d-inl0-512z": 12.3581,
-      "cpp-f-inl0-none": 4.594,
-      "cpp-f-inl0-sse4": 16.418,
-      "cpp-f-inl0-avx2": 31.1695,
-      "cpp-f-inl0-512y": 33.2729,
-      "cpp-f-inl0-512z": 26.4114,
-      "cpp-d-inl1-none": 16.9365,
-      "cpp-d-inl1-sse4": 29.3373,
-      "cpp-d-inl1-avx2": 38.2394,
-      "cpp-d-inl1-512y": 42.3531,
-      "cpp-d-inl1-512z": 29.4415,
-      "cpp-f-inl1-none": 25.7683,
-      "cpp-f-inl1-sse4": 62.7303,
-      "cpp-f-inl1-avx2": 86.377,
-      "cpp-f-inl1-512y": 89.2037,
-      "cpp-f-inl1-512z": 61.1574
+      "cuda-d-inl0": 1336.788,
+      "cuda-f-inl0": 3259.709,
+      "cuda-d-inl1": 1355.664,
+      "cuda-f-inl1": 3260.011,
+      "cpp-d-inl0-none": 1.3055,
+      "cpp-d-inl0-sse4": 2.5348,
+      "cpp-d-inl0-avx2": 4.5044,
+      "cpp-d-inl0-512y": 4.6305,
+      "cpp-d-inl0-512z": 3.4765,
+      "cpp-f-inl0-none": 1.2016,
+      "cpp-f-inl0-sse4": 4.5242,
+      "cpp-f-inl0-avx2": 8.2438,
+      "cpp-f-inl0-512y": 8.8604,
+      "cpp-f-inl0-512z": 7.117,
+      "cpp-d-inl1-none": 4.3259,
+      "cpp-d-inl1-sse4": 7.6645,
+      "cpp-d-inl1-avx2": 10.7028,
+      "cpp-d-inl1-512y": 11.6234,
+      "cpp-d-inl1-512z": 8.0197,
+      "cpp-f-inl1-none": 6.5605,
+      "cpp-f-inl1-sse4": 16.5001,
+      "cpp-f-inl1-avx2": 23.8756,
+      "cpp-f-inl1-512y": 24.8866,
+      "cpp-f-inl1-512z": 16.8995
     },
     "wl-stats": {
       "cuda-d-inl0": {
-        "avg": 1103.1825,
-        "median": 1099.0995,
-        "min": 1085.296,
-        "max": 1129.235,
-        "count": 4
+        "avg": 1336.788,
+        "median": 1336.788,
+        "min": 1336.788,
+        "max": 1336.788,
+        "count": 1
       },
       "cuda-f-inl0": {
-        "avg": 2112.5592,
-        "median": 2249.7535,
-        "min": 1363.266,
-        "max": 2587.464,
-        "count": 4
+        "avg": 3259.709,
+        "median": 3259.709,
+        "min": 3259.709,
+        "max": 3259.709,
+        "count": 1
       },
       "cuda-d-inl1": {
-        "avg": 1090.7865,
-        "median": 1091.33,
-        "min": 1073.924,
-        "max": 1106.562,
-        "count": 4
+        "avg": 1355.664,
+        "median": 1355.664,
+        "min": 1355.664,
+        "max": 1355.664,
+        "count": 1
       },
       "cuda-f-inl1": {
-        "avg": 1650.0921,
-        "median": 1964.5985,
-        "min": 407.2425,
-        "max": 2263.929,
-        "count": 4
+        "avg": 3260.011,
+        "median": 3260.011,
+        "min": 3260.011,
+        "max": 3260.011,
+        "count": 1
       },
       "cpp-d-inl0-none": {
-        "avg": 1.2254,
-        "median": 1.2268,
-        "min": 1.2097,
-        "max": 1.2384,
-        "count": 4
+        "avg": 1.3055,
+        "median": 1.3055,
+        "min": 1.3055,
+        "max": 1.3055,
+        "count": 1
       },
       "cpp-d-inl0-sse4": {
-        "avg": 2.3497,
-        "median": 2.3899,
-        "min": 2.2099,
-        "max": 2.4089,
-        "count": 4
+        "avg": 2.5348,
+        "median": 2.5348,
+        "min": 2.5348,
+        "max": 2.5348,
+        "count": 1
       },
       "cpp-d-inl0-avx2": {
-        "avg": 4.1261,
-        "median": 4.0816,
-        "min": 4.0197,
-        "max": 4.3217,
-        "count": 4
+        "avg": 4.5044,
+        "median": 4.5044,
+        "min": 4.5044,
+        "max": 4.5044,
+        "count": 1
       },
       "cpp-d-inl0-512y": {
-        "avg": 4.4325,
-        "median": 4.4327,
-        "min": 4.1853,
-        "max": 4.6794,
-        "count": 4
+        "avg": 4.6305,
+        "median": 4.6305,
+        "min": 4.6305,
+        "max": 4.6305,
+        "count": 1
       },
       "cpp-d-inl0-512z": {
-        "avg": 3.0895,
-        "median": 3.0666,
-        "min": 2.928,
-        "max": 3.2969,
-        "count": 4
+        "avg": 3.4765,
+        "median": 3.4765,
+        "min": 3.4765,
+        "max": 3.4765,
+        "count": 1
       },
       "cpp-f-inl0-none": {
-        "avg": 1.1485,
-        "median": 1.1516,
-        "min": 1.1204,
-        "max": 1.1704,
-        "count": 4
+        "avg": 1.2016,
+        "median": 1.2016,
+        "min": 1.2016,
+        "max": 1.2016,
+        "count": 1
       },
       "cpp-f-inl0-sse4": {
-        "avg": 4.1045,
-        "median": 4.0471,
-        "min": 3.9549,
-        "max": 4.3688,
-        "count": 4
+        "avg": 4.5242,
+        "median": 4.5242,
+        "min": 4.5242,
+        "max": 4.5242,
+        "count": 1
       },
       "cpp-f-inl0-avx2": {
-        "avg": 7.7924,
-        "median": 7.7917,
-        "min": 7.7387,
-        "max": 7.8474,
-        "count": 4
+        "avg": 8.2438,
+        "median": 8.2438,
+        "min": 8.2438,
+        "max": 8.2438,
+        "count": 1
       },
       "cpp-f-inl0-512y": {
-        "avg": 8.3182,
-        "median": 8.3606,
-        "min": 8.0657,
-        "max": 8.4859,
-        "count": 4
+        "avg": 8.8604,
+        "median": 8.8604,
+        "min": 8.8604,
+        "max": 8.8604,
+        "count": 1
       },
       "cpp-f-inl0-512z": {
-        "avg": 6.6029,
-        "median": 6.6135,
-        "min": 6.4677,
-        "max": 6.7166,
-        "count": 4
+        "avg": 7.117,
+        "median": 7.117,
+        "min": 7.117,
+        "max": 7.117,
+        "count": 1
       },
       "cpp-d-inl1-none": {
-        "avg": 4.2341,
-        "median": 4.2371,
-        "min": 4.2047,
-        "max": 4.2575,
-        "count": 4
+        "avg": 4.3259,
+        "median": 4.3259,
+        "min": 4.3259,
+        "max": 4.3259,
+        "count": 1
       },
       "cpp-d-inl1-sse4": {
-        "avg": 7.3343,
-        "median": 7.3373,
-        "min": 7.2728,
-        "max": 7.39,
-        "count": 4
+        "avg": 7.6645,
+        "median": 7.6645,
+        "min": 7.6645,
+        "max": 7.6645,
+        "count": 1
       },
       "cpp-d-inl1-avx2": {
-        "avg": 9.5598,
-        "median": 9.8387,
-        "min": 8.5139,
-        "max": 10.048,
-        "count": 4
+        "avg": 10.7028,
+        "median": 10.7028,
+        "min": 10.7028,
+        "max": 10.7028,
+        "count": 1
       },
       "cpp-d-inl1-512y": {
-        "avg": 10.5883,
-        "median": 10.5666,
-        "min": 10.4499,
-        "max": 10.7701,
-        "count": 4
+        "avg": 11.6234,
+        "median": 11.6234,
+        "min": 11.6234,
+        "max": 11.6234,
+        "count": 1
       },
       "cpp-d-inl1-512z": {
-        "avg": 7.3604,
-        "median": 7.3179,
-        "min": 7.3138,
-        "max": 7.4919,
-        "count": 4
+        "avg": 8.0197,
+        "median": 8.0197,
+        "min": 8.0197,
+        "max": 8.0197,
+        "count": 1
       },
       "cpp-f-inl1-none": {
-        "avg": 6.4421,
-        "median": 6.4191,
-        "min": 6.374,
-        "max": 6.556,
-        "count": 4
+        "avg": 6.5605,
+        "median": 6.5605,
+        "min": 6.5605,
+        "max": 6.5605,
+        "count": 1
       },
       "cpp-f-inl1-sse4": {
-        "avg": 15.6826,
-        "median": 15.6804,
-        "min": 15.6149,
-        "max": 15.7546,
-        "count": 4
+        "avg": 16.5001,
+        "median": 16.5001,
+        "min": 16.5001,
+        "max": 16.5001,
+        "count": 1
       },
       "cpp-f-inl1-avx2": {
-        "avg": 21.5943,
-        "median": 21.6456,
-        "min": 21.2261,
-        "max": 21.8598,
-        "count": 4
+        "avg": 23.8756,
+        "median": 23.8756,
+        "min": 23.8756,
+        "max": 23.8756,
+        "count": 1
       },
       "cpp-f-inl1-512y": {
-        "avg": 22.3009,
-        "median": 22.2852,
-        "min": 22.1744,
-        "max": 22.4589,
-        "count": 4
+        "avg": 24.8866,
+        "median": 24.8866,
+        "min": 24.8866,
+        "max": 24.8866,
+        "count": 1
       },
       "cpp-f-inl1-512z": {
-        "avg": 15.2894,
-        "median": 15.2395,
-        "min": 15.0886,
-        "max": 15.5898,
-        "count": 4
+        "avg": 16.8995,
+        "median": 16.8995,
+        "min": 16.8995,
+        "max": 16.8995,
+        "count": 1
       }
     },
     "log": "ok"
   },
   "app": {
-    "version": "ci-v0.6",
+    "version": "ci-v0.7",
     "description": "Madgraph5_aMCatNLO madgraph4gpu benchmark",
-    "cvmfs_checksum": "48a860dd15d851b2c6b82bfd568bfef7",
-    "bmkdata_checksum": "09778836b962d01072d52dc09c61007b",
-    "bmk_checksum": "9f37480cfd749d5fadbb610ca7235046",
+    "cvmfs_checksum": "e62fa439bdb4a8e6d4888a0789613b3f",
+    "bmkdata_checksum": "4eb8273563af22b829ddbf985ae0b0a9",
+    "bmk_checksum": "bebb6ecc27bce4bdc1f79862424d3faf",
     "containment": "docker"
   }
 }
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/doOneWrapper_1.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/doOneWrapper_1.log
index d1c4ecb86de8ff7d0834e22f733c701ca1c4bf6d..dc6ee338dd6164a436523ad5a142aa163230f7cc 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/doOneWrapper_1.log
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/doOneWrapper_1.log
@@ -1,10 +1,11 @@
-[doOneWrapper (1)] Sun Aug 14 15:08:42 CEST 2022 : process 1 configured
-[doOneWrapper (1)] HOME=/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/HOME
+[doOneWrapper (1)] Fri Aug 19 23:32:14 CEST 2022 : process 1 configured
+[doOneWrapper (1)] HOME=/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/HOME
 [doOneWrapper (1)] current process pid is 51
 [doOneWrapper (1)] run doOne as bmkuser
 
-[doOne (1)] Sun Aug 14 15:08:42 CEST 2022 starting in /results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1
-Sun Aug 14 15:08:42 2022       
+[doOne (1)] Fri Aug 19 23:32:14 CEST 2022 starting in /results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1
+[doOne (1)] EXTRA_ARGS='--both'
+Fri Aug 19 23:32:15 2022       
 +-----------------------------------------------------------------------------+
 | NVIDIA-SMI 515.65.01    Driver Version: 515.65.01    CUDA Version: 11.7     |
 |-------------------------------+----------------------+----------------------+
@@ -13,7 +14,7 @@ Sun Aug 14 15:08:42 2022
 |                               |                      |               MIG M. |
 |===============================+======================+======================|
 |   0  Tesla V100S-PCI...  Off  | 00000000:00:08.0 Off |                    0 |
-| N/A   33C    P0    36W / 250W |      0MiB / 32768MiB |      0%      Default |
+| N/A   32C    P0    36W / 250W |      0MiB / 32768MiB |      0%      Default |
 |                               |                      |                  N/A |
 +-------------------------------+----------------------+----------------------+
                                                                                
@@ -24,36 +25,36 @@ Sun Aug 14 15:08:42 2022
 |=============================================================================|
 |  No running processes found                                                 |
 +-----------------------------------------------------------------------------+
-[doOne (1)] Sun Aug 14 15:08:42 CEST 2022 madgraph4gpu not found in /bmk/build/madgraph4gpu: download it from github
-Cloning into '/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu'...
-Checking out files:  23% (3526/14797)   
Checking out files:  24% (3552/14797)   
Checking out files:  25% (3700/14797)   
Checking out files:  26% (3848/14797)   
Checking out files:  27% (3996/14797)   
Checking out files:  28% (4144/14797)   
Checking out files:  29% (4292/14797)   
Checking out files:  30% (4440/14797)   
Checking out files:  31% (4588/14797)   
Checking out files:  32% (4736/14797)   
Checking out files:  33% (4884/14797)   
Checking out files:  34% (5031/14797)   
Checking out files:  35% (5179/14797)   
Checking out files:  36% (5327/14797)   
Checking out files:  37% (5475/14797)   
Checking out files:  38% (5623/14797)   
Checking out files:  39% (5771/14797)   
Checking out files:  40% (5919/14797)   
Checking out files:  41% (6067/14797)   
Checking out files:  42% (6215/14797)   
Checking out files:  43% (6363/14797)   
Checking out files:  44% (6511/14797)   
Checking out files:  44% (6597/14797)   
Checking out files:  45% (6659/14797)   
Checking out files:  46% (6807/14797)   
Checking out files:  47% (6955/14797)   
Checking out files:  48% (7103/14797)   
Checking out files:  49% (7251/14797)   
Checking out files:  50% (7399/14797)   
Checking out files:  51% (7547/14797)   
Checking out files:  52% (7695/14797)   
Checking out files:  53% (7843/14797)   
Checking out files:  54% (7991/14797)   
Checking out files:  55% (8139/14797)   
Checking out files:  56% (8287/14797)   
Checking out files:  57% (8435/14797)   
Checking out files:  58% (8583/14797)   
Checking out files:  59% (8731/14797)   
Checking out files:  60% (8879/14797)   
Checking out files:  61% (9027/14797)   
Checking out files:  62% (9175/14797)   
Checking out files:  63% (9323/14797)   
Checking out files:  64% (9471/14797)   
Checking out files:  65% (9619/14797)   
Checking out files:  66% (9767/14797)   
Checking out files:  67% (9914/14797)   
Checking out files:  68% (10062/14797)   
Checking out files:  68% (10206/14797)   
Checking out files:  69% (10210/14797)   
Checking out files:  70% (10358/14797)   
Checking out files:  71% (10506/14797)   
Checking out files:  72% (10654/14797)   
Checking out files:  73% (10802/14797)   
Checking out files:  74% (10950/14797)   
Checking out files:  75% (11098/14797)   
Checking out files:  76% (11246/14797)   
Checking out files:  77% (11394/14797)   
Checking out files:  78% (11542/14797)   
Checking out files:  79% (11690/14797)   
Checking out files:  80% (11838/14797)   
Checking out files:  81% (11986/14797)   
Checking out files:  82% (12134/14797)   
Checking out files:  83% (12282/14797)   
Checking out files:  84% (12430/14797)   
Checking out files:  85% (12578/14797)   
Checking out files:  86% (12726/14797)   
Checking out files:  87% (12874/14797)   
Checking out files:  88% (13022/14797)   
Checking out files:  89% (13170/14797)   
Checking out files:  90% (13318/14797)   
Checking out files:  91% (13466/14797)   
Checking out files:  91% (13567/14797)   
Checking out files:  92% (13614/14797)   
Checking out files:  93% (13762/14797)   
Checking out files:  94% (13910/14797)   
Checking out files:  95% (14058/14797)   
Checking out files:  96% (14206/14797)   
Checking out files:  97% (14354/14797)   
Checking out files:  98% (14502/14797)   
Checking out files:  99% (14650/14797)   
Checking out files: 100% (14797/14797)   
Checking out files: 100% (14797/14797), done.
