Merge pull request madgraph5#857 from valassi/tmad

Fix SIGFPE crash (855) in rotxxx by adding volatile in aloha_functions.f

.github/workflows/testsuite_oneprocess.sh

@@ -272,6 +272,7 @@ function getinputfile()
   nevt=$(getnevt)
   tmp=$tmpdir/input_${proc%.mad}_${backend}
   iconfig=1
+  if [ "${proc%.mad}" == "gg_ttgg" ]; then iconfig=104; fi # test iconfig=104 on gg_ttgg (LHE color mismatch #856?)
   cat << EOF >> ${tmp}
 ${nevt} 1 1 ! Number of events and max and min iterations
 0.000001 ! Accuracy (ignored because max iterations = min iterations)
@@ -533,12 +534,12 @@ function bypassIssue(){
 if [ $BYPASS_KNOWN_ISSUES -eq 1 ] && [ $status -ne 0 ]; then
   # Known issues in tmad_test
   if [ "$stage" == "tmad_test" ]; then
-    # No cross section for susy_gg_t1t1 (#826)
+    # No cross section in susy_gg_t1t1 (#826)
     ###if [ "${proc%.mad}" == "susy_gg_t1t1" ]; then bypassIssue "No cross section in ${proc%.mad} for FPTYPE=d,f,m (#826)"; fi
-    # SIGFPE crashes in rotxxx (#855)
-    ###if [ "${proc%.mad}" == "gq_ttq" ]; then bypassIssue "SIGFPE crash in rotxxx in ${proc%.mad} for FPTYPE=d,f,m (#855)"; fi
-    ###if [ "${proc%.mad}" == "pp_tt012j" ]; then bypassIssue "SIGFPE crash in rotxxx in ${proc%.mad} for FPTYPE=d,f,m (#855)"; fi
-    ###if [ "${proc%.mad}" == "nobm_pp_ttW" ]; then bypassIssue "#SIGFPE crash in rotxxx in ${proc%.mad} for FPTYPE=d,f,m (#855)"; fi
+    # LHE color mismatch in gg_ttgg for iconfig=104 (#856)
+    ###if [ "${proc%.mad}" == "gg_ttgg" ]; then bypassIssue "LHE color mismatch for iconfig=104 in ${proc%.mad} for FPTYPE=d,f,m (#856)"; fi
+    # Cross section mismatch in pp_tt012j for P2_gu_ttxgu (#872)
+    ###if [ "${proc%.mad}" == "pp_tt012j" ]; then bypassIssue "Cross section mismatch for P2_gu_ttxgu in ${proc%.mad} for FPTYPE=d,f,m (#856)"; fi
     # Final printout
     if [ $status -ne 0 ]; then echo "[testsuite_oneprocess.sh] $stage ($proc) FPTYPE=${FPTYPE}: issue will not be bypassed, test has FAILED"; fi
   fi

