Skip to content

Commit 967e077

Browse files
valassivalassi
committed
[csm] rerun 144 tput tests on itscrd90 - all ok
With respect to the last rd90 logs for upstream/master (commit 4178974 in hack_ihel3p1): - Performance is around 5% better on CPU (mainly cppnone) and essentially the same everywhere else
1 parent 1ba0e92 commit 967e077

File tree

144 files changed

+7176
-8040
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

144 files changed

+7176
-8040
lines changed

‎epochX/cudacpp/tput/logs_eemumu_mad/log_eemumu_mad_d_inl0_hrd0.scaling‎

Lines changed: 57 additions & 63 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
MADGRAPH_CUDA_ARCHITECTURE=
1+
MADGRAPH_CUDA_ARCHITECTURE=70
22
MADGRAPH_HIP_ARCHITECTURE=
33

44
HASBLAS=hasBlas
@@ -16,35 +16,29 @@ make: Nothing to be done for 'gtestlibs'.
1616

1717
make USEBUILDDIR=1 BACKEND=cuda
1818
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
19-
make[1]: Nothing to be done for 'all'.
2019
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
2120

2221
make USEBUILDDIR=1 BACKEND=cppnone
2322
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
24-
make[1]: Nothing to be done for 'all'.
2523
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
2624

2725
make USEBUILDDIR=1 BACKEND=cppsse4
2826
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
29-
make[1]: Nothing to be done for 'all'.
3027
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
3128

3229
make USEBUILDDIR=1 BACKEND=cppavx2
3330
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
34-
make[1]: Nothing to be done for 'all'.
3531
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
3632

3733
make USEBUILDDIR=1 BACKEND=cpp512y
3834
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
39-
make[1]: Nothing to be done for 'all'.
4035
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
4136

4237
make USEBUILDDIR=1 BACKEND=cpp512z
4338
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
44-
make[1]: Nothing to be done for 'all'.
4539
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
4640

