diff --git a/epochX/cudacpp/tmad/logs_pptt012j_mad/log_pptt012j_mad_d_inl0_hrd0.txt b/epochX/cudacpp/tmad/logs_pptt012j_mad/log_pptt012j_mad_d_inl0_hrd0.txt deleted file mode 100644 index b109045263..0000000000 --- a/epochX/cudacpp/tmad/logs_pptt012j_mad/log_pptt012j_mad_d_inl0_hrd0.txt +++ /dev/null @@ -1,603 +0,0 @@ -Working directory (build): /data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu - - - - -make USEBUILDDIR=1 BACKEND=cuda -make USEBUILDDIR=1 BACKEND=cppnone -make USEBUILDDIR=1 BACKEND=cppsse4 -make USEBUILDDIR=1 BACKEND=cppavx2 - -make USEBUILDDIR=1 BACKEND=cpp512y -make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' - -make USEBUILDDIR=1 BACKEND=cpp512z -make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Nothing to be done for 'all'. -make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Nothing to be done for 'all'. -make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Nothing to be done for 'all'. -make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Nothing to be done for 'all'. -make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Nothing to be done for 'all'. -make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' -make[1]: Nothing to be done for 'all'. -make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu' - -OMP_NUM_THREADS= - -DATE: 2024-07-26_17:36:51 - -On itscrd90.cern.ch [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: 1x Tesla V100S-PCIE-32GB]: -Working directory (run): /data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/pp_tt012j.mad/SubProcesses/P2_gu_ttxgu - -*** (1) EXECUTE MADEVENT_FORTRAN (create results.dat) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./madevent_fortran < /tmp/avalassi/input_pptt012j_x1_fortran > /tmp/avalassi/output_pptt012j_x1_fortran' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702806] fbridge_mode=0 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 3.2439s - [COUNTERS] Fortran Overhead ( 0 ) : 1.0065s - [COUNTERS] Fortran MEs ( 1 ) : 2.2375s for 24576 events => throughput is 1.10E+04 events/s - -*** (1) EXECUTE MADEVENT_FORTRAN x1 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./madevent_fortran < /tmp/avalassi/input_pptt012j_x1_fortran > /tmp/avalassi/output_pptt012j_x1_fortran' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702806] fbridge_mode=0 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 3.2609s - [COUNTERS] Fortran Overhead ( 0 ) : 1.0253s - [COUNTERS] Fortran MEs ( 1 ) : 2.2356s for 24576 events => throughput is 1.10E+04 events/s - -*** (1) EXECUTE MADEVENT_FORTRAN x10 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -81920 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./madevent_fortran < /tmp/avalassi/input_pptt012j_x10_fortran > /tmp/avalassi/output_pptt012j_x10_fortran' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.212 [1.2120105567647810] fbridge_mode=0 - [UNWEIGHT] Wrote 773 events (found 3685 events) - [COUNTERS] PROGRAM TOTAL : 15.7262s - [COUNTERS] Fortran Overhead ( 0 ) : 4.5048s - [COUNTERS] Fortran MEs ( 1 ) : 11.2214s for 122880 events => throughput is 1.10E+04 events/s - -*** (2-none) EXECUTE MADEVENT_CPP x1 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.none_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x1_cudacpp > /tmp/avalassi/output_pptt012j_x1_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702806] fbridge_mode=1 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 4.0629s - [COUNTERS] Fortran Overhead ( 0 ) : 1.7772s - [COUNTERS] CudaCpp MEs ( 2 ) : 2.2857s for 24576 events => throughput is 1.08E+04 events/s - -*** (2-none) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.1841273083702806) and cpp (1.1841273083702806) differ by less than 3E-14 (0.0) - -*** (2-none) Compare MADEVENT_CPP x1 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.1 and events.lhe.ref.1 are identical - -*** (2-none) EXECUTE MADEVENT_CPP x10 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -81920 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.none_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x10_cudacpp > /tmp/avalassi/output_pptt012j_x10_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.212 [1.2120105567647805] fbridge_mode=1 - [UNWEIGHT] Wrote 773 events (found 3685 events) - [COUNTERS] PROGRAM TOTAL : 16.9802s - [COUNTERS] Fortran Overhead ( 0 ) : 5.3690s - [COUNTERS] CudaCpp MEs ( 