+[doOne (1)] Fri Aug 19 23:32:15 CEST 2022 madgraph4gpu not found in /bmk/build/madgraph4gpu: download it from github
+Cloning into '/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu'...
+Checking out files:  18% (2763/14797)   
Checking out files:  19% (2812/14797)   
Checking out files:  20% (2960/14797)   
Checking out files:  21% (3108/14797)   
Checking out files:  22% (3256/14797)   
Checking out files:  23% (3404/14797)   
Checking out files:  24% (3552/14797)   
Checking out files:  25% (3700/14797)   
Checking out files:  26% (3848/14797)   
Checking out files:  27% (3996/14797)   
Checking out files:  28% (4144/14797)   
Checking out files:  29% (4292/14797)   
Checking out files:  30% (4440/14797)   
Checking out files:  31% (4588/14797)   
Checking out files:  32% (4736/14797)   
Checking out files:  33% (4884/14797)   
Checking out files:  34% (5031/14797)   
Checking out files:  35% (5179/14797)   
Checking out files:  36% (5327/14797)   
Checking out files:  37% (5475/14797)   
Checking out files:  38% (5623/14797)   
Checking out files:  39% (5771/14797)   
Checking out files:  39% (5886/14797)   
Checking out files:  40% (5919/14797)   
Checking out files:  41% (6067/14797)   
Checking out files:  42% (6215/14797)   
Checking out files:  43% (6363/14797)   
Checking out files:  44% (6511/14797)   
Checking out files:  45% (6659/14797)   
Checking out files:  46% (6807/14797)   
Checking out files:  47% (6955/14797)   
Checking out files:  48% (7103/14797)   
Checking out files:  49% (7251/14797)   
Checking out files:  50% (7399/14797)   
Checking out files:  51% (7547/14797)   
Checking out files:  52% (7695/14797)   
Checking out files:  53% (7843/14797)   
Checking out files:  54% (7991/14797)   
Checking out files:  55% (8139/14797)   
Checking out files:  56% (8287/14797)   
Checking out files:  57% (8435/14797)   
Checking out files:  58% (8583/14797)   
Checking out files:  59% (8731/14797)   
Checking out files:  60% (8879/14797)   
Checking out files:  61% (9027/14797)   
Checking out files:  62% (9175/14797)   
Checking out files:  63% (9323/14797)   
Checking out files:  64% (9471/14797)   
Checking out files:  65% (9619/14797)   
Checking out files:  66% (9767/14797)   
Checking out files:  67% (9914/14797)   
Checking out files:  68% (10062/14797)   
Checking out files:  69% (10210/14797)   
Checking out files:  70% (10358/14797)   
Checking out files:  71% (10506/14797)   
Checking out files:  72% (10654/14797)   
Checking out files:  73% (10802/14797)   
Checking out files:  74% (10950/14797)   
Checking out files:  75% (11098/14797)   
Checking out files:  76% (11246/14797)   
Checking out files:  77% (11394/14797)   
Checking out files:  78% (11542/14797)   
Checking out files:  79% (11690/14797)   
Checking out files:  79% (11738/14797)   
Checking out files:  80% (11838/14797)   
Checking out files:  81% (11986/14797)   
Checking out files:  82% (12134/14797)   
Checking out files:  83% (12282/14797)   
Checking out files:  84% (12430/14797)   
Checking out files:  85% (12578/14797)   
Checking out files:  86% (12726/14797)   
Checking out files:  87% (12874/14797)   
Checking out files:  88% (13022/14797)   
Checking out files:  89% (13170/14797)   
Checking out files:  90% (13318/14797)   
Checking out files:  91% (13466/14797)   
Checking out files:  92% (13614/14797)   
Checking out files:  93% (13762/14797)   
Checking out files:  94% (13910/14797)   
Checking out files:  95% (14058/14797)   
Checking out files:  96% (14206/14797)   
Checking out files:  97% (14354/14797)   
Checking out files:  98% (14502/14797)   
Checking out files:  99% (14650/14797)   
Checking out files: 100% (14797/14797)   
Checking out files: 100% (14797/14797), done.
 HEAD is now at 2af1644 Merge pull request #240 from valassi/amd
-real	0m1.562s
-real	0m1.578s
-real	0m2.046s
-real	0m0.651s
-real	0m7.982s
-real	0m5.467s
-real	0m4.531s
-real	0m3.966s
-real	0m4.699s
-real	0m7.646s
-real	0m3.849s
-real	0m2.985s
-real	0m2.910s
-real	0m3.248s
-real	0m4.133s
-real	0m3.545s
-real	0m3.392s
-real	0m3.271s
-real	0m3.464s
-real	0m3.086s
-real	0m2.556s
-real	0m2.469s
-real	0m2.350s
-real	0m2.498s
-[doOne (1)] Sun Aug 14 15:19:37 CEST 2022 copy /results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu to /results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/build/madgraph4gpu
-[doOne (1)] Sun Aug 14 15:19:53 CEST 2022 completed (status=0)
+real	0m1.379s
+real	0m1.148s
+real	0m1.259s
+real	0m1.070s
+real	0m7.358s
+real	0m5.046s
+real	0m3.870s
+real	0m3.822s
+real	0m4.291s
+real	0m7.261s
+real	0m3.449s
+real	0m2.770s
+real	0m2.719s
+real	0m2.883s
+real	0m3.981s
+real	0m3.364s
+real	0m3.047s
+real	0m3.007s
+real	0m3.269s
+real	0m2.981s
+real	0m2.421s
+real	0m2.308s
+real	0m2.270s
+real	0m2.379s
+[doOne (1)] Fri Aug 19 23:40:02 CEST 2022 copy /results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu to /results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/build/madgraph4gpu
+[doOne (1)] Fri Aug 19 23:40:03 CEST 2022 completed (status=0)
 
-[doOneWrapper (1)] Sun Aug 14 15:19:53 CEST 2022 : process 1 (pid=51) completed ok
+[doOneWrapper (1)] Fri Aug 19 23:40:03 CEST 2022 : process 1 (pid=51) completed ok
 
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/out_1.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/out_1.log
index 08b400f302de3f94d9ea2891ada783121cfc45f2..4a9d353eaa1f88fcf8a89a64f25011e38e681e89 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/out_1.log
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_1/out_1.log
@@ -1,4 +1,4 @@
-/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/SubProcesses/P1_Sigma_sm_epem_mupmum
+/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/SubProcesses/P1_Sigma_sm_epem_mupmum
 AVX=none
 FPTYPE=d
 HELINL=0
@@ -6,14 +6,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.none_d_inl0 for tag=none_d_inl0_curdev
 make -C ../../src 
 AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_d_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl0/rambo.o
 ar cru ../lib/build.none_d_inl0/libmodel_sm.a build.none_d_inl0/Parameters_sm.o build.none_d_inl0/read_slha.o build.none_d_inl0/rambo.o 
 ranlib ../lib/build.none_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_d_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl0/gCPPProcess.o
@@ -29,14 +29,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.sse4_d_inl0 for tag=sse4_d_inl0_curdev
 make -C ../../src 
 AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_d_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl0/rambo.o
 ar cru ../lib/build.sse4_d_inl0/libmodel_sm.a build.sse4_d_inl0/Parameters_sm.o build.sse4_d_inl0/read_slha.o build.sse4_d_inl0/rambo.o 
 ranlib ../lib/build.sse4_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_d_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl0/gCPPProcess.o
@@ -52,14 +52,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.avx2_d_inl0 for tag=avx2_d_inl0_curdev
 make -C ../../src 
 AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_d_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl0/rambo.o
 ar cru ../lib/build.avx2_d_inl0/libmodel_sm.a build.avx2_d_inl0/Parameters_sm.o build.avx2_d_inl0/read_slha.o build.avx2_d_inl0/rambo.o 
 ranlib ../lib/build.avx2_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_d_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl0/gCPPProcess.o
@@ -75,14 +75,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512y_d_inl0 for tag=512y_d_inl0_curdev
 make -C ../../src 
 AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_d_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl0/rambo.o
 ar cru ../lib/build.512y_d_inl0/libmodel_sm.a build.512y_d_inl0/Parameters_sm.o build.512y_d_inl0/read_slha.o build.512y_d_inl0/rambo.o 
 ranlib ../lib/build.512y_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_d_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl0/gCPPProcess.o
@@ -98,14 +98,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512z_d_inl0 for tag=512z_d_inl0_curdev
 make -C ../../src 
 AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_d_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl0/rambo.o
 ar cru ../lib/build.512z_d_inl0/libmodel_sm.a build.512z_d_inl0/Parameters_sm.o build.512z_d_inl0/read_slha.o build.512z_d_inl0/rambo.o 
 ranlib ../lib/build.512z_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_d_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl0/gCPPProcess.o
@@ -121,14 +121,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.none_f_inl0 for tag=none_f_inl0_curdev
 make -C ../../src 
 AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_f_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl0/rambo.o
 ar cru ../lib/build.none_f_inl0/libmodel_sm.a build.none_f_inl0/Parameters_sm.o build.none_f_inl0/read_slha.o build.none_f_inl0/rambo.o 
 ranlib ../lib/build.none_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_f_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl0/gCPPProcess.o
@@ -144,14 +144,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.sse4_f_inl0 for tag=sse4_f_inl0_curdev
 make -C ../../src 
 AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_f_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl0/rambo.o
 ar cru ../lib/build.sse4_f_inl0/libmodel_sm.a build.sse4_f_inl0/Parameters_sm.o build.sse4_f_inl0/read_slha.o build.sse4_f_inl0/rambo.o 
 ranlib ../lib/build.sse4_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_f_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl0/gCPPProcess.o
@@ -167,14 +167,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.avx2_f_inl0 for tag=avx2_f_inl0_curdev
 make -C ../../src 
 AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_f_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl0/rambo.o
 ar cru ../lib/build.avx2_f_inl0/libmodel_sm.a build.avx2_f_inl0/Parameters_sm.o build.avx2_f_inl0/read_slha.o build.avx2_f_inl0/rambo.o 
 ranlib ../lib/build.avx2_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_f_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl0/gCPPProcess.o
@@ -190,14 +190,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512y_f_inl0 for tag=512y_f_inl0_curdev
 make -C ../../src 
 AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_f_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl0/rambo.o
 ar cru ../lib/build.512y_f_inl0/libmodel_sm.a build.512y_f_inl0/Parameters_sm.o build.512y_f_inl0/read_slha.o build.512y_f_inl0/rambo.o 
 ranlib ../lib/build.512y_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_f_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl0/gCPPProcess.o
@@ -213,14 +213,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512z_f_inl0 for tag=512z_f_inl0_curdev
 make -C ../../src 
 AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_f_inl0
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl0/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl0/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl0/rambo.o
 ar cru ../lib/build.512z_f_inl0/libmodel_sm.a build.512z_f_inl0/Parameters_sm.o build.512z_f_inl0/read_slha.o build.512z_f_inl0/rambo.o 
 ranlib ../lib/build.512z_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_f_inl0
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl0/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl0/gCPPProcess.o
@@ -236,14 +236,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.none_d_inl1 for tag=none_d_inl1_curdev
 make -C ../../src 
 AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_d_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl1/rambo.o
 ar cru ../lib/build.none_d_inl1/libmodel_sm.a build.none_d_inl1/Parameters_sm.o build.none_d_inl1/read_slha.o build.none_d_inl1/rambo.o 
 ranlib ../lib/build.none_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_d_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl1/gCPPProcess.o
@@ -259,14 +259,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.sse4_d_inl1 for tag=sse4_d_inl1_curdev
 make -C ../../src 
 AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_d_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl1/rambo.o
 ar cru ../lib/build.sse4_d_inl1/libmodel_sm.a build.sse4_d_inl1/Parameters_sm.o build.sse4_d_inl1/read_slha.o build.sse4_d_inl1/rambo.o 
 ranlib ../lib/build.sse4_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_d_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl1/gCPPProcess.o
@@ -282,14 +282,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.avx2_d_inl1 for tag=avx2_d_inl1_curdev
 make -C ../../src 
 AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_d_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl1/rambo.o
 ar cru ../lib/build.avx2_d_inl1/libmodel_sm.a build.avx2_d_inl1/Parameters_sm.o build.avx2_d_inl1/read_slha.o build.avx2_d_inl1/rambo.o 
 ranlib ../lib/build.avx2_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_d_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl1/gCPPProcess.o
@@ -305,14 +305,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512y_d_inl1 for tag=512y_d_inl1_curdev
 make -C ../../src 
 AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_d_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl1/rambo.o
 ar cru ../lib/build.512y_d_inl1/libmodel_sm.a build.512y_d_inl1/Parameters_sm.o build.512y_d_inl1/read_slha.o build.512y_d_inl1/rambo.o 
 ranlib ../lib/build.512y_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_d_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl1/gCPPProcess.o
@@ -328,14 +328,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512z_d_inl1 for tag=512z_d_inl1_curdev
 make -C ../../src 
 AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_d_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl1/rambo.o
 ar cru ../lib/build.512z_d_inl1/libmodel_sm.a build.512z_d_inl1/Parameters_sm.o build.512z_d_inl1/read_slha.o build.512z_d_inl1/rambo.o 
 ranlib ../lib/build.512z_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_d_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl1/gCPPProcess.o
@@ -351,14 +351,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.none_f_inl1 for tag=none_f_inl1_curdev
 make -C ../../src 
 AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_f_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl1/rambo.o
 ar cru ../lib/build.none_f_inl1/libmodel_sm.a build.none_f_inl1/Parameters_sm.o build.none_f_inl1/read_slha.o build.none_f_inl1/rambo.o 
 ranlib ../lib/build.none_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.none_f_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl1/gCPPProcess.o
@@ -374,14 +374,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.sse4_f_inl1 for tag=sse4_f_inl1_curdev
 make -C ../../src 
 AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_f_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl1/rambo.o
 ar cru ../lib/build.sse4_f_inl1/libmodel_sm.a build.sse4_f_inl1/Parameters_sm.o build.sse4_f_inl1/read_slha.o build.sse4_f_inl1/rambo.o 
 ranlib ../lib/build.sse4_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.sse4_f_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl1/gCPPProcess.o
@@ -397,14 +397,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.avx2_f_inl1 for tag=avx2_f_inl1_curdev
 make -C ../../src 
 AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_f_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl1/rambo.o
 ar cru ../lib/build.avx2_f_inl1/libmodel_sm.a build.avx2_f_inl1/Parameters_sm.o build.avx2_f_inl1/read_slha.o build.avx2_f_inl1/rambo.o 
 ranlib ../lib/build.avx2_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.avx2_f_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl1/gCPPProcess.o
@@ -420,14 +420,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512y_f_inl1 for tag=512y_f_inl1_curdev
 make -C ../../src 
 AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_f_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl1/rambo.o
 ar cru ../lib/build.512y_f_inl1/libmodel_sm.a build.512y_f_inl1/Parameters_sm.o build.512y_f_inl1/read_slha.o build.512y_f_inl1/rambo.o 
 ranlib ../lib/build.512y_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512y_f_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl1/gCPPProcess.o
@@ -443,14 +443,14 @@ RNDGEN=curdev
 Building in BUILDDIR=build.512z_f_inl1 for tag=512z_f_inl1_curdev
 make -C ../../src 
 AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_f_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl1/Parameters_sm.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl1/read_slha.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl1/rambo.o
 ar cru ../lib/build.512z_f_inl1/libmodel_sm.a build.512z_f_inl1/Parameters_sm.o build.512z_f_inl1/read_slha.o build.512z_f_inl1/rambo.o 