47-
DATE: 2025-10-11_15:39:36
41+
DATE: 2025-12-07_17:56:28
4842

4943
HASBLAS=hasBlas
5044
CUDACPP_RUNTIME_BLASCOLORSUM=
@@ -53,85 +47,85 @@ On itscrd90.cern.ch [CPU: Intel(R) Xeon(R) Silver 4216 CPU] [GPU: 1x Tesla V100S
5347
=========================================================================
5448
scalingTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.cuda_d_inl0_hrd0/check_cuda.exe
5549
### GPU: scaling test 256
56-
2.365880e+06 1 256
57-
4.932658e+06 2 256
58-
1.130330e+07 4 256
59-
2.221065e+07 8 256
60-
3.796917e+07 16 256
61-
8.093742e+07 32 256
62-
1.438543e+08 64 256
63-
2.092652e+08 128 256
64-
2.586706e+08 256 256
65-
3.166572e+08 512 256
66-
3.450925e+08 1024 256
50+
2.828292e+06 1 256
51+
6.114090e+06 2 256
52+
1.163359e+07 4 256
53+
2.245762e+07 8 256
54+
3.962197e+07 16 256
55+
8.583493e+07 32 256
56+
1.486765e+08 64 256
57+
1.985410e+08 128 256
58+
2.543566e+08 256 256
59+
3.191398e+08 512 256
60+
3.589952e+08 1024 256
6761
### GPU: scaling test 32
68-
3.615411e+05 1 32
69-
7.956340e+05 2 32
70-
1.534533e+06 4 32
71-
2.896550e+06 8 32
72-
5.416499e+06 16 32
73-
1.086184e+07 32 32
74-
2.239377e+07 64 32
75-
4.040723e+07 128 32
76-
8.109125e+07 256 32
77-
1.501315e+08 512 32
78-
2.161406e+08 1024 32
79-
2.736516e+08 2048 32
80-
3.294400e+08 4096 32
81-
3.666924e+08 8192 32
62+
3.460732e+05 1 32
63+
8.262329e+05 2 32
64+
1.484385e+06 4 32
65+
2.893636e+06 8 32
66+
6.066567e+06 16 32
67+
1.002968e+07 32 32
68+
2.349297e+07 64 32
69+
3.536186e+07 128 32
70+
8.757002e+07 256 32
71+
1.583867e+08 512 32
72+
2.115170e+08 1024 32
73+
2.787414e+08 2048 32
74+
3.346815e+08 4096 32
75+
3.729437e+08 8192 32
8276
=========================================================================
8377
scalingTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.hip_d_inl0_hrd0/check_hip.exe
8478
Not found: /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.hip_d_inl0_hrd0/check_hip.exe
8579
=========================================================================
8680
scalingTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.none_d_inl0_hrd0/check_cpp.exe
8781
### CPU: scaling test 256
88-
1.112163e+06 1 256
89-
1.095778e+06 2 256
90-
1.085622e+06 4 256
82+
1.031165e+06 1 256
83+
1.102961e+06 2 256
84+
1.103733e+06 4 256
9185
### CPU: scaling test 32
92-
9.838283e+05 1 32
93-
1.009336e+06 2 32
94-
1.104848e+06 4 32
86+
1.057187e+06 1 32
87+
1.090959e+06 2 32
88+
1.105583e+06 4 32
9589
=========================================================================
9690
scalingTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.sse4_d_inl0_hrd0/check_cpp.exe
9791
### CPU: scaling test 256
98-
1.791676e+06 1 256
99-
1.843126e+06 2 256
100-
1.850216e+06 4 256
92+
1.891169e+06 1 256
93+
1.848496e+06 2 256
94+
1.743854e+06 4 256
10195
### CPU: scaling test 32
102-
1.835283e+06 1 32
103-
1.487162e+06 2 32
104-
1.478777e+06 4 32
96+
1.596488e+06 1 32
97+
1.520335e+06 2 32
98+
1.732469e+06 4 32
10599
=========================================================================
106100
scalingTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.avx2_d_inl0_hrd0/check_cpp.exe
107101
### CPU: scaling test 256
108-
2.691677e+06 1 256
109-
2.725347e+06 2 256
110-
2.679688e+06 4 256
102+
2.232591e+06 1 256
103+
2.714884e+06 2 256
104+
2.690560e+06 4 256
111105
### CPU: scaling test 32
112-
2.224230e+06 1 32
113-
2.558465e+06 2 32
114-
2.649774e+06 4 32
106+
2.373535e+06 1 32
107+
2.562870e+06 2 32
108+
2.650872e+06 4 32
115109
=========================================================================
116110
scalingTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.512y_d_inl0_hrd0/check_cpp.exe
117111
### CPU: scaling test 256
118-
2.781551e+06 1 256
119-
2.448941e+06 2 256
120-
2.756282e+06 4 256
112+
2.776181e+06 1 256
113+
2.564552e+06 2 256
114+
2.732605e+06 4 256
121115
### CPU: scaling test 32
122-
2.377238e+06 1 32
123-
2.626719e+06 2 32
124-
2.722014e+06 4 32
116+
2.251618e+06 1 32
117+
1.672285e+06 2 32
118+
2.326723e+06 4 32
125119
=========================================================================
126120
scalingTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.512z_d_inl0_hrd0/check_cpp.exe
127121
### CPU: scaling test 256
128-
2.040101e+06 1 256
129-
2.059277e+06 2 256
130-
2.194331e+06 4 256
122+
2.043455e+06 1 256
123+
2.033861e+06 2 256
124+
2.149784e+06 4 256
131125
### CPU: scaling test 32
132-
1.410251e+06 1 32
133-
1.626347e+06 2 32
134-
1.877466e+06 4 32
126+
2.679618e+06 1 32
127+
1.606789e+06 2 32
128+
1.849230e+06 4 32
135129
=========================================================================
136130

137131
TEST COMPLETED

‎epochX/cudacpp/tput/logs_eemumu_mad/log_eemumu_mad_d_inl0_hrd0.txt‎

Lines changed: 44 additions & 50 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
MADGRAPH_CUDA_ARCHITECTURE=
1+
MADGRAPH_CUDA_ARCHITECTURE=70
22
MADGRAPH_HIP_ARCHITECTURE=
33

44
HASBLAS=hasBlas
@@ -16,35 +16,29 @@ make: Nothing to be done for 'gtestlibs'.
1616

1717
make USEBUILDDIR=1 BACKEND=cuda
1818
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
19-
make[1]: Nothing to be done for 'all'.
2019
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
2120

