[fpe] ** COMPLETE FPE ** Revert to upstream/master performance logs i…

…n tput and tmad for easier merging

This completes the fpe and namespace patches, addressing #701 and #725, respectively.

Unfortunately, I tested that this patch only fixes the IEEE_DIVIDE_BY_ZERO part of #701,
but there are still other issues remaining (being debugged in branch nobm).

Revert "[fpe] rerun 15 tmad - ggttgg tests fail again #655 as expected"
This reverts commit 9212960.

Revert "[fpe] rerun 78 tput alltees, all ok"
This reverts commit 9a68868.

epochX/cudacpp/tmad/logs_ggttgg_mad/log_ggttgg_mad_d_inl0_hrd0.txt

@@ -6,22 +6,22 @@ make USEBUILDDIR=1 AVX=none
 
 make USEBUILDDIR=1 AVX=sse4
 make USEBUILDDIR=1 AVX=avx2
+
+make USEBUILDDIR=1 AVX=512y
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
-
-make USEBUILDDIR=1 AVX=512y
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 
 make USEBUILDDIR=1 AVX=512z
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
+CUDACPP_BUILDDIR='build.512z_d_inl0_hrd0'
 CUDACPP_BUILDDIR='build.avx2_d_inl0_hrd0'
-CUDACPP_BUILDDIR='build.sse4_d_inl0_hrd0'
-CUDACPP_BUILDDIR='build.512y_d_inl0_hrd0'
 CUDACPP_BUILDDIR='build.none_d_inl0_hrd0'
+CUDACPP_BUILDDIR='build.512y_d_inl0_hrd0'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
-CUDACPP_BUILDDIR='build.512z_d_inl0_hrd0'
+CUDACPP_BUILDDIR='build.sse4_d_inl0_hrd0'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Nothing to be done for 'all'.
@@ -33,7 +33,7 @@ make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/
 
 OMP_NUM_THREADS=
 
-DATE: 2023-07-20_22:12:01
+DATE: 2023-06-16_23:38:42
 
 On itscrd80.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/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg
@@ -58,9 +58,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x1_fortran > /tmp/av
  [XSECTION] Configuration = 1
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [] fbridge_mode=0
- [COUNTERS] PROGRAM TOTAL          :    4.4298s
- [COUNTERS] Fortran Overhead ( 0 ) :    0.2555s
- [COUNTERS] Fortran MEs      ( 1 ) :    4.1743s for     8192 events => throughput is 1.96E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    4.3757s
+ [COUNTERS] Fortran Overhead ( 0 ) :    0.2161s
+ [COUNTERS] Fortran MEs      ( 1 ) :    4.1595s for     8192 events => throughput is 1.97E+03 events/s
 
 *** (1) EXECUTE MADEVENT_FORTRAN x1 (create events.lhe) ***
 --------------------
@@ -83,9 +83,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x1_fortran > /tmp/av
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [3.6277277311352982E-004] fbridge_mode=0
  [UNWEIGHT] Wrote 49 events (found 738 events)
- [COUNTERS] PROGRAM TOTAL          :    4.4495s
- [COUNTERS] Fortran Overhead ( 0 ) :    0.3001s
- [COUNTERS] Fortran MEs      ( 1 ) :    4.1494s for     8192 events => throughput is 1.97E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    4.4480s
+ [COUNTERS] Fortran Overhead ( 0 ) :    0.2999s
+ [COUNTERS] Fortran MEs      ( 1 ) :    4.1481s for     8192 events => throughput is 1.97E+03 events/s
 
 *** (1) EXECUTE MADEVENT_FORTRAN x10 (create events.lhe) ***
 --------------------
@@ -108,9 +108,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x10_fortran > /tmp/a
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.000158 [1.5803725748610604E-004] fbridge_mode=0
  [UNWEIGHT] Wrote 204 events (found 1633 events)
- [COUNTERS] PROGRAM TOTAL          :   47.4939s
- [COUNTERS] Fortran Overhead ( 0 ) :    1.8948s
- [COUNTERS] Fortran MEs      ( 1 ) :   45.5991s for    90112 events => throughput is 1.98E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :   47.5995s
+ [COUNTERS] Fortran Overhead ( 0 ) :    1.8973s
+ [COUNTERS] Fortran MEs      ( 1 ) :   45.7021s for    90112 events => throughput is 1.97E+03 events/s
 
 *** (2-none) EXECUTE MADEVENT_CPP x1 (create events.lhe) ***
 --------------------
@@ -133,9 +133,9 @@ Executing ' ./build.none_d_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_ggttgg_x
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [3.6277277311352998E-004] fbridge_mode=1
  [UNWEIGHT] Wrote 49 events (found 738 events)
- [COUNTERS] PROGRAM TOTAL          :    8.6209s
- [COUNTERS] Fortran Overhead ( 0 ) :    4.4153s
- [COUNTERS] CudaCpp MEs      ( 2 ) :    4.2055s for     8192 events => throughput is 1.95E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    8.5650s
+ [COUNTERS] Fortran Overhead ( 0 ) :    4.3703s
+ [COUNTERS] CudaCpp MEs      ( 2 ) :    4.1947s for     8192 events => throughput is 1.95E+03 events/s
 