2 ) : 11.6112s for 122880 events => throughput is 1.06E+04 events/s - -*** (2-none) Compare MADEVENT_CPP x10 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.2120105567647810) and cpp (1.2120105567647805) differ by less than 3E-14 (3.3306690738754696e-16) - -*** (2-none) Compare MADEVENT_CPP x10 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.10 and events.lhe.ref.10 are identical - -*** EXECUTE CHECK(8192) -p 256 32 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+BRDHST/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 1.098364e+04 ) sec^-1 - -*** EXECUTE CHECK(8192) -p 256 32 1 *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 1.096010e+04 ) sec^-1 - -*** (2-sse4) EXECUTE MADEVENT_CPP x1 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.sse4_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x1_cudacpp > /tmp/avalassi/output_pptt012j_x1_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702806] fbridge_mode=1 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 2.6763s - [COUNTERS] Fortran Overhead ( 0 ) : 1.4307s - [COUNTERS] CudaCpp MEs ( 2 ) : 1.2456s for 24576 events => throughput is 1.97E+04 events/s - -*** (2-sse4) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.1841273083702806) and cpp (1.1841273083702806) differ by less than 3E-14 (0.0) - -*** (2-sse4) Compare MADEVENT_CPP x1 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.1 and events.lhe.ref.1 are identical - -*** (2-sse4) EXECUTE MADEVENT_CPP x10 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -81920 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.sse4_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x10_cudacpp > /tmp/avalassi/output_pptt012j_x10_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.212 [1.2120105567647805] fbridge_mode=1 - [UNWEIGHT] Wrote 773 events (found 3685 events) - [COUNTERS] PROGRAM TOTAL : 11.2027s - [COUNTERS] Fortran Overhead ( 0 ) : 4.9541s - [COUNTERS] CudaCpp MEs ( 2 ) : 6.2487s for 122880 events => throughput is 1.97E+04 events/s - -*** (2-sse4) Compare MADEVENT_CPP x10 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.2120105567647810) and cpp (1.2120105567647805) differ by less than 3E-14 (3.3306690738754696e-16) - -*** (2-sse4) Compare MADEVENT_CPP x10 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.10 and events.lhe.ref.10 are identical - -*** EXECUTE CHECK(8192) -p 256 32 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+BRDHST/sse4+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 2.004214e+04 ) sec^-1 - -*** EXECUTE CHECK(8192) -p 256 32 1 *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/sse4+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 2.012289e+04 ) sec^-1 - -*** (2-avx2) EXECUTE MADEVENT_CPP x1 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.avx2_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x1_cudacpp > /tmp/avalassi/output_pptt012j_x1_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702801] fbridge_mode=1 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 1.7669s - [COUNTERS] Fortran Overhead ( 0 ) : 1.2029s - [COUNTERS] CudaCpp MEs ( 2 ) : 0.5640s for 24576 events => throughput is 4.36E+04 events/s - -*** (2-avx2) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.1841273083702806) and cpp (1.1841273083702801) differ by less than 3E-14 (3.3306690738754696e-16) - -*** (2-avx2) Compare MADEVENT_CPP x1 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.1 and events.lhe.ref.1 are identical - -*** (2-avx2) EXECUTE MADEVENT_CPP x10 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -81920 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.avx2_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x10_cudacpp > /tmp/avalassi/output_pptt012j_x10_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.212 [1.2120105567647796] fbridge_mode=1 - [UNWEIGHT] Wrote 773 events (found 3685 events) - [COUNTERS] PROGRAM TOTAL : 7.5515s - [COUNTERS] Fortran Overhead ( 0 ) : 4.7330s - [COUNTERS] CudaCpp MEs ( 2 ) : 2.8185s for 122880 events => throughput is 4.36E+04 events/s - -*** (2-avx2) Compare MADEVENT_CPP x10 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.2120105567647810) and cpp (1.2120105567647796) differ by less than 3E-14 (1.1102230246251565e-15) - -*** (2-avx2) Compare MADEVENT_CPP x10 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.10 and events.lhe.ref.10 are identical - -*** EXECUTE CHECK(8192) -p 256 32 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+BRDHST/avx2+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 4.488789e+04 ) sec^-1 - -*** EXECUTE CHECK(8192) -p 256 32 1 *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/avx2+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 4.518493e+04 ) sec^-1 - -*** (2-512y) EXECUTE MADEVENT_CPP x1 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.512y_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x1_cudacpp > /tmp/avalassi/output_pptt012j_x1_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702801] fbridge_mode=1 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 1.7068s - [COUNTERS] Fortran Overhead ( 0 ) : 1.1928s - [COUNTERS] CudaCpp MEs ( 2 ) : 0.5141s for 24576 events => throughput is 4.78E+04 events/s - -*** (2-512y) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.1841273083702806) and cpp (1.1841273083702801) differ by less than 3E-14 (3.3306690738754696e-16) - -*** (2-512y) Compare MADEVENT_CPP x1 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.1 and events.lhe.ref.1 are identical - -*** (2-512y) EXECUTE MADEVENT_CPP x10 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -81920 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.512y_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x10_cudacpp > /tmp/avalassi/output_pptt012j_x10_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.212 [1.2120105567647796] fbridge_mode=1 - [UNWEIGHT] Wrote 773 events (found 3685 events) - [COUNTERS] PROGRAM TOTAL : 7.2444s - [COUNTERS] Fortran Overhead ( 0 ) : 4.7013s - [COUNTERS] CudaCpp MEs ( 2 ) : 2.5430s for 122880 events => throughput is 4.83E+04 events/s - -*** (2-512y) Compare MADEVENT_CPP x10 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.2120105567647810) and cpp (1.2120105567647796) differ by less than 3E-14 (1.1102230246251565e-15) - -*** (2-512y) Compare MADEVENT_CPP x10 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.10 and events.lhe.ref.10 are identical - -*** EXECUTE CHECK(8192) -p 256 32 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+BRDHST/512y+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 4.949402e+04 ) sec^-1 - -*** EXECUTE CHECK(8192) -p 256 32 1 *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/512y+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 4.903224e+04 ) sec^-1 - -*** (2-512z) EXECUTE MADEVENT_CPP x1 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.512z_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x1_cudacpp > /tmp/avalassi/output_pptt012j_x1_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702801] fbridge_mode=1 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 1.8847s - [COUNTERS] Fortran Overhead ( 0 ) : 1.2325s - [COUNTERS] CudaCpp MEs ( 2 ) : 0.6522s for 24576 events => throughput is 3.77E+04 events/s - -*** (2-512z) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.1841273083702806) and cpp (1.1841273083702801) differ by less than 3E-14 (3.3306690738754696e-16) - -*** (2-512z) Compare MADEVENT_CPP x1 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.1 and events.lhe.ref.1 are identical - -*** (2-512z) EXECUTE MADEVENT_CPP x10 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -81920 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.512z_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_pptt012j_x10_cudacpp > /tmp/avalassi/output_pptt012j_x10_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.212 [1.2120105567647796] fbridge_mode=1 - [UNWEIGHT] Wrote 773 events (found 3685 events) - [COUNTERS] PROGRAM TOTAL : 8.0126s - [COUNTERS] Fortran Overhead ( 0 ) : 4.7636s - [COUNTERS] CudaCpp MEs ( 2 ) : 3.2490s for 122880 events => throughput is 3.78E+04 events/s - -*** (2-512z) Compare MADEVENT_CPP x10 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.2120105567647810) and cpp (1.2120105567647796) differ by less than 3E-14 (1.1102230246251565e-15) - -*** (2-512z) Compare MADEVENT_CPP x10 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cpp.10 and events.lhe.ref.10 are identical - -*** EXECUTE CHECK(8192) -p 256 32 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+BRDHST/512z+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 3.847403e+04 ) sec^-1 - -*** EXECUTE CHECK(8192) -p 256 32 1 *** -Process = SIGMA_SM_GU_TTXGU_CPP [gcc 