2221
make USEBUILDDIR=1 BACKEND=cppnone
2322
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
24-
make[1]: Nothing to be done for 'all'.
2523
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
2624

2725
make USEBUILDDIR=1 BACKEND=cppsse4
2826
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
29-
make[1]: Nothing to be done for 'all'.
3027
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
3128

3229
make USEBUILDDIR=1 BACKEND=cppavx2
3330
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
34-
make[1]: Nothing to be done for 'all'.
3531
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
3632

3733
make USEBUILDDIR=1 BACKEND=cpp512y
3834
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
39-
make[1]: Nothing to be done for 'all'.
4035
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
4136

4237
make USEBUILDDIR=1 BACKEND=cpp512z
4338
make[1]: Entering directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
44-
make[1]: Nothing to be done for 'all'.
4539
make[1]: Leaving directory '/data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum'
4640

47-
DATE: 2025-10-11_15:13:43
41+
DATE: 2025-12-07_17:31:39
4842

4943
HASBLAS=hasBlas
5044
CUDACPP_RUNTIME_BLASCOLORSUM=
@@ -55,14 +49,14 @@ runExe /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubPr
5549
Process = SIGMA_SM_EPEM_MUPMUM_CUDA [nvcc 12.0.140 (gcc 11.5.0)] [inlineHel=0] [hardcodePARAM=0]
5650
Workflow summary = CUD:DBL+THX:CURDEV+RMBDEV+MESDEV/none+NAVBRK
5751
FP precision = DOUBLE (NaN/abnormal=0, zero=0)
58-
EvtsPerSec[Rmb+ME] (23) = ( 6.456825e+07 ) sec^-1
59-
EvtsPerSec[MatrixElems] (3) = ( 3.020579e+08 ) sec^-1
60-
EvtsPerSec[MECalcOnly] (3a) = ( 3.872827e+08 ) sec^-1
52+
EvtsPerSec[Rmb+ME] (23) = ( 6.448256e+07 ) sec^-1
53+
EvtsPerSec[MatrixElems] (3) = ( 3.095942e+08 ) sec^-1
54+
EvtsPerSec[MECalcOnly] (3a) = ( 3.924818e+08 ) sec^-1
6155
MeanMatrixElemValue = ( 1.371706e-02 +- 3.270315e-06 ) GeV^0
62-
TOTAL : 0.693291 sec
63-
2,729,119,040 cycles # 2.827 GHz
64-
4,039,185,150 instructions # 1.48 insn per cycle
65-
1.043410313 seconds time elapsed
56+
TOTAL : 0.779644 sec
57+
2,832,606,212 cycles # 2.888 GHz
58+
4,254,803,118 instructions # 1.50 insn per cycle
59+
1.394928839 seconds time elapsed
6660
.........................................................................
6761
runNcu /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.cuda_d_inl0_hrd0/check_cuda.exe -p 2048 256 1
6862
==PROF== Profiling "calculate_jamps": launch__registers_per_thread 144
@@ -89,14 +83,14 @@ Process = SIGMA_SM_EPEM_MUPMUM_CPP [gcc 11.5.0] [inlineHel=0
8983
Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/none+NAVBRK
9084
FP precision = DOUBLE (NaN/abnormal=0, zero=0)
9185
Internal loops fptype_sv = SCALAR ('none': ~vector[1], no SIMD)
92-
EvtsPerSec[Rmb+ME] (23) = ( 1.019940e+06 ) sec^-1
93-
EvtsPerSec[MatrixElems] (3) = ( 1.187870e+06 ) sec^-1
94-
EvtsPerSec[MECalcOnly] (3a) = ( 1.187870e+06 ) sec^-1
86+
EvtsPerSec[Rmb+ME] (23) = ( 1.047604e+06 ) sec^-1