 ranlib ../lib/build.512z_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
+make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c1-e12-1660944734_1925/proc_1/madgraph4gpu/epoch1/cuda/ee_mumu/src'
 mkdir build.512z_f_inl1
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl1/gcheck.o
 /usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl1/gCPPProcess.o
@@ -459,45 +459,45 @@ mkdir build.512z_f_inl1
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_f_inl1/CPPProcess.o
 /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_f_inl1/check.o -o build.512z_f_inl1/check.exe build.512z_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
 
-On a114b5198179 [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: 1x Tesla V100S-PCIE-32GB]:
+On 536fbf720e51 [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: 1x Tesla V100S-PCIE-32GB]:
 =========================================================================
 Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 1.999102e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.129235e+09                 )  sec^-1
+EvtsPerSec[MatrixElems] (3) = ( 6.728480e+08                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.336788e+09                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     1.425864 sec
-real	0m1.562s
+TOTAL       :     1.074115 sec
+real	0m1.379s
 ==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
 ==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
 FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 4.833345e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 2.287084e+09                 )  sec^-1
+EvtsPerSec[MatrixElems] (3) = ( 1.462891e+09                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 3.259709e+09                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     1.329920 sec
-real	0m1.578s
+TOTAL       :     0.852491 sec
+real	0m1.148s
 ==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
 ==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 2.410485e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.073924e+09                 )  sec^-1
+EvtsPerSec[MatrixElems] (3) = ( 6.757855e+08                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.355664e+09                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     1.688511 sec
-real	0m2.046s
+TOTAL       :     0.952203 sec
+real	0m1.259s
 ==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
 ==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
 FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 1.175276e+09                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 2.263929e+09                 )  sec^-1
+EvtsPerSec[MatrixElems] (3) = ( 1.458574e+09                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 3.260011e+09                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     0.515417 sec
-real	0m0.651s
+TOTAL       :     0.778860 sec
+real	0m1.070s
 ==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
 ==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
 =========================================================================
@@ -505,199 +505,199 @@ Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.238398e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.305462e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     7.923429 sec
-real	0m7.982s
+TOTAL       :     7.334805 sec
+real	0m7.358s
 =Symbols in CPPProcess.o= (~sse4:  638) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.403727e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 2.534760e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     5.395669 sec
-real	0m5.467s
+TOTAL       :     5.023391 sec
+real	0m5.046s
 =Symbols in CPPProcess.o= (~sse4: 3291) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.019734e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 4.504376e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.405888 sec
-real	0m4.531s
+TOTAL       :     3.846931 sec
+real	0m3.870s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2792) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.679371e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 4.630512e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.910912 sec
-real	0m3.966s
+TOTAL       :     3.799703 sec
+real	0m3.822s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2690) (512y:   51) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 3.200657e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 3.476466e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.612186 sec
-real	0m4.699s
+TOTAL       :     4.268202 sec
+real	0m4.291s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1283) (512y:   64) (512z: 2125)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = FLOAT (NaN/abnormal=6, zero=0)
 Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.136825e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.201576e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     7.603291 sec
-real	0m7.646s
+TOTAL       :     7.246822 sec
+real	0m7.261s
 =Symbols in CPPProcess.o= (~sse4:  584) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = FLOAT (NaN/abnormal=6, zero=0)
 Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.133596e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 4.524210e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     3.698524 sec
-real	0m3.849s
+TOTAL       :     3.434261 sec
+real	0m3.449s
 =Symbols in CPPProcess.o= (~sse4: 3974) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = FLOAT (NaN/abnormal=5, zero=0)
 Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.738654e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 8.243796e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.947348 sec
-real	0m2.985s
+TOTAL       :     2.755453 sec
+real	0m2.770s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3130) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = FLOAT (NaN/abnormal=5, zero=0)
 Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 8.485874e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 8.860381e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.878749 sec
-real	0m2.910s
+TOTAL       :     2.704251 sec
+real	0m2.719s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3027) (512y:   26) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
 FP precision                = FLOAT (NaN/abnormal=5, zero=0)
 Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.467721e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 7.116999e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     3.189566 sec
-real	0m3.248s
+TOTAL       :     2.867407 sec
+real	0m2.883s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1724) (512y:   13) (512z: 2235)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.204699e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 4.325891e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.092482 sec
-real	0m4.133s
+TOTAL       :     3.958712 sec
+real	0m3.981s
 =Symbols in CPPProcess.o= (~sse4:  161) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.272770e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 7.664534e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.483759 sec
-real	0m3.545s
+TOTAL       :     3.341739 sec
+real	0m3.364s
 =Symbols in CPPProcess.o= (~sse4:  553) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 8.513935e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.070285e+07                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.355321 sec
-real	0m3.392s
+TOTAL       :     3.024524 sec
+real	0m3.047s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2:  571) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.047300e+07                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.162337e+07                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.189209 sec
-real	0m3.271s
+TOTAL       :     2.983950 sec
+real	0m3.007s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2:  503) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
 Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.491942e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 8.019675e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.401142 sec
-real	0m3.464s
+TOTAL       :     3.245718 sec
+real	0m3.269s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2:  165) (512y:    0) (512z:  350)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = FLOAT (NaN/abnormal=6, zero=0)
 Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.411111e+06                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 6.560495e+06                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     3.052499 sec
-real	0m3.086s
+TOTAL       :     2.966211 sec
+real	0m2.981s
 =Symbols in CPPProcess.o= (~sse4:  140) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = FLOAT (NaN/abnormal=6, zero=0)
 Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.568431e+07                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.650006e+07                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     2.514636 sec
-real	0m2.556s
+TOTAL       :     2.406755 sec
+real	0m2.421s
 =Symbols in CPPProcess.o= (~sse4:  634) (avx2:    0) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = FLOAT (NaN/abnormal=5, zero=0)
 Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.159980e+07                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 2.387562e+07                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.392792 sec
-real	0m2.469s
+TOTAL       :     2.292842 sec
+real	0m2.308s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2:  643) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = FLOAT (NaN/abnormal=5, zero=0)
 Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.245894e+07                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 2.488656e+07                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.331306 sec
-real	0m2.350s
+TOTAL       :     2.255883 sec
+real	0m2.270s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2:  572) (512y:    0) (512z:    0)
 -------------------------------------------------------------------------
 Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
 FP precision                = FLOAT (NaN/abnormal=5, zero=0)
 Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
 OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.531024e+07                 )  sec^-1
+EvtsPerSec[MECalcOnly] (3a) = ( 1.689948e+07                 )  sec^-1
 MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     2.466773 sec
-real	0m2.498s
+TOTAL       :     2.364481 sec
+real	0m2.379s
 =Symbols in CPPProcess.o= (~sse4:    0) (avx2:  240) (512y:    0) (512z:  375)
 =========================================================================
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_2/doOneWrapper_2.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_2/doOneWrapper_2.log
deleted file mode 100644
index 21d3ab2fee5e50262aeba8d244975c95455d192c..0000000000000000000000000000000000000000
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_2/doOneWrapper_2.log
+++ /dev/null
@@ -1,58 +0,0 @@
-[doOneWrapper (2)] Sun Aug 14 15:08:42 CEST 2022 : process 2 configured
-[doOneWrapper (2)] HOME=/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/HOME
-[doOneWrapper (2)] current process pid is 86
-[doOneWrapper (2)] run doOne as bmkuser
-
-[doOne (2)] Sun Aug 14 15:08:42 CEST 2022 starting in /results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2
-Sun Aug 14 15:08:42 2022       
-+-----------------------------------------------------------------------------+
-| NVIDIA-SMI 515.65.01    Driver Version: 515.65.01    CUDA Version: 11.7     |
-|-------------------------------+----------------------+----------------------+
-| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
-| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
-|                               |                      |               MIG M. |
-|===============================+======================+======================|
-|   0  Tesla V100S-PCI...  Off  | 00000000:00:08.0 Off |                    0 |
-| N/A   33C    P0    36W / 250W |      0MiB / 32768MiB |      0%      Default |
-|                               |                      |                  N/A |
-+-------------------------------+----------------------+----------------------+
-                                                                               
-+-----------------------------------------------------------------------------+
-| Processes:                                                                  |
-|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
-|        ID   ID                                                   Usage      |
-|=============================================================================|
-|  No running processes found                                                 |
-+-----------------------------------------------------------------------------+
-[doOne (2)] Sun Aug 14 15:08:42 CEST 2022 madgraph4gpu not found in /bmk/build/madgraph4gpu: download it from github
-Cloning into '/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu'...
-Checking out files:  23% (3530/14797)   
Checking out files:  24% (3552/14797)   
Checking out files:  25% (3700/14797)   
Checking out files:  26% (3848/14797)   
Checking out files:  27% (3996/14797)   
Checking out files:  28% (4144/14797)   
Checking out files:  29% (4292/14797)   
Checking out files:  30% (4440/14797)   
Checking out files:  31% (4588/14797)   
Checking out files:  32% (4736/14797)   
Checking out files:  33% (4884/14797)   
Checking out files:  34% (5031/14797)   
Checking out files:  35% (5179/14797)   
Checking out files:  36% (5327/14797)   
Checking out files:  37% (5475/14797)   
Checking out files:  38% (5623/14797)   
Checking out files:  39% (5771/14797)   
Checking out files:  40% (5919/14797)   
Checking out files:  41% (6067/14797)   
Checking out files:  42% (6215/14797)   
Checking out files:  43% (6363/14797)   
Checking out files:  44% (6511/14797)   
Checking out files:  45% (6659/14797)   
Checking out files:  46% (6807/14797)   
Checking out files:  47% (6955/14797)   
Checking out files:  48% (7103/14797)   
Checking out files:  49% (7251/14797)   
Checking out files:  50% (7399/14797)   
Checking out files:  51% (7547/14797)   
Checking out files:  52% (7695/14797)   
Checking out files:  52% (7780/14797)   
Checking out files:  53% (7843/14797)   
Checking out files:  54% (7991/14797)   
Checking out files:  55% (8139/14797)   
Checking out files:  56% (8287/14797)   
Checking out files:  57% (8435/14797)   
Checking out files:  58% (8583/14797)   
Checking out files:  59% (8731/14797)   
Checking out files:  60% (8879/14797)   
Checking out files:  61% (9027/14797)   
Checking out files:  62% (9175/14797)   
Checking out files:  63% (9323/14797)   
Checking out files:  64% (9471/14797)   
Checking out files:  65% (9619/14797)   
Checking out files:  66% (9767/14797)   
Checking out files:  67% (9914/14797)   
Checking out files:  68% (10062/14797)   
Checking out files:  69% (10210/14797)   
Checking out files:  70% (10358/14797)   
Checking out files:  71% (10506/14797)   
Checking out files:  72% (10654/14797)   
Checking out files:  73% (10802/14797)   
Checking out files:  74% (10950/14797)   
Checking out files:  75% (11098/14797)   
Checking out files:  76% (11246/14797)   
Checking out files:  77% (11394/14797)   
Checking out files:  78% (11542/14797)   
Checking out files:  79% (11690/14797)   
Checking out files:  79% (11736/14797)   
Checking out files:  80% (11838/14797)   
Checking out files:  81% (11986/14797)   
Checking out files:  82% (12134/14797)   
Checking out files:  83% (12282/14797)   
Checking out files:  84% (12430/14797)   
Checking out files:  85% (12578/14797)   
Checking out files:  86% (12726/14797)   
Checking out files:  87% (12874/14797)   
Checking out files:  88% (13022/14797)   
Checking out files:  89% (13170/14797)   
Checking out files:  90% (13318/14797)   
Checking out files:  91% (13466/14797)   
Checking out files:  92% (13614/14797)   
Checking out files:  93% (13762/14797)   
Checking out files:  94% (13910/14797)   
Checking out files:  95% (14058/14797)   
Checking out files:  96% (14206/14797)   
Checking out files:  97% (14354/14797)   
Checking out files:  98% (14502/14797)   
Checking out files:  99% (14650/14797)   
Checking out files: 100% (14797/14797)   
Checking out files: 100% (14797/14797), done.