11.3.1] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/512z+CXVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 3.855186e+04 ) sec^-1 - -*** (3-cuda) EXECUTE MADEVENT_CUDA x1 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -8192 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.cuda_d_inl0_hrd0/madevent_cuda < /tmp/avalassi/input_pptt012j_x1_cudacpp > /tmp/avalassi/output_pptt012j_x1_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.184 [1.1841273083702799] fbridge_mode=1 - [UNWEIGHT] Wrote 282 events (found 1084 events) - [COUNTERS] PROGRAM TOTAL : 1.4985s - [COUNTERS] Fortran Overhead ( 0 ) : 1.4770s - [COUNTERS] CudaCpp MEs ( 2 ) : 0.0215s for 24576 events => throughput is 1.14E+06 events/s - -*** (3-cuda) Compare MADEVENT_CUDA x1 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.1841273083702806) and cuda (1.1841273083702799) differ by less than 3E-14 (5.551115123125783e-16) - -*** (3-cuda) Compare MADEVENT_CUDA x1 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cuda.1 and events.lhe.ref.1 are identical - -*** (3-cuda) EXECUTE MADEVENT_CUDA x10 (create events.lhe) *** --------------------- -CUDACPP_RUNTIME_FBRIDGEMODE = (not set) -CUDACPP_RUNTIME_VECSIZEUSED = 8192 --------------------- -81920 1 1 ! Number of events and max and min iterations -0.000001 ! Accuracy (ignored because max iterations = min iterations) -0 ! Grid Adjustment 0=none, 2=adjust (NB if = 0, ftn26 will still be used if present) -1 ! Suppress Amplitude 1=yes (i.e. use MadEvent single-diagram enhancement) -0 ! Helicity Sum/event 0=exact -1 ! ICONFIG number (1-N) for single-diagram enhancement multi-channel (NB used even if suppress amplitude is 0!) --------------------- -Executing ' ./build.cuda_d_inl0_hrd0/madevent_cuda < /tmp/avalassi/input_pptt012j_x10_cudacpp > /tmp/avalassi/output_pptt012j_x10_cudacpp' - [OPENMPTH] omp_get_max_threads/nproc = 1/4 - [NGOODHEL] ngoodhel/ncomb = 32/64 - [XSECTION] VECSIZE_USED = 8192 - [XSECTION] MultiChannel = TRUE - [XSECTION] Configuration = 1 - [XSECTION] ChannelId = 1 - [XSECTION] Cross section = 1.212 [1.2120105567647812] fbridge_mode=1 - [UNWEIGHT] Wrote 773 events (found 3685 events) - [COUNTERS] PROGRAM TOTAL : 5.0710s - [COUNTERS] Fortran Overhead ( 0 ) : 4.9788s - [COUNTERS] CudaCpp MEs ( 2 ) : 0.0922s for 122880 events => throughput is 1.33E+06 events/s - -*** (3-cuda) Compare MADEVENT_CUDA x10 xsec to MADEVENT_FORTRAN xsec *** - -OK! xsec from fortran (1.2120105567647810) and cuda (1.2120105567647812) differ by less than 3E-14 (2.220446049250313e-16) - -*** (3-cuda) Compare MADEVENT_CUDA x10 events.lhe to MADEVENT_FORTRAN events.lhe reference (including colors and helicities) *** - -OK! events.lhe.cuda.10 and events.lhe.ref.10 are identical - -*** EXECUTE GCHECK(8192) -p 256 32 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURHST+RMBHST+BRDDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 1.154915e+06 ) sec^-1 - -*** EXECUTE GCHECK(8192) -p 256 32 1 *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURDEV+RMBDEV+MESDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 1.256320e+06 ) sec^-1 - -*** EXECUTE GCHECK(MAX) -p 16384 32 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURHST+RMBHST+BRDDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 2.926708e+06 ) sec^-1 - -*** EXECUTE GCHECK(MAX) -p 16384 32 1 *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURDEV+RMBDEV+MESDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 3.151437e+06 ) sec^-1 - -*** EXECUTE GCHECK(MAX128THR) -p 4096 128 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURHST+RMBHST+BRDDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 2.913448e+06 ) sec^-1 - -*** EXECUTE GCHECK(MAX128THR) -p 4096 128 1 *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURDEV+RMBDEV+MESDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 3.128627e+06 ) sec^-1 - -*** EXECUTE GCHECK(MAX8THR) -p 65536 8 1 --bridge *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURHST+RMBHST+BRDDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 2.910695e+06 ) sec^-1 - -*** EXECUTE GCHECK(MAX8THR) -p 65536 8 1 *** -Process = SIGMA_SM_GU_TTXGU_CUDA [nvcc 12.0.140 (gcc 11.3.1)] [inlineHel=0] [hardcodePARAM=0] -Workflow summary = CUD:DBL+THX:CURDEV+RMBDEV+MESDEV/none+NAVBRK -EvtsPerSec[MECalcOnly] (3a) = ( 8.933053e+05 ) sec^-1 - -*** (3-hip) WARNING! SKIP MADEVENT_HIP (hip is not supported on this node) *** - -TEST COMPLETED