 *** (2-none) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec ***
 

epochX/cudacpp/tmad/logs_ggttgg_mad/log_ggttgg_mad_f_inl0_hrd0.txt

@@ -4,8 +4,8 @@ CUDACPP_BUILDDIR='.'
 
 
 make USEBUILDDIR=1 AVX=none
-
 make USEBUILDDIR=1 AVX=sse4
+
 make USEBUILDDIR=1 AVX=avx2
 make USEBUILDDIR=1 AVX=512y
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
@@ -15,25 +15,25 @@ make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp
 
 make USEBUILDDIR=1 AVX=512z
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
+CUDACPP_BUILDDIR='build.sse4_f_inl0_hrd0'
 CUDACPP_BUILDDIR='build.none_f_inl0_hrd0'
 CUDACPP_BUILDDIR='build.512y_f_inl0_hrd0'
-CUDACPP_BUILDDIR='build.sse4_f_inl0_hrd0'
 CUDACPP_BUILDDIR='build.avx2_f_inl0_hrd0'
+CUDACPP_BUILDDIR='build.512z_f_inl0_hrd0'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
-CUDACPP_BUILDDIR='build.512z_f_inl0_hrd0'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 
 OMP_NUM_THREADS=
 
-DATE: 2023-07-20_22:13:09
+DATE: 2023-06-16_23:39:50
 
 On itscrd80.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/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg
@@ -58,9 +58,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x1_fortran > /tmp/av
  [XSECTION] Configuration = 1
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [] fbridge_mode=0
- [COUNTERS] PROGRAM TOTAL          :    4.3889s
- [COUNTERS] Fortran Overhead ( 0 ) :    0.2167s
- [COUNTERS] Fortran MEs      ( 1 ) :    4.1723s for     8192 events => throughput is 1.96E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    4.4077s
+ [COUNTERS] Fortran Overhead ( 0 ) :    0.2143s
+ [COUNTERS] Fortran MEs      ( 1 ) :    4.1934s for     8192 events => throughput is 1.95E+03 events/s
 
 *** (1) EXECUTE MADEVENT_FORTRAN x1 (create events.lhe) ***
 --------------------
@@ -83,9 +83,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x1_fortran > /tmp/av
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [3.6277277311352982E-004] fbridge_mode=0
  [UNWEIGHT] Wrote 49 events (found 738 events)
- [COUNTERS] PROGRAM TOTAL          :    4.4632s
- [COUNTERS] Fortran Overhead ( 0 ) :    0.3033s
- [COUNTERS] Fortran MEs      ( 1 ) :    4.1599s for     8192 events => throughput is 1.97E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    4.4393s
+ [COUNTERS] Fortran Overhead ( 0 ) :    0.3039s
+ [COUNTERS] Fortran MEs      ( 1 ) :    4.1354s for     8192 events => throughput is 1.98E+03 events/s
 
 *** (1) EXECUTE MADEVENT_FORTRAN x10 (create events.lhe) ***
 --------------------
@@ -108,9 +108,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x10_fortran > /tmp/a
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.000158 [1.5803725748610604E-004] fbridge_mode=0
  [UNWEIGHT] Wrote 204 events (found 1633 events)
- [COUNTERS] PROGRAM TOTAL          :   47.8293s
- [COUNTERS] Fortran Overhead ( 0 ) :    1.9068s
- [COUNTERS] Fortran MEs      ( 1 ) :   45.9225s for    90112 events => throughput is 1.96E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :   47.6726s
+ [COUNTERS] Fortran Overhead ( 0 ) :    1.8981s
+ [COUNTERS] Fortran MEs      ( 1 ) :   45.7745s for    90112 events => throughput is 1.97E+03 events/s
 
 *** (2-none) EXECUTE MADEVENT_CPP x1 (create events.lhe) ***
 --------------------
@@ -133,9 +133,9 @@ Executing ' ./build.none_f_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_ggttgg_x
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [3.6277396515517582E-004] fbridge_mode=1
  [UNWEIGHT] Wrote 49 events (found 738 events)
- [COUNTERS] PROGRAM TOTAL          :    8.3655s
- [COUNTERS] Fortran Overhead ( 0 ) :    4.2729s
- [COUNTERS] CudaCpp MEs      ( 2 ) :    4.0926s for     8192 events => throughput is 2.00E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    8.4058s
+ [COUNTERS] Fortran Overhead ( 0 ) :    4.2739s
+ [COUNTERS] CudaCpp MEs      ( 2 ) :    4.1319s for     8192 events => throughput is 1.98E+03 events/s
 
 *** (2-none) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec ***
 

epochX/cudacpp/tmad/logs_ggttgg_mad/log_ggttgg_mad_m_inl0_hrd0.txt

@@ -5,20 +5,21 @@ CUDACPP_BUILDDIR='.'
 