87+
EvtsPerSec[MatrixElems] (3) = ( 1.219439e+06 ) sec^-1
88+
EvtsPerSec[MECalcOnly] (3a) = ( 1.219439e+06 ) sec^-1
9589
MeanMatrixElemValue = ( 1.371706e-02 +- 3.270315e-06 ) GeV^0
96-
TOTAL : 6.588033 sec
97-
19,038,044,386 cycles # 2.888 GHz
98-
46,485,585,356 instructions # 2.44 insn per cycle
99-
6.596061286 seconds time elapsed
90+
TOTAL : 6.411794 sec
91+
19,014,240,782 cycles # 2.964 GHz
92+
46,485,315,191 instructions # 2.44 insn per cycle
93+
6.416861168 seconds time elapsed
10094
=Symbols in CPPProcess_cpp.o= (~sse4: 482) (avx2: 0) (512y: 0) (512z: 0)
10195
-------------------------------------------------------------------------
10296
runTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.none_d_inl0_hrd0/runTest_cpp.exe
@@ -116,14 +110,14 @@ Process = SIGMA_SM_EPEM_MUPMUM_CPP [gcc 11.5.0] [inlineHel=0
116110
Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/sse4+CXVBRK
117111
FP precision = DOUBLE (NaN/abnormal=0, zero=0)
118112
Internal loops fptype_sv = VECTOR[2] ('sse4': SSE4.2, 128bit) [cxtype_ref=YES]
119-
EvtsPerSec[Rmb+ME] (23) = ( 1.557129e+06 ) sec^-1
120-
EvtsPerSec[MatrixElems] (3) = ( 2.030035e+06 ) sec^-1
121-
EvtsPerSec[MECalcOnly] (3a) = ( 2.030035e+06 ) sec^-1
113+
EvtsPerSec[Rmb+ME] (23) = ( 1.598686e+06 ) sec^-1
114+
EvtsPerSec[MatrixElems] (3) = ( 2.089132e+06 ) sec^-1
115+
EvtsPerSec[MECalcOnly] (3a) = ( 2.089132e+06 ) sec^-1
122116
MeanMatrixElemValue = ( 1.371706e-02 +- 3.270315e-06 ) GeV^0
123-
TOTAL : 4.460811 sec
124-
12,939,620,485 cycles # 2.898 GHz
125-
31,810,901,247 instructions # 2.46 insn per cycle
126-
4.469139042 seconds time elapsed
117+
TOTAL : 4.346013 sec
118+
12,961,637,078 cycles # 2.979 GHz
119+
31,812,423,686 instructions # 2.45 insn per cycle
120+
4.352494980 seconds time elapsed
127121
=Symbols in CPPProcess_cpp.o= (~sse4: 1669) (avx2: 0) (512y: 0) (512z: 0)
128122
-------------------------------------------------------------------------
129123
runTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.sse4_d_inl0_hrd0/runTest_cpp.exe
@@ -143,14 +137,14 @@ Process = SIGMA_SM_EPEM_MUPMUM_CPP [gcc 11.5.0] [inlineHel=0
143137
Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/avx2+CXVBRK
144138
FP precision = DOUBLE (NaN/abnormal=0, zero=0)
145139
Internal loops fptype_sv = VECTOR[4] ('avx2': AVX2, 256bit) [cxtype_ref=YES]
146-
EvtsPerSec[Rmb+ME] (23) = ( 1.933537e+06 ) sec^-1
147-
EvtsPerSec[MatrixElems] (3) = ( 2.681631e+06 ) sec^-1
148-
EvtsPerSec[MECalcOnly] (3a) = ( 2.681631e+06 ) sec^-1
140+
EvtsPerSec[Rmb+ME] (23) = ( 1.995383e+06 ) sec^-1
141+
EvtsPerSec[MatrixElems] (3) = ( 2.769392e+06 ) sec^-1
142+
EvtsPerSec[MECalcOnly] (3a) = ( 2.769392e+06 ) sec^-1
149143
MeanMatrixElemValue = ( 1.371706e-02 +- 3.270315e-06 ) GeV^0
150-
TOTAL : 3.671840 sec
151-
10,104,892,452 cycles # 2.749 GHz
152-
19,727,697,375 instructions # 1.95 insn per cycle
153-
3.679095535 seconds time elapsed
144+
TOTAL : 3.556154 sec
145+
10,091,928,187 cycles # 2.835 GHz
146+