-HEAD is now at 2af1644 Merge pull request #240 from valassi/amd
-real	0m2.462s
-real	0m1.347s
-real	0m1.939s
-real	0m0.951s
-real	0m7.865s
-real	0m5.360s
-real	0m4.348s
-real	0m4.283s
-real	0m5.149s
-real	0m7.563s
-real	0m4.049s
-real	0m2.960s
-real	0m2.979s
-real	0m3.202s
-real	0m4.170s
-real	0m3.509s
-real	0m3.172s
-real	0m3.146s
-real	0m3.427s
-real	0m3.092s
-real	0m2.484s
-real	0m2.404s
-real	0m2.363s
-real	0m2.498s
-[doOne (2)] Sun Aug 14 15:19:37 CEST 2022 completed (status=0)
-
-[doOneWrapper (2)] Sun Aug 14 15:19:37 CEST 2022 : process 2 (pid=86) completed ok
-
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_2/out_2.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_2/out_2.log
deleted file mode 100644
index b92ef475841b1b4f5f3a7ba9b70604547d6edf11..0000000000000000000000000000000000000000
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_2/out_2.log
+++ /dev/null
@@ -1,703 +0,0 @@
-/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/SubProcesses/P1_Sigma_sm_epem_mupmum
-AVX=none
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.none_d_inl0 for tag=none_d_inl0_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl0/rambo.o
-ar cru ../lib/build.none_d_inl0/libmodel_sm.a build.none_d_inl0/Parameters_sm.o build.none_d_inl0/read_slha.o build.none_d_inl0/rambo.o 
-ranlib ../lib/build.none_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_d_inl0/gcheck.o -o build.none_d_inl0/gcheck.exe build.none_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_d_inl0/check.o -o build.none_d_inl0/check.exe build.none_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_d_inl0 for tag=sse4_d_inl0_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl0/rambo.o
-ar cru ../lib/build.sse4_d_inl0/libmodel_sm.a build.sse4_d_inl0/Parameters_sm.o build.sse4_d_inl0/read_slha.o build.sse4_d_inl0/rambo.o 
-ranlib ../lib/build.sse4_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_d_inl0/gcheck.o -o build.sse4_d_inl0/gcheck.exe build.sse4_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_d_inl0/check.o -o build.sse4_d_inl0/check.exe build.sse4_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_d_inl0 for tag=avx2_d_inl0_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl0/rambo.o
-ar cru ../lib/build.avx2_d_inl0/libmodel_sm.a build.avx2_d_inl0/Parameters_sm.o build.avx2_d_inl0/read_slha.o build.avx2_d_inl0/rambo.o 
-ranlib ../lib/build.avx2_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_d_inl0/gcheck.o -o build.avx2_d_inl0/gcheck.exe build.avx2_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_d_inl0/check.o -o build.avx2_d_inl0/check.exe build.avx2_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_d_inl0 for tag=512y_d_inl0_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl0/rambo.o
-ar cru ../lib/build.512y_d_inl0/libmodel_sm.a build.512y_d_inl0/Parameters_sm.o build.512y_d_inl0/read_slha.o build.512y_d_inl0/rambo.o 
-ranlib ../lib/build.512y_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_d_inl0/gcheck.o -o build.512y_d_inl0/gcheck.exe build.512y_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_d_inl0/check.o -o build.512y_d_inl0/check.exe build.512y_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_d_inl0 for tag=512z_d_inl0_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl0/rambo.o
-ar cru ../lib/build.512z_d_inl0/libmodel_sm.a build.512z_d_inl0/Parameters_sm.o build.512z_d_inl0/read_slha.o build.512z_d_inl0/rambo.o 
-ranlib ../lib/build.512z_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_d_inl0/gcheck.o -o build.512z_d_inl0/gcheck.exe build.512z_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_d_inl0/check.o -o build.512z_d_inl0/check.exe build.512z_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.none_f_inl0 for tag=none_f_inl0_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl0/rambo.o
-ar cru ../lib/build.none_f_inl0/libmodel_sm.a build.none_f_inl0/Parameters_sm.o build.none_f_inl0/read_slha.o build.none_f_inl0/rambo.o 
-ranlib ../lib/build.none_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_f_inl0/gcheck.o -o build.none_f_inl0/gcheck.exe build.none_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_f_inl0/check.o -o build.none_f_inl0/check.exe build.none_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_f_inl0 for tag=sse4_f_inl0_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl0/rambo.o
-ar cru ../lib/build.sse4_f_inl0/libmodel_sm.a build.sse4_f_inl0/Parameters_sm.o build.sse4_f_inl0/read_slha.o build.sse4_f_inl0/rambo.o 
-ranlib ../lib/build.sse4_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_f_inl0/gcheck.o -o build.sse4_f_inl0/gcheck.exe build.sse4_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_f_inl0/check.o -o build.sse4_f_inl0/check.exe build.sse4_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_f_inl0 for tag=avx2_f_inl0_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl0/rambo.o
-ar cru ../lib/build.avx2_f_inl0/libmodel_sm.a build.avx2_f_inl0/Parameters_sm.o build.avx2_f_inl0/read_slha.o build.avx2_f_inl0/rambo.o 
-ranlib ../lib/build.avx2_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_f_inl0/gcheck.o -o build.avx2_f_inl0/gcheck.exe build.avx2_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_f_inl0/check.o -o build.avx2_f_inl0/check.exe build.avx2_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_f_inl0 for tag=512y_f_inl0_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl0/rambo.o
-ar cru ../lib/build.512y_f_inl0/libmodel_sm.a build.512y_f_inl0/Parameters_sm.o build.512y_f_inl0/read_slha.o build.512y_f_inl0/rambo.o 
-ranlib ../lib/build.512y_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_f_inl0/gcheck.o -o build.512y_f_inl0/gcheck.exe build.512y_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_f_inl0/check.o -o build.512y_f_inl0/check.exe build.512y_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_f_inl0 for tag=512z_f_inl0_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl0/rambo.o
-ar cru ../lib/build.512z_f_inl0/libmodel_sm.a build.512z_f_inl0/Parameters_sm.o build.512z_f_inl0/read_slha.o build.512z_f_inl0/rambo.o 
-ranlib ../lib/build.512z_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_f_inl0/gcheck.o -o build.512z_f_inl0/gcheck.exe build.512z_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_f_inl0/check.o -o build.512z_f_inl0/check.exe build.512z_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.none_d_inl1 for tag=none_d_inl1_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl1/rambo.o
-ar cru ../lib/build.none_d_inl1/libmodel_sm.a build.none_d_inl1/Parameters_sm.o build.none_d_inl1/read_slha.o build.none_d_inl1/rambo.o 
-ranlib ../lib/build.none_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_d_inl1/gcheck.o -o build.none_d_inl1/gcheck.exe build.none_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_d_inl1/check.o -o build.none_d_inl1/check.exe build.none_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_d_inl1 for tag=sse4_d_inl1_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl1/rambo.o
-ar cru ../lib/build.sse4_d_inl1/libmodel_sm.a build.sse4_d_inl1/Parameters_sm.o build.sse4_d_inl1/read_slha.o build.sse4_d_inl1/rambo.o 
-ranlib ../lib/build.sse4_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_d_inl1/gcheck.o -o build.sse4_d_inl1/gcheck.exe build.sse4_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_d_inl1/check.o -o build.sse4_d_inl1/check.exe build.sse4_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_d_inl1 for tag=avx2_d_inl1_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl1/rambo.o
-ar cru ../lib/build.avx2_d_inl1/libmodel_sm.a build.avx2_d_inl1/Parameters_sm.o build.avx2_d_inl1/read_slha.o build.avx2_d_inl1/rambo.o 
-ranlib ../lib/build.avx2_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_d_inl1/gcheck.o -o build.avx2_d_inl1/gcheck.exe build.avx2_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_d_inl1/check.o -o build.avx2_d_inl1/check.exe build.avx2_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_d_inl1 for tag=512y_d_inl1_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl1/rambo.o
-ar cru ../lib/build.512y_d_inl1/libmodel_sm.a build.512y_d_inl1/Parameters_sm.o build.512y_d_inl1/read_slha.o build.512y_d_inl1/rambo.o 
-ranlib ../lib/build.512y_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_d_inl1/gcheck.o -o build.512y_d_inl1/gcheck.exe build.512y_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_d_inl1/check.o -o build.512y_d_inl1/check.exe build.512y_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_d_inl1 for tag=512z_d_inl1_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl1/rambo.o
-ar cru ../lib/build.512z_d_inl1/libmodel_sm.a build.512z_d_inl1/Parameters_sm.o build.512z_d_inl1/read_slha.o build.512z_d_inl1/rambo.o 
-ranlib ../lib/build.512z_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_d_inl1/gcheck.o -o build.512z_d_inl1/gcheck.exe build.512z_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_d_inl1/check.o -o build.512z_d_inl1/check.exe build.512z_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.none_f_inl1 for tag=none_f_inl1_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl1/rambo.o
-ar cru ../lib/build.none_f_inl1/libmodel_sm.a build.none_f_inl1/Parameters_sm.o build.none_f_inl1/read_slha.o build.none_f_inl1/rambo.o 
-ranlib ../lib/build.none_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_f_inl1/gcheck.o -o build.none_f_inl1/gcheck.exe build.none_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_f_inl1/check.o -o build.none_f_inl1/check.exe build.none_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_f_inl1 for tag=sse4_f_inl1_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl1/rambo.o
-ar cru ../lib/build.sse4_f_inl1/libmodel_sm.a build.sse4_f_inl1/Parameters_sm.o build.sse4_f_inl1/read_slha.o build.sse4_f_inl1/rambo.o 
-ranlib ../lib/build.sse4_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_f_inl1/gcheck.o -o build.sse4_f_inl1/gcheck.exe build.sse4_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_f_inl1/check.o -o build.sse4_f_inl1/check.exe build.sse4_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_f_inl1 for tag=avx2_f_inl1_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl1/rambo.o
-ar cru ../lib/build.avx2_f_inl1/libmodel_sm.a build.avx2_f_inl1/Parameters_sm.o build.avx2_f_inl1/read_slha.o build.avx2_f_inl1/rambo.o 
-ranlib ../lib/build.avx2_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_f_inl1/gcheck.o -o build.avx2_f_inl1/gcheck.exe build.avx2_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_f_inl1/check.o -o build.avx2_f_inl1/check.exe build.avx2_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_f_inl1 for tag=512y_f_inl1_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl1/rambo.o
-ar cru ../lib/build.512y_f_inl1/libmodel_sm.a build.512y_f_inl1/Parameters_sm.o build.512y_f_inl1/read_slha.o build.512y_f_inl1/rambo.o 
-ranlib ../lib/build.512y_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_f_inl1/gcheck.o -o build.512y_f_inl1/gcheck.exe build.512y_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_f_inl1/check.o -o build.512y_f_inl1/check.exe build.512y_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_f_inl1 for tag=512z_f_inl1_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl1/rambo.o
-ar cru ../lib/build.512z_f_inl1/libmodel_sm.a build.512z_f_inl1/Parameters_sm.o build.512z_f_inl1/read_slha.o build.512z_f_inl1/rambo.o 
-ranlib ../lib/build.512z_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_2/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_f_inl1/gcheck.o -o build.512z_f_inl1/gcheck.exe build.512z_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_f_inl1/check.o -o build.512z_f_inl1/check.exe build.512z_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-On a114b5198179 [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: 1x Tesla V100S-PCIE-32GB]:
-=========================================================================
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 2.577970e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.085296e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     2.079372 sec
-real	0m2.462s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 6.266196e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 2.212423e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     1.181799 sec
-real	0m1.347s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 3.132475e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.106562e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     1.564972 sec
-real	0m1.939s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 6.140443e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.902236e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     0.862181 sec
-real	0m0.951s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
-=========================================================================
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.209736e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     7.820560 sec
-real	0m7.865s
-=Symbols in CPPProcess.o= (~sse4:  638) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.408937e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     5.294158 sec
-real	0m5.360s
-=Symbols in CPPProcess.o= (~sse4: 3291) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.071613e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.263748 sec
-real	0m4.348s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2792) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.185342e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.249330 sec
-real	0m4.283s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2690) (512y:   51) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.927987e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     5.111235 sec
-real	0m5.149s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1283) (512y:   64) (512z: 2125)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.166365e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     7.423350 sec
-real	0m7.563s
-=Symbols in CPPProcess.o= (~sse4:  584) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 3.954943e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     3.996941 sec
-real	0m4.049s
-=Symbols in CPPProcess.o= (~sse4: 3974) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.847353e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.911704 sec
-real	0m2.960s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3130) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 8.363706e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.952384 sec
-real	0m2.979s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3027) (512y:   26) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.583364e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     3.043804 sec
-real	0m3.202s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1724) (512y:   13) (512z: 2235)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.229994e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.085894 sec
-real	0m4.170s
-=Symbols in CPPProcess.o= (~sse4:  161) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.373351e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.451636 sec
-real	0m3.509s
-=Symbols in CPPProcess.o= (~sse4:  553) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.004802e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.138098 sec
-real	0m3.172s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  571) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.077007e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.106596 sec
-real	0m3.146s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  503) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.313937e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.371148 sec
-real	0m3.427s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  165) (512y:    0) (512z:  350)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.427162e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     3.038274 sec
-real	0m3.092s
-=Symbols in CPPProcess.o= (~sse4:  140) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.575460e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     2.454237 sec
-real	0m2.484s
-=Symbols in CPPProcess.o= (~sse4:  634) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.122606e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.355435 sec
-real	0m2.404s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  643) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.239033e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.340685 sec
-real	0m2.363s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  572) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.516877e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     2.468331 sec
-real	0m2.498s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  240) (512y:    0) (512z:  375)
-=========================================================================
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_3/doOneWrapper_3.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_3/doOneWrapper_3.log
deleted file mode 100644
index 1f274057221c4000250a0b70d07d9492c8ec4566..0000000000000000000000000000000000000000
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_3/doOneWrapper_3.log
+++ /dev/null
@@ -1,58 +0,0 @@
-[doOneWrapper (3)] Sun Aug 14 15:08:42 CEST 2022 : process 3 configured
-[doOneWrapper (3)] HOME=/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/HOME
-[doOneWrapper (3)] current process pid is 128
-[doOneWrapper (3)] run doOne as bmkuser
-
-[doOne (3)] Sun Aug 14 15:08:42 CEST 2022 starting in /results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3
-Sun Aug 14 15:08:42 2022       
-+-----------------------------------------------------------------------------+
-| NVIDIA-SMI 515.65.01    Driver Version: 515.65.01    CUDA Version: 11.7     |
-|-------------------------------+----------------------+----------------------+
-| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
-| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
-|                               |                      |               MIG M. |
-|===============================+======================+======================|
-|   0  Tesla V100S-PCI...  Off  | 00000000:00:08.0 Off |                    0 |
-| N/A   33C    P0    36W / 250W |      0MiB / 32768MiB |      0%      Default |
-|                               |                      |                  N/A |
-+-------------------------------+----------------------+----------------------+
-                                                                               
-+-----------------------------------------------------------------------------+
-| Processes:                                                                  |
-|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
-|        ID   ID                                                   Usage      |
-|=============================================================================|
-|  No running processes found                                                 |
-+-----------------------------------------------------------------------------+
-[doOne (3)] Sun Aug 14 15:08:42 CEST 2022 madgraph4gpu not found in /bmk/build/madgraph4gpu: download it from github
-Cloning into '/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu'...
-Checking out files:  20% (2994/14797)   
Checking out files:  21% (3108/14797)   
Checking out files:  22% (3256/14797)   
Checking out files:  23% (3404/14797)   
Checking out files:  24% (3552/14797)   
Checking out files:  25% (3700/14797)   
Checking out files:  26% (3848/14797)   
Checking out files:  27% (3996/14797)   
Checking out files:  28% (4144/14797)   
Checking out files:  29% (4292/14797)   
Checking out files:  30% (4440/14797)   
Checking out files:  31% (4588/14797)   
Checking out files:  32% (4736/14797)   
Checking out files:  33% (4884/14797)   
Checking out files:  34% (5031/14797)   
Checking out files:  35% (5179/14797)   
Checking out files:  36% (5327/14797)   
Checking out files:  37% (5475/14797)   
Checking out files:  38% (5623/14797)   
Checking out files:  39% (5771/14797)   
Checking out files:  40% (5919/14797)   
Checking out files:  41% (6067/14797)   
Checking out files:  42% (6215/14797)   
Checking out files:  43% (6363/14797)   
Checking out files:  44% (6511/14797)   
Checking out files:  44% (6643/14797)   
Checking out files:  45% (6659/14797)   
Checking out files:  46% (6807/14797)   
Checking out files:  47% (6955/14797)   
Checking out files:  48% (7103/14797)   
Checking out files:  49% (7251/14797)   
Checking out files:  50% (7399/14797)   
Checking out files:  51% (7547/14797)   
Checking out files:  52% (7695/14797)   
Checking out files:  53% (7843/14797)   
Checking out files:  54% (7991/14797)   
Checking out files:  55% (8139/14797)   
Checking out files:  56% (8287/14797)   
Checking out files:  57% (8435/14797)   
Checking out files:  58% (8583/14797)   
Checking out files:  59% (8731/14797)   
Checking out files:  60% (8879/14797)   
Checking out files:  61% (9027/14797)   
Checking out files:  62% (9175/14797)   
Checking out files:  63% (9323/14797)   
Checking out files:  64% (9471/14797)   
Checking out files:  65% (9619/14797)   
Checking out files:  66% (9767/14797)   
Checking out files:  66% (9838/14797)   
Checking out files:  67% (9914/14797)   
Checking out files:  68% (10062/14797)   
Checking out files:  69% (10210/14797)   
Checking out files:  70% (10358/14797)   
Checking out files:  71% (10506/14797)   
Checking out files:  72% (10654/14797)   
Checking out files:  73% (10802/14797)   
Checking out files:  74% (10950/14797)   
Checking out files:  75% (11098/14797)   
Checking out files:  76% (11246/14797)   
Checking out files:  77% (11394/14797)   
Checking out files:  78% (11542/14797)   
Checking out files:  79% (11690/14797)   
Checking out files:  80% (11838/14797)   
Checking out files:  81% (11986/14797)   
Checking out files:  82% (12134/14797)   
Checking out files:  83% (12282/14797)   
Checking out files:  84% (12430/14797)   
Checking out files:  85% (12578/14797)   
Checking out files:  86% (12726/14797)   
Checking out files:  87% (12874/14797)   
Checking out files:  88% (13022/14797)   
Checking out files:  88% (13080/14797)   
Checking out files:  89% (13170/14797)   
Checking out files:  90% (13318/14797)   
Checking out files:  91% (13466/14797)   
Checking out files:  92% (13614/14797)   
Checking out files:  93% (13762/14797)   
Checking out files:  94% (13910/14797)   
Checking out files:  95% (14058/14797)   
Checking out files:  96% (14206/14797)   
Checking out files:  97% (14354/14797)   
Checking out files:  98% (14502/14797)   
Checking out files:  99% (14650/14797)   
Checking out files: 100% (14797/14797)   
Checking out files: 100% (14797/14797), done.