 make USEBUILDDIR=1 AVX=none
 make USEBUILDDIR=1 AVX=sse4
-
 make USEBUILDDIR=1 AVX=avx2
-make USEBUILDDIR=1 AVX=512y
-make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 
+make USEBUILDDIR=1 AVX=512y
+
 make USEBUILDDIR=1 AVX=512z
 make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
+make[1]: Entering directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 CUDACPP_BUILDDIR='build.sse4_m_inl0_hrd0'
+CUDACPP_BUILDDIR='build.none_m_inl0_hrd0'
+CUDACPP_BUILDDIR='build.512y_m_inl0_hrd0'
 CUDACPP_BUILDDIR='build.avx2_m_inl0_hrd0'
 CUDACPP_BUILDDIR='build.512z_m_inl0_hrd0'
-CUDACPP_BUILDDIR='build.none_m_inl0_hrd0'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Nothing to be done for 'all'.
@@ -27,13 +28,12 @@ make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
-CUDACPP_BUILDDIR='build.512y_m_inl0_hrd0'
 make[1]: Nothing to be done for 'all'.
 make[1]: Leaving directory '/data/avalassi/GPU2023/madgraph4gpuX/epochX/cudacpp/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg'
 
 OMP_NUM_THREADS=
 
-DATE: 2023-07-20_22:14:17
+DATE: 2023-06-16_23:40:57
 
 On itscrd80.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/gg_ttgg.mad/SubProcesses/P1_gg_ttxgg
@@ -58,9 +58,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x1_fortran > /tmp/av
  [XSECTION] Configuration = 1
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [] fbridge_mode=0
- [COUNTERS] PROGRAM TOTAL          :    4.4609s
- [COUNTERS] Fortran Overhead ( 0 ) :    0.2133s
- [COUNTERS] Fortran MEs      ( 1 ) :    4.2476s for     8192 events => throughput is 1.93E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    4.4170s
+ [COUNTERS] Fortran Overhead ( 0 ) :    0.2154s
+ [COUNTERS] Fortran MEs      ( 1 ) :    4.2016s for     8192 events => throughput is 1.95E+03 events/s
 
 *** (1) EXECUTE MADEVENT_FORTRAN x1 (create events.lhe) ***
 --------------------
@@ -83,9 +83,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x1_fortran > /tmp/av
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [3.6277277311352982E-004] fbridge_mode=0
  [UNWEIGHT] Wrote 49 events (found 738 events)
- [COUNTERS] PROGRAM TOTAL          :    4.4921s
- [COUNTERS] Fortran Overhead ( 0 ) :    0.3024s
- [COUNTERS] Fortran MEs      ( 1 ) :    4.1897s for     8192 events => throughput is 1.96E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    4.6794s
+ [COUNTERS] Fortran Overhead ( 0 ) :    0.3042s
+ [COUNTERS] Fortran MEs      ( 1 ) :    4.3752s for     8192 events => throughput is 1.87E+03 events/s
 
 *** (1) EXECUTE MADEVENT_FORTRAN x10 (create events.lhe) ***
 --------------------
@@ -108,9 +108,9 @@ Executing ' ./madevent_fortran < /tmp/avalassi/input_ggttgg_x10_fortran > /tmp/a
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.000158 [1.5803725748610604E-004] fbridge_mode=0
  [UNWEIGHT] Wrote 204 events (found 1633 events)
- [COUNTERS] PROGRAM TOTAL          :   47.5479s
- [COUNTERS] Fortran Overhead ( 0 ) :    1.9002s
- [COUNTERS] Fortran MEs      ( 1 ) :   45.6476s for    90112 events => throughput is 1.97E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :   47.7432s
+ [COUNTERS] Fortran Overhead ( 0 ) :    1.9035s
+ [COUNTERS] Fortran MEs      ( 1 ) :   45.8397s for    90112 events => throughput is 1.97E+03 events/s
 
 *** (2-none) EXECUTE MADEVENT_CPP x1 (create events.lhe) ***
 --------------------
@@ -133,9 +133,9 @@ Executing ' ./build.none_m_inl0_hrd0/madevent_cpp < /tmp/avalassi/input_ggttgg_x
  [XSECTION] ChannelId = 2
  [XSECTION] Cross section = 0.0003628 [3.6277277432965013E-004] fbridge_mode=1
  [UNWEIGHT] Wrote 49 events (found 738 events)
- [COUNTERS] PROGRAM TOTAL          :    8.7180s
- [COUNTERS] Fortran Overhead ( 0 ) :    4.4096s
- [COUNTERS] CudaCpp MEs      ( 2 ) :    4.3084s for     8192 events => throughput is 1.90E+03 events/s
+ [COUNTERS] PROGRAM TOTAL          :    8.6986s
+ [COUNTERS] Fortran Overhead ( 0 ) :    4.4272s
+ [COUNTERS] CudaCpp MEs      ( 2 ) :    4.2714s for     8192 events => throughput is 1.92E+03 events/s
 
 *** (2-none) Compare MADEVENT_CPP x1 xsec to MADEVENT_FORTRAN xsec ***