epochX/cudacpp/ee_mumu.mad/CODEGEN_mad_ee_mumu_log.txt

@@ -62,7 +62,7 @@ generate e+ e- > mu+ mu-
 No model currently active, so we import the Standard Model
 INFO: load particles 
 INFO: load vertices 
-DEBUG: model prefixing  takes 0.005818367004394531 
+DEBUG: model prefixing  takes 0.005683183670043945 
 INFO: Restrict model sm with file models/sm/restrict_default.dat . 
 DEBUG: Simplifying conditional expressions 
 DEBUG: remove interactions: u s w+ at order: QED=1 
@@ -154,7 +154,7 @@ INFO: Checking for minimal orders which gives processes.
 INFO: Please specify coupling orders to bypass this step. 
 INFO: Trying process: e+ e- > mu+ mu- WEIGHTED<=4 @1  
 INFO: Process has 2 diagrams 
-1 processes with 2 diagrams generated in 0.005 s
+1 processes with 2 diagrams generated in 0.004 s
 Total: 1 processes with 2 diagrams
 output madevent_simd ../TMPOUT/CODEGEN_mad_ee_mumu --hel_recycling=False --vector_size=32
 Load PLUGIN.CUDACPP_OUTPUT
@@ -176,8 +176,8 @@ INFO: Organizing processes into subprocess groups
 INFO: Generating Helas calls for process: e+ e- > mu+ mu- WEIGHTED<=4 @1 
 INFO: Processing color information for process: e+ e- > mu+ mu- @1 
 INFO: Creating files in directory P1_epem_mupmum 
-DEBUG:  kwargs[prefix] = 0 [model_handling.py at line 1148] 
-DEBUG:  process_exporter_cpp =  <PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_OneProcessExporter object at 0x7feecb09d3a0> [export_v4.py at line 6261] 
+DEBUG:  kwargs[prefix] = 0 [model_handling.py at line 1151] 
+DEBUG:  process_exporter_cpp =  <PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_OneProcessExporter object at 0x7f1c23eaf730> [export_v4.py at line 6261] 
 INFO: Creating files in directory . 
 FileWriter <class 'PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_CPPWriter'> for ././CPPProcess.h
 FileWriter <class 'PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_CPPWriter'> for ././CPPProcess.cc
@@ -194,19 +194,19 @@ INFO: Created files CPPProcess.h and CPPProcess.cc in directory ./.
 INFO: Generating Feynman diagrams for Process: e+ e- > mu+ mu- WEIGHTED<=4 @1 
 INFO: Finding symmetric diagrams for subprocess group epem_mupmum 
 Generated helas calls for 1 subprocesses (2 diagrams) in 0.004 s
-Wrote files for 8 helas calls in 0.104 s
+Wrote files for 8 helas calls in 0.099 s
 ALOHA: aloha starts to compute helicity amplitudes
 ALOHA: aloha creates FFV1 routines
 ALOHA: aloha creates FFV2 routines
 ALOHA: aloha creates FFV4 routines
-ALOHA: aloha creates 3 routines in  0.213 s
+ALOHA: aloha creates 3 routines in  0.197 s
 DEBUG:  Entering PLUGIN_ProcessExporter.convert_model (create the model) [output.py at line 205] 
 ALOHA: aloha starts to compute helicity amplitudes
 ALOHA: aloha creates FFV1 routines
 ALOHA: aloha creates FFV2 routines
 ALOHA: aloha creates FFV4 routines
 ALOHA: aloha creates FFV2_4 routines
-ALOHA: aloha creates 7 routines in  0.280 s
+ALOHA: aloha creates 7 routines in  0.258 s
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV2
@@ -228,6 +228,7 @@ If you want to make this value the default for future session, you can run 'save
 save configuration file to /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_mad_ee_mumu/Cards/me5_configuration.txt
 INFO: Use Fortran compiler gfortran 
 INFO: Use c++ compiler g++ 
+INFO: Generate jpeg diagrams 
 INFO: Generate web pages 
 DEBUG: cd /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_mad_ee_mumu; patch -p4 -i /data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/CODEGEN/PLUGIN/CUDACPP_SA_OUTPUT/MG5aMC_patches/PROD/patch.common
 patching file Source/genps.inc
@@ -250,9 +251,9 @@ Type "launch" to generate events from this process, or see
 Run "open index.html" to see more information about this process.
 quit
 
-real	0m1.985s
-user	0m1.747s
-sys	0m0.212s
+real	0m2.034s
+user	0m1.790s
+sys	0m0.231s
 Code generation completed in 2 seconds
 ************************************************************
 *                                                          *

epochX/cudacpp/ee_mumu.mad/Source/DHELAS/aloha_functions.f

@@ -1201,7 +1201,7 @@ subroutine rotxxx(p,q , prot)
 c
       implicit none
       double precision p(0:3),q(0:3),prot(0:3),qt2,qt,psgn,qq,p1
-
+      volatile qt, p1, qq ! prevent optimizations with -O3 (workaround for SIGFPE crashes in rotxxx: madgraph5/madgraph4gpu#855)
       double precision rZero, rOne
       parameter( rZero = 0.0d0, rOne = 1.0d0 )
 