-HEAD is now at 2af1644 Merge pull request #240 from valassi/amd
-real	0m1.536s
-real	0m1.395s
-real	0m1.349s
-real	0m0.925s
-real	0m7.880s
-real	0m6.056s
-real	0m4.020s
-real	0m4.280s
-real	0m4.570s
-real	0m7.948s
-real	0m3.736s
-real	0m2.981s
-real	0m3.008s
-real	0m3.070s
-real	0m4.118s
-real	0m3.577s
-real	0m3.206s
-real	0m3.137s
-real	0m3.448s
-real	0m3.107s
-real	0m2.573s
-real	0m2.454s
-real	0m2.341s
-real	0m2.434s
-[doOne (3)] Sun Aug 14 15:19:40 CEST 2022 completed (status=0)
-
-[doOneWrapper (3)] Sun Aug 14 15:19:40 CEST 2022 : process 3 (pid=128) completed ok
-
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_3/out_3.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_3/out_3.log
deleted file mode 100644
index 16ab0daa3bf72879dd7d09a39e7b467033dfd346..0000000000000000000000000000000000000000
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_3/out_3.log
+++ /dev/null
@@ -1,703 +0,0 @@
-/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/SubProcesses/P1_Sigma_sm_epem_mupmum
-AVX=none
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.none_d_inl0 for tag=none_d_inl0_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl0/rambo.o
-ar cru ../lib/build.none_d_inl0/libmodel_sm.a build.none_d_inl0/Parameters_sm.o build.none_d_inl0/read_slha.o build.none_d_inl0/rambo.o 
-ranlib ../lib/build.none_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_d_inl0/gcheck.o -o build.none_d_inl0/gcheck.exe build.none_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_d_inl0/check.o -o build.none_d_inl0/check.exe build.none_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_d_inl0 for tag=sse4_d_inl0_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl0/rambo.o
-ar cru ../lib/build.sse4_d_inl0/libmodel_sm.a build.sse4_d_inl0/Parameters_sm.o build.sse4_d_inl0/read_slha.o build.sse4_d_inl0/rambo.o 
-ranlib ../lib/build.sse4_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_d_inl0/gcheck.o -o build.sse4_d_inl0/gcheck.exe build.sse4_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_d_inl0/check.o -o build.sse4_d_inl0/check.exe build.sse4_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_d_inl0 for tag=avx2_d_inl0_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl0/rambo.o
-ar cru ../lib/build.avx2_d_inl0/libmodel_sm.a build.avx2_d_inl0/Parameters_sm.o build.avx2_d_inl0/read_slha.o build.avx2_d_inl0/rambo.o 
-ranlib ../lib/build.avx2_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_d_inl0/gcheck.o -o build.avx2_d_inl0/gcheck.exe build.avx2_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_d_inl0/check.o -o build.avx2_d_inl0/check.exe build.avx2_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_d_inl0 for tag=512y_d_inl0_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl0/rambo.o
-ar cru ../lib/build.512y_d_inl0/libmodel_sm.a build.512y_d_inl0/Parameters_sm.o build.512y_d_inl0/read_slha.o build.512y_d_inl0/rambo.o 
-ranlib ../lib/build.512y_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_d_inl0/gcheck.o -o build.512y_d_inl0/gcheck.exe build.512y_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_d_inl0/check.o -o build.512y_d_inl0/check.exe build.512y_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_d_inl0 for tag=512z_d_inl0_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl0/rambo.o
-ar cru ../lib/build.512z_d_inl0/libmodel_sm.a build.512z_d_inl0/Parameters_sm.o build.512z_d_inl0/read_slha.o build.512z_d_inl0/rambo.o 
-ranlib ../lib/build.512z_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_d_inl0/gcheck.o -o build.512z_d_inl0/gcheck.exe build.512z_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_d_inl0/check.o -o build.512z_d_inl0/check.exe build.512z_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.none_f_inl0 for tag=none_f_inl0_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl0/rambo.o
-ar cru ../lib/build.none_f_inl0/libmodel_sm.a build.none_f_inl0/Parameters_sm.o build.none_f_inl0/read_slha.o build.none_f_inl0/rambo.o 
-ranlib ../lib/build.none_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_f_inl0/gcheck.o -o build.none_f_inl0/gcheck.exe build.none_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_f_inl0/check.o -o build.none_f_inl0/check.exe build.none_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_f_inl0 for tag=sse4_f_inl0_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl0/rambo.o
-ar cru ../lib/build.sse4_f_inl0/libmodel_sm.a build.sse4_f_inl0/Parameters_sm.o build.sse4_f_inl0/read_slha.o build.sse4_f_inl0/rambo.o 
-ranlib ../lib/build.sse4_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_f_inl0/gcheck.o -o build.sse4_f_inl0/gcheck.exe build.sse4_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_f_inl0/check.o -o build.sse4_f_inl0/check.exe build.sse4_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_f_inl0 for tag=avx2_f_inl0_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl0/rambo.o
-ar cru ../lib/build.avx2_f_inl0/libmodel_sm.a build.avx2_f_inl0/Parameters_sm.o build.avx2_f_inl0/read_slha.o build.avx2_f_inl0/rambo.o 
-ranlib ../lib/build.avx2_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_f_inl0/gcheck.o -o build.avx2_f_inl0/gcheck.exe build.avx2_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_f_inl0/check.o -o build.avx2_f_inl0/check.exe build.avx2_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_f_inl0 for tag=512y_f_inl0_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl0/rambo.o
-ar cru ../lib/build.512y_f_inl0/libmodel_sm.a build.512y_f_inl0/Parameters_sm.o build.512y_f_inl0/read_slha.o build.512y_f_inl0/rambo.o 
-ranlib ../lib/build.512y_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_f_inl0/gcheck.o -o build.512y_f_inl0/gcheck.exe build.512y_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_f_inl0/check.o -o build.512y_f_inl0/check.exe build.512y_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_f_inl0 for tag=512z_f_inl0_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl0/rambo.o
-ar cru ../lib/build.512z_f_inl0/libmodel_sm.a build.512z_f_inl0/Parameters_sm.o build.512z_f_inl0/read_slha.o build.512z_f_inl0/rambo.o 
-ranlib ../lib/build.512z_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_f_inl0/gcheck.o -o build.512z_f_inl0/gcheck.exe build.512z_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_f_inl0/check.o -o build.512z_f_inl0/check.exe build.512z_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.none_d_inl1 for tag=none_d_inl1_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl1/rambo.o
-ar cru ../lib/build.none_d_inl1/libmodel_sm.a build.none_d_inl1/Parameters_sm.o build.none_d_inl1/read_slha.o build.none_d_inl1/rambo.o 
-ranlib ../lib/build.none_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_d_inl1/gcheck.o -o build.none_d_inl1/gcheck.exe build.none_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_d_inl1/check.o -o build.none_d_inl1/check.exe build.none_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_d_inl1 for tag=sse4_d_inl1_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl1/rambo.o
-ar cru ../lib/build.sse4_d_inl1/libmodel_sm.a build.sse4_d_inl1/Parameters_sm.o build.sse4_d_inl1/read_slha.o build.sse4_d_inl1/rambo.o 
-ranlib ../lib/build.sse4_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_d_inl1/gcheck.o -o build.sse4_d_inl1/gcheck.exe build.sse4_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_d_inl1/check.o -o build.sse4_d_inl1/check.exe build.sse4_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_d_inl1 for tag=avx2_d_inl1_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl1/rambo.o
-ar cru ../lib/build.avx2_d_inl1/libmodel_sm.a build.avx2_d_inl1/Parameters_sm.o build.avx2_d_inl1/read_slha.o build.avx2_d_inl1/rambo.o 
-ranlib ../lib/build.avx2_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_d_inl1/gcheck.o -o build.avx2_d_inl1/gcheck.exe build.avx2_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_d_inl1/check.o -o build.avx2_d_inl1/check.exe build.avx2_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_d_inl1 for tag=512y_d_inl1_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl1/rambo.o
-ar cru ../lib/build.512y_d_inl1/libmodel_sm.a build.512y_d_inl1/Parameters_sm.o build.512y_d_inl1/read_slha.o build.512y_d_inl1/rambo.o 
-ranlib ../lib/build.512y_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_d_inl1/gcheck.o -o build.512y_d_inl1/gcheck.exe build.512y_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_d_inl1/check.o -o build.512y_d_inl1/check.exe build.512y_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_d_inl1 for tag=512z_d_inl1_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl1/rambo.o
-ar cru ../lib/build.512z_d_inl1/libmodel_sm.a build.512z_d_inl1/Parameters_sm.o build.512z_d_inl1/read_slha.o build.512z_d_inl1/rambo.o 
-ranlib ../lib/build.512z_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_d_inl1/gcheck.o -o build.512z_d_inl1/gcheck.exe build.512z_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_d_inl1/check.o -o build.512z_d_inl1/check.exe build.512z_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.none_f_inl1 for tag=none_f_inl1_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl1/rambo.o
-ar cru ../lib/build.none_f_inl1/libmodel_sm.a build.none_f_inl1/Parameters_sm.o build.none_f_inl1/read_slha.o build.none_f_inl1/rambo.o 
-ranlib ../lib/build.none_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_f_inl1/gcheck.o -o build.none_f_inl1/gcheck.exe build.none_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_f_inl1/check.o -o build.none_f_inl1/check.exe build.none_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_f_inl1 for tag=sse4_f_inl1_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl1/rambo.o
-ar cru ../lib/build.sse4_f_inl1/libmodel_sm.a build.sse4_f_inl1/Parameters_sm.o build.sse4_f_inl1/read_slha.o build.sse4_f_inl1/rambo.o 
-ranlib ../lib/build.sse4_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_f_inl1/gcheck.o -o build.sse4_f_inl1/gcheck.exe build.sse4_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_f_inl1/check.o -o build.sse4_f_inl1/check.exe build.sse4_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_f_inl1 for tag=avx2_f_inl1_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl1/rambo.o
-ar cru ../lib/build.avx2_f_inl1/libmodel_sm.a build.avx2_f_inl1/Parameters_sm.o build.avx2_f_inl1/read_slha.o build.avx2_f_inl1/rambo.o 
-ranlib ../lib/build.avx2_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_f_inl1/gcheck.o -o build.avx2_f_inl1/gcheck.exe build.avx2_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_f_inl1/check.o -o build.avx2_f_inl1/check.exe build.avx2_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_f_inl1 for tag=512y_f_inl1_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl1/rambo.o
-ar cru ../lib/build.512y_f_inl1/libmodel_sm.a build.512y_f_inl1/Parameters_sm.o build.512y_f_inl1/read_slha.o build.512y_f_inl1/rambo.o 
-ranlib ../lib/build.512y_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_f_inl1/gcheck.o -o build.512y_f_inl1/gcheck.exe build.512y_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_f_inl1/check.o -o build.512y_f_inl1/check.exe build.512y_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_f_inl1 for tag=512z_f_inl1_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl1/rambo.o
-ar cru ../lib/build.512z_f_inl1/libmodel_sm.a build.512z_f_inl1/Parameters_sm.o build.512z_f_inl1/read_slha.o build.512z_f_inl1/rambo.o 
-ranlib ../lib/build.512z_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_3/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_f_inl1/gcheck.o -o build.512z_f_inl1/gcheck.exe build.512z_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_f_inl1/check.o -o build.512z_f_inl1/check.exe build.512z_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-On a114b5198179 [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: 1x Tesla V100S-PCIE-32GB]:
-=========================================================================
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 2.962831e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.105280e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     1.400650 sec
-real	0m1.536s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 8.778684e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 2.587464e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     1.033079 sec
-real	0m1.395s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 2.672479e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.097317e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     1.183585 sec
-real	0m1.349s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 7.583405e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 2.026961e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     0.854344 sec
-real	0m0.925s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
-=========================================================================
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.232785e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     7.829481 sec
-real	0m7.880s
-=Symbols in CPPProcess.o= (~sse4:  638) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.209872e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     5.957770 sec
-real	0m6.056s
-=Symbols in CPPProcess.o= (~sse4: 3291) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.321659e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.977917 sec
-real	0m4.020s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2792) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.195284e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.161716 sec
-real	0m4.280s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2690) (512y:   51) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 3.296903e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.527197 sec
-real	0m4.570s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1283) (512y:   64) (512z: 2125)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.120378e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     7.817854 sec
-real	0m7.948s
-=Symbols in CPPProcess.o= (~sse4:  584) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.368796e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     3.682062 sec
-real	0m3.736s
-=Symbols in CPPProcess.o= (~sse4: 3974) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.786969e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.917415 sec
-real	0m2.981s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3130) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 8.357556e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.898510 sec
-real	0m3.008s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3027) (512y:   26) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.716639e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     2.994031 sec
-real	0m3.070s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1724) (512y:   13) (512z: 2235)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.257542e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.073433 sec
-real	0m4.118s
-=Symbols in CPPProcess.o= (~sse4:  161) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.390008e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.525420 sec
-real	0m3.577s
-=Symbols in CPPProcess.o= (~sse4:  553) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 9.952651e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.169572 sec
-real	0m3.206s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  571) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.066017e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.085146 sec
-real	0m3.137s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  503) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.313752e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.388988 sec
-real	0m3.448s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  165) (512y:    0) (512z:  350)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.374023e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     3.071523 sec
-real	0m3.107s
-=Symbols in CPPProcess.o= (~sse4:  140) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.561494e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     2.553010 sec
-real	0m2.573s
-=Symbols in CPPProcess.o= (~sse4:  634) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.169132e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.384856 sec
-real	0m2.454s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  643) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.217435e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.322627 sec
-real	0m2.341s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  572) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.558979e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     2.417825 sec
-real	0m2.434s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  240) (512y:    0) (512z:  375)
-=========================================================================
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_4/doOneWrapper_4.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_4/doOneWrapper_4.log
deleted file mode 100644
index e7109ccb4ea59ecab1814cfa65a8aa97ae671dd8..0000000000000000000000000000000000000000
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_4/doOneWrapper_4.log
+++ /dev/null
@@ -1,58 +0,0 @@
-[doOneWrapper (4)] Sun Aug 14 15:08:42 CEST 2022 : process 4 configured
-[doOneWrapper (4)] HOME=/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/HOME
-[doOneWrapper (4)] current process pid is 177
-[doOneWrapper (4)] run doOne as bmkuser
-
-[doOne (4)] Sun Aug 14 15:08:42 CEST 2022 starting in /results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4
-Sun Aug 14 15:08:42 2022       
-+-----------------------------------------------------------------------------+
-| NVIDIA-SMI 515.65.01    Driver Version: 515.65.01    CUDA Version: 11.7     |
-|-------------------------------+----------------------+----------------------+
-| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
-| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
-|                               |                      |               MIG M. |
-|===============================+======================+======================|
-|   0  Tesla V100S-PCI...  Off  | 00000000:00:08.0 Off |                    0 |
-| N/A   33C    P0    36W / 250W |      0MiB / 32768MiB |      0%      Default |
-|                               |                      |                  N/A |
-+-------------------------------+----------------------+----------------------+
-                                                                               
-+-----------------------------------------------------------------------------+
-| Processes:                                                                  |
-|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
-|        ID   ID                                                   Usage      |
-|=============================================================================|
-|  No running processes found                                                 |
-+-----------------------------------------------------------------------------+
-[doOne (4)] Sun Aug 14 15:08:42 CEST 2022 madgraph4gpu not found in /bmk/build/madgraph4gpu: download it from github
-Cloning into '/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu'...
-Checking out files:  36% (5467/14797)   
Checking out files:  37% (5475/14797)   
Checking out files:  38% (5623/14797)   
Checking out files:  39% (5771/14797)   
Checking out files:  40% (5919/14797)   
Checking out files:  41% (6067/14797)   
Checking out files:  42% (6215/14797)   
Checking out files:  43% (6363/14797)   
Checking out files:  44% (6511/14797)   
Checking out files:  45% (6659/14797)   
Checking out files:  46% (6807/14797)   
Checking out files:  47% (6955/14797)   
Checking out files:  48% (7103/14797)   
Checking out files:  49% (7251/14797)   
Checking out files:  50% (7399/14797)   
Checking out files:  51% (7547/14797)   
Checking out files:  52% (7695/14797)   
Checking out files:  53% (7843/14797)   
Checking out files:  54% (7991/14797)   
Checking out files:  55% (8139/14797)   
Checking out files:  56% (8287/14797)   
Checking out files:  57% (8435/14797)   
Checking out files:  58% (8583/14797)   
Checking out files:  59% (8731/14797)   
Checking out files:  60% (8879/14797)   
Checking out files:  61% (9027/14797)   
Checking out files:  62% (9175/14797)   
Checking out files:  63% (9323/14797)   
Checking out files:  64% (9471/14797)   
Checking out files:  65% (9619/14797)   
Checking out files:  66% (9767/14797)   
Checking out files:  67% (9914/14797)   
Checking out files:  68% (10062/14797)   
Checking out files:  69% (10210/14797)   
Checking out files:  70% (10358/14797)   
Checking out files:  71% (10506/14797)   
Checking out files:  72% (10654/14797)   
Checking out files:  73% (10802/14797)   
Checking out files:  74% (10950/14797)   
Checking out files:  75% (11098/14797)   
Checking out files:  76% (11246/14797)   
Checking out files:  77% (11394/14797)   
Checking out files:  78% (11542/14797)   
Checking out files:  78% (11618/14797)   
Checking out files:  79% (11690/14797)   
Checking out files:  80% (11838/14797)   
Checking out files:  81% (11986/14797)   
Checking out files:  82% (12134/14797)   
Checking out files:  83% (12282/14797)   
Checking out files:  84% (12430/14797)   
Checking out files:  85% (12578/14797)   
Checking out files:  86% (12726/14797)   
Checking out files:  87% (12874/14797)   
Checking out files:  88% (13022/14797)   
Checking out files:  89% (13170/14797)   
Checking out files:  90% (13318/14797)   
Checking out files:  91% (13466/14797)   
Checking out files:  92% (13614/14797)   
Checking out files:  93% (13762/14797)   
Checking out files:  94% (13910/14797)   
Checking out files:  95% (14058/14797)   
Checking out files:  96% (14206/14797)   
Checking out files:  97% (14354/14797)   
Checking out files:  98% (14502/14797)   
Checking out files:  99% (14650/14797)   
Checking out files: 100% (14797/14797)   
Checking out files: 100% (14797/14797), done.