19,729,979,199 instructions # 1.96 insn per cycle
147+
3.561316676 seconds time elapsed
154148
=Symbols in CPPProcess_cpp.o= (~sse4: 0) (avx2: 1917) (512y: 0) (512z: 0)
155149
-------------------------------------------------------------------------
156150
runTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.avx2_d_inl0_hrd0/runTest_cpp.exe
@@ -170,14 +164,14 @@ Process = SIGMA_SM_EPEM_MUPMUM_CPP [gcc 11.5.0] [inlineHel=0
170164
Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/512y+CXVBRK
171165
FP precision = DOUBLE (NaN/abnormal=0, zero=0)
172166
Internal loops fptype_sv = VECTOR[4] ('512y': AVX512, 256bit) [cxtype_ref=YES]
173-
EvtsPerSec[Rmb+ME] (23) = ( 1.989488e+06 ) sec^-1
174-
EvtsPerSec[MatrixElems] (3) = ( 2.781185e+06 ) sec^-1
175-
EvtsPerSec[MECalcOnly] (3a) = ( 2.781185e+06 ) sec^-1
167+
EvtsPerSec[Rmb+ME] (23) = ( 2.071101e+06 ) sec^-1
168+
EvtsPerSec[MatrixElems] (3) = ( 2.895575e+06 ) sec^-1
169+
EvtsPerSec[MECalcOnly] (3a) = ( 2.895575e+06 ) sec^-1
176170
MeanMatrixElemValue = ( 1.371706e-02 +- 3.270315e-06 ) GeV^0
177-
TOTAL : 3.576826 sec
178-
9,900,381,139 cycles # 2.765 GHz
179-
19,380,047,753 instructions # 1.96 insn per cycle
180-
3.585735108 seconds time elapsed
171+
TOTAL : 3.437536 sec
172+
9,847,578,789 cycles # 2.862 GHz
173+
19,380,355,138 instructions # 1.97 insn per cycle
174+
3.442447176 seconds time elapsed
181175
=Symbols in CPPProcess_cpp.o= (~sse4: 0) (avx2: 1655) (512y: 180) (512z: 0)
182176
-------------------------------------------------------------------------
183177
runTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.512y_d_inl0_hrd0/runTest_cpp.exe
@@ -197,14 +191,14 @@ Process = SIGMA_SM_EPEM_MUPMUM_CPP [gcc 11.5.0] [inlineHel=0
197191
Workflow summary = CPP:DBL+CXS:CURHST+RMBHST+MESHST/512z+CXVBRK
198192
FP precision = DOUBLE (NaN/abnormal=0, zero=0)
199193
Internal loops fptype_sv = VECTOR[8] ('512z': AVX512, 512bit) [cxtype_ref=YES]
200-
EvtsPerSec[Rmb+ME] (23) = ( 1.671348e+06 ) sec^-1
201-
EvtsPerSec[MatrixElems] (3) = ( 2.193135e+06 ) sec^-1
202-
EvtsPerSec[MECalcOnly] (3a) = ( 2.193135e+06 ) sec^-1
194+
EvtsPerSec[Rmb+ME] (23) = ( 1.773261e+06 ) sec^-1
195+
EvtsPerSec[MatrixElems] (3) = ( 2.352997e+06 ) sec^-1
196+
EvtsPerSec[MECalcOnly] (3a) = ( 2.352997e+06 ) sec^-1
203197
MeanMatrixElemValue = ( 1.371706e-02 +- 3.270315e-06 ) GeV^0
204-
TOTAL : 4.184170 sec
205-
8,626,596,296 cycles # 2.060 GHz
206-
15,802,085,882 instructions # 1.83 insn per cycle
207-
4.189889070 seconds time elapsed
198+
TOTAL : 3.952172 sec
199+
8,636,738,592 cycles # 2.183 GHz
200+
15,800,904,624 instructions # 1.83 insn per cycle
201+
3.957156027 seconds time elapsed
208202
=Symbols in CPPProcess_cpp.o= (~sse4: 0) (avx2: 873) (512y: 156) (512z: 1263)
209203
-------------------------------------------------------------------------
210204
runTest /data/avalassi/GPU2025/test-madgraph4gpu/epochX/cudacpp/ee_mumu.mad/SubProcesses/P1_epem_mupmum/build.512z_d_inl0_hrd0/runTest_cpp.exe

0 commit comments

Comments
 (0)