epochX/cudacpp/ee_mumu.sa/CODEGEN_cudacpp_ee_mumu_log.txt

@@ -62,7 +62,7 @@ generate e+ e- > mu+ mu-
 No model currently active, so we import the Standard Model
 INFO: load particles 
 INFO: load vertices 
-DEBUG: model prefixing  takes 0.005814552307128906 
+DEBUG: model prefixing  takes 0.005368709564208984 
 INFO: Restrict model sm with file models/sm/restrict_default.dat . 
 DEBUG: Simplifying conditional expressions 
 DEBUG: remove interactions: u s w+ at order: QED=1 
@@ -154,7 +154,7 @@ INFO: Checking for minimal orders which gives processes.
 INFO: Please specify coupling orders to bypass this step. 
 INFO: Trying process: e+ e- > mu+ mu- WEIGHTED<=4 @1  
 INFO: Process has 2 diagrams 
-1 processes with 2 diagrams generated in 0.005 s
+1 processes with 2 diagrams generated in 0.004 s
 Total: 1 processes with 2 diagrams
 output standalone_cudacpp ../TMPOUT/CODEGEN_cudacpp_ee_mumu
 Load PLUGIN.CUDACPP_OUTPUT
@@ -184,7 +184,7 @@ ALOHA: aloha creates FFV1 routines
 ALOHA: aloha creates FFV2 routines
 ALOHA: aloha creates FFV4 routines
 ALOHA: aloha creates FFV2_4 routines
-ALOHA: aloha creates 4 routines in  0.285 s
+ALOHA: aloha creates 4 routines in  0.265 s
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV2
@@ -203,7 +203,7 @@ INFO: Created files Parameters_sm.h and Parameters_sm.cc in directory
 INFO: /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_cudacpp_ee_mumu/src/. and /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_cudacpp_ee_mumu/src/. 
 quit
 
-real	0m0.689s
-user	0m0.639s
-sys	0m0.045s
+real	0m0.653s
+user	0m0.596s
+sys	0m0.048s
 Code generation completed in 1 seconds

epochX/cudacpp/gg_tt.mad/CODEGEN_mad_gg_tt_log.txt

@@ -62,7 +62,7 @@ generate g g > t t~
 No model currently active, so we import the Standard Model
 INFO: load particles 
 INFO: load vertices 
-DEBUG: model prefixing  takes 0.0058248043060302734 
+DEBUG: model prefixing  takes 0.00545048713684082 
 INFO: Restrict model sm with file models/sm/restrict_default.dat . 
 DEBUG: Simplifying conditional expressions 
 DEBUG: remove interactions: u s w+ at order: QED=1 
@@ -155,7 +155,7 @@ INFO: Please specify coupling orders to bypass this step.
 INFO: Trying coupling order WEIGHTED<=2: WEIGTHED IS QCD+2*QED 
 INFO: Trying process: g g > t t~ WEIGHTED<=2 @1  
 INFO: Process has 3 diagrams 
-1 processes with 3 diagrams generated in 0.009 s
+1 processes with 3 diagrams generated in 0.008 s
 Total: 1 processes with 3 diagrams
 output madevent_simd ../TMPOUT/CODEGEN_mad_gg_tt --hel_recycling=False --vector_size=32
 Load PLUGIN.CUDACPP_OUTPUT
@@ -177,8 +177,8 @@ INFO: Organizing processes into subprocess groups
 INFO: Generating Helas calls for process: g g > t t~ WEIGHTED<=2 @1 
 INFO: Processing color information for process: g g > t t~ @1 
 INFO: Creating files in directory P1_gg_ttx 
-DEBUG:  kwargs[prefix] = 0 [model_handling.py at line 1148] 
-DEBUG:  process_exporter_cpp =  <PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_OneProcessExporter object at 0x7ffabb64e610> [export_v4.py at line 6261] 
+DEBUG:  kwargs[prefix] = 0 [model_handling.py at line 1151] 
+DEBUG:  process_exporter_cpp =  <PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_OneProcessExporter object at 0x7f038426f850> [export_v4.py at line 6261] 
 INFO: Creating files in directory . 
 FileWriter <class 'PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_CPPWriter'> for ././CPPProcess.h
 FileWriter <class 'PLUGIN.CUDACPP_OUTPUT.model_handling.PLUGIN_CPPWriter'> for ././CPPProcess.cc
@@ -194,16 +194,16 @@ INFO: Created files CPPProcess.h and CPPProcess.cc in directory ./.
 INFO: Generating Feynman diagrams for Process: g g > t t~ WEIGHTED<=2 @1 
 INFO: Finding symmetric diagrams for subprocess group gg_ttx 
 Generated helas calls for 1 subprocesses (3 diagrams) in 0.006 s
-Wrote files for 10 helas calls in 0.108 s
+Wrote files for 10 helas calls in 0.103 s
 ALOHA: aloha starts to compute helicity amplitudes
 ALOHA: aloha creates VVV1 set of routines with options: P0
 ALOHA: aloha creates FFV1 routines
-ALOHA: aloha creates 2 routines in  0.154 s
+ALOHA: aloha creates 2 routines in  0.144 s
 DEBUG:  Entering PLUGIN_ProcessExporter.convert_model (create the model) [output.py at line 205] 
 ALOHA: aloha starts to compute helicity amplitudes
 ALOHA: aloha creates VVV1 set of routines with options: P0
 ALOHA: aloha creates FFV1 routines
-ALOHA: aloha creates 4 routines in  0.139 s
+ALOHA: aloha creates 4 routines in  0.132 s
 <class 'aloha.create_aloha.AbstractRoutine'> VVV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
@@ -221,6 +221,7 @@ If you want to make this value the default for future session, you can run 'save
 save configuration file to /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_mad_gg_tt/Cards/me5_configuration.txt
 INFO: Use Fortran compiler gfortran 
 INFO: Use c++ compiler g++ 
+INFO: Generate jpeg diagrams 
 INFO: Generate web pages 
 DEBUG: cd /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_mad_gg_tt; patch -p4 -i /data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/CODEGEN/PLUGIN/CUDACPP_SA_OUTPUT/MG5aMC_patches/PROD/patch.common
 patching file Source/genps.inc
@@ -239,10 +240,10 @@ Type "launch" to generate events from this process, or see
 Run "open index.html" to see more information about this process.
 quit
 