-HEAD is now at 2af1644 Merge pull request #240 from valassi/amd
-real	0m1.530s
-real	0m1.422s
-real	0m1.671s
-real	0m1.353s
-real	0m7.946s
-real	0m5.387s
-real	0m4.311s
-real	0m4.028s
-real	0m4.923s
-real	0m7.545s
-real	0m3.896s
-real	0m3.016s
-real	0m3.003s
-real	0m3.157s
-real	0m4.125s
-real	0m3.551s
-real	0m3.475s
-real	0m3.206s
-real	0m3.465s
-real	0m3.047s
-real	0m2.544s
-real	0m2.377s
-real	0m2.362s
-real	0m2.503s
-[doOne (4)] Sun Aug 14 15:19:37 CEST 2022 completed (status=0)
-
-[doOneWrapper (4)] Sun Aug 14 15:19:37 CEST 2022 : process 4 (pid=177) completed ok
-
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_4/out_4.log b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_4/out_4.log
deleted file mode 100644
index 6b9c7f2c4d08c7575dbdb0283a827be41c9d6811..0000000000000000000000000000000000000000
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/proc_4/out_4.log
+++ /dev/null
@@ -1,703 +0,0 @@
-/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/SubProcesses/P1_Sigma_sm_epem_mupmum
-AVX=none
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.none_d_inl0 for tag=none_d_inl0_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl0/rambo.o
-ar cru ../lib/build.none_d_inl0/libmodel_sm.a build.none_d_inl0/Parameters_sm.o build.none_d_inl0/read_slha.o build.none_d_inl0/rambo.o 
-ranlib ../lib/build.none_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_d_inl0/gcheck.o -o build.none_d_inl0/gcheck.exe build.none_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_d_inl0/check.o -o build.none_d_inl0/check.exe build.none_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_d_inl0 for tag=sse4_d_inl0_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl0/rambo.o
-ar cru ../lib/build.sse4_d_inl0/libmodel_sm.a build.sse4_d_inl0/Parameters_sm.o build.sse4_d_inl0/read_slha.o build.sse4_d_inl0/rambo.o 
-ranlib ../lib/build.sse4_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_d_inl0/gcheck.o -o build.sse4_d_inl0/gcheck.exe build.sse4_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_d_inl0/check.o -o build.sse4_d_inl0/check.exe build.sse4_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_d_inl0 for tag=avx2_d_inl0_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl0/rambo.o
-ar cru ../lib/build.avx2_d_inl0/libmodel_sm.a build.avx2_d_inl0/Parameters_sm.o build.avx2_d_inl0/read_slha.o build.avx2_d_inl0/rambo.o 
-ranlib ../lib/build.avx2_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_d_inl0/gcheck.o -o build.avx2_d_inl0/gcheck.exe build.avx2_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_d_inl0/check.o -o build.avx2_d_inl0/check.exe build.avx2_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_d_inl0 for tag=512y_d_inl0_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl0/rambo.o
-ar cru ../lib/build.512y_d_inl0/libmodel_sm.a build.512y_d_inl0/Parameters_sm.o build.512y_d_inl0/read_slha.o build.512y_d_inl0/rambo.o 
-ranlib ../lib/build.512y_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_d_inl0/gcheck.o -o build.512y_d_inl0/gcheck.exe build.512y_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_d_inl0/check.o -o build.512y_d_inl0/check.exe build.512y_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=d
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_d_inl0 for tag=512z_d_inl0_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl0/rambo.o
-ar cru ../lib/build.512z_d_inl0/libmodel_sm.a build.512z_d_inl0/Parameters_sm.o build.512z_d_inl0/read_slha.o build.512z_d_inl0/rambo.o 
-ranlib ../lib/build.512z_d_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_d_inl0/gcheck.o -o build.512z_d_inl0/gcheck.exe build.512z_d_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_d_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_d_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_d_inl0/check.o -o build.512z_d_inl0/check.exe build.512z_d_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_d_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.none_f_inl0 for tag=none_f_inl0_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl0/rambo.o
-ar cru ../lib/build.none_f_inl0/libmodel_sm.a build.none_f_inl0/Parameters_sm.o build.none_f_inl0/read_slha.o build.none_f_inl0/rambo.o 
-ranlib ../lib/build.none_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_f_inl0/gcheck.o -o build.none_f_inl0/gcheck.exe build.none_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_f_inl0/check.o -o build.none_f_inl0/check.exe build.none_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_f_inl0 for tag=sse4_f_inl0_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl0/rambo.o
-ar cru ../lib/build.sse4_f_inl0/libmodel_sm.a build.sse4_f_inl0/Parameters_sm.o build.sse4_f_inl0/read_slha.o build.sse4_f_inl0/rambo.o 
-ranlib ../lib/build.sse4_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_f_inl0/gcheck.o -o build.sse4_f_inl0/gcheck.exe build.sse4_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_f_inl0/check.o -o build.sse4_f_inl0/check.exe build.sse4_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_f_inl0 for tag=avx2_f_inl0_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl0/rambo.o
-ar cru ../lib/build.avx2_f_inl0/libmodel_sm.a build.avx2_f_inl0/Parameters_sm.o build.avx2_f_inl0/read_slha.o build.avx2_f_inl0/rambo.o 
-ranlib ../lib/build.avx2_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_f_inl0/gcheck.o -o build.avx2_f_inl0/gcheck.exe build.avx2_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_f_inl0/check.o -o build.avx2_f_inl0/check.exe build.avx2_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_f_inl0 for tag=512y_f_inl0_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl0/rambo.o
-ar cru ../lib/build.512y_f_inl0/libmodel_sm.a build.512y_f_inl0/Parameters_sm.o build.512y_f_inl0/read_slha.o build.512y_f_inl0/rambo.o 
-ranlib ../lib/build.512y_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_f_inl0/gcheck.o -o build.512y_f_inl0/gcheck.exe build.512y_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_f_inl0/check.o -o build.512y_f_inl0/check.exe build.512y_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=f
-HELINL=0
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_f_inl0 for tag=512z_f_inl0_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl0
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl0/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl0/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl0/rambo.o
-ar cru ../lib/build.512z_f_inl0/libmodel_sm.a build.512z_f_inl0/Parameters_sm.o build.512z_f_inl0/read_slha.o build.512z_f_inl0/rambo.o 
-ranlib ../lib/build.512z_f_inl0/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl0
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl0/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl0/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_f_inl0/gcheck.o -o build.512z_f_inl0/gcheck.exe build.512z_f_inl0/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_f_inl0/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_f_inl0/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_f_inl0/check.o -o build.512z_f_inl0/check.exe build.512z_f_inl0/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_f_inl0 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.none_d_inl1 for tag=none_d_inl1_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_d_inl1/rambo.o
-ar cru ../lib/build.none_d_inl1/libmodel_sm.a build.none_d_inl1/Parameters_sm.o build.none_d_inl1/read_slha.o build.none_d_inl1/rambo.o 
-ranlib ../lib/build.none_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_d_inl1/gcheck.o -o build.none_d_inl1/gcheck.exe build.none_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_d_inl1/check.o -o build.none_d_inl1/check.exe build.none_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_d_inl1 for tag=sse4_d_inl1_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_d_inl1/rambo.o
-ar cru ../lib/build.sse4_d_inl1/libmodel_sm.a build.sse4_d_inl1/Parameters_sm.o build.sse4_d_inl1/read_slha.o build.sse4_d_inl1/rambo.o 
-ranlib ../lib/build.sse4_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_d_inl1/gcheck.o -o build.sse4_d_inl1/gcheck.exe build.sse4_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_d_inl1/check.o -o build.sse4_d_inl1/check.exe build.sse4_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_d_inl1 for tag=avx2_d_inl1_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_d_inl1/rambo.o
-ar cru ../lib/build.avx2_d_inl1/libmodel_sm.a build.avx2_d_inl1/Parameters_sm.o build.avx2_d_inl1/read_slha.o build.avx2_d_inl1/rambo.o 
-ranlib ../lib/build.avx2_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_d_inl1/gcheck.o -o build.avx2_d_inl1/gcheck.exe build.avx2_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_d_inl1/check.o -o build.avx2_d_inl1/check.exe build.avx2_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_d_inl1 for tag=512y_d_inl1_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_d_inl1/rambo.o
-ar cru ../lib/build.512y_d_inl1/libmodel_sm.a build.512y_d_inl1/Parameters_sm.o build.512y_d_inl1/read_slha.o build.512y_d_inl1/rambo.o 
-ranlib ../lib/build.512y_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_d_inl1/gcheck.o -o build.512y_d_inl1/gcheck.exe build.512y_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_d_inl1/check.o -o build.512y_d_inl1/check.exe build.512y_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=d
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_d_inl1 for tag=512z_d_inl1_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_d_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_d_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_d_inl1/rambo.o
-ar cru ../lib/build.512z_d_inl1/libmodel_sm.a build.512z_d_inl1/Parameters_sm.o build.512z_d_inl1/read_slha.o build.512z_d_inl1/rambo.o 
-ranlib ../lib/build.512z_d_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_d_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_d_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_d_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_d_inl1/gcheck.o -o build.512z_d_inl1/gcheck.exe build.512z_d_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_d_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_d_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_d_inl1/check.o -o build.512z_d_inl1/check.exe build.512z_d_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_DOUBLE -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_d_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=none
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.none_f_inl1 for tag=none_f_inl1_curdev
-make -C ../../src 
-AVX=none
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.none_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.none_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.none_f_inl1/rambo.o
-ar cru ../lib/build.none_f_inl1/libmodel_sm.a build.none_f_inl1/Parameters_sm.o build.none_f_inl1/read_slha.o build.none_f_inl1/rambo.o 
-ranlib ../lib/build.none_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.none_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.none_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.none_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.none_f_inl1/gcheck.o -o build.none_f_inl1/gcheck.exe build.none_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.none_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.none_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.none_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.none_f_inl1/check.o -o build.none_f_inl1/check.exe build.none_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math   -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.none_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=sse4
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.sse4_f_inl1 for tag=sse4_f_inl1_curdev
-make -C ../../src 
-AVX=sse4
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.sse4_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.sse4_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.sse4_f_inl1/rambo.o
-ar cru ../lib/build.sse4_f_inl1/libmodel_sm.a build.sse4_f_inl1/Parameters_sm.o build.sse4_f_inl1/read_slha.o build.sse4_f_inl1/rambo.o 
-ranlib ../lib/build.sse4_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.sse4_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.sse4_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.sse4_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.sse4_f_inl1/gcheck.o -o build.sse4_f_inl1/gcheck.exe build.sse4_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.sse4_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.sse4_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.sse4_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.sse4_f_inl1/check.o -o build.sse4_f_inl1/check.exe build.sse4_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=nehalem  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.sse4_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=avx2
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.avx2_f_inl1 for tag=avx2_f_inl1_curdev
-make -C ../../src 
-AVX=avx2
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.avx2_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.avx2_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.avx2_f_inl1/rambo.o
-ar cru ../lib/build.avx2_f_inl1/libmodel_sm.a build.avx2_f_inl1/Parameters_sm.o build.avx2_f_inl1/read_slha.o build.avx2_f_inl1/rambo.o 
-ranlib ../lib/build.avx2_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.avx2_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.avx2_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.avx2_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.avx2_f_inl1/gcheck.o -o build.avx2_f_inl1/gcheck.exe build.avx2_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.avx2_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.avx2_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.avx2_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.avx2_f_inl1/check.o -o build.avx2_f_inl1/check.exe build.avx2_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=haswell  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.avx2_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512y
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512y_f_inl1 for tag=512y_f_inl1_curdev
-make -C ../../src 
-AVX=512y
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512y_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512y_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512y_f_inl1/rambo.o
-ar cru ../lib/build.512y_f_inl1/libmodel_sm.a build.512y_f_inl1/Parameters_sm.o build.512y_f_inl1/read_slha.o build.512y_f_inl1/rambo.o 
-ranlib ../lib/build.512y_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512y_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512y_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512y_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512y_f_inl1/gcheck.o -o build.512y_f_inl1/gcheck.exe build.512y_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512y_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512y_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512y_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512y_f_inl1/check.o -o build.512y_f_inl1/check.exe build.512y_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -mprefer-vector-width=256  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512y_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-AVX=512z
-FPTYPE=f
-HELINL=1
-RNDGEN=curdev
-Building in BUILDDIR=build.512z_f_inl1 for tag=512z_f_inl1_curdev
-make -C ../../src 
-AVX=512z
-make[1]: Entering directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl1
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c Parameters_sm.cc -o build.512z_f_inl1/Parameters_sm.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c read_slha.cc -o build.512z_f_inl1/read_slha.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3 -std=c++17 -I.  -Wall -Wshadow -Wextra -fopenmp  -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c rambo.cc -o build.512z_f_inl1/rambo.o
-ar cru ../lib/build.512z_f_inl1/libmodel_sm.a build.512z_f_inl1/Parameters_sm.o build.512z_f_inl1/read_slha.o build.512z_f_inl1/rambo.o 
-ranlib ../lib/build.512z_f_inl1/libmodel_sm.a
-make[1]: Leaving directory `/results/mg5amc-madgraph4gpu-c4-e12-1660482521_6582/proc_4/madgraph4gpu/epoch1/cuda/ee_mumu/src'
-mkdir build.512z_f_inl1
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gcheck.cu -o build.512z_f_inl1/gcheck.o
-/usr/local/cuda-11.6/bin/nvcc  -O3  -lineinfo -std=c++14 -I. -I../../src -I../../../../../tools -I/usr/local/cuda-11.6/include/ -DUSE_NVTX -arch=compute_70 -use_fast_math -ccbin /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -c gCPPProcess.cu -o build.512z_f_inl1/gCPPProcess.o