-real	0m1.797s
-user	0m1.551s
-sys	0m0.228s
-Code generation completed in 1 seconds
+real	0m1.891s
+user	0m1.634s
+sys	0m0.251s
+Code generation completed in 2 seconds
 ************************************************************
 *                                                          *
 *                      W E L C O M E to                    *

epochX/cudacpp/gg_tt.mad/Source/DHELAS/aloha_functions.f

@@ -1201,7 +1201,7 @@ subroutine rotxxx(p,q , prot)
 c
       implicit none
       double precision p(0:3),q(0:3),prot(0:3),qt2,qt,psgn,qq,p1
-
+      volatile qt, p1, qq ! prevent optimizations with -O3 (workaround for SIGFPE crashes in rotxxx: madgraph5/madgraph4gpu#855)
       double precision rZero, rOne
       parameter( rZero = 0.0d0, rOne = 1.0d0 )
 

epochX/cudacpp/gg_tt.sa/CODEGEN_cudacpp_gg_tt_log.txt

@@ -62,7 +62,7 @@ generate g g > t t~
 No model currently active, so we import the Standard Model
 INFO: load particles 
 INFO: load vertices 
-DEBUG: model prefixing  takes 0.005820274353027344 
+DEBUG: model prefixing  takes 0.00540924072265625 
 INFO: Restrict model sm with file models/sm/restrict_default.dat . 
 DEBUG: Simplifying conditional expressions 
 DEBUG: remove interactions: u s w+ at order: QED=1 
@@ -155,7 +155,7 @@ INFO: Please specify coupling orders to bypass this step.
 INFO: Trying coupling order WEIGHTED<=2: WEIGTHED IS QCD+2*QED 
 INFO: Trying process: g g > t t~ WEIGHTED<=2 @1  
 INFO: Process has 3 diagrams 
-1 processes with 3 diagrams generated in 0.009 s
+1 processes with 3 diagrams generated in 0.008 s
 Total: 1 processes with 3 diagrams
 output standalone_cudacpp ../TMPOUT/CODEGEN_cudacpp_gg_tt
 Load PLUGIN.CUDACPP_OUTPUT
@@ -183,7 +183,7 @@ Generated helas calls for 1 subprocesses (3 diagrams) in 0.006 s
 ALOHA: aloha starts to compute helicity amplitudes
 ALOHA: aloha creates VVV1 set of routines with options: P0
 ALOHA: aloha creates FFV1 routines
-ALOHA: aloha creates 2 routines in  0.153 s
+ALOHA: aloha creates 2 routines in  0.144 s
 <class 'aloha.create_aloha.AbstractRoutine'> VVV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
 <class 'aloha.create_aloha.AbstractRoutine'> FFV1
@@ -198,7 +198,7 @@ INFO: Created files Parameters_sm.h and Parameters_sm.cc in directory
 INFO: /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_cudacpp_gg_tt/src/. and /data/avalassi/GPU2023/madgraph4gpuX/MG5aMC/TMPOUT/CODEGEN_cudacpp_gg_tt/src/. 
 quit
 
-real	0m0.635s
-user	0m0.505s
-sys	0m0.048s
-Code generation completed in 0 seconds
+real	0m0.547s
+user	0m0.473s
+sys	0m0.059s
+Code generation completed in 1 seconds