-/usr/local/cuda-11.6/bin/nvcc build.512z_f_inl1/gcheck.o -o build.512z_f_inl1/gcheck.exe build.512z_f_inl1/gCPPProcess.o -arch=compute_70 -L../../lib/build.512z_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c check.cc -o build.512z_f_inl1/check.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -I/usr/local/cuda-11.6/include/ -c CPPProcess.cc -o build.512z_f_inl1/CPPProcess.o
-/cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/bin/g++ build.512z_f_inl1/check.o -o build.512z_f_inl1/check.exe build.512z_f_inl1/CPPProcess.o  -O3  -std=c++17 -I. -I../../src -I../../../../../tools -DUSE_NVTX -Wall -Wshadow -Wextra -fopenmp -ffast-math  -march=skylake-avx512 -DMGONGPU_PVW512  -DMGONGPU_FPTYPE_FLOAT -DMGONGPU_INLINE_HELAMPS -DMGONGPU_CURAND_ONDEVICE -ldl -pthread -L../../lib/build.512z_f_inl1 -lmodel_sm -L/usr/local/cuda-11.6/lib64/ -lcurand 
-
-On a114b5198179 [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: 1x Tesla V100S-PCIE-32GB]:
-=========================================================================
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 3.657896e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.092919e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     1.405076 sec
-real	0m1.530s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 4.434958e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.363266e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     1.194377 sec
-real	0m1.422s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 2.858217e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 1.085343e+09                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270300e-06 )  GeV^0
-TOTAL       :     1.296410 sec
-real	0m1.671s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 122
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CUDA [nvcc 11.6.124 (gcc 10.2.0)] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=2, zero=0)
-EvtsPerSec[MatrixElems] (3) = ( 2.543406e+08                 )  sec^-1
-EvtsPerSec[MECalcOnly] (3a) = ( 4.072425e+08                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371687e-02 +- 3.270204e-06 )  GeV^0
-TOTAL       :     1.007192 sec
-real	0m1.353s
-==PROF== Profiling "sigmaKin": launch__registers_per_thread 48
-==PROF== Profiling "sigmaKin": sm__sass_average_branch_targets_threads_uniform.pct 100%
-=========================================================================
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.220826e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     7.898007 sec
-real	0m7.946s
-=Symbols in CPPProcess.o= (~sse4:  638) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.376095e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     5.288358 sec
-real	0m5.387s
-=Symbols in CPPProcess.o= (~sse4: 3291) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.091540e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.260385 sec
-real	0m4.311s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2792) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.670090e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.941023 sec
-real	0m4.028s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 2690) (512y:   51) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.932580e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.822003 sec
-real	0m4.923s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1283) (512y:   64) (512z: 2125)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.170390e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     7.453922 sec
-real	0m7.545s
-=Symbols in CPPProcess.o= (~sse4:  584) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 3.960673e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     3.826406 sec
-real	0m3.896s
-=Symbols in CPPProcess.o= (~sse4: 3974) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.796526e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.956571 sec
-real	0m3.016s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3130) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 8.065736e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.968650 sec
-real	0m3.003s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 3027) (512y:   26) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=0]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.643695e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     3.091892 sec
-real	0m3.157s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2: 1724) (512y:   13) (512z: 2235)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 4.244237e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     4.084002 sec
-real	0m4.125s
-=Symbols in CPPProcess.o= (~sse4:  161) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.301184e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.500425 sec
-real	0m3.551s
-=Symbols in CPPProcess.o= (~sse4:  553) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 9.724770e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.435483 sec
-real	0m3.475s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  571) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.044986e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.152053 sec
-real	0m3.206s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  503) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = DOUBLE (NaN/abnormal=0, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 7.321836e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270315e-06 )  GeV^0
-TOTAL       :     3.416972 sec
-real	0m3.465s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  165) (512y:    0) (512z:  350)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = SCALAR ('none': ~vector[1], no SIMD)
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 6.556002e+06                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371707e-02 +- 3.270376e-06 )  GeV^0
-TOTAL       :     3.000875 sec
-real	0m3.047s
-=Symbols in CPPProcess.o= (~sse4:  140) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=6, zero=0)
-Internal loops fptype_sv    = VECTOR[4] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.567642e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371706e-02 +- 3.270375e-06 )  GeV^0
-TOTAL       :     2.512180 sec
-real	0m2.544s
-=Symbols in CPPProcess.o= (~sse4:  634) (avx2:    0) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.185983e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.332417 sec
-real	0m2.377s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  643) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[8] ('512y': AVX512, 256bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 2.218009e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270339e-06 )  GeV^0
-TOTAL       :     2.343634 sec
-real	0m2.362s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  572) (512y:    0) (512z:    0)
--------------------------------------------------------------------------
-Process                     = EPOCH1_EEMUMU_CPP [gcc 10.2.0] [inlineHel=1]
-FP precision                = FLOAT (NaN/abnormal=5, zero=0)
-Internal loops fptype_sv    = VECTOR[16] ('512z': AVX512, 512bit) [cxtype_ref=YES]
-OMP threads / `nproc --all` = 1 / 4
-EvtsPerSec[MECalcOnly] (3a) = ( 1.508861e+07                 )  sec^-1
-MeanMatrixElemValue         = ( 1.371705e-02 +- 3.270340e-06 )  GeV^0
-TOTAL       :     2.477933 sec
-real	0m2.503s
-=Symbols in CPPProcess.o= (~sse4:    0) (avx2:  240) (512y:    0) (512z:  375)
-=========================================================================
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/version.json b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/version.json
index 29cadcc815bda8e5f7572660b58c9cf684090982..8714099c3c9bc1febb3b35bc40f5c313bbf604b8 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/version.json
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/jobs/good_1/version.json
@@ -1 +1 @@
-{"version":"ci-v0.6","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"48a860dd15d851b2c6b82bfd568bfef7","bmkdata_checksum":"09778836b962d01072d52dc09c61007b","bmk_checksum":"9f37480cfd749d5fadbb610ca7235046"}
+{"version":"ci-v0.7","description":"Madgraph5_aMCatNLO madgraph4gpu benchmark","cvmfs_checksum":"e62fa439bdb4a8e6d4888a0789613b3f","bmkdata_checksum":"4eb8273563af22b829ddbf985ae0b0a9","bmk_checksum":"bebb6ecc27bce4bdc1f79862424d3faf"}
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/mg5amc-madgraph4gpu-bmk.sh b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/mg5amc-madgraph4gpu-bmk.sh
index b6ea7de0be2691f81e1d6691dc04dee6dee2c41e..227a8d268b2b6fb8b185579a7144bb4de8850fbb 100755
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/mg5amc-madgraph4gpu-bmk.sh
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/mg5amc-madgraph4gpu-bmk.sh
@@ -11,40 +11,75 @@
 # Function doOne must be defined in each benchmark
 # Input argument $1: process index (between 1 and $NCOPIES)
 # Return value: please return 0 if this workload copy was successful, 1 otherwise
-# The following variables are guaranteed to be defined and exported: NCOPIES, NTHREADS, NEVENTS_THREAD, BMKDIR, DEBUG
+# The following variables are guaranteed to be defined and exported: NCOPIES, NTHREADS, NEVENTS_THREAD, EXTRA_ARGS, BMKDIR, DEBUG
 # The function is started in process-specific working directory <basewdir>/proc_$1:
 # please store here the individual log files for each of the NCOPIES processes
 function doOne(){
-  if [ "$1" == "" ] || [ "$2" != "" ]; then echo "[doOne] ERROR! Invalid arguments '$@' to doOne"; return 1; fi
+  if [ "$1" == "" ] || [ "$2" != "" ]; then echo "[doOne] INTERNAL ERROR! Invalid arguments '$@' to doOne"; return 1; fi
   echo "[doOne ($1)] $(date) starting in $(pwd)"
-  # Temporary workaround (sanity check)
+  # Choose whether to run benchmarks on CPU only, GPU only or both (BMK-1018 and BMK-1017)
+  # Based on EXTRA_ARGS passed via '--args' (BMK-1014)
+  echo "[doOne ($1)] $(date) EXTRA_ARGS='$EXTRA_ARGS'"
+  status=0
+  action=
+  for arg in $EXTRA_ARGS; do
+    if [ "$action" != "" ]; then
+      echo "[doOne ($1)] $(date) ERROR! EXTRA_ARGS can be at most one argument"
+      status=1; echo "[doOne ($1)] $(date) completed (status=$status)"; return $status
+    elif [ "$arg" == "--cpu" ] || [ "$arg" == "--gpu" ] || [ "$arg" == "--both" ]; then
+      action=$arg
+    else
+      echo "[doOne ($1)] $(date) ERROR! Invalid argument '$arg'"
+      status=1; echo "[doOne ($1)] $(date) completed (status=$status)"; return $status
+    fi
+  done
+  if [ "$action" == "" ]; then
+    action="--both" # default is both CPU and GPU benchmarks
+    echo "[doOne ($1)] $(date) No EXTRA_ARGS were specified: assume '--both' (CPU and GPU benchmarks)"
+  fi
+  # Temporary workaround (sanity check for BMK-813)
   # The CI nodes became unresponsive and had to be rebooted using NEVTS_THREAD=120 (out of memory?)
   if [ $NEVENTS_THREAD -gt 12 ]; then
-    echo "ERROR! Please keep NEVENTS_THREAD <= 12 to avoid resource starvation"
-    status=1
-    echo "[doOne ($1)] $(date) completed (status=$status)"
-    return $status
+    echo "[doOne ($1)] $(date) WARNING! NEVENTS_THREAD=$NEVENTS_THREAD : keeping NEVENTS_THREAD <= 12 is recommended!"
+    echo "[doOne ($1)] $(date) WARNING! Memory usage grows in time proportional to NEVENTS_THREAD (bug BMK-813)"
+    ###echo "[doOne ($1)] $(date) ERROR! Please keep NEVENTS_THREAD <= 12 to avoid resource starvation"
+    ###status=1; echo "[doOne ($1)] $(date) completed (status=$status)"; return $status
   fi
-  # Test if a GPU exists (BMK-983)
-  # [NB: if a GPU physically exists but nvidia-smi fails, then /dev will still contain nvidia*]
-  # See https://docs.nvidia.com/datacenter/tesla/mig-user-guide/#device-nodes
-  nocuda=
-  if ls /dev | grep nvidia > /dev/null; then
-    # Test access to the GPU
-    # [NB: this will always be tested in the CI (assuming the GPU is correctly exposed via docker)
-    if ! nvidia-smi; then
-      echo "ERROR! nvidia-smi failed: cannot access the GPU"
-      status=1
-      echo "[doOne ($1)] $(date) completed (status=$status)"
-      return $status
-    fi
+  # Configure GPU tests if needed
+  if [ "$action" == "--cpu" ]; then
+    nocuda="-avxall -nocuda"
   else
-    echo "WARNING! no GPU is installed on this system (no /dev/nvidia* found)"
-    nocuda=-nocuda
+    nocuda="-avxall"
+    if [ "$action" == "--gpu" ]; then nocuda="-nocpp"; fi
+    # Test if a GPU exists (BMK-983)
+    # [NB: if a GPU physically exists but nvidia-smi fails, then /dev will still contain nvidia*]
+    # See https://docs.nvidia.com/datacenter/tesla/mig-user-guide/#device-nodes
+    if ls /dev | grep nvidia > /dev/null; then
+      echo "[doOne ($1)] $(date) a GPU is installed on this system (/dev/nvidia* found)"
+      if [ "$NCOPIES" != "1" ]; then
+        echo "[doOne ($1)] $(date) ERROR! GPU tests can only be run with one copy (NCOPIES=$NCOPIES)"
+        status=1; echo "[doOne ($1)] $(date) completed (status=$status)"; return $status
+      fi
+      # Test access to the GPU
+      # [NB: this will always be tested in the CI (assuming the GPU is correctly exposed via docker)
+      if ! nvidia-smi; then
+        echo "[doOne ($1)] $(date) ERROR! nvidia-smi failed: cannot access the GPU"
+        status=1; echo "[doOne ($1)] $(date) completed (status=$status)"; return $status
+      fi
+    else
+      echo "[doOne ($1)] $(date) WARNING! no GPU is installed on this system (no /dev/nvidia* found)"
+      if [ "$action" == "--both" ]; then
+        echo "[doOne ($1)] $(date) WARNING! --both was specified: run only the CPU benchmarks"
+        nocuda="-avxall -nocuda"
+      else
+        echo "[doOne ($1)] $(date) ERROR! --gpu was specified: there is no benchmark to run"
+        status=1; echo "[doOne ($1)] $(date) completed (status=$status)"; return $status
+      fi
+    fi
+    # Set up CUDA
+    export CUDA_HOME=/usr/local/cuda-11.6
+    export PATH=${CUDA_HOME}/bin:${PATH}
   fi
-  # Set up CUDA
-  export CUDA_HOME=/usr/local/cuda-11.6
-  export PATH=${CUDA_HOME}/bin:${PATH}
   # Configure WL copy
   procdir=$(pwd)
   download=0
@@ -69,8 +104,10 @@ function doOne(){
   chmod +x ./throughput12.sh
   # Execute WL copy
   cd $procdir/madgraph4gpu/epoch1/cuda/ee_mumu/SubProcesses/P1_Sigma_sm_epem_mupmum
-  source /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/setup.sh > $procdir/out_$1.log 2> >(tee -a $procdir/out_$1.log >&2) \
-    && ./throughput12.sh -avxall -inl -flt $nocuda >> $procdir/out_$1.log 2> >(tee -a $procdir/out_$1.log >&2)
+  echo "[doOne ($1)] $(date) Will execute './throughput12.sh -inl -flt $nocuda' from $(pwd)"
+  source /cvmfs/sft.cern.ch/lcg/releases/gcc/10.2.0-c44b3/x86_64-centos7/setup.sh \
+    > $procdir/out_$1.log 2> >(tee -a $procdir/out_$1.log >&2) \
+    && ./throughput12.sh -inl -flt $nocuda >> $procdir/out_$1.log 2> >(tee -a $procdir/out_$1.log >&2)
   status=${?}
   # Copy madgraph4gpu including newly built binaries to $procdir/build (only in process 1!)
   # The bmk-driver.sh will then copy this to /results/build
@@ -92,6 +129,16 @@ function doOne(){
   return $status
 }
 
+# Optional function usage_detailed may be defined in each benchmark
+# Input arguments: none
+# Return value: none
+function usage_detailed(){
+  echo "Optional EXTRA_ARGS can have at most one argument:"
+  echo "  --cpu  : run only the C++ benchmarks on CPU (1 or more copies)"
+  echo "  --gpu  : run only the CUDA benchmarks on GPU (1 copy)"
+  echo "  --both : run both the C++ benchamrks on CPU and the CUDA benchmarks on GPU (1 copy)"
+}
+
 # Default values for NCOPIES, NTHREADS, NEVENTS_THREAD must be set in each benchmark
 NCOPIES=$(nproc)
 NTHREADS=1 # cannot be changed by user input (single-threaded single-process WL)
diff --git a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/parseResults.sh b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/parseResults.sh
index 0736f5562727b7d2afb71f67343a21efe627296b..18a188464bdfa76b566dd288f8a8ea4e5e185463 100644
--- a/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/parseResults.sh
+++ b/mg5amc/madgraph4gpu/mg5amc-madgraph4gpu/parseResults.sh
@@ -14,12 +14,12 @@ function parseResults(){
   #-----------------------
   echo -e "\n[parseResults] python parser starting using $(python3 -V &> /dev/stdout)"
   local resJSON # declare 'local' separately to avoid masking $? (https://stackoverflow.com/a/4421282)
-  resJSON=$(PYTHONPATH=${parseResultsDir} python3 -c "from parseResults import *; out=parseBmkDir('.', debug=False); print(out)") # same directory as parseResults.sh
+  resJSON=$(PYTHONPATH=${parseResultsDir} python3 -c "from parseResults import *; parseBmkDir('.')") # same directory as parseResults.sh
   pystatus=$?
   [ "$pystatus" != "0" ] && return $pystatus
   echo $resJSON > $baseWDir/parser_output.json
   cat $baseWDir/parser_output.json
   echo "[parseResults] python parser completed (status=$pystatus)"
- return $pystatus
+  return $pystatus
 }
 
diff --git a/test/ci/test-ci.spec b/test/ci/test-ci.spec
index 761e0edddfd2afafba1c53639a7cc3d4c607f677..a666bf4555b9287b05dc94f6b1112be849ab6144 100644
--- a/test/ci/test-ci.spec
+++ b/test/ci/test-ci.spec
@@ -5,10 +5,11 @@ HEPWL_BMKDESCRIPTION="DUMMY benchmark for CI tests (based on LHCb setup)"
 HEPWL_BMKOS="gitlab-registry.cern.ch/linuxsupport/cc7-base:20220801-1.x86_64" # test the CI using cc7-base:latest (default is slc6-base:latest)
 
 # These variables are needed in main.sh
-###HEPWL_BMKOPTS="-c 1 -e 1" # DUMMY HalloWorld TEST (FASTER)
-HEPWL_BMKOPTS="-c 1 -e 2" # DUMMY LHCb setup TEST (DEFAULT)
-###HEPWL_BMKOPTS="-c 1 -e 3" # DUMMY LHCb Gauss setup TEST (SLOWER)
-#####HEPWL_BMKOPTS="-c 200 -e 2" # Debug BMK-247 using 200 copies of the default test (DEBUG)
+###HEPWL_BMKOPTS="-c 1" # DUMMY LHCb setup TEST (DEFAULT)
+###HEPWL_BMKOPTS="-c 1 --args '--hallo-world'" # DUMMY HalloWorld TEST (FASTER)
+HEPWL_BMKOPTS="-c 1 --args '--setup-lhcb'" # DUMMY LHCb setup TEST (DEFAULT)
+###HEPWL_BMKOPTS="-c 1 --args '--setup-gauss'" # DUMMY LHCb Gauss setup TEST (SLOWER)
+###HEPWL_BMKOPTS="-c 200" # Debug BMK-247 using 200 copies of the default test (DEBUG)
 HEPWL_DOCKERIMAGENAME=test-ci-bmk
-HEPWL_DOCKERIMAGETAG=ci0.2 # versions >= ci0.1 use common bmk driver
+HEPWL_DOCKERIMAGETAG=ci0.3 # versions >= ci0.3 use optional EXTRA_ARGS
 HEPWL_CVMFSREPOS=lhcb.cern.ch
diff --git a/test/ci/test-ci/jobs/good_1/cishortlog.txt b/test/ci/test-ci/jobs/good_1/cishortlog.txt
index 3661535b9ac256a3d7cdfd835a1c78bf7e4c7c5a..1716c2b576d6c2251c8b9120c96c5abcd138098d 100644
--- a/test/ci/test-ci/jobs/good_1/cishortlog.txt
+++ b/test/ci/test-ci/jobs/good_1/cishortlog.txt
@@ -1,12 +1,12 @@
 [parseResults] --> Parsing results from proc_1/out_1.log
     "copies": 1,
-{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":2},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.2","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"2e3e551d7b0c0948ee726821c52bd471","bmkdata_checksum":"8da736fb1aff8f14366ca61bc1e42835","bmk_checksum":"d4899ed04d2d6aee998dfb0639aeff41","containment":"docker"}}
-{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":2},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.2","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"2e3e551d7b0c0948ee726821c52bd471","bmkdata_checksum":"8da736fb1aff8f14366ca61bc1e42835","bmk_checksum":"d4899ed04d2d6aee998dfb0639aeff41","containment":"docker"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":1,"extra_arguments":"--setup-lhcb"},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.3","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"475c4dee4cba24fc9300329981e75d2f","bmkdata_checksum":"bebf340195cab22b2ce2f3e47f21b5c4","bmk_checksum":"caff46585d50edf748ea274cb2370c44","containment":"docker"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":1,"extra_arguments":"--setup-lhcb"},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.3","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"475c4dee4cba24fc9300329981e75d2f","bmkdata_checksum":"bebf340195cab22b2ce2f3e47f21b5c4","bmk_checksum":"caff46585d50edf748ea274cb2370c44","containment":"docker"}}
 [parseResults] --> Parsing results from proc_1/out_1.log
     "copies": 1,
-{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":2},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.2","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"2287dd9bc80425d8a0396c51c54b86ca","bmkdata_checksum":"8da736fb1aff8f14366ca61bc1e42835","bmk_checksum":"85b10dfb20eb5da129b5fe24b5d8e204","containment":"docker"}}
-{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":2},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.2","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"2e3e551d7b0c0948ee726821c52bd471","bmkdata_checksum":"8da736fb1aff8f14366ca61bc1e42835","bmk_checksum":"d4899ed04d2d6aee998dfb0639aeff41","containment":"docker"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":1,"extra_arguments":"--setup-lhcb"},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.3","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"642d6c1ee9bafb3ba2be2ecdc3c52013","bmkdata_checksum":"bebf340195cab22b2ce2f3e47f21b5c4","bmk_checksum":"2b6ba0894eddaa0ac3673978f596b025","containment":"docker"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":1,"extra_arguments":"--setup-lhcb"},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.3","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"475c4dee4cba24fc9300329981e75d2f","bmkdata_checksum":"bebf340195cab22b2ce2f3e47f21b5c4","bmk_checksum":"caff46585d50edf748ea274cb2370c44","containment":"docker"}}
 [parseResults] --> Parsing results from proc_1/out_1.log
     "copies": 1,
-{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":2},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.2","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"2287dd9bc80425d8a0396c51c54b86ca","bmkdata_checksum":"8da736fb1aff8f14366ca61bc1e42835","bmk_checksum":"85b10dfb20eb5da129b5fe24b5d8e204","containment":"singularity"}}
-{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":2},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.2","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"2e3e551d7b0c0948ee726821c52bd471","bmkdata_checksum":"8da736fb1aff8f14366ca61bc1e42835","bmk_checksum":"d4899ed04d2d6aee998dfb0639aeff41","containment":"docker"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":1,"extra_arguments":"--setup-lhcb"},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.3","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"642d6c1ee9bafb3ba2be2ecdc3c52013","bmkdata_checksum":"bebf340195cab22b2ce2f3e47f21b5c4","bmk_checksum":"2b6ba0894eddaa0ac3673978f596b025","containment":"singularity"}}
+{"run_info":{"copies":1,"threads_per_copy":1,"events_per_thread":1,"extra_arguments":"--setup-lhcb"},"report":{"wl-scores":{"gen-sim":1},"log":"ok"},"app":{"version":"ci0.3","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"475c4dee4cba24fc9300329981e75d2f","bmkdata_checksum":"bebf340195cab22b2ce2f3e47f21b5c4","bmk_checksum":"caff46585d50edf748ea274cb2370c44","containment":"docker"}}
diff --git a/test/ci/test-ci/jobs/good_1/inputs.log b/test/ci/test-ci/jobs/good_1/inputs.log
index 7f6cbd8c30a5141eb78619f626225645ce601a97..7204dabe89bc064231349945cb3017027139fb28 100644
--- a/test/ci/test-ci/jobs/good_1/inputs.log
+++ b/test/ci/test-ci/jobs/good_1/inputs.log
@@ -1,3 +1,4 @@
 NCOPIES=1
 NTHREADS=1
-NEVENTS_THREAD=2
+NEVENTS_THREAD=1
+EXTRA_ARGS='--setup-lhcb'
diff --git a/test/ci/test-ci/jobs/good_1/joburl.txt b/test/ci/test-ci/jobs/good_1/joburl.txt
index 26d6e4416d7ef861ed732fbec94694799b6fc24f..2eae9a02e28bb5225c1b54ea32691276e6afdde4 100644
--- a/test/ci/test-ci/jobs/good_1/joburl.txt
+++ b/test/ci/test-ci/jobs/good_1/joburl.txt
@@ -1 +1 @@
-https://gitlab.cern.ch/valassi/hep-workloads/-/jobs/23944545
+https://gitlab.cern.ch/valassi/hep-workloads/-/jobs/24061308
diff --git a/test/ci/test-ci/jobs/good_1/proc_1/doOneWrapper_1.log b/test/ci/test-ci/jobs/good_1/proc_1/doOneWrapper_1.log
index a03566bb80d04260f02924bb5fe47df19b045d3d..425a5565e2282bd87d8460ab060e0e776c447948 100644
--- a/test/ci/test-ci/jobs/good_1/proc_1/doOneWrapper_1.log
+++ b/test/ci/test-ci/jobs/good_1/proc_1/doOneWrapper_1.log
@@ -1,10 +1,11 @@
-[doOneWrapper (1)] Sun Aug 14 12:28:41 CEST 2022 : process 1 configured
-[doOneWrapper (1)] HOME=/results/test-ci-c1-e2-1660472921_6718/proc_1/HOME
+[doOneWrapper (1)] Fri Aug 19 23:11:59 CEST 2022 : process 1 configured
+[doOneWrapper (1)] HOME=/results/test-ci-c1-e1-1660943519_8241/proc_1/HOME
 [doOneWrapper (1)] current process pid is 50
 [doOneWrapper (1)] run doOne as bmkuser
 
-[doOne (1)] Sun Aug 14 12:28:41 CEST 2022 starting in /results/test-ci-c1-e2-1660472921_6718/proc_1
-[doOne (1)] Sun Aug 14 12:28:42 CEST 2022 completed (status=0)
+[doOne (1)] Fri Aug 19 23:11:59 CEST 2022 starting in /results/test-ci-c1-e1-1660943519_8241/proc_1
+[doOne (1)] EXTRA_ARGS='--setup-lhcb'
+[doOne (1)] Fri Aug 19 23:12:00 CEST 2022 completed (status=0)
 
-[doOneWrapper (1)] Sun Aug 14 12:28:42 CEST 2022 : process 1 (pid=50) completed ok
+[doOneWrapper (1)] Fri Aug 19 23:12:00 CEST 2022 : process 1 (pid=50) completed ok
 
diff --git a/test/ci/test-ci/jobs/good_1/proc_1/out_1.log b/test/ci/test-ci/jobs/good_1/proc_1/out_1.log
index 12a605bb3f0aab3cc59668b86a6b5641973b1557..046eec6794cace9f416b6a1fb1024297001d36a4 100644
--- a/test/ci/test-ci/jobs/good_1/proc_1/out_1.log
+++ b/test/ci/test-ci/jobs/good_1/proc_1/out_1.log
@@ -2,7 +2,7 @@
 *                         ---- LHCb Login v9r2p6 ----                          *
 *    Building with gcc8 on centos7 x86_64 system (x86_64-centos7-gcc8-opt)     *
 ********************************************************************************
- --- User_release_area is set to /results/test-ci-c1-e2-1660472921_6718/proc_1/HOME/cmtuser
+ --- User_release_area is set to /results/test-ci-c1-e1-1660943519_8241/proc_1/HOME/cmtuser
  --- LHCBPROJECTPATH is set to:
     /cvmfs/lhcb.cern.ch/lib/lhcb
     /cvmfs/lhcb.cern.ch/lib/lcg/releases
diff --git a/test/ci/test-ci/jobs/good_1/test-ci_summary.json b/test/ci/test-ci/jobs/good_1/test-ci_summary.json
index 957f0b1220fc82e6a0b4ea76a3ad8265c1f3861c..382ac6ebe3508298a36c34e8c0c693bc33137423 100644
--- a/test/ci/test-ci/jobs/good_1/test-ci_summary.json
+++ b/test/ci/test-ci/jobs/good_1/test-ci_summary.json
@@ -2,7 +2,8 @@
   "run_info": {
     "copies": 1,
     "threads_per_copy": 1,
-    "events_per_thread": 2
+    "events_per_thread": 1,
+    "extra_arguments": "--setup-lhcb"
   },
   "report": {
     "wl-scores": {
@@ -11,11 +12,11 @@
     "log": "ok"
   },
   "app": {
-    "version": "ci0.2",
+    "version": "ci0.3",
     "description": "DUMMY benchmark for CI tests (based on LHCb setup)",
-    "cvmfs_checksum": "2e3e551d7b0c0948ee726821c52bd471",
-    "bmkdata_checksum": "8da736fb1aff8f14366ca61bc1e42835",
-    "bmk_checksum": "d4899ed04d2d6aee998dfb0639aeff41",
+    "cvmfs_checksum": "475c4dee4cba24fc9300329981e75d2f",
+    "bmkdata_checksum": "bebf340195cab22b2ce2f3e47f21b5c4",
+    "bmk_checksum": "caff46585d50edf748ea274cb2370c44",
     "containment": "docker"
   }
 }
diff --git a/test/ci/test-ci/jobs/good_1/version.json b/test/ci/test-ci/jobs/good_1/version.json
index 2507d49cb2d50325ee822b283f014e06b4b40299..f0f7f6a8ef61aed2e56e04261f282f8fe8200a2d 100644
--- a/test/ci/test-ci/jobs/good_1/version.json
+++ b/test/ci/test-ci/jobs/good_1/version.json
@@ -1 +1 @@
-{"version":"ci0.2","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"2e3e551d7b0c0948ee726821c52bd471","bmkdata_checksum":"8da736fb1aff8f14366ca61bc1e42835","bmk_checksum":"d4899ed04d2d6aee998dfb0639aeff41"}
+{"version":"ci0.3","description":"DUMMY benchmark for CI tests (based on LHCb setup)","cvmfs_checksum":"475c4dee4cba24fc9300329981e75d2f","bmkdata_checksum":"bebf340195cab22b2ce2f3e47f21b5c4","bmk_checksum":"caff46585d50edf748ea274cb2370c44"}
diff --git a/test/ci/test-ci/test-ci-bmk.sh b/test/ci/test-ci/test-ci-bmk.sh
index e9223a5635d8cb1c695050ddfe38e3289c168369..781d335f9719bd631ff52ca0b643e91a4937a307 100755
--- a/test/ci/test-ci/test-ci-bmk.sh
+++ b/test/ci/test-ci/test-ci-bmk.sh
@@ -11,24 +11,43 @@
 # Function doOne must be defined in each benchmark
 # Input argument $1: process index (between 1 and $NCOPIES)
 # Return value: please return 0 if this workload copy was successful, 1 otherwise
-# The following variables are guaranteed to be defined and exported: NCOPIES, NTHREADS, NEVENTS_THREAD, BMKDIR, DEBUG
+# The following variables are guaranteed to be defined and exported: NCOPIES, NTHREADS, NEVENTS_THREAD, EXTRA_ARGS, BMKDIR, DEBUG
 # The function is started in process-specific working directory <basewdir>/proc_$1:
 # please store here the individual log files for each of the NCOPIES processes
 function doOne(){
   if [ "$1" == "" ] || [ "$2" != "" ]; then echo "[doOne] ERROR! Invalid arguments '$@' to doOne"; return 1; fi
   # Configure WL copy
+  echo "[doOne ($1)] $(date) starting in $(pwd)"  
+  echo "[doOne ($1)] EXTRA_ARGS='$EXTRA_ARGS'"
+  status=0
+  action=
+  for arg in $EXTRA_ARGS; do
+    if [ "$action" != "" ]; then
+      echo "ERROR! EXTRA_ARGS can be at most one argument"; status=1; break
+    elif [ "$arg" == "--hallo-world" ]; then
+      action=1
+    elif [ "$arg" == "--setup-lhcb" ]; then
+      action=2
+    elif [ "$arg" == "--setup-gauss" ]; then
+      action=3
+    else
+      echo "ERROR! Invalid argument '$arg'"; status=1; break
+    fi
+  done
+  if [ "$action" == "" ]; then action=2; fi # default is LHCb setup
   # Execute WL copy
-  echo "[doOne ($1)] $(date) starting in $(pwd)"
-  if [ "$NEVENTS_THREAD" == "1" ]; then
-    echo "Hallo World! (FASTER DUMMY TEST)" >out_$1.log 2>&1
-  elif [ "$NEVENTS_THREAD" == "2" ]; then
-    source /cvmfs/lhcb.cern.ch/lib/LbLogin.sh >out_$1.log 2>&1 \
-      && echo "LHCb setup (DEFAULT DUMMY TEST)" >>out_$1.log 2>&1
-  else
-    source /cvmfs/lhcb.cern.ch/lib/LbLogin.sh >out_$1.log 2>&1 \
-      && lb-run --use-sp -c x86_64-slc6-gcc48-opt --use="AppConfig v3r335" --use="DecFiles v30r11" --use="ProdConf" Gauss/v49r9 echo "LHCb Gauss setup (SLOWER DUMMY TEST)" >>out_$1.log 2>&1
+  if [ "$status" == "0" ]; then
+    if [ "$action" == "1" ]; then
+      echo "Hallo World! (FASTER DUMMY TEST)" >out_$1.log 2>&1
+    elif [ "$action" == "2" ]; then
+      source /cvmfs/lhcb.cern.ch/lib/LbLogin.sh >out_$1.log 2>&1 \
+	&& echo "LHCb setup (DEFAULT DUMMY TEST)" >>out_$1.log 2>&1
+    else
+      source /cvmfs/lhcb.cern.ch/lib/LbLogin.sh >out_$1.log 2>&1 \
+	&& lb-run --use-sp -c x86_64-slc6-gcc48-opt --use="AppConfig v3r335" --use="DecFiles v30r11" --use="ProdConf" Gauss/v49r9 echo "LHCb Gauss setup (SLOWER DUMMY TEST)" >>out_$1.log 2>&1
+    fi
+    status=${?}
   fi
-  status=${?}
   echo "[doOne ($1)] $(date) completed (status=$status)"
   # Return 0 if this workload copy was successful, 1 otherwise
   return $status
@@ -38,17 +57,18 @@ function doOne(){
 # Input arguments: none
 # Return value: none
 function usage_detailed(){
-  echo "NEVENTS_THREAD =1 : dummy HalloWorld test (faster)"
-  echo "               =2 : dummy LHCb setup test (default)"
-  echo "               >2 : dummy LHCb Gauss setup test (slower)"
+  echo "NEVENTS_THREAD is ignored"
+  echo
+  echo "Optional EXTRA_ARGS can have at most one argument:"
+  echo "  --hallo-world : dummy HalloWorld test (faster)"
+  echo "  --setup-lhcb  : dummy LHCb setup test (DEFAULT)"
+  echo "  --setup-gauss : dummy LHCb Gauss setup test (slower)"
 }
 
 # Default values for NCOPIES, NTHREADS, NEVENTS_THREAD must be set in each benchmark
 NTHREADS=1 # cannot be changed by user input (single-threaded single-process WL)
 NCOPIES=$(nproc)
-###NEVENTS_THREAD=1 # DUMMY HalloWorld TEST (FASTER)
-NEVENTS_THREAD=2 # DUMMY LHCb setup TEST (DEFAULT)
-###NEVENTS_THREAD=3 # DUMMY LHCb Gauss setup TEST (SLOWER)
+NEVENTS_THREAD=1 # ignored
 
 # Source the common benchmark driver
 if [ -f $(dirname $0)/bmk-driver.sh ]; then